Phenotypic diversity of Methylobacterium associated with rice landraces in North-East India

Abstract

The ecology and distribution of many bacteria is strongly associated with specific eukaryotic hosts. However, the impact of such host association on bacterial ecology and evolution is not well understood. Bacteria from the genus Methylobacterium consume plant-derived methanol, and are some of the most abundant and widespread plant-associated bacteria. In addition, many of these species impact plant fitness. To determine the ecology and distribution of Methylobacterium in nature, we sampled bacteria from 36 distinct rice landraces, traditionally grown in geographically isolated locations in NorthEast (NE) India. These landraces have been selected for diverse phenotypic traits by local communities, and we expected that the divergent selection on hosts may have also generated divergence in associated Methylobacterium strains. We determined the ability of 91 distinct rice-associated Methylobacterium isolates to use a panel of carbon sources, finding substantial variability in carbon use profiles. Consistent with our expectation, across spatial scales this phenotypic variation was largely explained by host landrace identity rather than geographical factors or bacterial taxonomy. However, variation in carbon utilisation was not correlated with sugar exudates on leaf surfaces, suggesting that bacterial carbon use profiles do not directly determine bacterial colonization across landraces. Finally, experiments showed that at least some rice landraces gain an early growth advantage from their specific phyllosphere-colonizing Methylobacterium strains. Together, our results suggest that landrace-specific hostmicrobial relationships may contribute to spatial structure in rice-associated Methylobacterium in a natural ecosystem. In turn, association with specific bacteria may provide new ways to preserve and understand diversity in one of the most important food crops of the world.

IMPACT OF PHYLLOSPHERE METHYLOBACTERIUM ON HOST RICE LANDRACES

“…The copyright holder for this this version posted February 8, 2022. ; https://doi.org/10.1101/2022.02.07.479485 doi: bioRxiv preprint preliminary greenhouse experiments indicating landrace-specific benefits for plant growth [19]. These results suggested that host selection might have driven specific Methylobacterium-rice landrace associations in paddy fields; and that this could be utilized for increasing rice yields.…”

“…We also note that due to logistical limitations we only included one “other” and “own” strain in our study, and in future work it will be important to include more strains of each category to be able to generalize the effect of native vs. non-native bacteria. We also note the possibility that the “own” strains that we identified as native strains colonizing Phouren-mubi and Phou-ngang [19] were perhaps only transiently dominant and are not consistently associated with these landraces; or that their association is relatively recent. Lastly, variability in the impact of bacteria on host fitness may be driven by more complex interactions between multiple causal factors.…”

“…To test whether the sprayed bacteria successfully colonized the plants, we collected flag leaves from three plants per treatment, before and after spraying. We isolated Methylobacterium by imprinting the leaves on Hypho minimal agar with 0.5% methanol, and identified the strains by sequencing the 16S rRNA gene as described earlier [19]. As the rice plants grew, we measured traits such as plant height, flag leaf width and length, and the number of tillers and panicles [22] at three different timepoints (Fig.…”

“…In 2016, we had isolated hundreds of Methylobacterium strains from the phyllosphere of multiple rice landraces from north east India [19]. In this study, we analysed the impact of the association of four of these strains on host rice plants in the field, choosing strains that were dominant in the respective host landrace.…”

“…In this region, distinct rice landraces have been cultivated locally for many generations, and show substantial genotypic and phenotypic divergence [18]. Previously, we showed that host rice landrace was the best predictor of phenotypic variation in epiphytic Methylobacterium strains, with different landraces associating with distinct Methylobacterium strains [19]. Preliminary greenhouse experiments further showed that some rice landraces gain an early growth advantage when inoculated with Methylobacterium strains isolated from their leaves.…”