Detection of Changes in Soil Microbial Community Physiological Profiles in Relation to Forest Types and Presence of Antibiotics Using BIOLOG EcoPlate

Abstract

Soil is home to microbiota with diverse metabolic activities. These microorganisms play vital roles in many ecological processes. Thus, the assessment of microbial functional diversity is an important quality indicator of soil ecosystems. In this study, we collected soil samples from three distinct forest habitats, i.e., an agroforest, a primary forest (PF), and a secondary forest, within the Angat Watershed Reservation in Bulacan, Northern Philippines. Community-level physiological profiling (CLPP) was done with the BIOLOG EcoPlate™ to analyze the responses of the soil microbial communities from the three forest habitats in the absence or presence of antibiotics. The BIOLOG EcoPlate represents 31 utilizable carbon sources. Based on the CLPP analysis, soil samples from the PF showed significantly higher utilization of most carbon sources than the other forest types (p < 0.05). Thus, less disturbed forest types constitute more functionally diverse microbial communities. The presence of antibiotics significantly decreased the carbon utilization patterns of the soil microbial communities (p < 0.05), indicating the possible use of CLPP in monitoring contamination in soil.

Introduction

Soil microorganisms perform many ecological and physicochemical processes such as soil structure formation, nutrient cycling, and improvement of plant health, thus making them an important indicator of soil health [1, 2]. A single gram of soil may contain up to billions of bacterial cells and an estimated diversity of between 4000 and 5000 species [3]. The microbial diversity in soil affects its functional diversity [4]. Traditionally, selective plating and direct viable counts are used to evaluate microbial communities in soil. However, many soil microbes remained unculturable as laboratory culture media may fail to mimic their natural growth environment due to a lack of information regarding factors such as multiplication period, appropriate temperature, and nutrient requirement [5–7]. Biochemical-based techniques may therefore be advantageous to measure changes in microbial diversity, particularly in response to pollutants or hazardous chemicals, and thus can be easily and routinely employed as a monitoring tool. An example of biochemical-based techniques is the evaluation of sole carbon source utilization patterns, also called community-level physiological profiling. In this approach, large amounts of information regarding mixed microbial community function and functional adaptations over space and time can be easily obtained. Currently, CLPP is done almost exclusively with BIOLOG™ microplates. The BIOLOG EcoPlate has been successfully used in different environmental research, e.g., microbial community analysis of soil including the rhizosphere, water, wastewater, sludge, compost, and industrial waste [4]. It also has an application in agricultural studies, particularly in the evaluation of soil health as good quality soil was shown to have higher enzyme activity as compared with dystric soils [2]. The effect of organic matter amendment on the rhizosphere microbial community and root-infecting pathogen of rice was also determined through CLPP [8]. Microbial communities can be compared between the O horizon and the uppermost mineral layers of mine soils afforested with different tree species [9] or at different altitudinal mountain gradients [10]. The effects of organic matter removal on microbial communities in a temperate deciduous forest were also explored with this method [11]. Other studies simply used CLPP to characterize soil microbial community profiles in red mud-affected soils [12], human-impacted soils of ancient settlements [13], and in a shrubland ecosystem exposed to experimental fire [14]. The method was optimized for the study of functional diversity of soil microbial communities [15]. Therefore, in this study, we used BIOLOG EcoPlate to measure changes in microbial community profiles and functions across different forest types and in response to simulated antibiotic contamination. One of the recognized drivers of increased antimicrobial resistance is the continuous use of antibiotics in agricultural practices which could find their way to the soil. Antibiotics have been shown to impact microbial metabolic activities in soils [16–18]. Thus, it is important to monitor the presence of antibiotic contamination in soil samples and this can also be easily done with the use of BIOLOG EcoPlate to check any changes in microbial community physiological profiles.

We selected three forest types for the study: a primary forest, a secondary forest, and an agroforest within the Angat Watershed Reservation. The primary forest is mainly a lowland dipterocarp forest with zero disturbance. The secondary forest was part of the rehabilitation of Angat Forests in which trees were planted around 1–3 years ago prior to soil collection and anthropogenic activities such as slash-and-burn, and illegal logging were previously recorded. The agroforest has woody perennials and planted fruit-bearing trees with livestock in the area being taken care of by the local communities. In each forest type, three sampling points were randomly chosen, within which a 1 × 1-m quadrant was set up for the collection of five soil samples per sampling point and at a depth of 0–15 cm. The litter layer was initially removed prior to soil collection with a hand shovel. The collected five soil samples were then pooled in a plastic bag as composite soil samples per sampling point, three soil samples per forest type (n = 9). The composite soil samples were initially sieved with a two-mm-sized sieve to remove any visible plant roots, stones, litter, and other debris. Finally, the physiological profiles expressed as sole carbon source utilization by the soil microbial communities were evaluated with the BIOLOG EcoPlates (Biolog Inc., Hayward, CA, USA). The BIOLOG EcoPlate system has 31 different carbon sources, plus a blank well, in three replications, and with substrates subdivided into six groups: carbohydrates (n = 7), carboxylic acids (n = 9), amines and amides (n = 2), amino acids (n = 6), polymers (n = 4), and miscellaneous (n = 3). These compounds occur naturally in the soil environment while some are products or exudates of plant roots [19]. One gram of composite soil samples from each of the three forest types was suspended in a flask containing 99 mL sterile water, followed by hand shaking for 20 min at room temperature. The soil suspension was left to settle for 30 min at 4 °C. Then, 120 µL of the soil suspension was inoculated into each well of the BIOLOG EcoPlates. To check if CLPP can be used to detect changes in soil microbial communities in response to antibiotic contamination, 90 µL of soil suspension was added to each well, in which 30 µL of the antibiotic cocktail containing streptomycin (10 µL, Sigma-Aldrich S6501), chlortetracycline (10 µL, Sigma-Aldrich T7660), and penicillin (10 µL, Sigma-Aldrich 13752) at 10 mg/ml concentration was added. These antibiotics and their derivatives have been used in agriculture [20] or in other CLPP studies [17, 18] while a similar concentration was also used in the study of da Silva et al. [21]. All EcoPlates were incubated at 28 °C for 168 h and the absorbance at 590 nm was measured every 24 h using FLUOstar Omega Microplate Reader. For the CLPP analysis, readings at 72 h were chosen as the most suitable, representing the optimum optical density. The average well color development (AWCD) after 72 h of incubation was calculated for each plate as a mean of the optical densities (OD590) from the 31 wells as similarly described in Garlan [22]. The absorbance of each well was corrected by the subtraction of the optical density (OD590) of the well-containing water. The OD590 = 0.25 was assumed as a threshold value, below which a substrate was considered unmetabolized. One-way analysis of variance (ANOVA) and the post-hoc Tukey’s HSD (honest significant difference) test at p ≤ 0.05 were performed to evaluate significant differences in AWCD among soil types. T-test at p ≤ 0.05 determines any significant difference between the carbon utilization with and without antibiotics.

The AWCD index for all soil samples had the best variation at 72 h of incubation, but clear differences could be observed between soil samples treated with or without antibiotics (Fig. 1). The added cocktails of three antibiotics had significantly decreased the AWCD (p < 0.05), thereby indicating the impact of any antibiotic contamination on the soil microbial communities. The combined application of the antibiotics used in this study, i.e., streptomycin, chlortetracycline, and penicillin at 10 mg/mL concentrations, has so far not been studied for CLPP as most studies used single antibiotic [17, 18] and in combination with other treatments [16, 21], but the results follow the observation with other antibiotics. Pinna et al. [16] revealed that the addition of 53.6 μg/g sulfamethazine to the manured soil samples had a significant negative impact on readily culturable bacteria and its potential metabolic activity via BIOLOG CLPP. The AWCD of soil microbial communities taken from manured soils, regardless of the presence of sulfamethazine, remained significantly similar at day 1 of incubation but were distinct after 7 days. In the study of Xu et al. [23], high concentration of combined sulfadiazine (SDZ) and copper (Cu) significantly reduced carbon utilization rates and values of Shannon Index (p < 0.05) in a 28-day incubation period, while in low concentrations, its inhibiting effect was observed only after 14 days.

Fig. 1.

AWCD of the soil samples in the presence and absence of antibiotics after 24, 48, 72, up to 168 h of incubation using the BIOLOG EcoPlate assay

However, when the AWCD values of the soil microbial communities for each replication were averaged and statistically compared, we saw no significant differences between the different forest types (p > 0.05; Fig. 2), although we observed a decreasing trend from primary forest to agro-forest to secondary forest. Similarly, when the analysis was done per substrate group, we also did not see any significant differences between the forest types.

Fig. 2.

The mean AWCD values of the soil microbial communities (A) and per substrate group (B) between the three forest types: agroforest (AF), secondary forest (SF), and primary forest (PF)

Interestingly, when we compared the utilization of the individual substrate types, we observed some differences between forest types. For example, we observed better utilization of putrescine, phenylethylamine, arginine, asparagine, and serine by microbial communities in the untouched primary forest soil as compared with the disturbed agro- and secondary forest soils (Fig. 3). In the study of Sang et al. [24], six major bacterial genera found in the grasshopper shrimp paste were found to be positively correlated with the production of biogenic amines, including putrescine and phenylethylamine. Beaumont et al. [25] also stated the addition of some amino acids, including arginine, asparagine, and serine, to the gut microbiome of mice increased the abundance of beneficial bacteria. These studies showed the ready utilization of these substrates by microbial communities as shown in this study. Our findings also suggest the selection of amides, amines and the six amino acids as potential substrates for monitoring any changes in microbial communities in different soil environments.

Fig. 3.

Utilization of amine and amide and six amino acids by the soil microbial communities in different forest types

In virgin brigalow soils, microbial communities had a higher utilization of carbohydrates as compared to the cultivated soils [26]. The same pattern was observed in this study. Utilization of β-methyl glucoside, mannitol, N-acetyl-D-glucosamine, and cellobiose were significantly better in PF than the other forest types (Fig. 4). These carbohydrate compounds can promote the growth of microbial communities in a variety of habitats. For example, β-glucans have been shown to promote the growth of good bacteria [27], while the study of Groisilier et al. [28] identified mannitol as an important source of carbon for marine heterotrophic bacteria. N-acetyl-D-glucosamine was found to enhance tomato plant growth by promoting microbial growth found in tomato root area [29]. In the study of Ucar et al. [30], it was mentioned that cellobiose can be broken down into glucose subunits by several microbes.

Fig. 4.

Utilization of carbohydrates and carboxylic acids by the soil microbial communities of the different forest types. AF agroforest, SF secondary forest, PF primary forest. Tukey’s Test, p < 0.05

High utilization of carboxylic acids may suggest that these substrates are also important energy sources for soil microbes [31]. The carboxylic acids utilized by the microbial communities of the soil samples from primary forest that were significantly different from the other forest types are galacturonic acid, 4-hydroxy-benzoic acid, γ-hydroxy-butyric acid, itaconic acid and D-malic acid (p < 0.05) as shown in Fig. 4. These carboxylic acids are considered organic acids. Adeleke et al. [32] observed that organic acids help poorly soluble nutrients more soluble and mobile to promote microbial growth.

The utilization of polymers and miscellaneous carbon sources in different forest types are also shown in Fig. 5. No significant difference between forest types was observed in all miscellaneous carbon sources. However, for polymers, bacterial communities of the soil samples from AF utilized α-cyclodextrin significantly higher than the samples from the other forest types (p < 0.05) and microbial communities of the soil samples from PF utilized Tween 40 and Tween 80 significantly higher than the samples from the other forest types (p < 0.05). In the study of Zhu et al. [33], the application of α-cyclodextrin, along with other cyclodextrins, has shown a remodeled gut microbiota, thereby showing the utilization of this substrates by microorganisms. Tween 80 is also added to microbial growing culture media as it provides bacterial cells with an external supply of oleic acid, the lipophilic component of Tween 80 [34].

Fig. 5.

Utilization of polymers and miscellaneous carbon sources by the soil microbial communities of the different forest types. AF agroforest, SF secondary forest, PF primary forest. Tukey’s Test, p < 0.05

The CLPP approach assumes that a functionally diverse microbial community will be able to utilize a wide range of structurally different organic molecules at more similar rates than a less functionally diverse community [35]. In this study, samples from primary forest showed significantly higher utilization of most carbon sources, e.g., phenylethylamine, arginine, asparagine, serine, hydroxy-butyric acid, cellobiose, mannitol, β-methyl-D-glucoside, N-acetyl-D-glucosamine, and Tween 80. This suggested higher functionality of the soil microbial communities present in PF as compared to the other forest types. The metabolic activity of soil microbial communities can be associated with the composition and bioavailability of soil organic matter [36]. Hence, utilizing the CLPP method in understanding the microbiota in different environments could be a good basis to monitor their responses to environmental changes and could possibly be a good indicator of soil health. For example, CLLP method has been standardized to estimate bacterial functional diversity as a function of soil fertility. This was then applied to evaluate a Mediterranean olive orchard managed by different sustainable agricultural practices [37] or to compare between organic and conventional crop production systems [38]. Furthermore, CLLP can evaluate crop cover and fertilization regimes as farming practices [39, 40]. BIOLOG EcoPlate method can therefore be routinely used to study the variability of physiological profiling of the microbial communities from different forest types including those soil affected by pollutants and other contaminants [41, 42], as statistically significant results were obtained in this study. We also suggest the validation of this observation by comparing different habitat types or ecoregions with or without chemical pollutants. In conclusion, CLPP can differentiate soil microbial functional diversity in different forest types and in soil contaminated with antibiotic pollutants. A combined treatment of different antibiotics as opposed to a single application can drastically impact the physiological profiles of the microbial communities in soil.