Research Article |
Corresponding author: Shoko Sakai ( shokosakai0910@gmail.com ) Academic editor: Hugo de Boer
© 2022 Nuria Jiménez Elvira, Masayuki Ushio, Shoko Sakai.
This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Citation:
Jiménez Elvira N, Ushio M, Sakai S (2022) Are microbes growing on flowers evil? Effects of old flower microbes on fruit set in a wild ginger with one-day flowers, Alpinia japonica (Zingiberaceae). Metabarcoding and Metagenomics 6: e84331. https://doi.org/10.3897/mbmg.6.84331
|
Flowers are colonized and inhabited by diverse microbes. Flowers have various mechanisms to suppress microbial growth, such as flower volatiles, reactive oxygen and secondary compounds. Besides, plants rapidly replace flowers that have a short lifespan, and old flowers senesce. They may contribute to avoiding adverse effects of the microbes. In this study, we investigate if the flower microbial community on old flowers impedes fruit and seed production in a wild ginger with one-day flowers. We focus on microbes on old flowers because they may be composed of microbes that would grow during flowering if the flowers did not have mechanisms to suppress microbial growth. We inoculated newly opened flowers with old flower microbes, and monitored the effects on fruit and seed set. We also assessed prokaryotic communities on the flowers using 16S rRNA amplicon sequencing. We found six bacterial amplicon sequence variants (ASVs) whose proportions were increased on the inoculated flowers. These ASVs were also found on flower buds and flowers that were bagged by net or paper during anthesis, suggesting that they had been present in small numbers prior to flowering. Fruit set was negatively associated with the proportions of these ASVs, while seed set was not. The results suggest that old flowers harbor microbial communities different from those at anthesis, and that the microbes abundant on old flowers negatively affect plant reproduction. Although it has received little attention, antagonistic microbes that rapidly proliferate on the flowers may have affected the evolution of various flower characteristics such as flower volatiles and life span.
anthosphere, flower longevity, fruit set, microbiome, pollination
Flowers are an elaborate device to exchange pollen among conspecific individuals. Instead of moving around to find a potential mate, the plants disperse and receive pollen relying on different vectors such as insects and wind to achieve sexual reproduction. As a consequence, flowers are also exposed to microbial colonization under natural conditions (
Today, flower microbes are increasingly recognized as essential components in the ecology and evolution of plant reproduction (reviewed in
In general, even newly opened flowers already have microbes at least on some flower tissues. Their abundance increases over time on individual flowers (
In this study, we examine if the flower microbes on old flowers impede fruit and seed production in a wild ginger, Alpinia japonica (Zingiberaceae) by inoculating microbes from old flowers onto newly opened flowers in the morning. This perennial herb grows on the humid forest floor of evergreen and deciduous forests in western Japan. It flowers in June, when the climate is hot and humid. Its flowers open in the morning and wilt at sunset within the same day. Therefore, its flowering coincides with the optimal season for microbial growth, while it may avoid the negative effects of microbes by renewing flowers daily. Besides, fresh flowers emit volatiles that are known to have antimicrobial activity (
Alpinia japonica (Zingiberaceae) is a wild ginger distributed in temperate and subtropical regions of eastern Asia. It is a perennial herb 0.5–0.7 m in height occurring mostly in broad-leaved evergreen forests. Inflorescences are compound racemose with 10–60 flowers (37 on average in this population). Flowers are zygomorphic with a prominent labellum 1.1–1.2 cm in length, and white with red stripes at the margin of the labellum (Fig.
Studies were conducted in May to July of 2018, in Seta Park, Otsu, Shiga Prefecture, Japan (34°50'N, 135°50'E). Seta Park has a young secondary deciduous forest where streams are abundant, which provides an optimal habitat for A. japonica. The annual mean temperature is 14.9 °C, and annual total rainfall is ~1530 mm in Otsu city (Japan Meteorological Agency, https://www.data.jma.go.jp/). Mean daily maximum temperatures in May, June and July 2018 were 24.0 °C, 27.3 °C and 33.7 °C, respectively.
Alpinia japonica typically presents 2–12 inflorescences per plant at the study site. We selected six plants with at least eight inflorescences with more than 10 flowers on each. Unfortunately, one of them had been damaged, probably by park visitors, before the experiments and sampling. Therefore, we used the remaining five plants (Plant ID = N1, N2, N4, N5, N6). In each target plant, eight inflorescences were tagged and covered with a fine net (Cloth Cabin, Suminoe Teijin Techno, Osaka, Japan) prior to flowering to exclude colonization of microbes by flower visitors so that we can detect colonization of microbes better.
We applied one of five treatments to part of the flowers that opened on the inflorescences between 22nd May and 7th June. The flowers under these treatments were used for either identification of microbes on the flowers or for observation of fruit set (Fig.
To identify microbial communities, we sampled 4–5 flowers with each treatment from each plant. The flowers were cropped at ~1700 of the day of flower opening and separately put into a 5 ml plastic tube after removing the stigma (to use for another study) and petal lobes, which protect the showy and sexual parts of the flower (staminodes, anther and stamen) before anthesis (Fig.
To detach microbes from the flower surface, we added 3 ml of phosphate-buffered saline solution (PBS buffer, pH 7.4, Nippon Gene Co., Ltd.) to the tube and sonicated it using an Ultrasonic Disruptor Handy Sonic UR-21P (Digital Biology) at Level 5 power for 30 sec. Microbes in the PBS solution were then filtered using a Sterivex Millipore filter (Sterivex-HV 0.22, Merck). Residual PBS on the Sterivex filter was removed using a Vac-Man Laboratory Vacuum Manifold (Promega).
Microbial DNA was extracted with DNeasy Blood & Tissue Kits (QIAGEN) following the manufacturer’s instructions. A portion of the 16S small-subunit ribosomal gene was PCR-amplified using the primer pairs 515F (5’-GTG YCA GCM GCC GCG GTA A-3’) and 806R (5’-GGA CTA CNV GGG TWT CTA AT-3’) (
We processed the sequence data following
Taxonomic identification was performed for ASVs (Amplicon Sequence Variants; proxy for Operational Taxonomic Units) inferred using DADA2 based on the query‐centric auto‐k‐nearest‐neighbor (QCauto) method (
The treated flowers for the observation of fruit set under all the treatments were hand-pollinated around 1700 using outcrossing pollen to ensure that the fruit set was not limited by pollen. Pollen was taken with a wood toothpick from a cut flower from a pollen donor plant, of which inflorescences had been covered with a mesh bag prior to anthesis to prevent removal of pollen by flower visitors. In total, 251 pollinated flowers were monitored for fruit and seed set (Suppl. material
To examine differences in the prokaryotic diversity among the treatments, we evaluated the Shannon diversity index of each sample using the phyloseq::estimate_richness() function. Significant differences in the index among treatments were tested using the Generalized Linear Mixed Model (GLMM). Prior to the analysis, we confirmed that distributions of the index did not significantly deviate from the normal distribution (Shapiro-Wilk test, W = 0.983, p = 0.0759). We constructed a GLMM with the treatment as a fixed term and Plant ID as a random factor using the R package lme4 (
We examined the variation in the structure of the prokaryotic communities using the ASVs that were present in 20 or more flower samples (36 ASVs satisfied this criterion). The frequency of each ASV was divided by the number of the total sequences of the 36 ASVs in each sample to convert it to a proportion.
To compare microbial communities among flowers under the different treatments, we calculated the Bray–Curtis dissimilarity index using the package vegan (
To find ASVs that tend to co-occur in the samples, we conducted hierarchical clustering of ASVs based on Bray–Curtis dissimilarity using Ward’s minimum variance algorithm using the hclust procedure. The algorithm merges clusters that minimize the increase in the sum of squared distances from the cluster centroid. We chose the number of clusters that maximize the silhouette index (
To assess the factors associated with fruit set and seed production, we used generalized linear mixed models (GLMMs). To evaluate differences in fruit set among the treatments after controlling for the difference among the individuals, we modeled fruit set (retained/aborted) with the treatments as a fixed factor and the individuals and flowering dates as random factors using a binomial error distribution. Then we calculated and plotted estimated marginal means of fruit set for the treatments using the ggpredict() function of the R package ggeffects (
We also modeled seed set (retained/aborted) and tested its association with microbial composition. The model included the proportion of Cluster 6 as a fixed factor and the individuals and fruit as nested random factors. We fitted the model with a binominal error distribution. As for seed set, we could not include it into a model, because some fruits on the same inflorescence under the same treatments were harvested together without distinguishing flowering dates. The GLMM analyses for fruit and seed sets were performed using the R package lme4 (
We obtained 839,113 reads of prokaryotic sequences, which were clustered into 2,828 ASVs. We have removed samples that had < 250 prokaryotic reads (12 samples). This reduced the number of samples to 134 (Suppl. material
Composition of prokaryotic communities on the flowers of Alpinia japonica. Proportions in the samples of the same treatment in the same individuals were averaged. See Suppl. material
The 36 ASVs that appeared in 20 or more samples were identified as bacteria, and accounted for 75.7% of the prokaryotic sequences. NMDS plots of the microbial communities revealed that the flowers under different treatments in each plant individual had different microbial communities, while the same treatment on different plant individuals did not always result in microbial communities with similar compositions. For example, the OT flowers are plotted on the right of the flower bud samples in the plant N1, while the OT flowers are on the left side of the buds in N4 and N5 (Suppl. material
The cluster analyses grouped the 36 ASVs into 11 clusters (Fig.
Heatmap of the prokaryotic communities of flower buds and flowers under different treatments. Samples were vertically arranged and grouped by the treatments as shown on the right. ASVs were clustered hierarchically based on Bray–Curtis dissimilarity using Ward’s minimum variance algorithm. The class of each ASV is shown by the color below the tree. The color intensity in each panel shows the proportion in a sample, referring to the color key on the left.
The Claident pipeline identified the most abundant four ASVs out of the six in Cluster 6 as belonging to Pseudomonas (T00005, T00006 and T0007) and Erwinia (T0004). The three Pseudomonas ASV sequences were different for five to eight bases among 253 (2.0–3.0%). Blast search further identified the other two ASVs (T00028 and T00031) in the dataset as Luteibacter (Rhodanobacteraceae, Gammaproteobacteria) and Massilia (Oxalobacteraceae, Betaproteobacteria) (Table
The 6 ASVs in Cluster 6. Taxonomic information was based on Claident (shown by normal face) supplemented by BLAST search (shown by bold face with percent of identical bases). The results of BLAST search of the ASVs in Cluster 6, and their percent of matches.
ASV | Phylum | Class | Family | Genus | Percent identity1 | Average proportion among samples2 |
---|---|---|---|---|---|---|
T0004 | Proteobacteria | Gammaproteobacteria | Erwiniaceae | Erwinia | 100.0 | 0.086 |
T0005 | Proteobacteria | Gammaproteobacteria | Pseudomonadaceae | Pseudomonas | 0.062 | |
T0006 | Proteobacteria | Gammaproteobacteria | Pseudomonadaceae | Pseudomonas | 0.040 | |
T0007 | Proteobacteria | Gammaproteobacteria | Pseudomonadaceae | Pseudomonas | 0.086 | |
T0028 | Proteobacteria | Betaproteobacteria | Oxalobacteraceae | Massilia | 99.2 | 0.017 |
T0031 | Proteobacteria | Gammaproteobacteria | Rhodanobacteraceae | Luteibacter | 99.2 | 0.015 |
The probability of fruit set of OT predicted by the first GLMM with the treatment as the fixed factor was 57.3%, and that of MI was the lowest (34.7%) (Fig.
Each fruit that remained on the plants under the treatments had 12–24 ovules, 55.6 ± 20% of which had developed. In contrast to fruit set, seed set was not significantly associated with the microbial composition (GLMM, z = 0.685, p > 0.05).
Alpinia flowers harbored diverse prokaryote communities comparable with those on flowers of other plant species reported so far (
Although these major taxa were shared among the samples, prokaryotic community compositions were significantly different among these plant individuals. The differences already existed in the buds prior to anthesis. Since flower microbes are often shared by leaves of the same plant (
We identified a group of ASVs that showed substantial changes after inoculation of old-flower microbes. The cluster analyses grouped the ASVs into 11 clusters based on the co-occurrence among the samples. The microbial inoculation treatment drastically increased Cluster 6 (Figs
Association between the proportion of Cluster 6 and predicted probability of fruit set. a. Variation in the total proportions of the bacteria among the treatments. b. The predicted probabilities of fruit set six weeks after flowering under the different treatments by the GLMM with the treatment as the fixed factor. The thick lines and the boxes indicate the predicted values and their 95% confidential intervals. Distribution of flowers that set fruits (top) or were aborted (bottom) are shown by dots. Colors indicate the five individual plants).
Although the number of studies that investigate microbial communities on flowers is rapidly increasing, no study has examined microbial communities on old flowers as far as we know. Our results suggested that old A. japonicus flowers in our field site were dominated by bacteria of Erwinia and Pseudomonas. Many bacterial species of these genera are frequently found both on flowers and inside the plant and/or on leaves (
Do the microbes that become dominant on old flowers negatively affect the reproductive success of the plant? We observed the lowest fruit set in the microbial inoculated treatment, which recorded the highest abundance of Cluster 6. Besides, we found significant negative associations between the proportion of the ASVs of Cluster 6 and fruit set based on the GLMM analysis, while we did not find such associations for seed set. Since Erwinia and Pseudomonas include plant pathogens that trigger abortion of flowers and fruits (
It is increasingly recognized that flowers have characteristic microbial communities, but the effects and implications of these communities have just started to be explored and our knowledge about them is still very fragmentary (
Tables S1, S2 and Figures S1–S4
Data type: Tables and figures (pdf file)
Explanation note: Table S1. The number of the samples used to monitor fruit set for the five treatments and to analyze microbial communities. Table S2. Characteristics of the 11 clusters. Figure S1. Rarefaction curves of the samples. Figure S2. Variation in the prokaryotic diversity among the treatments and samples. Figure S3. Variation of the prokaryotic communities on the flowers under different treatments. Figure S4. Variation in the total proportions of the 10 ASV clusters among the flower bud and flowers under the five treatments.
Fruit and seed set datasets
Data type: Excel file
Explanation note: Fruit and seed set data.