An Exploratory Comparison of Target-Gene Signals for Selected Bacterial and Fungal Taxa in Agricultural and Forest Soils in Panchkhal, Nepal

The intensification of global agriculture has led to a heavy reliance on agrochemicals, mainly synthetic fertilizers and pesticides, to maximize crop yields and protect plants from biological threats. The use of these chemicals has become a primary input to tackle pests and prevent crop failures, with consumption increasing at an annual rate of 10–20% in Nepal (G​h​i​m​i​r​e​ ​&​ ​G​C​,​ ​2​0​1​8). While these inputs contribute to national food security by reducing harvest losses, their indiscriminate and heavy application has brought about substantial negative impacts on human health, the ecosystem, and soil microbial communities (P​a​u​d​e​l​ ​e​t​ ​a​l​.​,​ ​2​0​2​4). In regions like Panchkhal, Kavre district of Nepal, farmers have reported acute health issues such as respiratory problems, skin rashes, and chest pain, while chronic exposure is associated with potential adverse health effects, including cancer, neurotoxicity, and birth defects (G​C​ ​&​ ​N​e​u​p​a​n​e​,​ ​2​0​1​9; S​a​p​k​o​t​a​ ​e​t​ ​a​l​.​,​ ​2​0​2​0). Environmentally, the overuse of synthetic fertilizers can lead to chemical leaching into groundwater and a fundamental disruption of the soil’s biological and chemical properties (S​r​i​v​a​s​t​a​v​,​ ​2​0​2​0).

Microbial diversity is the cornerstone of soil health, representing the “living skin” of the earth and one of the most complex habitats on the planet. These microorganisms are the primary drivers of biogeochemical cycles, decomposing organic matter to release vital nutrients like nitrogen, phosphorus, and carbon back into the biosphere (O​s​i​n​u​g​a​ ​e​t​ ​a​l​.​,​ ​2​0​2​3). Beyond nutrient cycling, the soil microbiome controls plant immune responses, suppresses pathogens, and increases host resistance to abiotic stresses such as drought. Furthermore, soil microorganisms contribute to soil carbon sequestration through the accumulation of microbial biomass and necromass (S​a​e​-​T​u​n​ ​e​t​ ​a​l​.​,​ ​2​0​2​2). However, the heavy application of agrochemicals can lead to soil dysbiosis, where high levels of synthetic nitrogen and pesticides potentially inhibit beneficial nitrogen-fixing bacteria and fungi, thereby altering microbial abundance and community composition (N​é​m​e​t​h​o​v​á​ ​e​t​ ​a​l​.​,​ ​2​0​2​6).

Maintaining soil health through sustainable agriculture is essential for long-term productivity, as the prolonged overuse of agrochemicals can eventually reduce the efficiency of crop yields (O​s​i​n​u​g​a​ ​e​t​ ​a​l​.​,​ ​2​0​2​3). While traditional soil mapping in Nepal, conducted by the National Soil Science Research Center (NSSRC) under Nepal Agricultural Research Council (NARC), has focused on physico-chemical parameters like pH and nutrient levels, there is an urgent need to prioritize soil biodiversity as a functional, living infrastructure. Understanding differences in soil microbial communities between agricultural and forest sites can provide preliminary evidence to inform future investigations of soil health and agricultural management. Despite this imperative, studies comparing quantitative polymerase chain reaction (qPCR)-derived signals for selected soil microbial taxa between agricultural and forest sites in Nepal remain limited. Although high-intensity farming areas like Panchkhal are known for heavy pesticide consumption (P​a​u​d​e​l​ ​e​t​ ​a​l​.​,​ ​2​0​2​4), molecular evidence describing soil microbial communities in these agroecosystems remains scarce. To our knowledge, no published study has compared target-gene signals for selected bacterial and fungal taxa between an agrochemically managed agricultural field and a forest site in Nepal.

Consequently, this exploratory study aimed to compare qPCR-derived signals for selected bacterial and fungal taxa between one agrochemically managed agricultural field and one nearby forest site in Panchkhal, Nepal. The study employed qPCR to measure target-specific amplification signals and estimate differences between the two soil samples using the Ct-difference approach (2⁻^ΔCt). The findings provide preliminary, site-specific data that may serve as a foundation for future investigations incorporating multiple independent sampling sites, comprehensive soil physicochemical characterization, pesticide residue analyses, and high-resolution microbial community profiling.

Soil samples were collected at a depth of approximately 15 cm below the soil surface at a forest site (27°39'34.9"N, 85°37'05.5"E; elevation 886 m) and an agrochemically treated cauliflower farm (27°38'59.7"N, 85°37'04.8"E; elevation 862 m) covering approximately 0.9 acres of land in Panchkhal, Nepal. Soil sampling was conducted on 25 November 2025, one day after the application of a commercial insecticide, Cora Plus^TM +++ (Emamectin Benzoate 2% suspension concentrate + Indoxacarb 8%), at a concentration of 0.5 mL/L of water, and two fungicides, Bavistin (carbendazim 50% wettable powder) and Indofil M-45 (mancozeb 75% wettable powder), each applied at a concentration of 2 g/L of water, to the agricultural farm.

Soil samples were collected from five randomly selected locations within each sampling site (forest and agricultural field). Each soil subsample was placed in a sterile zip-lock bag, appropriately labeled, transported to the laboratory in an ice box, and processed immediately upon arrival. For each composite sample, 30 g of soil from each of the five subsamples was weighed and thoroughly homogenized to produce a 150 g composite sample. This sampling procedure was repeated independently using a different set of five randomly selected locations, resulting in two separately collected composite samples (within-site sampling replicates) for each site.

Total soil DNA was extracted from 0.2 ± 0.05 g of each composite soil sample using the PureLink™ Microbiome DNA Purification Kit (Invitrogen, Thermo Fisher Scientific India Pvt Ltd., Mumbai, India Cat#A29790), following the manufacturer’s instructions for DNA extraction. The extraction process included homogenization of soil with a bead-beating lysis buffer, removal of inhibitors using a cleanup buffer and centrifugation, binding of DNA to a spin column, washing with appropriate buffers, and final elution of purified DNA. DNA concentrations were measured using a Qubit™ 4 Fluorometer with the Qubit™ 1X dsDNA BR Assay Kit (Invitrogen, Thermo Fisher Scientific India Pvt. Ltd., Mumbai, India; Cat. No. Q33265), following the manufacturer's instructions. Briefly, 2 µL of each DNA sample was mixed with 198 µL of Qubit™ 1X dsDNA BR Working Solution in a Qubit™ assay tube. The mixture was vortexed for 3–5 s and incubated at room temperature for 2 min before fluorescence measurement using the 1X dsDNA BR assay mode. The DNA samples were subsequently diluted 1:10 with molecular biology-grade water and used as templates for endpoint PCR and qPCR.

Endpoint PCR was performed before qPCR to optimize annealing temperatures. A total of five bacterial primer sets targeting the 16S rRNA gene (Table 1) and three fungal primer sets targeting the internal transcribed spacer (ITS) region (Table 2) were selected to detect selected bacterial and fungal targets. Each 50 µL endpoint PCR reaction contained 5.0 µL of 10× iTaq buffer, 1.5 µL of 50 mM MgCl₂, 1.0 µL of 10 mM dNTP mix, 0.25 µL of iTaq DNA polymerase (5 U/µL), forward and reverse primers at final concentrations of 0.3 µM each, and 5 ng of DNA template. Molecular biology-grade water was added to a final reaction volume of 50 µL. Thermal cycling comprised an initial denaturation at 95 °C for 5 min, followed by 40 cycles of denaturation at 95 °C for 15 s, annealing at a gradient from 47 °C to 56 °C for 15 s, and extension at 72 °C for 20 s, with a final extension at 72 °C for 5 min. Amplified products were resolved on 2% agarose gels at 90 V for 60 min and visualized using a gel documentation system to confirm amplicon size and determine the optimal annealing temperature for each primer pair.

The qPCR was performed using all eight primer sets to compare qPCR-derived target-gene signals between the sampled forest and agrochemically treated agricultural soils. The reaction mixture was 20 μl, consisting of 10 µL of 2× iTaq™ Universal SYBR® Green Supermix, 0.6 µL each of 10 µM forward and reverse primers (final concentration of 0.3 µM each), and approximately 5 ng of template DNA, with molecular biology-grade water added to make up the final volume. For each primer set, five reaction tubes were assembled: two replicates for the agrochemically treated soil, two replicates for the forest soil, and one no-template control. Thermal cycling began with a 5-min denaturation step at 95 °C, followed by 40 amplification cycles. Each cycle consisted of 15 s at 95 °C, 15 s at the primer-specific optimized annealing temperature, and 20 s at 72 °C for extension. A melting curve was generated from 65 °C to 95 °C in 0.5 °C increments (5 s per step) to assess amplification specificity.

The qPCR data were analyzed by comparing cycle threshold (Ct) values for each target between agricultural and forest soil samples. For each target, ΔCt was calculated as the difference between the mean Ct values of the two sampled sites: . The estimated fold difference in qPCR-derived target-gene signals was calculated as 2⁻^ΔCt. This calculation assumes approximately 100% amplification efficiency, which was not experimentally verified in the present study. Values greater than 1 indicated higher target-gene signals, whereas values less than 1 indicated lower target-gene signals in agricultural soil relative to forest soil. A similar calculation approach has been used in soil microbial qPCR research (P​e​i​ ​e​t​ ​a​l​.​,​ ​2​0​2​6). Ct values from the two composite soil samples per site were used for exploratory statistical comparisons. Differences between agricultural and forest soil samples were evaluated using Welch’s two-sample t-test, with statistical significance defined as p ≤ 0.05. Because each land-use type was represented by only one sampling site, these comparisons were restricted to the two sampled sites and do not establish generalizable land-use effects.

Annealing temperatures for all primer sets were empirically determined by conducting independent PCR runs across a temperature gradient ranging from 47 °C to 56 °C (1 °C increment). The optimal annealing temperature for each primer pair was defined as the lowest temperature yielding a single, specific amplicon of the expected size with minimal non-specific amplification and primer-dimer formation (Table 3). The primer sets targeting Ascomycota, Basidiomycota, and total fungi showed the best amplification at 50 °C, with reduced non-specific products and primer-dimer bands at this temperature. Similarly, the primer pairs for all bacteria and Alphaproteobacteria demonstrated optimal specificity at 50 °C, while the primer pair targeting Actinobacteria produced discrete bands of the anticipated size (~300 bp) at 53 °C across both agrochemically treated and forest soil samples. Betaproteobacteria and Acidobacteria primers, which possess comparatively lower melting temperatures, required a reduced annealing temperature of 47 °C to obtain specific amplification. Basidiomycota-specific amplification was not detected in the agricultural soil samples at any of the annealing temperatures tested during endpoint PCR, a finding consistent with the subsequent qPCR results. Although residual non-specific amplification and primer-dimer formation could not be eliminated, these were reduced following optimization. Collectively, these optimizations successfully mitigated non-specific amplification and primer-dimer formation, thereby supporting the subsequent qPCR analysis.

The qPCR analysis was employed to compare the qPCR-derived signals of selected bacterial and fungal taxa between soil samples from the agrochemically treated agricultural site and the nearby forest site. Relative differences in target-gene signals were estimated using the 2^⁻ΔCt method, as described in Section 2.5. Values greater than 1 indicate higher target-gene signals, and values less than 1 indicate lower target-gene signals, in the sampled agricultural soil relative to the forest soil. Statistical significance of inter-group differences in mean Ct values was assessed using Welch’s t-test, with significance defined at p ≤ 0.05.

The total bacterial qPCR-derived signal in sampled agricultural soil was slightly lower (0.84-fold) than that in forest soil; this difference was not statistically significant (p = 0.531). Among the selected bacterial taxa, Alphaproteobacteria showed a lower qPCR signal in the agricultural soil (0.57-fold; p = 0.002), as did Acidobacteria, which exhibited the most pronounced decline among the bacterial taxa (0.26-fold; p = 0.008). In contrast, Betaproteobacteria showed a higher qPCR signal in the agrochemically treated soil (1.97-fold; p = 0.001). Although the signal for Actinobacteria was also slightly higher (1.09-fold) in sampled agricultural soil, this difference was not statistically significant between the two sampled sites (p = 0.768). The corresponding results for the selected bacterial taxa are summarized in Table 4.

The selected fungal targets exhibited larger qPCR-derived differences between the two sampled sites than most of the bacterial targets. The total fungal qPCR signal was substantially lower in sampled agricultural soil (0.06-fold; p = 0.007), representing the largest magnitude of change among the targets with quantifiable Ct values. The signal for Ascomycota was also lower in the agricultural soil (0.40-fold; p = 0.0503). Basidiomycota displayed a distinct detection pattern: target-specific amplification was detected in sampled forest soil (mean Ct = 31.20) but was not detected under the applied qPCR conditions in the agricultural soil replicates, precluding quantitative fold-change comparison. The corresponding results for the selected fungal taxa are summarized in Table 5.

The overall differences in qPCR-derived target-gene signals between the sampled agricultural and forest soils are illustrated in Figure 1 using log₂ fold changes and in Figure 2 using mean Ct values across the selected bacterial and fungal taxa.

In this study, qPCR-derived target-gene signals for selected bacterial and fungal taxa were compared between an agrochemically managed agricultural field and a forest site at a similar elevation. The qPCR-derived results indicated lower target-gene signals for total bacteria and total fungi in the agricultural soil samples relative to the forest samples. However, the difference in the total bacterial signal was not statistically significant. Soil microbial communities vary across different land-use types, possibly due to anthropogenic activities and differences in land management (S​u​n​ ​e​t​ ​a​l​.​,​ ​2​0​2​4). In particular, reduced soil tillage intensity has been associated with greater fungal biomass (S​a​e​-​T​u​n​ ​e​t​ ​a​l​.​,​ ​2​0​2​2). The agricultural site investigated in this study was a farmer-managed cauliflower farm where insecticides, fungicides, herbicides, chemical fertilizers, and organic manure were commonly applied, whereas the forest site was undisturbed and had not been subjected to cultivation or agrochemical inputs. These contrasting management practices may have contributed to the differences in qPCR-derived target-gene signals observed between the two sites. However, because the present study did not quantify pesticide residues, soil physicochemical properties, and other environmental variables known to influence microbial communities, it is not possible to attribute the observed differences directly to agrochemical use. Therefore, although the findings are consistent with previous reports that land-use and management practices influence soil microbial communities, further studies incorporating multiple independent sampling sites, soil physicochemical data, and measurements of agrochemical residues are required to identify the specific factors responsible for the observed differences.

The qPCR-derived signal for Acidobacteria was lower (0.26-fold), while that for Actinobacteria was slightly higher (1.09-fold) at the sampled agricultural site compared to the forest site, although the latter difference was not statistically significant. This is consistent with findings from previous studies. X​u​ ​e​t​ ​a​l​.​ ​(​2​0​2​5​) found that Acidobacteria and Proteobacteria had higher relative abundances in forest lands, while Actinobacteria had higher abundance in croplands. Acidobacteria are known to be oligotrophic bacteria and can respond to changes in nutrient availability and soil pH (L​l​a​d​ó​ ​e​t​ ​a​l​.​,​ ​2​0​1​7). The abundance of Actinobacteria is associated with land-use practices and soil moisture conditions (S​u​n​ ​e​t​ ​a​l​.​,​ ​2​0​2​4). A meta-analysis of tropical land-use conversion studies reported decreases in the relative abundances of Acidobacteria and Proteobacteria and an increase in Actinobacteria in agricultural soils relative to forest soils (P​e​t​e​r​s​e​n​ ​e​t​ ​a​l​.​,​ ​2​0​1​9). Although our results showed similar patterns, the observed differences cannot be causally attributed to land-use type or agrochemical use, as this study compared only one agricultural site with one forest site, and soil physicochemical properties, vegetation, management history, and agrochemical exposure were not systematically characterized or controlled.

The qPCR-derived signal for Alphaproteobacteria was higher in the sampled forest soil, whereas that for Betaproteobacteria was higher in the sampled agricultural soil. Previous research has shown that fertilization practices can be associated with shifts in bacterial community composition (F​r​a​n​c​i​o​l​i​ ​e​t​ ​a​l​.​,​ ​2​0​1​6). However, because soil nutrient availability and microbial functional traits were not assessed in this study, these ecological interpretations should be regarded as literature-based hypotheses rather than mechanisms demonstrated by the present data.

Lower qPCR-derived signals for total fungi and Ascomycota were observed in sampled agricultural soil. Notably, Basidiomycota-specific amplification was not detected in the agricultural samples under the applied qPCR conditions. Substantial reductions in Basidiomycota following the conversion of forest land to agricultural land and under intensive agricultural management have been reported (N​a​v​a​r​r​o​-​N​o​y​a​ ​e​t​ ​a​l​.​,​ ​2​0​2​1). Information obtained from the farmer in this study confirmed that fungicides formed part of the routine crop management practices in the sampled agricultural field. However, fungal community composition is known to be influenced by multiple factors, including vegetation type, land-use practices, soil physicochemical properties, and agrochemical use (B​o​d​d​y​ ​e​t​ ​a​l​.​,​ ​2​0​0​7; W​i​g​h​t​w​i​c​k​ ​e​t​ ​a​l​.​,​ ​2​0​1​0). Therefore, although fungicide application may have contributed to the observed non-detection of Basidiomycota in the agricultural samples, this relationship cannot be confirmed by the present study because pesticide residues, soil physicochemical properties, and other management variables were not evaluated.

This research provides preliminary evidence of differences in qPCR-derived target-gene signals between the sampled agrochemically managed agricultural field and nearby forest site in the Panchkhal region of Nepal. However, the findings should be interpreted with some limitations. The study was conducted at only one site for each land-use type, and subsamples were pooled into two composite samples per site, which may have obscured within-site spatial heterogeneity. The absence of independent site-level replication also limits inference about land-use types. Furthermore, pesticide residues, microbial physiological activity, and soil physicochemical properties (including pH, moisture content, soil organic carbon, total nitrogen, available phosphorus, electrical conductivity, and soil texture) were not measured in the present study. Consequently, the mechanisms underlying the observed differences could not be directly determined. The use of taxon-specific qPCR limited the analysis to selected microbial groups and did not capture the full diversity or functional potential of the soil microbiome, and the possibility of residual PCR inhibition cannot be completely excluded. Because standard-curve slopes, amplification efficiencies, coefficients of determination (R²), analytical sensitivity, inhibition assessments, and extraction controls were not assessed during the present study, our results should be interpreted as exploratory comparisons of qPCR-derived target-gene signals between the two sampled sites rather than validated estimates of absolute gene-copy numbers or soil microbial abundance.

This exploratory study identified differences in qPCR-derived target-gene signals for selected bacterial and fungal taxa between one agrochemically managed agricultural site and one nearby forest site in Panchkhal, Nepal. Within the sampled sites, the forest soil showed higher qPCR-derived signals for total fungi, Alphaproteobacteria, Acidobacteria, and Ascomycota, while Basidiomycota-specific amplification was detected only in the forest samples. The agricultural soil showed higher qPCR-derived signals for Betaproteobacteria and Actinobacteria, although the latter difference was not statistically significant. Because the study was based on a single agricultural site and a single forest site, these findings should be interpreted as site-specific observations rather than representative of agricultural and forest soils throughout Panchkhal or Nepal. Furthermore, the observational design and the absence of measurements of pesticide residues, soil physicochemical properties, and other environmental variables preclude conclusions regarding the specific effects of agrochemical management on the observed qPCR-derived signals. Future studies involving multiple independent sampling sites, comprehensive environmental measurements, and high-throughput sequencing approaches are needed to distinguish the effects of land-use, agrochemical management, and other environmental factors on soil microbial communities.