Bromeliad populations perform distinct ecological strategies across a tropical elevation gradient
Data files
Feb 02, 2024 version files 47.64 MB
-
Dryad_compac2.zip
-
README.md
Abstract
- The effect of environmental gradients on the remarkable diversity of mountain-associated plants and on the species’ abilities to cope with climate change transcends species-specific strategies. For instance, our understanding of the impact of thermal gradients on ecological divergences in populations of widely distributed species is limited, although it could provide important insights regarding species’ response to climate change.
- Here, we investigated whether populations of an endemic species broadly distributed across an elevation gradient employ unique or multiple divergent ecological strategies according to specific environmental conditions. We hypothesized that populations employ distinct strategies, producing a tolerance-avoidance trade-off related to the thermal conditions they experience across elevations.
- We conducted our research with 125 individuals of Pitcairnia flammea (Bromeliaceae) sampled from various elevations spanning from sea level to ~2,200 meters and cultivated under the same conditions. To assess specific ecological strategies of P. flammea populations across elevations, we examined leaf temperature, heat and cold tolerances, as well as other structural/morphological, optical, physiological, and biochemical leaf traits.
- We majorly observed that water-saving traits diminish as elevation increases while membrane fluidity, majorly associated with unsaturated and very-long-chain lipids, enhances. Low-elevation individuals of P. flammea invest in water storage tissues, which likely prevent excessive water loss through the intense transpiration rates under warming periods. Conversely, high-elevation plants exhibit increased membrane fluidity, a possible response to the stiffening induced by low temperature.
- Our results revealed a tolerance-avoidance trade-off related to thermal strategies of populations distributed across an elevation gradient. Low-elevation plants avoid excessive leaf temperature by investing in water-saving traits to maintain transpiration rates. High-elevation individuals, in turn, tend to invest in membrane properties to tolerate thermal variations, particularly cold events.
- Our findings challenge the conventional notion that plants' vulnerability to warming depends on species-specific thermal tolerance by showing diverse thermal strategies on populations across an elevation gradient.
README
This README file was generated on 2024-01-31 by Cleber Chaves.
GENERAL INFORMATION
- Title of Dataset: Bromeliad populations perform distinct ecological strategies across a tropical elevation gradient
https://doi.org/10.5061/dryad.6m905qg71
Author Information
A. Principal Investigator Contact Information
Name: Cleber J. N. Chaves
Institution: Universidade Estadual de Campinas
Email: cleberchaves@gmail.comB. Associate or Co-investigator Contact Information
Name: Clarisse Palma-Silva
Institution: Universidade Estadual de Campinas
Email: cpalma@unicamp.brDate of data collection (single date, range, approximate date): 2019-2023
Geographic location of data collection: Brazilian Atlantic Forest
Information about funding sources that supported the collection of the data: FAPESP 2020/14805-4 FAPESP 2020/16696-8 FAPESP 2020/16696-8 FAPESP 2015/24351-2 FAPESP 2023/01800-2 FAPESP 2021/10639-5 CAPES 88887.361034/2019-00
CNPq 163094/2020-9 CNPq 308260/2021-0 CNPq 306662/2022-1 CNPq 304295/2022-1 CNPq 302962/2022
SHARING/ACCESS INFORMATION
Licenses/restrictions placed on the data: CC0 1.0 Universal (CC0 1.0) Public Domain
Links to publications that cite or use the data:
CHAVES, C. J. N.; CACOSSI, T. C.; MATOS, T. S.; BENTO, J. P. S. P.; SILVA, S. F.; SILVA-FERREIRA, M. V.; BARBIN, D. F.; MAYER, J. L. S.; SUSSULINI, A.; RIBEIRO, R. V.; PALMA-SILVA, C. Bromeliad populations perform distinct ecological strategies across a tropical elevation gradient. Functional Ecology, 2024
Links to other publicly accessible locations of the data: None
Links/relationships to ancillary data sets: None
Was data derived from another source? No
A. If yes, list source(s): NARecommended citation for this dataset:
Chaves, Cleber et al. (Forthcoming 2024). Bromeliad populations perform distinct ecological strategies across a tropical elevation gradient [Dataset]. Dryad. https://doi.org/10.5061/dryad.6m905qg71
DATA & FILE OVERVIEW
- File List: Root 'all_traits.csv' 'script_leaf_physiology.R'
Folder 'Thermal Tolerance'
'Calc_t15t50.R'
'Cold_g1.csv'
'Cold_g2.csv'
'Cold_g3.csv'
'Cold_g4.csv'
'Heat_g1.csv'
'Heat_g2.csv'
'Heat_g3.csv'
'Heat_g4.csv'
Folder 'Lipidome'
'script_lipid_pops.R'
'Area_0_202262927.xlsx'
'Area_0_20226241518 POS BRO ALT.xlsx'
Relationship between files, if important: According to folders
Additional related data collected that was not included in the current data package: None
Are there multiple versions of the dataset? No
A. If yes, name of file(s) that was updated: NA
i. Why was the file updated? NA
ii. When was the file updated? NA
#########################################################################
ROOT
DATA-SPECIFIC INFORMATION FOR: 'script_leaf_physiology.R': R script for all general data analysis
DATA-SPECIFIC INFORMATION FOR: 'all_traits.csv'
- Number of variables: 90
- Number of cases/rows: 126
- Variable List:
- ind: individual ID
- pop: population of origin
- latitude (°)
- longitude (°)
- altitude: elevation (m.a.s.l.)
- temp_mean: average environmental tempearature (extracted from WorldClim) (°C)
- temp_breadth: maximum minus minimum environmental temperature (extracted from WorldClim) (°C)
- temp_min: minimum environmental tempearature (extracted from WorldClim) (°C)
- temp_max: maximum environmental tempearature (extracted from WorldClim) (°C)
- group_heat: methodological group identification to aclimatation for heat tolerance measurement
- group_cold: methodological group identification to aclimatation for cold tolerance measurement
- tc_cold: critical cold temperature estimated from cold tolerance measurement (°C)
- max_temp_cold: minimum cold temperature with estimated with non-null Fv/Fm estimated from cold tolerance measurement (°C)
- critical_zone_cold: max_temp_cold - tc_cold (°C)
- t15_cold: temperature in which the sample reduces 15% of its original Fv/Fm under cold tolerance measurement (°C)
- t50_cold: temperature in which the sample reduces 50% of its original Fv/Fm under cold tolerance measurement (°C)
- tc_heat critical heat temperature estimated from heat tolerance measurement (°C)
- max_temp_heat: maximum heat temperature with estimated with non-null Fv/Fm estimated from heat tolerance measurement (°C)
- critical_zone_heat: max_temp_heat - tc_heat (°C)
- t15_heat: temperature in which the sample reduces 15% of its original Fv/Fm under heat tolerance measurement (°C)
- t50_heat: temperature in which the sample reduces 50% of its original Fv/Fm under heat tolerance measurement (°C)
- tc_diff: tc_heat - tc_cold (°C)
- max_temp_diff: max_temp_heat - max_temp_cold (°C)
- critical_zone_diff: critical_zone_heat - critical_zone_cold (°C)
- t15_diff: t15_heat - t15_cold (°C)
- t50_diff: t50_heat - t50_cold (°C)
- chr: relative water content (RWC)
- is: succulence index
- afe: specific leaf area (SLA) (cm²/g)
- cmfs: dry leaf matter content (DMC) (g)
- af: leaf area (LA) (cm²)
- de: stomata density (units/mm²)
- dt_aba: trichome density at adaxial leaf face (units/mm²)
- dt_ada: trichome density at abaxial leaf face (units/mm²)
- te_larg: stomata width (µm)
- te_comp: stomata length (µm)
- epidermis_ad: adaxial epidermis width (µm)
- epidermis_ab: abaxial epidermis width (µm)
- parenchyma_ch: chlorenchyma width (µm)
- parenchyma_aq: water parenchyma width (µm)
- mesophyll: mesophyll width (µm)
- parenchyma_ch_prop: relative chlorenchyma width (considering mesophyll)
- parenchyma_aq_prop: relative water parenchyma width (considering mesophyll)
- vascular_bundle_length (µm)
- vascular_bundle_width (µm)
- vascular_bundle_area (µm)
- vascular_bundle_perimeter (µm)
- length_phloem (µm)
- length_xylem (µm)
- length_phloem_prop (µm)
- length_xylem_prop (µm)
- reflectance: leaf reflectance value (W m-2 nm-¹)
- absorbance: leaf absorbance value (W m-2 nm-¹)
- transmittance: leaf transmittance value (W m-2 nm-¹)
- Photo: Photosynthetic rate (µmol CO2 m-² s-¹)
- Cond: Stomatal conductance (mol H2O m-² s-¹)
- Trmmol: Transpiration rate (mmol H2O m-² s-¹)
- Ci: Intercelular CO2 concentration (µmol CO2 mol air-¹)
- Fo': Minimal F, dark adapted
- Fm': Maximal F, dark adapted
- Fs: Steady state F
- Fv'/Fm': 1-Fo/Fm
- PhiPS2: 1-Fs/Fm’
- ETR: Electron transport rate (µmol s-1)
- Tair_normal: air temperature (°C)
- Tleaf_normal: leaf temperature (°C)
- T_airleaf_diff_normal: Tleaf_normal - Tair_normal (°C)
- VpdL_normal: Vapor pressure deficit at leaf temp (kPa)
- Tair_vaseline: air temperature on leaf samples with vaseline (°C)
- Tleaf_vaseline: temperature of leaf sample with vaseline (°C)
- T_airleaf_diff_vaseline: Tleaf_vaseline - Tleaf_vaseline (°C)
- VpdL_vaseline: Vapor pressure deficit at leaf with vaseline temp (kPa)
- Tdiff_norm_vaseline: Tleaf_vaseline - Tleaf_normal (°C)
- Tleaf_air_vas_norm_ratio: Tair_vaseline - Tair_normal (°C)
- abs_HexCer: Hexosylceramide content
- rel_HexCer: relative Hexosylceramide content
- abs_PG: Phosphatidylglycerol content
- rel_PG: relative Phosphatidylglycerol content
- abs_VLCFA: Very Long Chain Fatty Acids content
- rel_VLCFA: Very Long Chain Fatty Acids relative content
- abs_sat: saturated lipids content
- rel_sat:: saturated lipids relative content
- abs_monounsat: monounsaturated lipids content
- rel_monounsat: monounsaturated lipids relative content
- temp_IR_left: infrared temperature of the left side of the Leaf (with vaseline) (°C)
- temp_IR_right: infrared temperature of the right side of the Leaf (°C)
- temp_IR_black: infrared temperature of a black paper square (for comparison) (°C)
- temp_IR_left_right: temp_IR_left - temp_IR_right (°C)
- temp_IR_left_black: temp_IR_left - temp_IR_black (°C)
- temp_IR_right_black: temp_IR_right - temp_IR_black (°C)
- Missing data codes: NA
- Specialized formats or other abbreviations used: None
FOLDER 'Thermal Tolerance'
DATA-SPECIFIC INFORMATION FOR: 'Calc_t15t50.R': R script for the estimation of thermal tolerance metrics
DATA-SPECIFIC INFORMATION FOR: 'Cold_g1.csv'/'Cold_g2.csv'/'Cold_g3.csv'/'Cold_g4.csv': raw data of Fv/Fm values of leaf samples under cooling temperatures (cold tolerance measurement).
- Number of variables (individuals): 16/19/17/18
- Number of cases/rows (temperatures): 13/10/10/10
Column names: individuals' IDs; Rows: temperature conditions
Plants were randomly splitted in groups (g1-g4) to reduce time between measurements
DATA-SPECIFIC INFORMATION FOR: 'Heat_g1.csv'/'Heat_g2.csv'/'Heat_g3.csv'/'Heat_g4.csv'/'Heat_g5.csv': raw data of Fv/Fm values of leaf samples under warming temperatures (heat tolerance measurement).
- Number of variables (individuals): 18/18/17/13/9
- Number of cases/rows (temperatures): 12/14/14/14/13
Column names: individuals' IDs; Rows: temperature conditions
Plants were randomly splitted in groups (g1-g5) to reduce time between measurements
FOLDER 'Lipidome'
DATA-SPECIFIC INFORMATION FOR: 'script_lipid_pops.R': R script for the general lipidome analysis
DATA-SPECIFIC INFORMATION FOR: 'Area_0_202262927.xlsx': raw non-normalized data of the intensity of each compound observed in each leaf samples in the negative mode
- Number of variables: 65
- Number of cases/rows: 99994
Column names: individuals' IDs; Rows: each observed compound in the negative mode(lipid)
DATA-SPECIFIC INFORMATION FOR: 'Area_0_20226241518 POS BRO ALT.xlsx' raw non-normalized data of the intensity of each compound observed in each leaf samples in the positive mode
- Number of variables: 55
- Number of cases/rows: 34773
Column names: individuals' IDs; Rows: each observed compound in the positive mode(lipid)
Methods
Plant sampling
To characterize ecological strategies employed by distinct P. flammea populations, over two years, we collected ~20 individuals of P. flammea per population (according to population abundance) from eight localities along an elevation gradient from 0 to ~2,200 m a.s.l. (Fig. 2) for greenhouse cultivation, under the same environmental conditions. To acclimate the same genotype (genets) to warm and cold conditions for the thermal tolerance test (see the following subsection), we prioritized collecting individuals (i.e., genets) with at least two ramets. These genets were split and cultivated in a substrate mixture of equal parts of expanded clay and potting soil for at least one year in randomized blocks under sprinkler irrigation for five minutes, five times daily. The temperature inside the greenhouse during the cultivation averaged 19 °C, the air relative humidity averaged 57.7%, and the maximum photosynthetic photon flux density (PAR) was ~500 µmol m−2 s−1. Below, we outline the methods for evaluating the thermal sensitivity, thermal tolerance, and each plant trait. For methodological clarity, we have categorized the plant traits into biochemical (lipidome), structural/morphological, optical, and physiological groups (Table 1).
Leaf temperature and thermal tolerance
To test whether leaves from distinct P. flammea populations growing under the same conditions sense the environmental temperature and how this difference affects the plants, we selected five individuals per population to be maintained for one hour (at noon) under 30 °C inside a controlled temperature chamber. After, we measured, in a random order, the temperature of the youngest fully expanded leaf of each individual, using an infrared camera (thermal imaging camera, model 871, Testo, Germany), recording 7.5–14 µm infrared spectrum with a thermal sensitivity <0.08 °C (80 mK). We opted to exclusively measure the youngest leaves as they were developed under the greenhouse cultivation period. The leaf emissivity was set to 0.98 (Lopez et al., 2012). We estimated the leaf temperature from the thermal images considering the average temperature measured on the whole leaf using the IRSoft software v. 4.8 (Testo, Germany).
To infer the thermal tolerance range of each sampled P. flammea population while avoiding possible variability depending on the time of day or the season of the year (e.g., Chaves et al., 2018), we measured the heat and cold tolerance of eight to 13 individuals per population (according to the availability of ramets split at the beginning of cultivation) during the summer and winter, respectively. Before the heat and cold tolerance tests, the selected plants were grown for three days inside a controlled chamber with a 12-hour photoperiod and continuous temperatures of 30 °C and 15 °C, respectively (~30% RH). To prevent daily variation in thermal tolerance measures (e.g., Chaves et al., 2018), we measured the heat and cold tolerances shortly after noon and after sunrise, respectively, following Godoy et al. (2011). Leaf discs of 1.5 cm² extracted from the middle of the youngest fully expanded leaf of each individual were adapted to dark for 10 min at room temperature (~25 °C). Then, we proceeded with the first measurement of the potential quantum efficiency of PSII (Fv/Fm) using a pulse amplitude-modulated fluorometer (FMS1, Hansatech, UK). For heat tolerance, we exposed the samples to increasing temperatures from 27 to 60 °C, with a rate of 1 °C increase every 3 minutes, while performing Fv/Fm measurements every 3 °C of temperature increase. For cold tolerance, we put the samples within individual plastic bags, which were exposed to decreasing temperatures from 20 to -22 °C at a similar rate while performing Fv/Fm measurements every 5 °C of temperature decrease. Before the Fv/Fm measurements, samples were maintained at the target temperatures for about one minute, to equalize all leaf sample temperatures and to complete at least 10 min between each measurement. We used an ultra thermostatic water bath (2050, Thoth, Brazil) to increase and decrease temperature. To ensure accurate leaf sample temperatures, we attached K-type thermocouples connected to a digital thermometer to some samples (TH-096, Instrutherm, Brazil). To enable comparisons across populations, we first employed a sigmoid curve fitting approach to the acquired Fv/Fm measurements at each temperature using the 'drc' R-package (Ritz et al., 2015), following Knight and Ackerly (2003), Gimeno et al. (2009), and Godoy et al. (2011). This method allowed us to estimate critical temperatures (Tc) at which there was a significant decrease in Fv/Fm, as well as temperatures corresponding to 15% (T15) and 50% (T50) reduction of the maximum Fv/Fm (refer to Fig. 3A).
Structural/morphological traits
To characterize the leaf morphology, anatomical structure, and leaf surface, we measured 25 traits (see Table 1 for references) from 1.5 cm² leaf discs extracted from the base, middle, and top regions of the youngest fully expanded leaf from eight to 21 individuals per population, all of which were healthy genets. For leaf area (LA) measurement, in particular, we estimated the total area of the sampled leaves, using the WinRHIZO image analysis system (Expression 12000XL, Regent Instruments, Canada). Please refer to the Supplementary Information for details regarding the sample preparation for the measurement of leaf structure and surface.
Physiological and optical traits
To characterize the leaf spectrometry, gas exchange, and PSII photochemistry, we measured nine traits (see Table 1) at noon, from the mid-region of the youngest fully expanded leaf of four to 14 healthy individuals per population. For leaf spectrometry, we estimated the reflectance, absorbance, and transmittance at noon using a handheld portable spectroradiometer (SpectraPen SP 256, PSI, Czech Republic). For leaf gas exchange, we measured the transpiration rate (E), stomatal conductance (gs), photosynthetic rate (Pn), and intercellular CO2 concentration (Ci) using an infrared gas analyzer (LI-6400F, LI-COR, USA), under the air CO2 concentration of ~400 μmol mol−1 and PAR of 800 μmol m−2 s−1. Lastly, we estimated the potential (Fv/Fm, after 15 min under dark conditions) and effective (ΦPS2, simultaneously to leaf gas exchange) quantum efficiency of PSII by using a modulated fluorometer (LCF6400–40, LI-COR, USA) attached to the LI-6400F system, following the pulse saturation method (λ = 630 nm, PAR ~6,000 µmol m-2 s-1, 0.8 s).
Biochemical traits (lipidome)
To characterize the lipid composition of P. flammea leaves, we harvested the mid-region of the youngest fully expanded leaves from five to seven healthy individuals per population in summer morning. This timing was chosen to ensure high air humidity and minimize the likelihood of individuals experiencing drought or thermal stress. The samples were immediately frozen in liquid nitrogen and ground into a fine powder. We used the extraction method from Hummel et al. (2011) adjusted for P. flammea samples according to Matos et al. (2023). We performed ultra-high performance liquid chromatography coupled to mass spectrometry analysis operating a Thermo Scientific UltiMateTM 3000 RSLCnano system equipped with a Titan C18 column (100 mm x 2.1 mm x 1.9 µm particle size, Supelco Sigma-Aldrich, Bellefonte PA, USA), coupled to a Thermo Scientific Orbitrap Q-Exactive (Waltham MA, USA) mass spectrometer. We used MS full-scan followed by MS/MS analysis in the DDA mode for the five most intense peaks. We acquired full scan data between m/z 100 and 1500 in profile mode at 70000 resolution (at m/z 200) with automatic gain control set at 1 x 106 and injection time set to 100 ms. For ESI (+), the spray voltage was 3.5 kV, and for ESI (-), it was 3.2 kV. The ion optics setting was S-Lens RF level 50, S-Lens 25 V; skimmer 15 V, and C-Trap RF 1010 V. For data preprocessing, we converted the raw MS data to .mzXML format using MSConvert 3.0 from ProteoWizard. MS-DIAL 4.9 software was used for data preprocessing, and alignment was done according to Matos et al. (2023). The lipidomics dataset is deposited in the Metabolomics Workbench (Sud et al., 2016) under project ID PR001712 (http://dx.doi.org/10.21228/M8DM8V).
We identified the lipids with an identification score cut-off of 80%. We filtered the peak spot viewer to display only the peaks that matched with the reference libraries and had MS/MS information. The internal lipid annotation in MS-DIAL is based on LipidBlast (Kind et al., 2013). To perform lipid enrichment analysis, we used the Lipid Mini-On software (Clair et al., 2019) that creates lipid ontology (LO) terms from the list of annotated lipids by using a text mining process that categorizes the lipids by their main classes and subclasses, and their total number of chain carbon and double bonds (unsaturations). The LO terms generated and their abundances are used as a query list to compare with the available lipid universe and generate a network of the over-representative LOs (see Fig. S5) based on Fisher’s exact test (p < 0.05; Clair et al., 2019). Finally, to further compare across the elevation gradient, we analyzed the absolute and relative abundance of lipids from each over-represented class (i.e., hexosylceramide, phosphatidylglycerol, and very long chain fatty acids, a.k.a. VLCFA), as well as the abundance of lipids with distinct levels of unsaturation (including saturated, monounsaturated, diunsaturated, and polyunsaturated).