Location and plant growth conditions
The experimental farm “Cañada Honda” is located near the village of Librilla (Region of Murcia, Spain). It is a semi-arid area (Mediterranean climate), with an average annual rainfall of 300–350 l/m2 and average annual temperatures of 18.8 ºC (Elvira-Rendueles et al. 2019).
Seven-year-old lemon trees of the variety "Fino" (Citrus x limon) with drip irrigation were used. This variety is characterized by a large harvest from November to March. The lemon trees are grown with organic methods; being certified 100% organic in 2021. The soil showed a pH of 8.66 ± 0.02 and an electrical conductivity (EC) of 139.33 ± 23.26 µS cm− 1. Each tree was irrigated with 2 drippers twice a week, for 90–120 minutes and the water flow rate was set at 4 l/h per dripper. The crop was organic with manure and organic fertilization as reported previously (Olmos et al. 2024).
The experiment was divided into four different lines with ten trees per treatment, with a 3x5 crop frame, comparing soil cover with white/black plastic film mulch, mulching with crushed dry pruning and an outdoor control over a period of six months. Samples were collected between March-July 2023. To make the sampling representative of the crop, samples were randomly taken from five different trees, avoiding crop edge.
The white and black plastic (130 g m−²) were composed of semi-impermeable polypropylene geotextile; while the dry pruning was obtained from the pruning of lemon trees. Dry pruning was used after open air dried, crushed with a tract, and stored for 3 months in the open air before being used as mulch.
Vegetative growth
Ten measurements of the new leaf area were made with an interval of 15 days between each measurement from March to July, selected as points represented on the graph in the months of March, May and July. Three leaves on each plant from each of the treatments were previously selected and marked. This procedure was carried out in situ by drawing the outline of the leaf on a sheet of paper and subsequently using the ImageJ program (Tsung-Luo 2017) to calculate the area, and then obtain the relative growth rate with the following equation: RGR= (Ln A2-Ln A1) / (t1-t2), (cm2 cm− 2 day −¹). The measurement of lemon fruit growth was calculated by measuring its caliber with a caliper, followed by the procedure of the aforementioned programme.
Stomata content
Stomatal printing on the new leaves of the tree was carried out to count how many stomata were open or closed on the leaf per unit area. To do this, the surface of the underside of the leaf was impress on a slide with adhesive tape (composed of cellulose acetate), digested with a drop of acetone. In this way, the entire surface of the was impressed on the adhesive tape. Subsequently, the impressions were observed under an optical microscope (OLYMPUS U-CMAD3, Olympus Corporation, Tokyo, Japan) and a counted with an ImageJ analysis program (Jinn 2017). Five plates of new leaves were obtained for each of five trees of each treatment.
Photosynthetic parameters
Photosynthetic capacity (An, µmol CO2 m− 2 s− 1), stomatal conductance (Gs, mol H2O m− 2 s− 1), internal content of CO2 (Ci, µmol mol⁻¹) were measured in fully-expanded new leaves using a TPS-2 Portable Photosynthesis System (PP Systems, Inc., Amesbury, MA, USA). Intrinsic water use efficiency (WUEi µmol CO2 mol− 1 H2O) was calculated by dividing the net photosynthetic rate and the stomatal conductance. Five new leaves were measured for each of the five trees in each treatment.
Soil and leaves mineral content
The macro and micro mineral contents were analysed using Inductively Coupled Plasma-Optical Emission Spectrometry (ICP-OES) on a Thermo ICAP 6500 Duo instrument (Thermo Fisher Scientific, Waltham, MA, USA). Leaves and soil samples were collected, dried, and ground into a fine powder. A total of 200 mg of each sample was added to a 25 mL tube along with a mixture of 4 mL of 68% purity HNO3 and 1 mL of 33% purity H2O2 for digestion. Additionally, a Teflon reactor contained 300 mL of high-purity de-ionized water, 30 mL of 33% purity H2O2, and 2 mL of 98% purity H2SO4 was added. The microwave heating digestion program consisted of three steps: starting at 20 ºC and 40 bar, increasing by 10 bar per minute for30 min until reaching 220 ºC, and maintaining the temperature at 220 ºC for 20 min. After cooling, the mineralized samples were transferred to 10 mL (for micro minerals) and 25 mL (for macro minerals) double gauge tubes, and the volume was adjusted using high-purity de-ionized water. Calibration standards were prepared using a multi-mineral standard solution supplied by SCP Science (Quebec, Canada).
Soil temperature and moisture
The soil temperature was measured using a precision thermometer (Precision Plus, ETI Ltd, Worthing, West Sussex, United Kingdom) obtaining six measurements (three being in the superficial part of the soil under the tarp and the other three at 15 cm from depth) from each of the selected sampling point of each of the five selected trees in each treatment, at 15 day-intervals between each measurement from March to July.
To determine the moisture content, a determined quantity of soil was weighted and placed in an oven with temperature range of 110 ± 5 ºC for about 24 hours. After that, the difference in the wet mass and dry mass of the soil was the water content of the soil. It was determined by triplicate per treatment.
Dehydrogenase activity
Dehydrogenase activity in the soil was determined using a colorimetric procedure according to (von Mersi and Schinner 1991). Briefly, 2 mL of 2-(4-iodophenyl)-3-(4-nitrophenyl)-5-phenyltetrazolium chloride (INT, 0.015 M) were added to 2 g of soil and then homogenized and incubated at 25 ºC for 4 hours in the dark. Subsequently, 8 mL of acetone were added to all samples and put them on an orbital shaker (250 rpm) for 1 hour in the dark. Iodonitrotetrazolium formazan (INTF) was determined in the centrifuged extracts by measurement at 485 nm spectrophotometrically. The dehydrogenase activity was expressed as nmol INTF g− 1 h− 1.
Metabarcoding: DNA Extraction, Amplification, High-Throughput Sequencing, and Library Processing
DNA was extracted from soil samples (500 mg) using the DNeasy Power Soil Pro Kit (Qiagen) following the manufacture’s protocol. The quantity and quality of the DNA extracts were quantified using a Nano Drop 2000 fluorospectrometer (Thermo Fisher Scientific, Waltham, MA, USA).
As stablished in the Molecular Biology Service at the University of Murcia, purified DNA was used as the template for generating a 16S rRNA gene library. The oligonucleotide primers used for this experiment were 5′-TCGTCGGCAGCGTCAGATGTGTATAAGAGACAGCCTACGGGNGGCWGCAG-3′ and 5′-GTCTCGTGGGCTCGGAGATGTGTATAAGAGACAGGACTACHVGGGTATCTAATCC3′, where the underlined regions are the Illumina adapter overhang nucleotide sequences, while the non-underline sequences are locus-specific sequences targeting conserved regions within the V3 and V4 domains of prokaryotic 16S rRNA genes. The amplified fragments were quantified with the Qubit dsDNA HS Assay Kit (Invitrogen, Merelbeke, Belgium) on a Qubit 2.0 Fluorometer prior to sequencing. Paired-end sequencing of the library was performed on an Illumina MiSeq sequencer (San Diego, CA, USA) using the MiSeq Reagent Kit (v3) with the longest read length set to 2 × 300 base pairs (bp). Library qualities were estimated using the Bioanalyzer High Sensitivity DNA Analysis Kit (Agilent).
The 16S-V4 sequencing library was first reviewed with FastQC (Andrews and Others, 2010) for overall quality assessment, and the libraries were processed in R package DADA2 (v.1.8.0) (Callahan et al. 2016). Reads were quality trimmed with the “filterAndTrim” function with “maxEE (2,5)”. Reads below 165 bp after the trimming process were discarded. Errors learned from all samples were used for sample inference with the dada2 algorithm by employing an evaluation of 1E8 bases. Forward and reverse reads are merged below to generate a table of sequences, and the resulting Amplicon Sequence Variants (ASVs) were subjected to de novo chimera detection, using DADA2 and any artifacts were removed.
ASV table generation and taxonomic characterization
For bacteria taxonomic assignment, ASVs were queried against the SILVA database v.132 using IDTAXA (Murali et al. 2018) implemented in the R package DECIPHER (Wright 2016) with a threshold 40. Sequences identified as non-bacterial were discarded. Similarly, to numerous recently published studies, we chose to forego rarefaction of our samples as it increases uncertainty in relative abundances (McMurdie and Holmes 2013).
Alpha and beta diversity analysis
The abundance matrix, the taxonomy assignment and the metadata obtained from each samples were merged and imported with the phyloseq v3.12 package (McMurdie and Holmes, 2013). The "prune_taxa" function was used to keep only subsystems with absolute abundance > 0.01%. Counts were normalized in each sample using the median sequencing depth, and phylum and class level plots were created using the ggplot2 and ggpubr packages. Alpha diversity was calculated in R using the phyloseq package, and several alpha indices were generated, such as Shannon and Simpson, using the function “plot_richness”. Beta diversity was calculated using weighted and unweighted Unifrac distances (Lozupone and Knight 2005). To test for significant differences in community composition among different seasons, permutational multivariate analysis of variance using distance matrices (PERMANOVA) was conducted using the Adonis function in the R package vegan with 999 permutations, and the results were visualized by Principal Coordinates Analysis (PCoA).
UpSet plots were created using the upsetR package version 1.4.0. This was done by transforming the data frame of average counts for each soil sample into a data frame that exclusively contained 0 and 1 values. Subsequently, the data was arranged based on the frequency of intersection size.
RDA analysis
Distance-based redundancy analysis (db-RDA) was conducted to identify soil physicochemical properties with a significant impact on soil bacterial communities across different factors using the dbrda function of the R vegan package vegan v2.6-1. Parameters that significantly explained variation in the bacterial community were identified using forward selection (the ordistep function of the vegan package) with p value < 0.05.
Statistical analysis
Statistical analyses were performed using the SPSS 29.0.0.0 (241). Significant differences between the values from all parameters were determined at p ≤ 0.05, according to a one-way ANOVA followed by Duncan’s test. For the studies of alpha diversity, also Student test was used and for the beta diversity, also Adonis was used. All the results are presented as the mean ± SE.