Biorefining of Anaerobic Digestates for the Recovery of Biostimulants and Bioelicitors for Immune Priming and Plant Protection

Biorefining of Anaerobic Digestates for the Recovery of Biostimulants and Bioelicitors for Immune Priming and Plant Protection Published as part of Environmental Science &Technology special issue “Advancing a Circular Economy”. Marco Greco, Daniele Coculo, Angela Conti, Savino Agresti, Daniela Pontiggia, Hugo Mélida, Lorenzo Favaro, and Vincenzo Lionetti* Cite This: Environ. Sci. Technol. 2025, 59, 21700−21714 Read Online ACCESS Metrics & More Article Recommendations * sı Supporting Information ABSTRACT: Olive oil production is a major global agricultural industry that generates significant waste, particularly olive pomace, which poses environmental and economic challenges. Anaerobic digestion emerges as a promising solution for its valorization into biogas. However, the resulting digestate remains underutilized and its long-term environmental impact is uncertain. Traditional disposal methods are costly and inefficient, underscoring the need for more sustainable approaches. In this study, olive pomace digestate was biorefined, and its components were upcycled into soil amendments and plant immunostimulants. Metagenomic analysis revealed a diverse microbial community in the liquid fraction. A microbial-enriched protein extract (MIPE) was obtained, containing precursors of microbeand damage-associated molecular patterns, including Flagellin, Elongation Factor Tu, and the plant phytocytokine Golven. Plant treatment with MIPE triggered a rapid immune response, characterized by oxidative burst, mitogen-activated protein kinase activation, and the upregulation of defense-related genes such as CYP81F2,FRK1, and WRKY53. MIPE-induced priming enhanced Arabidopsis and tomato resistance to Botrytis cinerea and Pseudomonas syringae. Our findings highlight olive pomace digestate as a valuable growth biostimulant, with its liquid fraction also representing a promising resource of immunity bioelicitors. This refinement valorizes olive mill waste, providing a sustainable alternative to chemical fertilizers and pesticides and supporting sustainable agriculture. KEYWORDS: olive pomace valorization, digestate metagenomics, plant immunity stimulation, circular waste upcycling, sustainable agriculture 1. INTRODUCTION A significant global increase in olive oil extraction and consumption has been recorded, driven by its appealing qualities and the growing recognition of its health benefits. 1 Compared to traditional olive oil extraction methods, such as pressing and three-phase centrifugation, the two-phase system produces only two products, oil and wet pomace, without generating a separate wastewater phase. 2 This process is more efficient and sustainable, yielding more oil, using less water, minimizing waste, and preserving the oil quality and nutritional properties. Two-phase olive pomace, the main byproduct of this system, is an acidic, organic-rich material consisting primarily of olive skin, pulp, stone fragments, water, and water-soluble compounds including oligosaccharides, polyphenols, and mineral elements. 3,4 Its management remains challenging, as conventional methods such as composting and specialized treatments are costly. 5 Consequently, olive pomace is often improperly disposed of on soil, which can exacerbate soil degradation and contribute to water eutrophication. 6,7 Developing sustainable methods to convert olive pomace into valuable bioproducts is therefore crucial to reduce costs and minimize environmental impact. 3,4 Anaerobic digestion (AD) is gaining attention as a promising biological solution for valorizing olive pomace by converting it into biogas and bioenergy. 8,9 This process enhances the profitability of the olive oil industry while mitigating greenhouse gas emissions, as the CO2released is biogenic and displaces petroleum-based fuels. 10−12 As a result, more olive oil mills are adopting biogas plants to leverage olive Received: March 19, 2025 Revised: September 20, 2025 Accepted: September 23, 2025 Published: September 30, 2025 Articlepubs.acs.org/est © 2025 The Authors. Published by American Chemical Society 21700 https://doi.org/10.1021/acs.est.5c03321 Environ. Sci. Technol. 2025, 59, 21700−21714 This article is licensed under CC-BY 4.0 Downloaded via TU GRAZ on November 12, 2025 at 18:32:40 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles. pomace as a valuable biomass resource in anaerobic digestion systems. 13,14 AD converts organic matter into biogas through hydrolysis, acidogenesis, acetogenesis, and methanogenesis, driven by bacteria and archaea. 15,16 Alongside biogas, the process generates a semisolid or liquid residue known as digestate. While research has optimized AD process, digestate valorization remains underexplored. 17 Digestates pose challenges due to its high moisture content, which increases volume and weight, complicates handling, transport, and nutrient utilization, and raises transportation costs. 18 Incineration releases harmful carbon gases, contributing to pollution. Digestates could hinder the biogas industry’s growth without sustainable solutions and harm the environment. 19,20 Digestates were proposed as a fertilizer because it contained partially degraded plant-derived organic matter, water, and essential nutrients. 21,22 However, the nutrient composition of the digestate may be unbalanced. Its soil application can cause environmental issues, including ammonia and greenhouse gas emissions, nutrient leaching, pathogen spread, and micropollutant contamination. 19,23 However, the digestate can harbor a diverse community of agronomically beneficial microorganisms, including plant growth-promoting bacteria, denitrifying and nitrifying bacteria, and nitrogen-fixing bacteria. 24 Arbuscular mycorrhizal fungi and saprophytic fungi can also be present in digestates due to their resilience as spores or contamination from raw materials and the environment. 25 However, the digestate may also contain pathogenic bacteria, opportunistic fungi, and antibiotic-resistant microbes, which could pose risks to plant health, soil quality, and even human and animal safety if not properly managed. 26 The effects of olive pomace digestate on plant growth and productivity have not yet been investigated. Interestingly, microbes and plant organic matter present in digestates may serve as elicitors to enhance plant immunity. This digestate valorization could be relevant for reducing reliance on chemical pesticides and minimizing health risks to workers and consumers. 27 The plant immune system detects pathogenic microbes through pattern-recognition receptors (PRRs), which are either receptor kinases (RKs) or receptor proteins (RPs) located on the cell surface. 28 These PRRs recognize conserved molecules from microbes, known as microbe-associated molecular patterns (MAMPs), triggering pattern-triggered immunity (PTI). 29,30 This recognition initiates a signaling cascade that can result in the production of antimicrobial compounds, the reinforcement of the plant cell wall (CW), and activation of additional defense mechanisms to limit pathogen spread. 31−33 MAMPs include fragments of flagellin (e.g., flg22), translation elongation factor EF-Tu, β-glucans, chitin, ergosterol, lipopolysaccharide, elicitin, and harpin. 34−39 Importantly, danger signals can also arise from immunogenic plant host factors. 40 The PRRs also detect endogenous danger molecules, including phytocytokines, cytosolic proteins, peptides, nucleotides, amino acids, and damage-associated molecular patterns (DAMPs), such as the oligosaccharides, from the degradation of the plant CW. 41 Pectin fragments, such as oligogalacturonides (OG), are known elicitors. 42 MAMPs’ and DAMPs’ application to plants can confer a greater ability to detect pathogens and activate defense responses faster than untreated crops. 43 This can be a consequence of priming, a sensory state that enables plants to “remember” previous stress exposures and mount faster, stronger, and less energy-demanding defense responses improving plant resistance. 44,45 This “primed state”, based on partial preactivation of immune pathways, can be induced by microbes, bioactive compounds, or chemicals, offering a sustainable way to enhance crop resilience. 43 Nevertheless, excessive applications of such elicitors may induce a hyperimmune response, causing the plant to strike a costly growthdefense trade-off. 4,46 Collectively, these considerations highlight both the environmental impact of olive pomace and its digestate and the largely unexplored potential of digestates as reservoirs of nutrients and bioactive compounds. Digestate upcycling could offer a novel route to integrate this underutilized byproduct into a circular economy pathway, converting it into valuable agricultural inputs while reducing dependence on synthetic fertilizers and pesticides. Therefore, this study aimed to develop a biorefining pipeline to recover and characterize bioactive fractions from olive pomace digestate, assess their nutrient composition, and evaluate their capacity to enhance plant growth and immunity. Specifically, this study proposes a biorefining approach to separate digestates into liquid and solid fractions, exploring their use as soil amendments, fertilizers, and immunostimulants. We characterized the biomass and nutrients and compared their ability to stimulate growth in Arabidopsis and tomato plants. The microbial communities were characterized by DNA metabarcoding. The liquid digestate was examined as a reservoir of lowcost MAMPs/DAMPs-based elicitors. A microbial-enriched protein extract (MIPE) was obtained from the liquid digestate and analyzed by proteomics to identify key immunogenic factors. MIPE was subsequently assessed for its ability to induce defense responses and protect Arabidopsis and tomatoes against pathogens. 2. MATERIALS AND METHODS 2.1. Olive Pomace Digestate Collection and Fractionation. The raw digestate (RD) was collected from the AGROLIO-AGROENERGY biogas plant (Andria, BT, Italy), the first AD plant in the world to use olive pomace as sole feedstock. RD was generated under mesophilic conditions (45 °C) for 60 days using pomace derived from a two-phase olive oil extraction system processing olive (Olea europaea L., cultivar Coratina). RD (1L) was subjected to sedimentation overnight at 4°C in a glass beaker to separate it into liquid digestate (LD) and solid digestate (SD). For testing the effects of digestatederived microbes on plant growth and for metabarcoding analysis, LD was filtered through 25 μm Miracloth to remove residual plant particles and the filtrate was centrifuged at 5000g for 30 min at 4 °C, yielding a supernatant called microbialdepleted LD (MD-LD) and a microbial-enriched pellet (M). The M pellet was washed with sterile 1×PBS and used for the following experiments. The presence of microbial cells in M was assessed by light and fluorescence microscopy using fluorescein isothiocyanate and trypan blue staining (Nikon Eclipse E200, 40×). 2.2. MIPE Extraction and Protein Quantification. The protein/peptide fraction MIPE was obtained through sedimentation, centrifugation, and sonication steps, as outlined in a Patent publication. 47 Briefly, the microbial pellet was resuspended in sterile water, and proteins were extracted via sonication (Sonics Vibra Cell VCX 130, Sonics & Materials, Inc., Newtown, CT, United States) at 70% amplitude, 5 min, 20 s ON/OFF pulses in ice bath. The lysate was centrifuged (15,000g, 30 min, 4 °C) to remove microbial and plant residues, and proteins were precipitated from supernatant with 20% trichloroacetic acid (Sigma Chemical Co, St. Louis, MO), 48 centrifuged (6000g, 10 min, 4 °C), washed with hydrochloric Environmental Science & Technology pubs.acs.org/est Article https://doi.org/10.1021/acs.est.5c03321 Environ. Sci. Technol. 2025, 59, 21700−21714 21701 acid-ethanol, lyophilized, and stored at −20 °C. MIPE yield was expressed as milligrams of dry weight per liter of LD. Lyophilized MIPE was dissolved in sterile distilled water at 5 mg/mL and used for protein analysis and biological assays. Protein concentration was determined by the Bradford method. 49 The presence of residual nucleic acids in the extract was ruled out by spectrophotometry and agarose gel electrophoresis (1.5%, 60 V, 25 min). Images were captured using a Gel Doc XR + System (BioRad, Hercules, CA, USA) (Figure S5). 2.3. Plant Growth Conditions and Dose−Effect Treatments. Arabidopsis thaliana (ecotype Columbia, Col-0) seeds were surface-sterilized with 20% NaClO (5 min) and washed four times. After 2 days of stratification at 4 °C, seeds were germinated in multiwell plates (10 seeds/well) with 1 mL of liquid MS/2 medium (2.2 g/L MS, 0.5% sucrose, pH 5.7). 50 To evaluate the dose effect of MIPE on Arabidopsis seedling growth, 7 day-old seedlings were treated with sterile water or MIPE (1, 10, or 100 μg dry mass/mL of distilled water) in MS/2 medium, where the extracts were sterilized using a MCE membrane filter (0.22 μm pore size). For measurements of shoot growth in adult plants, Arabidopsis seeds were initially germinated on solid MS/ 2 medium (2.2 g/L MS, 1% sucrose, 0.8% plant agar, pH 5.7) at 22 °C under a 16 h light/8 h dark cycle. After 7 days, seedlings were transferred to sterile soil and grown at 22 °C with a 12 h light/12 h dark cycle (photosynthetic active radiation 100 μmol m−2s−1). A commercial soil was sterilized by autoclaving and Figure 1. Biorefining of olive pomace digestate improves its biostimulant performance on Arabidopsis and tomato. (A) Diagram showing the process of preparation of the different fractions from the olive pomace raw digestate (RD). After sedimentation, the RD was fractionated into in solid digestate (SD) and liquid digestate (LD). After centrifugation, LD was fractionated in a microbe-depleted liquid digestate (MD-LD) and a microbial-enriched pellet (M). (B) Effects of soil amended with RD (45 g/kg), SD (15 g/kg), or LD (30 g/kg) fractions on Arabidopsis shoot growth, measured as rosette fresh weight. (C,D) Effects of soil amended with RD (20 kg/m2), SD (6 kg/m2), or LD (14 kg/m2) on height and fruit production of S. lycopersicum, grown on the field. (E) Effects of soil amended with LD, MD-LD (both 30 g/kg), or M (1.5 g/kg) on Arabidopsis shoot growth, measured as relative rosette fresh weight compared to mock. The values are expressed as percentages relative to plant grown on water-soaked soil used as mock. Icons next to the graphs indicate the plant speciesused for the analysis. Data shown represent the mean±SD (n= 10). The experiments were repeated three times with similar results. The different letters indicate significantly different data sets according to ANOVA followed by Tukey’s test (p≤0.05). Environmental Science & Technology pubs.acs.org/est Article https://doi.org/10.1021/acs.est.5c03321 Environ. Sci. Technol. 2025, 59, 21700−21714 21702 amended with 45, 90, or 180 g/kg of RD (Figure 1A). Additional experiments tested 90 g/kg of RD vs 30 g/kg of SD or 60 g/kg of LD. Further treatments compared 30 g/kg of LD, 30 g/kg of MD-LD, and 1.5 g/kg of M. In all the experiments, rosette fresh and dry weights were measured after 2 weeks. For dry weight determination, the aerial vegetative portion of plants were weighed after treatment at 80 °C for 6 h. Tomato seeds (Solanum lycopersicum, cultivar Minibel) were purchased from Mascarell Semillas S.L., (Benissoda, Spain). Minibel is a determinate-growing cultivar producing small-sized fruits and well suited for laboratory-scale cultivation. 51 Seeds were germinated on wet paper, transferred to soil, and grown ina greenhouse at 23 °C with a 16 h light/8 h dark cycle (photosynthetic active radiation 75 μmol m−2s−1, 35−40% humidity). Five-week-old tomato plants were transplanted in soil with 20, 40, or 80 kg/m2of RD. Tomato plants were also grown on the field with 20 kg/m2of RD, 6 kg/m2of SD, or 14 kg/m2of LD. Height and fruit number were measured four months after transplantation. For both Arabidopsis and tomato experiments water-soaked soil was used as mock. Data shown represent the mean ±SD (n= 10). The experiments were repeated three times with similar results. 2.4. DNA Extraction and 16S and ITS rRNA Sequencing. The genomic DNA of microbial communities in M was extracted by using the DNeasy PowerSoil kit (QIAGEN, Germany). M was resuspended in UREA buffer, incubated at 60 °C, centrifuged, and processed according to the kit instructions. DNA concentration was assessed with NanoDrop1000 (ThermoFisher Scientific, MA, USA). Metagenomic DNA served as a template for 16S rRNA amplification using primers 8F/1492R. 52 Polymerase chain reaction (PCR) was performed with Platinum SuperFi II PCR Master Mix (Invitrogen) under specific cycling conditions, and amplicons were validated on a 1% agarose gel. 53 A tagging step with modified primers followed the Oxford Nanopore protocol (SQK-LSK114), with additional amplification, purification with Ampure XP, and barcoding. Sequencing libraries were prepared using a NEBNext Ultra DNA library preparation kit and loaded onto an R10.4.1 flow cell. Reads were processed with Guppy (v6.4.6), filtered (400−1800 bp), and aligned to the SILVA 16S database using minimap2. Taxonomic abundances were normalized using the geometric mean of pairwise ratios (GMPR) implemented in the microeco R package. Abundance data from 10 biological replicates were averaged (n= 10) and visualized with ggplot2. Community composition was assessed at the Genus, Family, Order, and Phylum levels. Sequencing data are available under BioProject ID PRJNA1211516 (https:// www.ncbi.nlm.nih.gov/bioproject/?term=PRJNA1211516). 2.5. Protein Identification by LC−MS/MS Analysis. MIPE (15 μg of proteins) were prepared in 20 μL with 4× Laemmli buffer, denatured at 100 °C, and loaded onto a 12% acrylamide gel for 1D-SDS-PAGE. Proteins were visualized with Coomassie Brilliant Blue, each lane excised into seven gel slices, and subjected to in-gel trypsin digestion. 54 Peptides were analyzed using ultrahigh-performance liquid chromatography and high-resolution mass spectrometry. 55 Peptides were separated using a 75 μm C18 column (ES800-PepMap RSLC C18, 150 mm ×75 μm) and nano-LC (UltiMate 3000 RSLC, Thermo Fisher Scientific) with a 100 min gradient (4% to 90% eluent B, 80% acetonitrile, 0.1% formic acid, 0.3 μL/min flow). The analysis used a Q Exactive Plus Hybrid QuadrupoleOrbitrap mass spectrometer (Thermo Fisher Scientific) with data-dependent acquisition, selecting the 15 most intense ions. Full-scan spectra (m/z350.0 to 1700.0, 70,000 ppm resolution) and higher-energy collisional dissociation (HCD) fragmentation (17,500 ppm) were used. Protein identification was performed with MaxQuant software v. 2.2.0.0 as previously described. 56,57 A custom database (157,496 protein counts) was created for protein identification, combining UniProtKB reference proteomes for bacteria, fungi, and olives (released 09/05/2024), as detailed in Supporting Information Datasheet S1. The mass spectrometry data are available on the ProteomeXchange Consortium via the jPOST repository [http://repository.jpostdb.org], with data set identifier JPST003545/PXD059822. 58 The proteomic analysis was performed on three biological replicates. 2.6. Digestate Characterization. Dry residue (total solids) was determined gravimetrically by drying fresh samples at 105 °C for 12 h to constant weight and expressed as percentage of the initial wet mass. Organic dry matter was measured by ignition of dried samples in a muffle furnace at 550 °C until complete incineration; ash content was calculated gravimetrically, and organic dry matter was obtained by difference (100− ash %). The pH of the LD was determined by using a CRISON GLP21 pH-meter (Hach Lange Spain, S.L.U., Barcelona, Spain). The content of nitrogen, ammonium, and ammonium nitrogen was evaluated according to the CNR IRSA methods. 59 The contents of potassium, phosphorus, sulfur, magnesium, iron, copper, boron, manganese, nickel, chromium hexavalent, lead, cadmium, mercury, and zinc were measured by inductively coupled plasma−optical emission spectroscopy (ICP-OES) according to the UNI EN 16174:2012 + UNI EN 16170:2016 60,61 methods. The detection of Salmonella spp. was performed according to APAT 3/Man 20. 62 The monosaccharide composition and the oligosaccharide characterization are described in Text S1. 2.7. Hydrogen Peroxide Quantification. Four millimeter diameter leaf discs from four-week-old Arabidopsis plants and five-week-old S. lycopersicum were used to determine hydrogen peroxide (H2O2) production. Specifically, leaf discs were placed per well in a white 96-well plate and were incubated overnight with 150 μL of distilled water at room temperature. The following day, distilled water was replaced by 100 μL of 10 nM L-012, a chemical analogue of luminol (FUJIFILM Wako Pure Chemical Corporation), and 10 μg/mL horseradish peroxidase (HRP; Sigma-Aldrich), and incubated for 2 h. Subsequently, leaf discs were treated with 1 μg/mL MIPE or 1 μM flg22 (synthetic 22−amino acid flagellin-derived peptide, PhytoTech laboratories). Sterile distilled water was used as a mock. H2O2 production was measured for 200 min by determining the luminescence produced by the luminol-peroxidase reaction in a Varioskan Lux luminescence reader (Thermo Scientific, Waltham, MA, USA). 63 For each experiment, six leaf discs were used, each collected from a different adult plant (n= 6). The experiment was performed three times with similar results. 2.8. Immunoblot Analysis for Mitogen-Activated Protein Kinases Activation. Ten-day-old Arabidopsis seedlings were treated with MIPE (2 μg/mL), flg22 (1 μM), or sterile distilled water was used as a mock for 5, 10, and 20 min and then immediately frozen in liquid nitrogen. Protein extraction and detection of activated mitogen-activated protein kinases (MAPKs) were performed as previously described. 63 Briefly, frozen seedlings were homogenized, and proteins were extracted using a Tris-HCl-based buffer with protease and phosphatase inhibitors. Total proteins were quantified by the Bradford assay, separated by SDS-PAGE, and transferred onto Environmental Science & Technology pubs.acs.org/est Article https://doi.org/10.1021/acs.est.5c03321 Environ. Sci. Technol. 2025, 59, 21700−21714 21703 nitrocellulose membranes. Membranes were blocked, incubated with anti-Phospho-p44/42 MAPK primary antibody (Cell Signaling Technology Danvers, MA, USA) overnight at 4 °C, washed, incubated with HRP-conjugated secondary antibody (Thermo Fisher Scientific, Waltham, MS, USA), and developed using ECL. Equal loading was confirmed by Ponceau S staining. The experiment was performed three times with similar results. 2.9. Arabidopsis and Tomato Infection Assay. Botrytis cinerea (strain SF1 64 ) was grown in the dark before conidial collection at 23 °C and 70% relative humidity for 20 days on malt extract agar (20 g/L) with mycological peptone (10 g/L) and micro agar (12 g/L). Conidia of B. cinerea were collected by washing the mycelium from agar plates with sterile water, filtered, and counted using a Thoma counting chamber. Fourweek-old plants were pretreated with 2 mL of flg22 (1 μM) or MIPE (2 μg/mL) and sterile distilled water. After 24 h, each leaf was infected with six 5 μL droplets containing 1 ×106conidia/ mL in Potato Dextrose Broth (PDB). Mock was leaves pretreated with water and inoculated with fungus in PDB. Plants were incubated at 24 °C with a 12 h/12 h light/dark cycle (PAR level of 100 μmol m−2s−1). Lesion size at 48 h post infection (hpi) was measured using ImageJ software to assess fungal susceptibility. Pseudomonas syringae pv tomato DC3000 was cultured from a frozen glycerol stock on King Agar B (KB) supplemented with 20 mg/mL proteose peptone, 1.5 mg/mL K2HPO4, 1.5 mL/mL glycerol, 1.5 mg/mL agarose, 25 μg/mL rifampicin, and 5 mM MgSO4. The culture was incubated at 28 °C in the dark for 3 days before inoculum preparation. Fourweek-old Arabidopsis and five-week-old tomato plants were sprayed with 2 mL of 1 μM flg22 or 2 μg/mL MIPE using adjuvants (0.05% Tween 24 MBAL for Arabidopsis; 2.5% Tween 24 MBAL + 2.5% UEP-100 for tomato; Croda, Snaith, UK). Corresponding adjuvant solutions were used as mock. Infection occurred 24 h after pretreatment, as previously described. After 24 h, Arabidopsis plants were infected with P. syringae pv tomato DC3000, 4 spraying leaves with a bacterial concentration of OD600 = 0.1 added with 0.001% of Silwet L-77. Tomato plants were infected, with leaves sprayed with a bacterial concentration of OD600 = 0.5 added with 0.002% of Silwet L-77. Leaf discs from both plants were collected at 0 dpi (corresponding to 3 hpi) and 3 dpi to quantify bacterial colonies as cfu per leaf area. Data shown represent mean ±SE (n= 6). The experiments were performed three times with similar results. 2.10. Analysis of Defense Gene Expression. To evaluate the defense gene expression after MIPE treatment, ten-day-old Arabidopsis seedlings were treated for 1 h with flg22 (1 μM), MIPE (2 μg/mL), or water (mock). To assess the MIPEinduced priming effect, B. cinerea-infected Arabidopsis leaves were collected at 8 hpi. Tissues were homogenized in liquid nitrogen and total RNA was extracted using NucleoZol (Macherey-Nagel, Duren, Germany), treated with DNase (Promega, Southampton, UK), and reverse-transcribed into cDNA. Quantitative reverse transcription PCR (RT-qPCR) was performed with a CFX96 Real-Time System (Bio-Rad, Hercules, CA, USA) using GoTaq qPCR Master Mix (Promega, Southampton, UK). Gene expression was analyzed using the Pfaffl method with Ubiquitin 5 (UBQ5) and Beta-tubulin 4 (TUB4) as reference genes. 3 Primer sequences were generated with Primer3 software (https://primer3.ut.ee/) (Supporting Information Table S1). Data shown represent mean ±SE (n= 3). The experiment was repeated three times with similar results. 2.11. Data Analysis. Data were presented as mean ± standard deviation (SD) or standard error (SE) as indicated in the figure legends. The significant differences were evaluated by Student’s t-test or ANOVA analysis followed by Tukey’s test (p ≤0.05), as indicated in the figure legends. Statistical analyses were performed using GraphPad Prism 8.0.1 software (GraphPad Software, San Diego, CA, USA). 3. RESULTS 3.1. Two-Phase Olive Pomace Digestate Is Enriched in Mineral Nutrients and Characterized by Low Heavy Metal Content. To evaluate the plant fertilization potential of the two-phase olive pomace digestate, biochemical characterization was performed on raw digestate (RD) from a biodigester fed with the two-phase olive pomace. The digestate had a pH of 8.0 ±0.14, which is suitable for soil compatibility and plant growth. The total solids indicating inorganic and nonvolatile content was 6.69 ±0.04% (n= 3), confirming that it was predominantly aqueous. Organic dry matter was 78.09 ±0.02% (n= 3), indicating that the majority of the dry mass is of organic origin and therefore potentially biodegradable or biologically active. The remaining fraction consisted of inorganic and mineral residues. The C/N ratio of 8.0 ±0.5 (n= 3) reflects a relatively high nitrogen availability compared to carbon, which is consistent with the nitrogen-rich nature of the matrix. Among the macronutrients, potassium (K; 6.8%) was the most abundant, followed by nitrogen (N; 4.9%), ammonium (NH4+; 2.0%), ammonium nitrogen (NH4−N; 1.5%). Minor amounts of phosphorus (P), sulfur (S), magnesium (Mg) (<1% each) were detected (Figure S1A). Micronutrients such as copper (Cu), boron (B), manganese (Mn), zinc (Zn), nickel (Ni) were detected at <0.01% (Figure S1B). Next, we evaluated the potential presence of heavy metals in pomace digestate, which can be present in olives and, consequently, in two-phase olive pomace, as a result of uptake by trees grown in contaminated soil and from foliar applications of fertilizers and pesticides. 65 Heavy metals, including Cr(VI) (hexavalent chromium), Hg (mercury), Pb (lead), and Cd (cadmium), were detected only at trace levels (<0.00015%) (Figure S1C). Salmonella spp. was not detected, confirming the digestate’s safety as an agricultural amendment. To determine the most plant-beneficial fraction, RD was separated into liquid (LD) and solid (SD) digestates (Figure 1A). The LD comprised 68.5 ± 4.4% of RD (fresh weight), while SD accountedfor 32.15 ±4.3% (w/w). The LD showed similar chemical parameters, nutrient levels, and heavy metal content compared to RD (Figure S1D− F). Since residual CW-derived polysaccharides in waste biomasses can influence their effectiveness as soil conditioners, they were extracted as alcohol-insoluble solids (AIS) and the monosaccharide composition characterized by high-performance anion-exchange chromatography with pulsed amperometry detection (HPAEC-PAD). 66,67 LD contained a low monosaccharide content (0.018 ±0.002%, w/v of LD) indicating a low level of polysaccharides. These were primarily composed of galacturonic acid (∼30%), rhamnose (∼21%), and glucose (∼18%), with minor amounts of arabinose, galactose, mannose, xylose, fucose, and glucuronic acids (Figure S2A). Instead, in SD, the monosaccharide content accounted for 80.2 ±9.7% w/ w SD dry weight, indicating a high content of undigested polysaccharides. The monosaccharide composition of the SDAIS revealed a high xylose content (∼80%), with lower levels (<10% each) of glucose, galacturonic acid, arabinose, mannose, galactose, and rhamnose, as well as trace amounts of glucuronic acid and fucose (Figure S2B). The high xylose content suggests Environmental Science & Technology pubs.acs.org/est Article https://doi.org/10.1021/acs.est.5c03321 Environ. Sci. Technol. 2025, 59, 21700−21714 21704 that SD is rich in hemicelluloses (e.g., xylans) and contains minimal amounts of pectins. 3.2. Soil Amendment with Olive Pomace Digestate Positively Influences Plant Growth. The effects of RD as a soil amendment and fertilizer were evaluated in a dose− response experiment with Arabidopsis shoot grown in commercial soil, sterilized, and amended with RD at 45, 90, or 180 g/kg. Water-soaked soil was used as a mock. After 2 weeks, rosette fresh and dry weights were measured and compared to the mock (Figure S3A). The 45 g/kg dose increased rosette fresh weight by about 25%, while higher doses (90 and 180 g/ kg) resulted in only a 10% increase, with a maximum enhancement of 36%. Similar results were observed for the rosette dry weight (Figure S3B). A dose−response pattern was also observed with tomato, grown in soil amended with RD values of 20, 40, or 80 kg/m2. After four months, tomato height was measured, and the 20 kg/m2dose led to a 49% increase in height, with higher doses resulting in smaller increases of around 26%, yielding a total growth increase of 75% (Figure S3C). These findings suggest that higher concentrations reduce the growth response, with 45 and 20 kg/m2being optimal doses for further study in Arabidopsis and tomato, respectively. To compare the biostimulant potential of RD with that of SD and LD, Arabidopsis plants were grown in soil amended with RD Figure 2. Abundance analysis of 16S rDNA of bacterial cells and rDNA internal transcribed spacer (ITS) region of fungal cells from the digestate. (A) Donut chart describing the proportion of reads assigned to each genus of bacterial community. (B) Bar plot showing the composition of the ≪uncultured≫fraction of the bacterial community. Each percentage is calculated considering a restricted data set, represented by the hits classified as ≪uncultured≫at the genus level. (C) Donut chart describing the proportion of reads assigned to each genus of fungal community, represented with different colors. Percentage of abundance is indicated by the integer within each slice. Taxonomic abundances were averaged across replicates (n= 10). Environmental Science & Technology pubs.acs.org/est Article https://doi.org/10.1021/acs.est.5c03321 Environ. Sci. Technol. 2025, 59, 21700−21714 21705 (45 g/kg), SD (15 g/kg), or LD (30 g/kg). After 2 weeks, rosette fresh and dry weights were measured and compared to a control. All treatments significantly stimulated Arabidopsis shoot growth (Figure 1B). RD caused a 43% increase in growth compared to the control, while SD induced a 57% growth increase, and LD exhibited only a 12% increase. Similar results were observed for the rosette dry weight (Figure S3D). The study extended to tomato plants grown in soil amended with an RD (20 kg/m2), LD (14 kg/m2), or SD (6 kg/m2). After four months, growth parameters were quantified and compared to untreated controls (Figure 1D,E). RD amendment significantly increased tomato height and fruit number by 34% and 232%, respectively. Both digestate fractions enhanced tomato height and productivity compared with the control. As seen with Arabidopsis, SD produced a more substantial increase in height (63%) and fruit number (315%) than did LD (19% and 150%, respectively). These results indicate that two-phase olive pomace RD can serve as an effective biostimulant for both Arabidopsis and tomato, with SD being more effective than LD in stimulating plant growth and fruit production. 3.3. Microbial Fraction Reduces Digestate Biostimulant Efficiency. We hypothesized that specific components in the LD fraction might reduce its biostimulatory potential. The presence of galacturonic acid-based carbohydrates in LD (Figure S2A) suggests a possible trade-off between growth and defense responses induced by OG. To explore this hypothesis, pectic fragments were precipitated from LD using ethanol fractionation, and the oligosaccharide profile was analyzed using HPAEC-PAD 4,68 (Figure S4). However, no OG or other carbohydrate-based elicitor peaks were detected in LD. The LD fraction contained proteins (1.8 mg/mL). Given that many protein-derived MAMPs are shared across various microbial species and considering that digestates can be enriched with diverse bacterial and fungal populations as well as residual plant biomass, we hypothesized that digestates may function as a continuously proliferating, low-cost reservoir for MAMPs/DAMPs-based phytovaccines. To test this assumption, LD was centrifuged, generating two fractions: a microbialdepleted LD (MD-LD) and a microbial-enriched pellet (M) (Figure 1A). Arabidopsis plants were then cultivated in soil amended with varying concentrations of LD, MD-LD (both at 30 g/kg), or M (1.5 g/kg), with M applied at the relative proportion found in LD. After 2 weeks, the rosette fresh and dry weights were measured and compared to the control. Notably, an increase in Arabidopsis shoot growth was observed in plants grown in both LDand MD-LD-amended soil. The increase in rosette fresh weight was significantly higher in the MD-LD treatment compared to the LD treatment, with a difference of approximately 16.7% (Figure 1C). Conversely, the M treatment led to a decrease in plant rosette growth by about 19.2%. Similar results were observed for rosette dry weight (Figure S3E). Plants did not display visible disease symptoms, such as chlorosis, necrosis, or wilting, after M treatment. These data indicate that the microbes present in LD negatively affect its fertilizing potential for both Arabidopsis and tomato vegetative growth. 3.4. Several Bacterial and Fungal Species Were Identified in the Liquid Fraction of Olive Mill Digestate. Detailed information on the microbial component isolated in the M fraction was obtained through taxonomic characterization of the microbial community. 16S and ITS rRNA metabarcoding was performed to identify bacteria and fungi populations, respectively. The analysis revealed a complex bacterial genuslevel community structure, with nearly 50% of sequenced reads assigned to uncultured organisms (Figure 2A). Predominant genera (relative abundance >3%) included Luteimonas, Planomicrobium,Caldicoprobacter,Pseudomonas, and HNHF0106. Sequence data indicated also the presence of anaerobic bacteria such as Tissierella and Sedimentibacter (Peptostreptococcaceae) and genera within the Ruminococcaceae family (UCG010 and Ruminiclostridium). Notably, most uncultured organisms were associated with DTU014 (nearly 40%), followed by Chloroflexi (22%), Firmicutes (14%), Collierbacteria (6%), SAR324, and Bacillaceae (3%) (Figure 2B). Taxonomic assignment of the fungal community was challenging, likely due to the low DNA concentration in the samples (Figure 2C). This limitation may explain why 71% of the sequences were identified as Fungi gen. incertae sedis, indicating undefined broader taxonomic relationships. Despite the low abundance of ITS sequences, notable findings included 7% of the sequences assigned to the genus Pichia and Basidiomycota and >3% assigned to Paraglomerales,Brettanomyces, and Glomus. No Salmonella spp. were detected, indicating the digestate’s potential safety as an agricultural amendment. However, the Pseudomonas genera may contain species with pathogenic potential for plants and humans. The genus-level identification performed does not allow definitive exclusion of these pathogenic species, and future studies using species-level metagenomics or culture-based assays will be necessary to fully assess potential risks. 3.5. Immunogenic Proteins Are Present in Olive Pomace Digestate Fractions. LD was investigated as a potential cost-effective source of MAMPs/DAMPs. A MIPE was obtained from the M fraction of the LD (20 mg powder/L). Proteomic analysis of MIPE by mass spectrometry identified 1169 proteins (Supporting Information Datasheet S2), filtered to 67 consistently detected across biological replicates (Supporting Information Datasheet S3). The identified proteins originated from bacteria (50.7%), fungi (40.3%), and olive plants (8.9%) (Figure S6A). Functional gene ontology (GO) analysis grouped proteins into seven categories, with bacterial proteins linked to host interaction (35.7%), cellular organization (29.2%), metabolism (16.7%), localization (8.3%), and stimulus response (4.2%). Fungal proteins were mainly metabolic (64.3%) or involved in host interactions (21.4%) (Figure S6B). Olive-derived proteins were associated with defense, metabolism, and the cell cycle (33.3% each). The presence of proteins linked to host interaction and defense suggests a role in plant−pathogen interactions. MIPE contained proteins of various sizes, predominantly <300−400 amino acids, indicative of potential elicitor activity, alongside high-molecular-weight proteins (Figure S6C). Intriguingly, microbial functional proteins involved in plant immune responses were detected (Table 1). In particular, precursors of bacterial MAMPs were observed, including elongation factor Tu, releasing a fragment of 18 amino acids (elf18) that can be recognized by Elongation Factor Tu Receptor (EFR) in Arabidopsis 69 and flagellin, whose immunogenic peptides can be perceived via Arabidopsis Flagellin Sensitive 2 (FLS2). 70 Enzymatic MAMPs derived from bacterial and fungal communities were also detected, including endo-1,4beta-xylanases A, pectate lyases, and histidine kinases whose elicitor activities can be independent of their enzyme activities. 71−73 A homologue to golven 1−2A, a phytocytokine known as inducible peptidic DAMP, was also identified. 74,75 The proteomic analysis also showed plant and microbial enzymes that could produce signaling molecules from hemicellulose Environmental Science & Technology pubs.acs.org/est Article https://doi.org/10.1021/acs.est.5c03321 Environ. Sci. Technol. 2025, 59, 21700−21714 21706 components of the plant CW, such as xylanases, and from pectins such as the pectinases rhamnosidases, polygalacturonases, and pectin methylesterases. 32,76−79 3.6. MIPE Induced PTI Hallmarks in Arabidopsis and Tomato Plants. We examined whether MIPE can act as an elicitor of plant immune responses. A dose−response analysis was performed on Arabidopsis seedling growth with MIPE at 1, 10, or 100 μg dry mass/mL of distilled water (Figure S7A). Seedlings treated with MIPE at 1 and 10 μg/mL showed no significant growth difference from mock-treated seedlings. However, MIPE at 100 μg/mL caused a 35% inhibition in seedling growth. Next, we tested MIPE’s ability to activate PTI hallmarks, such as hydrogen peroxide (H2O2) production, MAPKs phosphorylation, and upregulation of defense genes. Arabidopsis leaf-discs were treated with 1 μg/mL MIPE or with 1 μM of the know elicitor flg22, and H2O2production was subsequently quantified. Distilled sterile water was used as mock. The MIPE treatment induced a transient H2O2burst (Figure S7B), which was slower than the response to 1 μM flg22 (Figure 3A). Flg22 peaked at 10 min, while MIPE peaked at 20 min. Immunoblot analysis of MAPKs phosphorylation revealed that MIPE induced phosphorylation of MAPK6 and MAPK3 after 5 min, with MAPK4/11 showing slight phosphorylation after 10 min, which increased at 20 min. The phosphorylation pattern of MIPE was similar to flg22, though flg22 induced higher MAPKs phosphorylation earlier (10 min). Next, we explored the expression of three PTI-reporter genes, CYP81F2, FRK1, and WRKY53, after 1 h of Arabidopsis seedling treatment with MIPE (1 μg/mL), flg22 (1 μM) or mock (Figure 3C). CYP81F2 (CYTOCHROME P450,FAMILY 81) encodes a cytochrome P450 monooxygenase involved in the biosynthesis of indole glucosinolates; 80 FRK1 (FLG22-INDUCED RECEPTOR-LIKE KINASE 1) encodes a leucine-rich repeat receptor kinase that functions in defense signaling pathways; 81 WRKY53 (WRKY DNA-BINDING PROTEIN 33) encodes a transcription factor contributing to basal resistance against P. syringae. 82 Both MIPE and flg22 treatments strongly stimulated the expression of all plant immunity genes. Flg22 was more effective in stimulating CYP81F2 expression, while MIPE induced higher levels of FRK1 expression. Both MIPE and flg22 treatments showed similar levels of WRKY53 expression. Overall, our findings suggest that the exogenous application of a specific MIPE concentration to Arabidopsis can induce multiple early immune responses. To investigate the effect of MIPE in tomato, leaf discs were treated with 1 μg/mL MIPE, 1 μM flg22, or distilled sterile water as mock, and H2O2production was subsequently quantified (Figure 3D). MIPE induced a transient increase in H2O2levels (Figure 3E), highlighting its capacity to trigger immune responses in tomato, as also observed for flg22. MIPE reached its maximum at 30 min post-treatment, whereas flg22 induced an earlier peak at 10 min, suggesting that distinct kinetics of signaling activation may be involved. 3.7. MIPE Primes Immune Responses and Reduced Arabidopsis and Tomato Disease Symptoms Caused by Bacterial and Fungal Pathogens. MIPE’s ability to induce priming responses and protect against B. cinerea in Arabidopsis was assessed. Four-week-old plants were pretreated with MIPE (1 μg/mL), flg22 (1 μM), or mock, and 24 h later, leaves were inoculated with B. cinerea spores (Figure 4A). To test MIPE’s priming effect, the expression of PTI genes CYP81F2 and PAD3 (PHYTOALEXIN DEFICIENT 3) was evaluated at 8 h post infection (hpi) (Figure 4B). PAD3 encodes a key biosynthetic enzyme involved in the biosynthesis of the antimicrobial compound camalexin and is essential for elicitor-induced resistance to B. cinerea. 83,84 Both MIPEand flg22-pretreated plants showed significantly higher expression of CYP81F2 and PAD3 compared to mock, with similar induction levels between the two elicitors. At 48 hpi, MIPE and flg22 pretreatments improved Arabidopsis resistance to B. cinerea, with MIPE-treated plants showing a greater reduction in lesion area (69%) compared to that of flg22 (54%) (Figure 4C). The MIPE induction of WRKY53 expression previously observed in Arabidopsis seedlings suggested a potential priming and protective effect of MIPE against P. syringae. 85 MIPE’s priming effect was also tested in Arabidopsis against P. syringae. Adult Arabidopsis plants were pretreated with MIPE (1 μg/mL), flg22 (1 μM), or mock, and after 24 h, they were inoculated with the bacterium (Figure S8A). The primed state was assessed by monitoring the expression of PTI genes CYP81F2 and WRKY53 at 8 hpi (Figure S8B). Both MIPE and flg22 treatments induced a significant upregulation of these genes compared to the mock pretreatment. Typical Pseudomonas symptoms, including small necrotic spots with yellow halos, were visible in mock-pretreated plants but absent in pretreated plants (Figure S7C). Bacterial quantification at 0 and 3 days post infection (dpi) (Figure S8C) showed no significant difference at 0 dpi, but at 3 dpi, bacterial growth increased by 39% in mock-pretreated plants, whereas MIPEand flg22-pretreated plants showed no bacterial growth increase. When pretreated tomato plants were inoculated with P. syringae (Figure 4D), bacterial colonies increased by 21% in mock-treated plants at 3 dpi, while MIPEand flg22-pretreated plants showed no growth increase (Figure 4E). These findings demonstrate that MIPE primes immune responses and protects Arabidopsis and tomato against B. cinerea and P. syringae. 4. DISCUSSION This study demonstrates that two-phase olive pomace digestate is a promising biostimulant for plant growth and productivity, extending the findings and discussion presented in a preprint. 86 The digestate exhibited a pH of 8.0 ±0.14, which is slightly alkaline and therefore compatible with various soil types, potentially helping to counteract soil acidification, a relevant factor for sustaining plant growth. 87 The high organic dry matter content emphasizes its richness in organic carbon, contributing to improved soil structure, water retention, and microbial activity, thus positioning it as an excellent soil amendment. 88,89 The relatively low C/N ratio ensures balanced carbon and nitrogen availability, promoting nutrient cycling and making nitrogen readily accessible to plants. 19 The high potassium and Table 1. List of Proteins and Peptides Identified in MIPE, Known to Act as MAMPs, Phytocytokines, or Enzymes That Release DAMPS in Plant Immunity MAMPs origin references elongation factor Tu bacteria 35,69 flagellin bacteria 70,105,106 endo-1,4-beta-xylanase A bacteria/fungi 71 pectate lyase bacteria/fungi 73 histidine kinase bacteria/fungi 72 Phytocytokines homologue to GOLVEN 1−2 olive 74,75 Enzymes Potentially Releasing DAMPs α-L-rhamnosidase fungi 76 endo-1,4-β-xylanase A bacteria/fungi 78 pectate lyase bacteria/fungi 100 Environmental Science & Technology pubs.acs.org/est Article https://doi.org/10.1021/acs.est.5c03321 Environ. Sci. Technol. 2025, 59, 21700−21714 21707 ammonium content enhances its value as a nutrient-rich fertilizer. 22 The absence of heavy metals or harmful bacteria, like Salmonella spp., ensures safer agricultural use compared to digestates from livestock waste, crop residues, or urban waste, which may contain pollutants and pathogens. 90 This digestate can be applied directly to the soil as a slurry, either by drenching or irrigation, and may optionally be lightly incorporated into the soil via shallow tillage, depending on agronomic practices. We highlight the need to refine digestates to enhance its agricultural performance. Separating raw digestate (RD) into solid (SD) and liquid (LD) fractions revealed that SD had a stronger impact on plant growth. SD’s advantage may stem from its higher hemicellulose content, which is more likely to stimulate microbial activity, promote exoenzyme release, and enhance organic matter mineralization. 91 LD’s lower efficacy was not linked to its mineral content or organic compounds. Our findings indicate that plants grown in soil treated with microbialdepleted liquid digestate (MD-LD) exhibited improved shoot growth compared with those treated with untreated LD, whereas the isolated microbial-enriched pellet (M) caused shoot growth inhibition. Some plant growth-promoting microbes enhance shoot growth over root development to improve plant health Figure 3. MIPE activated immunity hallmarks in Arabidopsis tissues. (A) Measurement of H2O2production by luminol reaction after treatment with flg22 (1 μM), or MIPE (1 μg/mL) in four-week-old Arabidopsis leaf-discs. The values are reported as ratio of relative luminescence units (RLU) with respect to total RLU at 0, 10, 20, 30, 40, and 50 min. (B) MAPK activationin Arabidopsis seedlings in response to sterile distilled water (mock), flg22 (1 μM), or MIPE (1 μg/mL) treatments. MAPKs phosphorylation was determined by Western blot using the phospho-p44/42 MAPKs antibody at different time points (5, 10, and 20 min). Equal protein loading in the gel was confirmed by Ponceau staining. MW = Molecular weight marker. (C) Expression of CYP81F2, FRK1, and WRKY53 after mock, flg22 (1 μM), or MIPE (1 μg/mL) treatments. The expression of defense genes was analyzed by quantitative RT-PCR at 1 h after treatments on 10 days-old Arabidopsis seedlings. The expression levels were normalized to UBQ5 and TUB4 expression levels. Data represent the mean ±SE (n= 3). (D) MIPE activated H2O2production in tomato. H2O2production measured by luminol reaction for 200 min after treatment with distilled water (mock) or MIPE (1 μg/mL) in five-week-old tomato leaf-discs. Data represent mean ±SE (n= 6). (E) Ratio of RLU respect to the total RLU after flg22 (1 μM), or MIPE (1 μg/mL) at 0, 10, 20, 30, 40, and 50 min. Data in A and E are presented as box plots (n= 6), with the center line showing the median, the box limits showing the 25th and 75th percentiles, and the whiskers showing the full range of data (minimum to maximum values). Icons next to the graphs indicate the plant species used for the analysis. All the experiments were performed three times with similar results. Different letters indicate significant differences according to ANOVA followed by Tukey’s test (p≤0.05). Environmental Science & Technology pubs.acs.org/est Article https://doi.org/10.1021/acs.est.5c03321 Environ. Sci. Technol. 2025, 59, 21700−21714 21708