Zooming in on the temporal dimensions of plant-soil feedback: plant sensitivity and microbial dynamics
Data files
Oct 02, 2024 version files 3.89 MB
-
metadata_XL.xlsx
3.88 MB
-
README.md
6.09 KB
Abstract
The magnitude of plant-soil feedback (PSF) can depend on the time of conditioning as well as the length of feedback. Understanding the temporal variation in PSF requires insight into the response of both soil characteristics and the plant. We examined how conspecific PSF varies with the length of conditioning and the size of the response plant using Jacobaea vulgaris, a species known to experience negative conspecific PSF. Together with a reanalysis of an existing microbial sequence dataset, we tested whether the temporal variation in PSF is due to size-dependent plant sensitivity to conditioned soil or due to compositional changes in microbial communities of conditioned soil. Further, by reanalyzing another existing dataset, we examined the temporal dynamics of the relative growth rates (RGR) of J. vulgaris during the feedback phase. Testing varying conditioning lengths, uncovered that J. vulgaris exhibited the strongest negative PSF at 5 weeks of conditioning, after which PSF gradually attenuated. Plant sensitivity to conditioned soil decreased with increasing plant age/size of the response plant. In the feedback phase, the RGR of J. vulgaris was first higher, then lower, and at the end similar in “away” soil compared to “home” soil. The dissimilarity in bacterial and fungal communities in “home” and “away” soil significantly decreased during the feedback phase. When J. vulgaris grew in “away” soil, the relative abundance of 10 (out of 80) bacterial OTUs that positively correlated with plant growth decreased over time, while 5 (out of 86) OTUs that negatively correlated with plant growth the relative abundance increased over time. Additionally, only one (out of 10) fungal OTU that was negatively associated with plant growth increased over time in “away” soil. Our findings illustrate that PSF varies with the duration of soil conditioning and the feedback phase. During the feedback phase, changes in PSF can be attributed to both the size-dependent plant sensitivity to conditioned soil and temporal changes in the microbial community of conditioned soil. This highlights the importance of considering the reconditioning of soil microbial communities by the test plants during the feedback phase for understanding temporal variation in PSF.
README: Zooming in on the temporal dimensions of plant-soil feedback: plant sensitivity and microbial dynamics
[Access this dataset on Dryad] (https://doi.org/10.5061/dryad.41ns1rnq6)
The metadata (metadata_XL.xlsx) contains data from three plant-soil feedback experiments with the focal plant, Jacobaea vulgaris.
Description of the data structure
Sheet1: Biomass data from experiment 1
- ColumnA: Soil.type (soil conditioned by which plant, 3 categories), JV represents soil conditioned by J. vulgaris, HL represents soil conditioned by H. lanatus, C represents unplanted soil.
- ColumnB: Conditioning.time (duration of conditioning, 3 levels), 2, 5, and 8 indicate soil that was conditioned for 2, 5 and 8 weeks.
- ColumnC: Response.plant (plant species that used as test plant, 1 level), JV indicates J. vulgaris plant.
- ColumnD: Response.plant.age (age/size of response plant, 4 levels), 0, 2, 5, 8 indicate seedlings, 2-week-old, 5-week-old and 8-week-old plants, respectively.
- ColumnE: Replicate (number of replicates), there were 5 replicates in this experiment.
- ColumnF: Total.fresh.biomass.t0 (unit gram) means total fresh biomass of individual test plant (J. vulgaris) before transplanting to different condition soils.
- ColumnG: Total.fresh.biomass.4.weeks (unit gram) means total fresh biomass of individual test plant (J. vulgaris) after 4 weeks of response to condition soils.
- Column H: Water.content (unit %) means water content of 2-week-old, 5-week-old and 8-week-old plants before transplanting to different condition soils. Because seedlings were homogeneous and too small to measure water content, water content of seedlings was NA.
- Column I: Total.dry.biomass.t0 (unit gram) means total dry biomass of seedlings, 2-week-old, 5-week-old, and 8-week-old plants before transplanting to different condition soils.
- Column J: Shoot.dry.mass (unit gram) means shoot dry mass of individual plant after 4 weeks of response to condition soils.
- Column K: Root.dry.mass (unit gram) means the root dry mass of an individual plant after 4 weeks of response to condition soils.
- Column L: Total.dry.mass (unit gram) means the total dry mass of an individual plant after 4 weeks of response to condition soils.
- Column M: derta.total.dry.mass (unit gram) represents the difference in total dry mass of an individual plant before and after the 4 weeks of feedback.
- Column N: derta.total.fresh.mass (unit gram) represents the difference in total fresh mass of an individual plant before and after the 4 weeks of feedback.
NA means no data available. NAs in biomass data (Columns F-G, Columns I-N) either because the plant died or we lost the plant or the measurement.
Sheet2: nutrient data from experiment 1
- ColumnA: conditioning plant (plants that conditioned soil, 3 categories), JV represents soil conditioned by J. vulgaris, HL represents soil conditioned by H. lanatus, and C represents unplanted soil.
- ColumnB: Duration of conditioning (duration of conditioning, 3 levels), 2, 5, and 8 indicate soil that was conditioned for 2, 5, and 8 weeks.
- ColumnC: Replicate (number of replicates), there were 5 replicates (1 to 5) in this experiment.
- ColumnD: SOM (soil organic matter), numeric values, NA means no data available.
- ColumnE: pH (soil pH), numeric values.
- ColumnF: NH4(mg/kg) (Ammonium, mg/Kg is unit), numeric values.
- ColumnG: NO3(mg/kg) (Nitrate, mg/Kg is unit), numeric values, NA means no data available.
- Column H: P(mg/kg) (phosphate (PO43-), mg/Kg is unit), numeric values, NA means no data available.
NAs due to there not being enough soil samples left from Phase 1.
Sheet3: Total.dry.mass from experiment2
- ColumnA: focal.plant means test plant in the feedback (1 category), JV represents J. vulgaris.
- ColumnB: Time.point means the length of feedback phase, numeric values, the unit is in days.
- ColumnC: soil.type (soil conditioned by which plant, 3 categories), JV represents soil conditioned by J. vulgaris, HL represents soil conditioned by H. lanatus, and C represents unplanted soil.
- ColumnD: replicate means the number of replicates of this experiment, there were 3 replicates.
- ColumnE: total.dry.mass unit in grams, numeric values.
Sheet4: Bacteria OTU in Exp3
- ColumnA: OTU names
- ColumnB-BA (52 columns): there were 52 samples left after microbial data processes (described in data analysis). The values were relative abundance.
Sheet5: Bacteria Tax table
- ColumnA: OTU names
- ColumnB-H: Ranks 1-7 are kingdom, phylum, class, order, family, genus, and species.
Sheet6: Bacteria sample data
- ColumnA: Alias indicates the name of samples.
- ColumnB: pot indicates the label of the pot.
- ColumnC: soil type indicates soil conditioned by which plant, 2 categories, JV represents soil conditioned by J. vulgaris, HL represents soil conditioned by H. lanatus.
- ColumnD: plant species indicates the response plant, 1 category, JV represents J. vulgaris.
- ColumnE: time means at which time point the soil was sampled.
- ColumnF: rep means replicate, there were 3 replicates in this experiment.
Sheet7: Fungal OTU in Exp3
- ColumnA: phylotype names
- ColumnB-BB: (53 columns): there were 53 samples left after microbial data processes (described in data analysis). The values were relative abundance.
Sheet8: Fungal Tax table
- ColumnA: phylotype names
- ColumnB-H: Ranks 1-7 are kingdom, phylum, class, order, family, genus, and species.
Sheet9: Fungal sample data
- ColumnA: Sample.ID indicates the name of the samples.
- ColumnB: pot indicates the label of the pot.
- ColumnC: soil type indicates soil conditioned by which plant, 2 categories, JV represents soil conditioned by J. vulgaris, HL represents soil conditioned by H. lanatus.
- ColumnD: plant species indicates the response plant, 1 category, JV represents J. vulgaris.
- ColumnE: time means at which time point the soil was sampled.
- ColumnF: rep means replicate, there were 3 replicates in this experiment.
Methods
Experiment 1: Duration of conditioning and response plant size
Phase 1
A large amount of soil was collected from a grassland near Driebergen, The Netherlands, and was stored on a concrete base and covered with plastic sheeting for several years at the Netherlands Institute of Ecology, Wageningen, The Netherlands. The soil is characterized as holt podzol sandy loam with a particle size distribution: 2% < 0.002 mm, 11% 0.002-0.063 mm, 84% > 0.063 mm, with ~3 % organic matter, 1,150 mg kg-1 N, 61 mg P2O5 100 g-1, 2.4 mmol K kg-1 and pH 5.9. The soil was sieved (1 cm mesh size) to remove stones and large root fragments. Seeds from both species were sterilized for 20 minutes in 5 % sodium hypochlorite solution and rinsed with sterilized MillQ water afterward. Seeds of both species were germinated on sterile glass beads in a climate room with 400 W high-pressure lamps at a 16/8 h light-dark regime and a 20/15°C temperature regime. Seedlings were grown on glass beads for one week after germination and then stored at 4°C until further use.
In the first phase, the soil was conditioned by growing either J. vulgaris or H. lanatus in monocultures (5 plants per pot) in pots (13 cm × 13 cm × 13 cm) filled with 1.45 L homogenized and sieved soil. A microbial soil inoculum was added to each pot as the live soil originated from the pile where no plants were grown and had been stored in 10 kg bags for several weeks before the setup of the first phase. The soil for the inoculum was collected from a grassland area in Leiden, The Netherlands. The soil was sieved through a 1 cm mesh to remove stones and large root fragments, mixed with water (1 soil: 2 water), and stirred. The soil slurry was then sieved through a 500 µm sieve and sieved liquid was stored at 4 °C. We added 50 ml soil inoculum and 50 ml Steiner nutrient solution to each pot. A third set of pots was kept unplanted. All planted seedlings were similar in size. Seedlings that emerged from the soil seed bank were removed. The preparation was initiated at different times so that plants were grown for eight, five, and two weeks to create soils that were conditioned for different periods of time (Fig. S2). In total, there were 117 pots (eight weeks of conditioning: three treatments [J. vulgaris monocultures, H. lanatus monocultures or “unconditioned” soil] × 15 replicates + five weeks conditioning: three treatments × 12 replicates + two weeks conditioning: three treatments × 12 replicates). All pots were watered once every two days when needed. After plants had grown for the designated period, all aboveground biomass was harvested from each pot, and large root fragments were removed from the soil using a 4 mm sieve. All pots were harvested on the same day. Soil from different pots that belonged to the same treatment was homogenized and mixed. Subsequently, there were five batches of soil conditioned for eight weeks (each batch was a mixture of soil from three pots), six batches of soil conditioned for five weeks, and six batches of soil conditioned for two weeks (each batch was a mixture of soil from two pots). To arrive at five soil batches, for the 5- and 2-week conditioned soil, the sixth batch was divided equally and mixed with the other five batches of soil. In total, there were 45 batches of soil (three conditioning treatments, three conditioning durations, and five replicates).
Preparation of plants of different ages
Prior to starting the feedback phase, we grew J. vulgaris alone in the gamma-irradiated sterilized soil (> 25 kGray; Synergy Health, Ede, The Netherlands). The soil was the same as described above. Plants were grown for eight, five, and two weeks. Seedlings of J. vulgaris were prepared as described above. Eight weeks before starting the feedback phase, a single seedling was planted in each of 17 pots (8 cm × 8 cm × 9 cm) filled with 500 g sterilized soil. Five weeks and two weeks before starting the feedback phase, a single seedling was planted in 50 pots (8 cm × 8 cm × 9 cm) filled with 500 g sterilized soil. Ten days before starting the feedback phase, seeds were germinated as described before. The seeds germinated after five days. Therefore, plants were eight, five, or two weeks old and seedlings were one week old when planted in the feedback phase (Fig. S2). Two extra 8-week-old and five extra 5- and 2-week-old plants were harvested, weighed, and oven-dried at 70°C for at least 72 hours to determine the fresh weight / dry weight ratio and water content.
Phase 2
A total of 150 pots (8 cm × 8 cm × 9 cm) were filled with 500 g soil from the conditioning phase. To test whether PSF is influenced by the age (size) of the response plant, we carefully washed the roots of the 8-, 5- and 2-week-old plants from the soil and transplanted them in pots with three types of soils. Water was absorbed from the roots using tissue paper and the total fresh biomass of each plant was determined before planting. Seedlings grown on sterile glass beads were also planted in the soils. A separate set of seedlings was weighed to determine fresh weight. For soil conditioned for eight weeks, there were 60 pots: four different response plant ages × three types of soil (conditioned by H. lanatus, conditioned by J. vulgaris, no plant previously) × five replicates). For 5- and 2-week conditioning soils, there were not enough 8-week-old plants and hence there were 45 pots for those sets (three response plant ages [without 8-week-old plants] × three types of soil (conditioned by H. lanatus, conditioned by J. vulgaris, no plant previously) × five replicates). To improve the probability of survival after transplanting, plants were watered daily when needed. The climate conditions in the plant growth room were as described above. Pots were randomly placed in the growth room. After four weeks of growth, the soil was carefully rinsed and the roots were cleaned in water. Hereafter, we removed adherent water from the plant with tissue paper and recorded total fresh biomass. The root and shoot material of each plant was then separated and oven-dried at 70°C for at least 72 hours and weighed.
Besides the experiment described above, we reanalyzed data from two published experiments on the response of J. vulgaris to soil conditioned by H. lanatus and J. vulgaris (Bezemer et al., 2018; Steinauer et al., 2023). The experimental details and methods of these two experiments are described in the original publications. A summary of the experimental conditions of all three PSF experiments and the total dry mass of J. vulgaris in “home” and “away” soil over the overlapping seven weeks in these two existing PSF experiments are presented in the supplementary information (Table S1, Fig. S3).
Soil properties
Soil samples collected from each batch (45 in total) at the end of Phase 1 were analyzed. Soil samples were oven-dried at 40°C and sieved through a 0.5 mm mesh. Ten g of dried soil was shaken for 3 hours in 25 ml 0.1 M calcium chloride (CaCl2) solution. Hereafter, soil pH was measured in the resulting soil: calcium chloride (CaCl2) (1: 2.5) suspension with a pH meter. Another 3 g of dried soil was shaken for 3 hours in 25 ml 0.1 M calcium chloride (CaCl2) solution, and the filtrated suspension was used to measure orthophosphate (PO43-) following the ascorbic acid method (Clesceri et al., 1998). Five g of soil was shaken for 3 hours with 25 ml 1 M potassium chloride (KCl) solution, and soil ammonium (NH4+) and nitrate (NO3-) were determined using the vanadium chloride method (Doane and Horwath, 2003). The absorbance was measured using a Tecan Spark 10M Microplate Reader (Männedorf, Switzerland) and the corresponding phosphorus, ammonium, and nitrate concentrations were calculated based on the standard curves (details described in the supplementary methods). Soil organic matter content was estimated by loss-on-ignition (LOI) analysis (Ball, 1964). Approximately 28 g dried soil samples were dried at 105°C for 16 h and weighed, then samples were heated at 550°C for 5 h and reweighed again. Soil organic matter was calculated as the percentage of weight loss.
Experiment 2: Changes in relative growth rates during the feedback phase
To examine the influence of soil conditioning effects on the relative growth rates of J. vulgaris, we used plants that were grown individually in this experiment. Briefly, one seedling of J. vulgaris was planted in pots (10 cm × 10 cm × 11 cm) filled with 1 kg of soil conditioned for eight weeks by monocultures of J. vulgaris or H. lanatus. Here we used data from 114 pots (two soil conditioning treatments × 57 pots). Nineteen times (twice a week), during the feedback phase three plants from both treatments were harvested (week 2–11 after transplanting). The total plant dry mass after oven drying at 70°C was then determined.
Experiment 3: Changes in rhizosphere bacterial and fungal communities during the feedback phase
Eighty pots (10 cm × 10 cm × 11 cm) were filled with 1 kg homogenized live soil (from the same source and pile as described above), and 40 J. vulgaris and 40 H. lanatus plants were grown individually in these pots for 10 weeks. After 10 weeks of soil conditioning, plants were removed from the soil. In the feedback phase, 54 pots (10 cm × 10 cm × 11 cm) were filled with a homogenized mixture of 1.62 kg of sterilized soil from the same field (gamma-irradiated > 25 kGray; Synergy Health, Ede, The Netherlands) and 0.18 kg of conditioned soil (10% inoculum). For the feedback phase, rooted J. vulgaris plantlets from a tissue culture were used. We used the data from 54 pots (two soil treatments × 27 pots). Plants were harvested weekly for nine times during the feedback phase (week 2-10). During each harvest, rhizosphere samples were collected for molecular identification of the soil microbial community (Illumina sequencing of bacteria and fungi), and the shoot and root biomass of each plant was determined. A detailed description of the DNA extraction and information about sequencing and primers are presented in the original paper (Steinauer et al., 2023).