Introduction

Pasture dieback (PDB) is a syndrome of grass deterioration spreading across Australia affecting several grass species including native grass. Despite its significant impact on the Australian cattle industry and the surrounding ecosystems, the causes of the condition have remained unknown for decades. The decline in pasture production due to PDB negatively impacts the Australian economy by reducing the quality and quantity of grazing pasture available for livestock. Queensland is the largest contributor to grazing land in Australia, with approximately 129 million hectares dedicated to grazing (Australian Bureau of Statistics 2018), as such pasture dieback is a significant threat to the productivity and sustainability of grazing practices.

A project report by the Department of Agriculture and Fisheries (DAF) (Buck 2017) estimated that approximately 35,000 hectares across 120 properties in Queensland were affected by PDB, with patches of pasture impacted across the area. It is important to note that the actual extent of damage was anticipated to be higher than what was reported. Alarmingly, the situation has further deteriorated over the past five years. The area affected by PDB has expanded significantly to approximately 4.4 million hectares spanning from Townsville in Queensland to Lismore in New South Wales (Hall et al. 2020).

The initial symptoms manifest on the older leaves, with the leaf tips turning red or yellow, and progressively affecting younger leaves (Makiela and Harrower 2008). The affected root system develops poorly and appears shorter, thinner, softer, and darker than those in unaffected plants, making them more susceptible to breakage. Ultimately, the plants affected by dieback show reduced biomass, produce fewer tillers, and possess shorter leaves compared to unaffected plants. The symptoms of PDB are similar across different grass species. Although PDB commonly appears in patches, it eventually progresses to affect the entire paddock (Future Beef 2021). The rate at which PDB is expanding suggests that it could cause significant damage to Australia’s pasture-based cattle production.

Despite undertaking several remediation efforts to restore PDB-affected soil via re-sowing, burning, fertilising, and slashing, none of these methods has proven successful. In most cases, after a minimal initial improvement, the situation reverts to dieback within a few weeks. The current understanding is that PDB is a complex problem that affects the grass ecosystem, causing disturbance to the harmony between soil, bacteria, fungi, helminths, insects, plants and other communities (Makiela and Harrower 2008; Ren et al. 2023).

Plant growth and development largely depend on their associated microbial communities. The soil microbiome comprises all the microorganisms that live in symbiosis as well as those living free around the plant. These include bacteria, fungi, mycoplasma, and viruses. Thus, the microbiome plays a vital role in the growth and development of plants by providing necessary nutrients, protecting the plant from pathogens and environmental stresses, including plant’s hormone signaling to enhance its overall health (Andreote and de Pereira e Silva 2017). Plants can recruit soil microbiomes to help control infections caused by soil pathogens. Despite this, plants experience various types of microbial diseases. Our recent studies (Ren et al. 2023; Whitton et al. 2022a, b, 2023) accentuated the critical role of soil remediation in the successful recovery from the PDB and restoration of the soil microbiome as a vital factor in controlling the destructive advancement of PDB.

Compared to pesticides and herbicides, that can contribute to global pollution and exhibit a risk to wildlife and human health, recent increases in the use of natural, organic and bioactive plant products are gaining popularity in agriculture (Gwinn 2018). Bioactive natural products improve the sustainability of crop production and introduce a vast range of compounds that can fight pathogens with lower risk to the environment (Gwinn 2018). Phytogenic metabolites are bioactive compounds derived from plants with various beneficial properties, including antimicrobial, antifungal, antiviral, and insecticidal activities. These compounds play a vital role in the plant’s defence against pathogens and pests while displaying a broad-spectrum activity. Phytogenic metabolites have also exhibited the ability to induce systemic acquired resistance (SAR) in plants, a defence mechanism triggered by the plant’s response to pathogen attack that activates defence pathways that provide long-term protection (Banani et al. 2018).

Recently, we investigated the ability of a phytogen-based product containing citric acid, carvacrol, and cinnamaldehyde to recover PDB (Ren et al. 2023). The phytogenic liquid significantly improved dieback-affected field study plots, and it reduced microbial genera associated with PDB-affected soils, promoted plant and soil-beneficial bacteria living in plant rhizosphere and increased soil microbial diversity. Additionally, the phytogenic treatment improved plant morphology and pasture productivity beyond 18 months post-application. Some improvements in pasture dieback symptoms were also noted via the introduction of sea minerals (Whitton et al. 2022a) and humic acid (Whitton et al. 2023). Both these studies demonstrated that the improvements were associated with alterations in the beneficial soil microbiome.

Amplicon sequencing methodologies based on the 16S rRNA marker gene allow the investigation of the structure of microbial communities; however, this methodology does not have adequate resolution to imply species and even genus-level taxa with certainty (Gupta et al. 2019). In contrast, shotgun metagenomic sequencing investigates a collection of all genomes from a mixed community of organisms. In addition to species-level taxonomy, this methodology provides insight into gene functions (Zhang et al. 2021a) and microbial metabolic pathways in the soil, allowing a much deeper investigation of the complete metagenome (Daniel 2005) from all living organisms in the soil rather than only primer-selected proportion of bacterial microbiota possible with 16S methodologies. Soil metagenomics involves sequencing and analysing all the genomes in soil samples, including DNA from microorganisms that are difficult to culture or isolate. This approach provides a comprehensive and unbiased view of the soil microbiome, giving insights into the complex microbial community’s taxonomic and functional diversity and ecological roles (Myrold et al. 2014).

In this study we built upon our previous manuscript (Ren et al. 2023), where we reported improvements in soil microbiota, microbiota – mineral interactions, pasture productivity and plant morphometrics; present functional metagenomic changes of pasture dieback-affected soil microbiome in response to phytogen treatment. We hypothesise that the application of phytogenic product will affect the function of the soil microbial community, leading to the results reported in our previous study. We conclude that the soil treated with phytogen had more functional diversity than untreated soil. Furthermore, the application of phytogenic liquid significantly improved the ability of soil microbial metabolism, which helped promote plant growth directly and indirectly.

Materials and methods

Field trial

The details of the trial were described previously (Ren et al. 2023). Briefly, the trial was performed on a PDB affected paddock using two treatments, each replicated three times. The treatments included a control (CTR) and a phytogenic liquid treatment (PHY) sprayed after dilution with water, randomly assigned to three plots each. The main components in the phytogenic liquid were cinnamaldehyde, carvacrol, and citric acid, sourced from Activo® Liquid (EW Nutrition GmbH, Visbek, Germany). The 5 m × 5 m labelled plots were separated by a 2 m buffer area to prevent cross-contamination. The phytogenic liquid was applied using watering cans and sprayed twice, one week apart, at the rate of 0.54 mL/m2 diluted in 50 L of water each time, resulting in a final application rate of 10.8 L/ha. The control plots were sprayed with water at the same rate as those treated with the phytogenic liquid.

Sampling

The soil cores were sampled to a depth of 15 cm using a T-bar auger. One sample was taken from a random spot from each plot before the phytogenic liquid was applied in week 0, and subsequently, one sample was collected weekly from week 1 to week 6, followed by fortnightly sampling from week 8 until week 20. All the samples were placed on dry ice and stored in a freezer at a temperature of − 80 ◦C for further analysis.

Soil chemistry analysis

Soil samples from two groups were outsourced to the University of Western Australia’s Earth and Environment Analysis Laboratory (EEAL). The soil properties including minerals (total carbon/nitrogen, Al, B, Ca, Cd, Co, Cr, Cu, Fe, K, Mg, Mn, Mo, Na, Ni, P, Pb, S, Zn), pH and Electric Conductivity (EC) were assessed by Inductively Coupled Plasma Optical Emission Spectroscopy (ICP – OES) (Khan et al. 2022).

Shotgun metagenomic sequencing

The DNA of soil microorganisms was obtained using a DNA soil kit (DNeasy PowerSoil Pro Kit, Qiagen, Hilden, Germany). The quality and quantity of the DNA were evaluated using Qubit3.0 (Thermo Fisher, Waltham, MA, USA) and agarose gel electrophoresis. DNA is fragmented by ultrasonication using Covaris S220 (Covaris, Woburn, Massachusetts, USA), followed by end repair, dA-tailing, adaptor ligation and purification. The purified DNA was then subjected to size selection before undergoing PCR amplification for library construction. The library was quantified using Qubit3.0 and diluted. The concentration of the resulting library and insert size were measured using Agilent 2100 (Agilent, Santa Clara, CA, USA). To ensure accurate sample concentration and reliable sequencing data, the effective concentration of each library in the library mixture was determined with qPCR prior to the sequencing. The sequencing was performed using Illumina Novaseq6000 with 150PE configuration.

A total of 3.11 billion quality filtered sequences with 43.15 ± 3.35 million (mean ± SD) sequences per sample with a minimum phred score of 35 were obtained from the shotgun metagenomic sequencing.

Data analysis and statistics

The integrity of the data was verified by cryptographic hash generated with Message Digest Algorithm 5 (Rivest 1992). The quality of the raw sequence files was analysed using FastQC v0.11.2 (Andrews 2010) and MultiQC v1.11 (Ewels et al. 2016). The preliminary quality control of the sequencing data, including the removal of the sequencing adaptors, was done using fastp v0.20.0 (Chen et al. 2018). Further quality control and processing of the data was done with KneadData v0.7.10 applying Trimmomatic (Bolger et al. 2014) for quality filtering and trimming with leading and trailing base quality equal to three and sliding window threshold of 4:15.

The quality-filtered clean sequences were analysed using HUMAnN v3.6 (Abubucker et al. 2012) with UniRef90 (Suzek et al. 2007) databases for profiling molecular functions and metabolic pathways. The resulting functional data were analysed and visualised with the R program using a range of packages and tools, including Phyloseq (McMurdie and Holmes 2013), Vegan (Dixon 2003), and Microeco (Liu et al. 2021). The statistical software GraphPad Prism v9 (GraphPad 2016) and Primer-e v7 (Anderson et al. 2008) were also used for statistical analysis and plotting. The difference between two groups at a particular sampling point was analysed using the unpaired T-test through GraphPad Prizm 10.1.0 with a significant level of 0.05. The Locally estimated scatterplot smoothing (LOESS) was used to visualize the trend of change under different groups. Subsequently, the difference across all the weeks between two groups was visualized in R using Bray–Curtis and Jaccard matrices. The group difference was calculated by using DIVERSE algorithm in Primer. A square root transformed (SQRT) relative abundance matrix was generated to perform Bray–Curtis similarity. Mixed design two-way PERMANOVA test was then employed to analyse the longitudinal temporal effect of treatment, time, and replicate plots on samples. The plot number was a random effect nested within “Treatment” (Plot(Tr)) as recommended for the longitudinal sampling of multiple plots for treatment and control (PERMANOVA-Manual (primer-e.com)). The group differences were then plotted into distance-based redundancy analysis (dbRDA).

The DistLM (distance-based linear modelling) was used to check if variations in soil chemistry (mineral quantity) are related to changes in microbial metabolic pathways. This model can use distance matrices like the Bray–Curtis distance matrix and assess the combined effect of multiple variables (different minerals) in metabolic pathways.

Results

Overview of soil functional pathways

Overall, most of the pathways were unmapped/unknown, meaning they have not been annotated yet in the reference database or they are not aligned or matched with the reference database. Despite this, using only annotated soil microbial pathways, both phytogen and control groups showed robust functional variation. The top twenty discovered pathways are shown in Fig. 1 and Supplementary Fig. 1. The top 20 functional pathways represent around 30% of annotated pathways.

Fig. 1
figure 1

Twenty most abundant soil microbial functional pathways. Legend abbreviations used: Superpathway = Spw; Adenosine = Ade; Guanosine = Gua; Nucleotides = Nucl; Biosynthesis = Bsynt; Deoxyribonucleotides = Doxrn; Ribonucleotide = Rn; Branched Chain Amino Acid = BCAA

Full size image

Functional richness and diversity

We used Primer-e v7 DIVERSE algorithm to perform functional alpha diversity analysis (Fig. 2). Generally, the phytogen significantly increased observed pathway richness compared to the control and showed a significant increase in Shannon pathway diversity (Fig. 2a and b). This increase suggested that the phytogenic treatment amplified the soil microbial community’s functional potential and metabolic capabilities.

Fig. 2
figure 2

Functional richness and diversity of soil microbial community. Panels (a) and (b) present overall differences in richness and diversity, considering all collected data, while panels (c) and (d) show the differences per week. The LOESS trend line underscores the overall change along the sampling time

Full size image

Additionally, the differences in functional richness and diversity between treatments at each sampling point are also presented in Fig. 2c and d. The t-test compared the differences in every week in diversity and richness (Supplementary Table 3). Overall, the LOESS trend line illustrated a clearly higher richness and diversity in phytogen treated samples over the period. Typically, from week 3 to week 5, the phytogen treated group revealed a remarkable increase in richness compared to the control (p = 0.044, p = 0.015, p = 0.021, respectively). In addition, the functional diversity in week 3 and week 5 also identified a significant variation between the two groups (p = 0.005, p = 0.047, respectively), highlighting the positive benefit from phytogen on improving soil microbial functional category (Supplementary Table 3). Notably, this field trial started during the drought season and only had two significant rain events during week 5 and week 7. Although phytogen no longer induced significant change after week 7, it recovered both richness and diversity faster after the rain until the last sampling time. In contrast, the control exhibited a decreasing trend of diversity and richness after week 12.

Beta diversity

To investigate distances between the two groups, square root transformed pathway abundance data was analysed using Bray–Curtis and Jaccard matrices (Supplementary Fig. 2). The group differences were visualized by dbRDA plot (Fig. 3). The three dimensions plot reveals the samples onto the first three coordinates against two groups, with each point representing one sample from each group. The samples from two groups are distinguished by yellow (control) and green (phytogen) dots, respectively. We can see a clear separation between the two groups, suggesting that phytogen had impact on the samples.

Fig. 3
figure 3

Beta diversity measure. dbRDA plot is generated in Primer 7-e using transformed square root abundance data and Bray–Curtis distance

Full size image

Mixed design PERMANOVA analysis (Table 1), using Bray–Curtis measure with “Week” and “Treatment” as crossed fixed effects and “Plot” as a random effect nested within “Treatment”, showed that phytogen didn’t have a significant effect on overall microbial functional profile. However, there was a significant temporal variation (p = 0.001) in microbial function. This agreed with the fluctuation of richness and diversity over the time (Fig. 2) that alteration induced by phytogen varied temporarily. Additionally, the spatial difference also had a significant impact on the community (p = 0.001), which underscored the necessity of random sampling.

Table 1 Mixed design longitudinal PERMANOVA test
Full size table

Pathways

The pathways showing significant differences in phytogen treated soil samples are shown in Figs. 4 and 5, Supplementary Fig. 3–7, and Supplementary Table 1. Among these pathways, HISTSYN-PWY: L-histidine biosynthesis (Fig. 4a), PWY-5494: pyruvate fermentation to propanoate II (acrylate pathway) (Supplementary Fig. 7p), PWY-5913: partial TCA cycle (obligate autotrophs) (Fig. 4c) and PWY-6969: TCA cycle V (2-oxoglutarate synthase) (Supplementary Fig. 4a) were among the most significantly altered metabolic pathways. Phytogen application promoted the biosynthesis of a number of amino acids, including lysine (Fig. 4b), histidine (Supplementary Fig. 3a), serine (Supplementary Fig. 3b), glycine (Supplementary Fig. 3b), arginine (Supplementary Fig. 3c, 3d, 3e), glutamine (Supplementary Fig. 3f), and methionine (Supplementary Fig. 3 g), while pathways involved in the synthesis of tryptophan (Supplementary Fig. 3 h, 3i) and cysteine (Supplementary Fig. 3j) were reduced by phytogen.

Fig. 4
figure 4

Pathways increased in phytogen-treated soil microbial community. The asterisk indicates significant level: * = p < 0.05, ** = p < 0.01, *** = p < 0.001, **** = p < 0.0001

Full size image
Fig. 5
figure 5

Pathways associated with soil detoxification. The asterisk indicates significant level: * = p < 0.05, ** = p < 0.01, *** = p < 0.001

Full size image

Phytogen also showed a significant influence on functions related to TCA cycle (Supplementary Table 1). The phytogen treated microbial community had increased abundance of TCA related pathways, including PWY-6969: TCA cycle V (2-oxoglutarate synthase), PWY-5913: partial TCA cycle (obligate autotrophs), P105-PWY: TCA cycle IV (2-oxoglutarate decarboxylase) (Supplementary Fig. 4b) and P23-PWY: reductive TCA cycle I (Supplementary Fig. 4c), which participates in energy metabolism. No similar pathways were downregulated.

Another group of functions elevated by phytogen treatment include nucleotide biosynthesis and recycling. In treatment samples, there was a significant increase in pathways that resulted in higher production of purines guanosine (PWY-7228) (Fig. 4d), (PWY-7234) (Supplementary Fig. 4d), recycling of pyrimidine deoxyribonucleoside (PWY-7199) (Supplementary Fig. 4e), and NAD de novo biosynthesis from aspartate (PYRIDNUCSYN-PWY) (Supplementary Fig. 4f) as well as pyrimidine deoxyribonucleoside biosynthesis (PWY-7198), degradation (PWY0-1298) (Supplementary Fig. 4 g), guanosine ribonucleotides de novo biosynthesis (Supplementary Fig. 4 h) and pyrimidine deoxyribonucleotides de novo biosynthesis IV (Supplementary Fig. 4i) indicating higher nucleotide turnover in microbial community in phytogen treated plots (Supplementary Table 1).

In phytogen-treated plots, 10 pathways involved in soil detoxification were enriched. They are: 14DICHLORBENZDEG-PWY pathway (Fig. 5a), which represents the 1,4-dichlorobenzene degradation pathway in soil bacteria, PWY-6084: 3,5-dichlorocatechol degradation (Fig. 5b), PROTOCATECHUATE-ORTHO-CLEAVAGE-PWY responsible for the degradation of protocatechuate, an intermediate compound in the breakdown of aromatic compounds. Additionally, some other pathways increased in phytogen supplemented group are indirectly involved in detoxification processes like PWY-5918 (Fig. 4e): the superpathway of heme b biosynthesis, which is a cofactor for enzymes involved in the detoxification of reactive oxygen species (ROS). Similarly, indirectly involved in detoxication is a PWY1G-0 (Fig. 5c): mycothiol biosynthesis pathway, PWY-6834: spermidine biosynthesis III (Fig. 5d), PWY-5918: superpathway of heme b biosynthesis (Fig. 5e) and PWY-3781: aerobic respiration I (cytochrome c) (Fig. 5f) also involved in ROS detoxification. Other three relevant pathways are presented in Fig. 5g, h and i.

There were 3 other pathways which increased in phytogen-treated soil related to increased carbon assimilation; these include all TCA cycle pathways mentioned above as well as PWY-241: C4 photosynthetic carbon assimilation cycle, NADP-ME type (Fig. 4i) as well as PWY-7117: C4 photosynthetic carbon assimilation cycle, PEPCK type (Supplementary Fig. 5a) and others that are shown in Supplementary Table 1.

Finally, the last group of pathways which increased in phytogen-treated soil microbial functions are those related to cell wall and membrane health improvements of plants and microbes alike. These include PWY-5667 pathway, which represents the CDP-diacylglycerol biosynthesis I pathway (Supplementary Fig. 5b) in soil bacteria, PWY0-1319: CDP-diacylglycerol biosynthesis II (Supplementary Fig. 5c), PWY-5136: fatty acid β-oxidation II (plant peroxisome) (Supplementary Fig. 5d), PHOSLIPSYN-PWY: superpathway of phospholipid biosynthesis I (bacteria) (Supplementary Fig. 5e), PWY-6282: palmitoleate biosynthesis I (from (5Z)-dodec-5-enoate) (Supplementary Fig. 5f), FASYN-ELONG-PWY: fatty acid elongation – saturated (Supplementary Fig. 5 g), PWY-7664: oleate biosynthesis IV (anaerobic) (Supplementary Fig. 5 h), UDPNAGSYN-PWY: UDP-N-acetyl-D-glucosamine biosynthesis I (Supplementary Fig. 5i), and indirectly PWY-6834: spermidine biosynthesis III which helps in regulating reactive oxygen species (ROS) levels and stabilising membranes.

While phytogen increased the overall number of pathways, only 11 pathways were reduced in phytogen-treated soil (Supplementary Table 1) and were higher in abundance in the untreated control. They include above mentioned 2 tryptophane and L cysteine biosynthesis pathways, synthesis of plant beneficial phosphopantothenate, a precursor in the biosynthesis of coenzyme A (CoA) (Supplementary Fig. 7a), PWY-5100: pyruvate fermentation to acetate and lactate II, one detoxification pathway (Fig. 5g), 12DICHLORETHDEG-PWY: 1,2-dichloroethane degradation (Fig. 5h) and two pathways involved in sulphate reduction SO4ASSIM-PWY: assimilatory sulphate reduction I (Supplementary Fig. 6a), PWY1ZNC-1: assimilatory sulphate reduction IV (Supplementary Fig. 6b) and these pathways are generally directly or indirectly beneficial to plants.

In summary, phytogenic treatment stimulated microbial pathways in the soil. These pathways are involved in amino acid, SCFA and cofactor production, energy-related TCA cycle pathways, detoxification from environmental pollutants and ROS, and cell wall and membrane protection.

Soil chemistry and pathways

There were no significant differences between CTR and PHY in any of the soil's analysed mineral concentrations or EC, either before the treatment was applied (time zero) or at 12 weeks post-application. Using the nonparametric Mann–Whitney test, the lowest P-value recorded in any comparisons was 0.10, confirming no difference in soil chemistry between the plots before the treatment was applied (Supplementary Table 2). PERMANOVA (Primer 7e) confirmed no significant effect of treatment (PHY, P = 0.648) or time (P = 0.211) on mineral concentrations. The data investigation was then continued in Primer 7e using the soil chemistry data matrix as the environmental and pathway matrix as the biological data measurements. Mineral data were square root normalised and transformed to the Euclidean distance, while pathway data were normalised and transformed to the Bray–Curtis matrix. We then used the DistLm algorithm to explain how well the environmental data patterns correlate with the pathway data patterns. Marginal tests indicated a significant correlation between concentrations of nitrogen (P = 0.02), carbon (P = 0.018) and iron (P = 0.018) with biological pathway data. Marginal tests observe each of the variables (minerals) independently. Sequential tests show that combined effects of N, Pb and K account for 40% of the variation, where sequential adding of each mineral to this combination resulted in significant changes.

Discussion

Due to the small number of sequenced and annotated soil microbial genomes, soil shotgun metagenomics can only use a fraction of generated sequences. Compared to the intestinal microbiome, the soil microbiome is still uncharted territory. This indicates the presence of massive untapped potential for soil microbiome characterisation that cannot yet be used for translational research, like novel enzyme development, until the databases are more complete. This encourages revisitation of the metagenomic soil studies as the databases continue to grow.

This project builds upon our previous work on phytogen application for PDB treatment (Ren et al. 2023). We showed that phytogen treatment enhanced alpha diversity and microbiota adaptation to rainfall (Ren et al. 2023). It also improved soil quality, plant health, and pasture productivity for over 18 months. Phytogen-treated plots had more biomass than control plots (Ren et al. 2023). The present study investigates capacity of the phytogen to enhance soil microbial community and pasture yield, particularly in soils that were severely affected by pasture dieback. This study confirmed our hypothesis that this phytogenic product had beneficial effects on soil microbial functions and metabolism.

The most abundant pathway in the soil (Fig. 1) was related to bacterial nutrient recycling: PWY-7208: Superpathway of pyrimidine nucleobases salvage. If the microbial community activates PWY-7208, it will increase available pyrimidine nucleobases in the soil, which can be used by plants for nucleic acid synthesis (Caspi et al. 2020). A similar role could be expected for the other 11 of the top 20 pathways involved in the microbial de-novo synthesis of nucleotides and nucleosides, which can ultimately end up in the soil and be used by the plants. The extent to which microbial production of basic biological molecules helps plants depends on a range of factors such as nutrient availability, environmental conditions, plant species and their health, plant-microbial interactions, and interaction networks between the soil microorganisms (Chauhan et al. 2023; Kumar et al. 2023). Four out of the top 20 soil functions provided by the microbiome were related to the microbial synthesis of amino acids, including L- valine, L-isoleucine, 5-aminoimidazole ribonucleotide biosynthesis and branched-chain amino acids. The high availability of amino acids in the soil is important for plants that absorb soil amino acids via active uptake or specific transport mechanisms in their root cells and passively via diffusion (Kolomazník et al. 2012).

Amino acids in the soil originate from the decomposition of organic matter, including microbial cells, root exudates, microbial metabolic activity, and fertilising (chemical fertilisers do not contain amino acids, but organic manure does). Microbial-facilitated fast access to branched-chain amino acids in the soil is important for stress tolerance, plant hormone biosynthesis and seed germination (Joshi et al. 2010). Two of the top 20 soil microbial pathways were related to assimilatory sulphate reduction, including PWY1ZNC-1: Assimilatory sulphate reduction IV and SO4ASSIM-PWY: Assimilatory sulphate reduction I. Plants depend on soil microbes to perform most of the sulphate reduction (Natasha et al. 2021). Molybdopterin is a cofactor required for molybdoenzymes, which are involved in various metabolic processes in plants, including nitrogen fixation. Thus, based on an overview of the most abundant soil microbial pathways, we can now confirm the relevance of the annotated portion of soil microbiome data for plant health and growth.

The alpha diversity analysis revealed a significant difference in pathway richness and diversity (Fig. 2), which can be attributed to several factors. The phytogen induced strong alterations in the pasture dieback soil microbial community (Ren et al. 2023), which would be expected to alter the microbiome in every aspect, especially its functional capacity. We have previously demonstrated that phytogen treatment restored microbial richness and diversity faster than untreated samples (Ren et al. 2023). We also must consider that both groups (CTR and PHY) can have functional redundancy in which different taxa could perform similar functional profiles (Biggs et al. 2020) and that there are many pathways and processes that can perform the same function; thus, the loss of one or a few similar pathways does not imply total loss of the specific soil function. However, in this case, the increase in soil functional capacity and diversity post-phytogen application is remarkable, and it is a long-term cumulative effect in both the number of functions (richness) and their functional contribution (diversity) as observed in Fig. 2. It should be noted that this trial was conducted during the dry season, and rain events occurred in weeks 5 and 7.

While we have previously shown a significant increase in microbial richness and diversity in phytogen-treated soil after the rain events (Ren et al. 2023), the functional capacity of phytogen-treated plots was already higher before the rain event. This indicates that functional enhancement effects of phytogen was not dependant on the rain. One possible reason is that phytogen affects the soil microbial functionality by altering the quantity and quality of soil microbial available resources, including carbon, nitrogen, etc. (Weng et al. 2021). These alterations might promote their metabolism without elevating microbial diversity, however, after a long time, the improved resource supply could enable more soil microorganisms to survive, leading to higher microbial richness and diversity (Li et al. 2022; Strecker et al. 2015).

Phytogenic treatment significantly altered 65 functional pathways in the soil environment, 54 increased, and 11 decreased in phytogen, many of which have been found to play an important role in plant health and development. Several altered pathways are involved in amino acid biosynthesis. As an essential amino acid, histidine (Fig. 4a) was increased in the soil environment, which positively influences plant growth by acting as a chelator and transporter of metal ions (Stepansky and Leustek 2006). Cadmium (Cd) can negatively impact plant growth and development at very low concentrations (Choppala et al. 2014). For example, most plants experience toxicity when leaf concentrations exceed 5–10 μg Cd/g dry matter (DM) (White and Brown 2010). Histidine may be involved in Cd resistance and accumulation by reducing oxidative damage (Zemanová et al. 2014). Increased L-histidine in soil enables plants to obtain adequate nutrients as L-histidine is involved in various gene regulation networks and signalling pathways, including but not limited to ethylene, cytokinin, osmosensing, and cold perception (Seo et al. 2016; Nongpiur et al. 2012).

L-histidine can elicit plant resistance against the bacterial pathogen Ralstonia solanacearum by partially activating ethylene signalling (Seo et al. 2016). Another essential amino acid which was affected by the phytogen treatment is lysine. Boosting the activity of the PWY-2941 pathway (Fig. 4b) can stimulate the production of L-lysine in soil bacteria, which has a beneficial influence on plant growth. L-lysine plays a critical role in protein synthesis and diverse metabolic processes within plants (Yang et al. 2020), including starch regulation, unfolded protein responses, lipid metabolism, glycolysis metabolism and nucleotide metabolism (Kiyota et al. 2015; Bernsdorff et al. 2016; Arruda and Barreto 2020). It is also involved in plant stress response in various forms, mainly catabolised through the saccharine pathway, which has been shown to play a role in abiotic and biotic stress responses (Yang et al. 2020).

Other amino acids produced by above-mentioned pathways also have benefits for plants. For example, L-serine (Supplementary Fig. 3b) performs catalytic functions in enzymes (Ros et al. 2014). Glycine (Supplementary Fig. 3b) enhances plant growth, nutrient uptake, and photosynthesis (Zargar Shooshtari et al. 2020). L-arginine (Supplementary Fig. 3d) plays an important role in nitrogen reserve as well as the biosynthesis of polyamines (Yang and Gao 2007). Glutamine (Supplementary Fig. 3f) is involved in nitrogen assimilation and carbon metabolism (Liao et al. 2022). Methionine (Supplementary Fig. 3 g) is essential for protein synthesis and responsible for S-adenosylmethionine (SAM), which has multiple roles in plant metabolism and regulation (Amir and Hacham 2008). Two amino acids that were reduced in the soil due to a reduction in functional capacity in the phytogen-treated soil, also play a role in plant health: L-tryptophan (Supplementary Figs. 3 h, 3i) and L-cysteine (Supplementary Fig. 3j).

On the other hand, the production of amino acids by soil microbes does not necessarily secure benefits for plants. The availability and accessibility of amino acids produced by soil microbes can depend on other factors, such as microbial use of the amino acids, nutrient competition and environmental conditions. All of this may influence the extent to which plants can access and use these amino acids.

Four pathways involved in TCA (the tricarboxylic acid) cycle were increased in phytogen soil. TCA is a fundamental component of respiratory metabolism in various plant organs. Phytogenic product has significantly increased PWY-5913 (obligate autotrophs) (Fig. 4C), PWY-6969 (2-oxoglutarate synthase) (Supplementary Fig. 4a), P105 PWY TCA cycle IV (2-oxoglutarate decarboxylase) (Supplementary Fig. 4b) and P23-PWY reductive TCA cycle I (Supplementary Fig. 4c). The more active PWY-5913 pathway in the soil microbial community makes higher TCA cycle flux, resulting in increased ATP production (Zhang and Fernie 2018). This process can promote energy availability to support plant growth. Other three pathways have similar functions. For example, PWY-6969 TCA cycle improves carbon metabolism (Caspi et al. 2020), P23-PWY pathway enhances nutrient availability (Caspi et al. 2013), and P105-PWY pathway helps the synthesis of metabolites.

Pathways involved in nucleotide biosynthesis PWY-7228 super pathway of guanosine nucleotides de novo biosynthesis I (Fig. 4d) (p < 0.05) together with PWY-7221 guanosine ribonucleotides de novo biosynthesis (Supplementary Fig. 4 h) (p < 0.05) were also altered. The PWY-7228 pathway is responsible for the de novo biosynthesis of guanosine nucleotides, including guanosine triphosphate (GTP), an energy carrier. Increasing the activity of this pathway can lead to more GTP production in soil bacteria. This improvement can potentially impact ATP synthesis, protein synthesis and other energy-required processes for plant growth (Nodop et al. 2008). PWY-7221 pathway plays a similar role in nucleotide synthesis and energy metabolism.

Intriguingly, some pathways increased in phytogen-treated soil are specifically associated with plant health. The upregulation of the PWY-5918 heme b pathway (Fig. 4e) in soil bacteria can have a beneficial influence on plant growth by enhancing the synthesis of chlorophyll to improve photosynthetic efficiency, leading to increased energy production and supporting plant growth (Zhang et al. 2021b). PWY1G-0 mycothiol biosynthesis pathway (Fig. 4f) helps plants in terms of plant resilience and interaction with soil microorganisms (Mokhtar et al. 2021). Mycothiol plays a crucial role as an antioxidant and helps protect bacterial cells against oxidative stress. By enhancing the activity of the PWY1G-0 pathway, the production of mycothiol by soil bacteria can be increased. This elevated production enhances the bacteria’s capacity to counteract oxidative stress. As a result, the stress tolerance of plants can be positively influenced by better management and resistance to oxidative damage (Zhang et al. 2021c). Increasing the PWY-6834 pathway in soil bacteria (Fig. 5d), leading to enhanced spermidine biosynthesis, can positively influence plant growth. It can result in enhanced polyamine biosynthesis, stress tolerance, cell division and growth, and gene expression and signalling. These effects can support essential cellular processes, stress tolerance mechanisms, and overall plant growth and development (Tabor et al. 1958).

In agreement with the data we presented previously (Ren et al. 2023) showing phytogen promoted a significant reduction in bacterial species known as biomarkers of polluted soils, here we detected a functional increase in pathways involved in detoxification as significantly improved by treatment. This can point towards complex mechanisms in soil detoxification by phytogen administration that involve altering both function and taxonomy in the same direction of soil detoxification. One of the pathways increased in phytogen soil was PROTOCATECHUATE-ORTHO-CLEAVAGE-PWY protocatechuate degradation II (ortho-cleavage pathway) (Fig. 4g). As a precursor, protocatechuate (PCA) is involved in other pathways where it is converted to other intermediates for the degradation of aromatic compounds, including various pollutants and plant secondary metabolites (Davis and Sello 2010). By breaking down these compounds, microorganisms can effectively eliminate toxins and prevent their accumulation to harmful levels and support plant health. Another pathway associated with detoxification is 14DICHLORBENZDEG-PWY (Fig. 5a), which is responsible for degradation of 1,4-dichlorobenzene, a toxic compound often found as an environmental pollutant (Rehfuss and Urban 2005). This pathway promotes pollutant detoxification, soil remediation, improved nutrient availability, reduced plant stress, and enhanced soil ecosystem dynamics (Rehfuss and Urban 2005). These effects can contribute to healthier plant growth, improved nutrient uptake, and overall enhanced plant productivity.

There are some pathways that regulate cofactors of critical enzymatic reactions. PWY-5198 pathway (Fig. 4h) and PWY-8112: factor 420 biosynthesis I (Archaea) (Supplementary Fig. 6c) represent cofactor 420 biosynthesis, positively impacting the enzymatic reactions that rely on factor 420, potentially enhancing important metabolic processes in the soil ecosystem that are crucial for plant growth (Bashiri and Baker 2020). Some enzymes requiring factor 420 are involved in the breakdown and transformation of organic compounds, contributing to nutrient availability for plants. Increasing the activity of this pathway can enhance the functioning of these enzymes, potentially improving the efficiency of nutrient cycling processes in the soil and positively impacting plant growth through increased nutrient availability. A range of functions involved in lipid production was increased exclusively in phytogen-treated soil. Higher biochemical lipid in the soil can benefit both plants and other microorganisms via increased lipid availability and better soil water-holding capacity (Khan et al. 2021).

Another essential function for plant growth is carbon assimilation, which could be influenced by two pathways significantly increased by phytogenic product, PWY-241 C4 photosynthetic carbon assimilation cycle (Fig. 4i), NADP-ME type and PWY-7117 C4 photosynthetic carbon assimilation cycle, PEPCK type (Supplementary Fig. 5a), and other pathways that improve this indirectly mentioned in the results section. The C4 photosynthetic pathway is a special metabolic process where carbon is fixed in non-optimal conditions, like drought season (Alvarez et al. 2019). This adaptative fixation of carbon might explain how plants tolerate extreme pasture dieback environments with phytogenic treatment.

Microbial photosynthesis in the soil has major ecological and climate change-related benefits. When microbes perform photosynthesis, they capture atmospheric carbon dioxide (CO2) and convert it into organic compounds, with sunlight as an energy source. This process contributes to carbon fixation in the soil, leading to carbon sequestration while producing organic compounds to be used as food for plants and other soil organisms. Several studies have highlighted the importance of microbial photosynthesis in soil ecosystems (Yang et al. 2022).

While the specified phytogen promoted several pathways and inhibited only a limited number of pathways, the significance of these inhibited pathways cannot be neglected. The favourable outcome of the treatment may be the net effects of enhanced and inhibited pathways, which raises the possibility that the mechanistic explanation of the treatment’s benefits might not be as straightforward as it may seem. This warrants a detailed study of individual affected pathways and their respective influence on soil and plant health.

The distance-based linear modelling revealed a strong and quantifiable association between some soil minerals and microbial metabolic pathways. The significant correlations with nitrogen, carbon, and iron individually, and the combined effects of nitrogen, lead, and potassium imply that soil minerals play an important role in influencing the soil microbiome.

Conclusion

In conclusion, this research demonstrated that a phytogenic product containing citric acid, carvacrol, and cinnamaldehyde had a positive effect on the soil microbial functional pathways, which are beneficial for soil health and plant growth. The treatment significantly increased soil functional capacity associated with amino acid biosynthesis, TCA cycle regulation, nucleotide biosynthesis, degradation of organic compounds, nutrient availability, stress tolerance, and detoxification, to name a few. Our results complement previously reported significant improvement in pasture productivity and plant morphology by phytogen and suggest that phytogenic products can enhance soil health and plant growth by modulating metabolic activities in the soil microbial community. This study provides new insights into the potential of applying phytogenic products to improve soil health with the view to achieving ecologically sustainable plant production.