Soil is alive: Biological, Physical, and Chemical Analysis

Introduction  

Soil Physical Properties 

The physical properties of soil are an important factor in determining which crops to plant. Soil’s physical properties are properties that can be seen and felt. Physical properties do not directly refer to chemical or biological properties but are still affected by both [1]. According to the Cornell University Agronomy Fact Sheet, “Soil texture determines the rate at which water drains through a saturated soil; water moves more freely through sandy soils than it does through clayey soils^” [2]. From the different experiments, we expect to find the soil to have a moisture content between 10%- 45% and an organic matter content between 5%-10% according to previous research done on soils. Moisture is thought to be one of the most important, perhaps because it is the most immediately tangible characteristic of soil. 

EC, pH, and TDS 

pH represents the intensity of acidity in soil; calcium-saturated organic soil typically has a pH of 7.2 to 7.8, which is mild to moderately alkaline [3]. Regarding EC values, “Apparent soil electrical conductivity (EC) has shown promise as a soil survey tool” [3]. This means monitoring over time is necessary. TDS (Total Dissolved Solids) levels are important because, “Irrigation can contribute a substantial amount of salt to a field over the season” [4].  

The accumulation of soluble salts may be an issue at 500 ppm for some sensitive plants, but is usually tolerated by most plants up to 1000 ppm. In these instances, soils are typically high in chlorides or sulfates. EC gives a more accurate measurement of soluble salts because electricity would not be conducted by pure water.  

pH is important because it may ultimately lead to crop failure. The accumulation of salts is known to be associated with high pH. Excessive salts can lead to fertilizer burn. The pH of the soil is the most important factor for plants in terms of their ability to incorporate soil nutrients. Even if the necessary nutrients are present in the soil, they are not available to plants if the pH is not in the optimal range for that nutrient [5]. A good example is iron which is not available at high pH. Many soil elements change form because of reactions in the soil. These reactions, controlled by pH, alter the solubility of nutrients [1]. 

Soil Nutrients 

The primary macronutrients, sometimes called fertilizer elements, are often not available in large enough amounts for optimal growth. The three primary nutrients— nitrogen, phosphorus, and potassium—are added to the soil by fertilization with synthetic mineral salts or organic amendments such as manure [1]. Visual cues may be used to detect excess or deficits but are unreliable for proper diagnosis. Many factors can cause issues such as chlorosis, and different species may take on different appearances. The fertilization regimen can become more efficient with soil testing, reducing the effects of contamination through runoff as well as reducing production costs. Keeping track of nitrogen, phosphorous, and potassium is important beyond the health of crops; it is also important in terms of protecting the environment.  

Nitrogen and phosphorus pollution can both cause serious damage to surface waters. Nitrates cause eutrophication; phosphates destroy habitat. Potassium may create an environment that is too salty. Results from field studies indicate that rational use of manure and mineral fertilizers can help reduce the pollution problems arising from livestock farming practices [6]. In such systems, defining N rates where yields are maximized, and environmental harm is minimized could benefit both food production, human health, and the environment [7]. Testing the soil is the best way to find out how much fertilizer needs to be added to grow a productive crop without adding too much, which can contribute to input costs and environmental damage [5]. 

Fluctuations in nutrient levels within the soil can act as catalysts for proliferation of specific bacterial strains, shaping the microbial landscape. The interplay between nutrient availability and microbial diversity offers a promising avenue for understanding the relationships driving bacterial evolution. Recognizing the intricate connections between nutrient levels and microbial diversity provides insights into sustainable agricultural practices. Unraveling the impact of soil nutrients on bacterial discovery contributes not only to scientific understanding but also to the development of eco-friendly farming approaches. 

Life in the Soil 

Industrial hemp (Cannabis sativa) is a fiber crop with impressive carbon sequestration potential. For example, 1 ton of hemp stalks contains about 0.445 tons of carbon absorbed as a gas or 1.63 tons of CO2 [8]. The cannabidiol varieties of hemp are also purported to produce products that have radical scavenging capabilities. Thus, it may be useful for building materials or be therapeutically relevant. The rootzone microbes of industrial hemp have been little studied in a field crop setting. Previous research showed that rhizosphere microbes in cannabis are cultivar-specific. However, it was unknown to what extent soil types or cropping systems influence these microbes.   

The technique used to assess this in our study was a comparative analysis of 16S soil metabarcoding results from 6 samples from each of two sites; in Camarillo, CA at Rodale Organic Institute where JinMa and CBD (cannabidiol) types of hemp were grown, and in Tulare, CA at Western Fiber’s south side field, where the YuMa fiber variety was grown and harvested. The Tulare site had a history of traditional synthetic farming methods.  

Purpose and Hypotheses 

The purpose of this study was to discover plant growth-promoting and antibiotic-producing bacteria from the rootzones of plants. The hypothesis was that each soil environment would have a unique microbial community associated with it. This hypothesis was developed by considering the differences in geography and the fact that the soil is sandier at the Tulare site and contains more clay at the Camarillo site. We also hypothesized that the organic farm in Camarillo would have more diverse and abundant microbes performing these essential functions, when compared to the traditionally farmed soil in Tulare. This is because organic farming is generally considered to be more biodynamic than traditional or synthetic farming methods. 

Overview and Importance 

The soil metagenome consists of the 16S ribosomal genes detected in a soil sample by Next-Generation Gene sequencing techniques (Massive Parallel Sequencing, such as Illumina). The metagenome is a snapshot, both qualitatively and quantitatively, of the microbial community present in the soil. Biologic activity is an important indicator of the soil’s health. The great multitude of functions of the soil’s microbiota is impressive. Major planetary nutrient cycles are involved and dependent on the microbiota of the soil and are of vast importance to both natural ecosystems as well as agricultural endeavors. Experiments that assess the metagenome are of great importance in informing best practices associated with organic and regenerative farming. 

The soil hosts billions of bacteria, but not all species can be readily cultured [9]. The number of culturable bacterial cells in soil is generally believed to be only about 1% of the total number of living cells present. However, in environmentally stressed soils, the number of culturable bacteria may be as low as 10^4 cells per gram [10]. This emphasizes the importance of metabarcoding and other cultivation-independent techniques. Metabarcoding as it now exists is novel, particularly in its ability to detect thousands of taxa quickly in a single sample and in an economically feasible fashion – opening gateways for research to further understand plant-microbe interactions, plant physiology, microbial characteristics, and anticipating potential threats from pathogens. There is also the potential to identify new species, which may be important to agricultural technologies.  

The Rodale Institute has been involved in the research and development of organic regenerative agriculture and has not applied synthetic fertilizers, pesticides, or herbicides to any of the fields that have been sampled. Furthermore, the Rodale Institute has experimented with various cropping systems and policies of leaving crop residues post-harvest, cover-cropping, green manuring, and not allowing fields to lie fallow. The recent history of Western Fiber’s field is in line with the current agricultural method of utilizing synthetic herbicides, insecticides, and fertilizers, with fields and orchards typically existing as monocultures and often left fallow during non-growing seasons. Conventional tillage systems have been shown to break down soil structure and diminish the quantity of soil bacteria [1].   

Methods 

Soil Physical and Chemical Properties  

During Fall 2021- Spring 2022, 51 soil samples from Los Angeles Pierce College farm and local farms were analyzed for soil chemical and physical properties. Industrial Hemp YuMa plots in Tulare grown at Western Fiber were sampled for soil. In Camarillo, JinMa fiber variety and CBD type rootzones were sampled for soil. Noxious weeds growing in Marquis field at Pierce College, soil from the Pierce Arboretum, and soil from fallow ground from a proposed vineyard site were also sampled.   

Moisture, Organic matter, EC, TDS, pH, N, P, and K were determined using the methods described in the St. Clair Soil Science Lab Manual. Moisture % was determined by the oven dry method. Organic matter percentage was determined by dry combustion. A 1:5 soil-to-water ratio was used for soil pH measurements. N, P, and K were determined using Hach color-changing reagents, and %T was measured using a Genesys 20 spectrophotometer. Soil physical and chemical properties from the sites were also measured, including EC, TDS, and texture by touch. Colony forming units (CFUs) were also quantified. Comparisons between fields were carried out in the R stats package. Since there was an indication of non-normality and unequal variance, the nonparametric Kruskal-Wallis regression method was used to analyze the data. 

Soil Metabarcoding 

During Fall 2021, 22 soil metabarcoding samples were collected from Los Angeles Pierce College farm and local farms, DNA was extracted, and samples were sequenced on Illumina. The samples and their associated locations are referred to as follows: JinMa and CBD are from Camarillo, Ventura County, and Industrial Hemp N_side, and S_side samples are from Lemoore, Tulare County. 

The data consists of 16S amplicons. DNA was extracted with the Qiagen (Hilden, Germany) Power Soil DNA kit. Quantification of the DNA extracted was achieved with spectrophotometry using the Thermo Fisher MultiSkan SkyHigh Microplate Spectrophotometer. DNAs were then sent to James Madison University for 16S amplification and library preparation, and 16S amplicon NGS was pooled on the Illumina (San Diego, CA, USA) MiniSeq platform. The V4 region of the bacterial 16S rRNA gene was amplified and barcoded for each sample using the primers developed by Kozich et al  [11]. Samples were pooled. A double-sided bead cleanup was carried out. The quality and concentration of the pooled library were checked using a Bioanalyzer (Agilent, Santa Clara, CA, USA) and NEB’s Library Quant Kit for Illumina. The library was sequenced on a MiniSeq using a mid-output reagent cartridge. Before loading, the library was combined with Illumina’s PhiX control (30:70 16s: PhiX) to ensure a high-quality run despite the problem of relatively low diversity in the 16S library. Metabarcoding data analysis was carried out using the DNA Subway Purple Line [12] using QIIME 2 [13] and Emperor [14].  DNA Subway was used for bioinformatics; the Purple Line analysis implemented DADA2 and QIIME2 for quality control, alpha rarefaction, and the output of the ASV table for taxonomic diversity analyses. Using MS Excel pivot tables, the relative frequency of reads from different functional categories of microbes was visualized [15]. 

Results and Discussion  

Soil Physical and Chemical Properties 

Data showed that the conditions between the fields differed significantly in values for organic matter, N, K, TDS, and pH (p<0.05); this offered a diverse panel of substrates for the discovery of bacteria performing beneficial functions to plants and humans. A visual representation of the correlation matrix is given in Figure 2. It is apparent that Potassium concentrations are making the highest contribution to EC and TDS from the elements measured, based on the Pearson correlation matrix. It also appears that the CFU concentrations for the 10^-5 dilution were mildly associated with higher N and P levels. As expected, EC and TDS values were 100% correlated. 

According to the results of nonparametric Kruskal-Wallis regression [16]. Field was associated with OM (p=0.02). There was a trend toward the Arboretum and Marquis C samples exhibiting the highest OM percentages. However, the results of the pairwise Wilcoxon rank sum test indicated that there was no significant difference in OM between the groups, based on the adjusted p-values using the BH correction (BH correction= Benjamini-Hochberg false discovery rate). 

According to regression results, Field was associated with N (p=0.02). There was a trend toward higher nitrogen levels at the proposed vineyard site, and the lowest N levels were from Rodale Institute. Most samples had low values for Nitrogen in ppm. No significant pairwise differences were detected when using the BH-corrected p-values and a cutoff of 0.10. 

Field was associated with K (p=3.24*10^-5). In the pairwise Wilcoxon rank sum test, the Potassium levels of Marquis A soil samples varied from the Arboretum soils, and Western Fiber soils from Tulare differed from Rodale soil K levels, using a marginally significant p-value cutoff of 0.10. According to the results of nonparametric Kruskal-Wallis regression, Field was associated with pH (p=0.002). In the pairwise Wilcoxon rank sum test, the pH of Marquis A soil samples varied from the Arboretum soils, and Western Fiber soils from N. side Tulare differed from Rodale soil pH levels, using a marginally significant adjusted p-value cutoff of 0.10. 

According to the results of nonparametric Kruskal-Wallis regression, Field was associated with TDS (p=8.49*10^-6). In the pairwise Wilcoxon rank sum test, the TDS of the Arboretum soils was significantly different than N. side Tulare field soils (p-adj<0.05). Furthermore, the Marquis C Drain soil samples varied significantly in TDS from the Marquis A soils (p-adj=0.03), and Marquis A was significantly different in TDS than N. Side Tulare fields. Marquis B and Rodale samples were also significantly different in TDS when compared with Tulare N. Side samples (p-adj=0.03). The highest TDS values were from Marquis B and the proposed vineyard site (Figure 2). The BH correction was applied. 

The tested sites, characterized by the lowest nitrogen levels among the fields, present an intriguing scenario. Nitrogen scarcity often acts as a selective pressure on microbes, encouraging the emergence of bacteria with specialized adaptations for nitrogen acquisition. Microbial communities, being highly adaptable, strategically adjust their physiology based on nutrient availability. Potassium influences microbial dynamics, and the correlation between potassium concentrations and Electrical Conductivity (EC) raises the question of whether microbial abundance may respond to varying potassium levels. Colony Forming Unit (CFU) concentrations for the 10^-5 dilution exhibit a subtle association with higher nitrogen (N) and phosphorus (P) levels. This suggests a microbial response to varying nutrient levels, hinting at a potential correlation between nutrient availability and microbial abundance. 

Some bacteria can increase iron availability in conditions of high pH. Bacillus, which was detected in our Tulare samples, is a genus known to catabolize and exude siderophores (compounds acting as iron chelators) into the rhizosphere, which create Fe availability in soils that would otherwise have too alkaline a pH for sufficient iron mobility in the soil, thus increasing the upper limit of the pH range tolerable to any specific crop; iron availability is naturally limited at high pH [17]. 

Beta Diversity Analysis 

In the beta diversity analysis using the unweighted UNIFRAC distance for a principal coordinates visualization of the 16S data, the Camarillo samples from both CBD and JinMa varieties appeared to be very similar based on their genetic distances (Figure 3). Meanwhile, the cluster of samples from Camarillo appeared to separate from the Tulare samples. The Tulare industrial hemp-planted and postharvest plots also appeared very similar, based on their projected distance on the PrinCoA graph. 

Principal coordinates analysis is a method commonly employed in the analysis of beta diversity, which is the similarity or dissimilarity between samples or groups of samples. The UNIFRAC distance has been shown to be an effective distance metric in simulation studies [18]. The first three axes of the principal coordinates accounted for 42.54% of the variation. Industrial hemp associated soil samples from Tulare tended to be low on Axis 2 and high on Axis 3, whereas the soil samples from Camarillo were high on Axis 2 and high on Axis 1. The postharvest samples from S. Side 6 appear to be like the industrial hemp plots. These samples, six samples from Camarillo and six samples from Tulare will be the focus of the remainder of the metabarcoding analysis, due to this interesting observation and the lack of sufficient replicates for the Pierce farm samples, some of which failed sequencing. 

Analysis of Bacterial Taxonomic Abundance 

To gain further insights into which genera may be performing essential ecological functions at the sites where industrial hemp was grown, Excel pivot table analysis was performed using the methods described in the Pipeline for Undergraduate Microbiome Analysis [15]. The bacterial genera found in the different samples were compared and visualized. The functional classification was based on their putative ecological functions from the literature.  

In terms of potential antibiotic producers, the genera Actinoplanes, Lysobacter, Bacillus, and Pseudomonas were well represented in the Camarillo samples. However, in Tulare samples, Actinomadura, Bacillus, and Streptomyces sp. had the highest abundance of reads. Low counts were also present for Lysinibacillus. The genus Lysinibacillus includes entomopathogenic bacteria that produce antifungal and antibacterial compounds [19]. 

Serratia species such as S. marcescens have been shown to promote growth in wheat seedlings and have a full suite of active nitroreductases [20]. Streptomyces are known for denitrification abilities, which is part of its suite of plant growth promotion characteristics [21]. Vibrio is a pathogen that uses nitrate reduction as a strategy. Vibrio cholerae uses nitrate reduction to determine how much it should grow its population under anaerobic and changing pH conditions[22]. Similarly, Vibrio sp. collected from sediments of Scottish estuaries have been shown to produce nitrite and ammonia from nitrate [23]. Pseudomonas such as P. aeruginosa are able to denitrify soil under anaerobic conditions [24]; a new nitrate-reducing species, Pseudomonas oligotropha, was recently discovered [25]. Achromobacter has denitrifying bacteria such as A. denitrificans, which lives only in aerobic conditions and was isolated from soil [26]. Bacilli such as B. subtilis are valued for their biofertilizer applications, including their ability to reduce nitrogen in agricultural soils, which reduces the loss of nitrate that typically occurs readily following a precipitation event [27]. Nitrogen reducers were represented by Achromobacter, Bacillus, Pseudomonas, Serratia, Streptomyces, and Vibrio in the Camarillo samples. In the Tulare samples, nitrogen-reducing bacteria were represented by Bacillus and Streptomyces.  

In terms of potential plant growth-promoting bacteria and nitrogen fixers, 100% of the Brevibacillus reads identified in the study were in the Camarillo soil samples. Greater than 80% of the Chitinophaga and 80% of the Agromyces reads identified in the study were from Camarillo, as well as more than 80% of the reads for Pseudomonas and greater than 60% of the Massilia sp. reads. Furthermore, more than 70% of the reads for Flavobacterium sp. belonged to the Camarillo samples (Figure 4).  

More than 60% of the reads from Pedobacter were from Tulare. It is interesting because Pedobacter and Flavobacterium are related, and known for antioxidant production but were not found to be equally abundant at both sites. Greater than 55% of Arthrobacter sp. reads were from Tulare. Paenibacillus reads were split evenly between sites; Mesorhizobium reads were also found in similar proportions at both soil sampling sites. Among the remaining suspected nitrogen fixing genera, reads were more prevalent in the Tulare samples for Arthrobacter, Bradyrhizobium, and Devosia. Conversely, Sinorhizobium sequences were only found in Camarillo soil samples, and of the sequences that were assigned to Rhizobium, 60% belonged to samples from Camarillo. 

Potential prokaryotic contributors to essential ecological functions in the industrial hemp rootzone were identified. Although there were differences in the most prevalent species at each site, all communities had players in each of the functional roles studied here: Antibiotic production, nitrogen fixation, denitrification, plant growth promotion, and pesticide and insecticide degradation. Variations in the community may be explained by both the plants grown at the location and their associated rhizosphere, as well as the soil conditions and current cultural practices of agriculture, performed at those specific locations.  

Bacteria from the plant rhizosphere alter plant growth, recruiting or repelling certain species. Soil biota from bulk soil provides the seed bank for the rhizosphere biota, creating a cycle. In this consideration, simply growing hemp or any other plant may act as a bioinoculant, which may modify edaphic soil conditions, which will, in turn, become an inoculant to future rhizospheres. This stresses the importance of not allowing a field to lay fallow for an extended period. As performed in another study [28], future work could involve the collection of bulk soil samples (edaphic soil), rhizosphere samples, and endosphere samples. A bulk soil sample would be useful to serve as a control by holding the fallow condition constant and reflecting the land use and soil factors. Future planned studies should work on the isolation of beneficial microbes that were unclassified to the species level from the organic farm in order to provide more tangible evidence of the claims made in this study. This would also lend itself to the potential discovery of novel soil bacteria. Including fungi in the next round of sequencing using an ITS (internal transcribed spacer) region [29] is also of interest.   

Conclusion  

The main results indicated that there were significant differences in nitrogen, potassium, pH, and TDS between pairs of fields. There was a subtle association between higher N and P levels and CFUs for the 10^-5 dilution. Therefore, the hypothesis was partially supported, particularly for organic matter and phosphorus levels, which were similar across farms. The hypothesis that each environment would have a distinct community was partially supported. The hypothesis that the organic farm would have a higher abundance and more variety of microbes performing these essential functions was supported. A surprising result was that the microbiome of industrial hemp seemed to be more influenced by soil type and cropping method rather than the genetic background of the cultivars. 

This comparative analysis contributes to the literature a novel investigation of soil microbes in the industrial hemp rootzone. The limitations of the study involve the need to have a more rigorous experimental design and allow for more statistical analysis and biochemical testing. The sites with the unique combination of low nitrogen, significant potassium, and high pH have likely shaped a distinctive microbial community. This discovery emphasizes the intricate interplay between soil nutrient levels and microbial adaptations, paving the way for sustainable agricultural practices and microbial-based solutions for soil management. Soil needs to be conditioned, with cover crops and amendments over time to retain their fertility. Periodic monitoring before, during, and after cropping are necessary. In order to draw a strong conclusion about the significance and interaction between soil conditions and cultivar types, a design that tests soils from the same cultivar at multiple sites would be helpful in controlling for some of the variables. 

Acknowledgments This work was supported by training and resources from the National Science Foundation IUSE grant, NSF DUE-1821657, awarded to Bruce Nash and Ray Enke in part to support Community College faculty. This material is based upon work supported by the National Science Foundation under Grant Nos. DBI-0735191, DBI-1265383, and DBI-1743442, which supported iPlant Collaborative and DNA Subway development. Metabarcoding reads are from public project 7929 on Cyverse.org DNA Subway Purple Line. 

Disclosures The authors declare no conflicts of interest.