Assessment of bacterial diversity during composting of agricultural byproducts

Physicochemical characteristics of compost

The pile and environmental temperatures were monitored during the entire period of composting (Figure 1). Initial temperature of the heap after mixing was 30°C. Within a week, the pile temperature reached to 37°C. However, the temperature increased to 40°C after 15 days and remained the same for four days, thereafter, which it rose to 50°C on 20^th day and remained static for next few days. However, as composting proceeded, the temperature of the pile dropped to 45°C by the 30^th day and fell further, but stabilized at 27°C (near to ambient) by the sixth week. After that, the pile was left uncovered for cooling for the next ten days.

During the present study, the substrates mixtures showed an initial electrical conductivity (EC) of 3.8 dS m^-1. It reached upto 4.9 dS m^-1 with progressive degradation upto 40 days. The pH of the compost heap remained 7.5 during first 30 days of the process, and thereafter it declined to 7.0 and continued till 50^th day.

Chemical characteristics

Total micronutrients

There was a significant increase in nutrients e.g. Na, Cu, Zn, Mg, S, Mn, Fe and Ca during composting. The respective average values of various metal contents varied considerably from the beginning to end of composting (Table 1).

Changes in viable bacterial population during composting

The number of mesophilic bacteria increased rapidly in first ten days, the count of mesophilic bacterial count was 1.7- 2.84 × 10⁹cfu g^-1. However, the thermophilic bacteria were dominant from 11–32 days of composting, with count in between 10⁸ to10⁷cfu g^-1. Finally, mesophilic population stabilized between 10⁶ to 10⁵ cfu g^-1 during the cooling and maturation phase (33–40 days).

Morphological, biochemical and molecular characterization of isolates

The most predominant bacterial isolates were picked up and morphologically different colonies were selected for further studies (Table 2). A total of thirty-three bacteria were subsequently purified and subjected to morphological, biochemical and molecular characterization. Interestingly, 84.8% isolates were Gram-positive, out of which 85.7% were rods and 14.3% cocci, whereas, the remaining 15.2% of the isolates were Gram-negative and all them were rods (Figure 2). The bacterial cultures were tentatively identified on the basis of Bergey’s Manual of Systematic Bacteriology (Table 3).

Medium	Nutrient agar
Laboratory designation/ No.	red	17	H	J, G, BC	32	26	M	QR	D, 30, Q, 21	actin 1, Y, 3	actin 5, actin 2, actin 6	X, IN,8, 38, N	B, A, Q	L, 14, PQ	31, 31 (a)
Shape	circular	round	circular	circular	round	round	round	circular	circular	irregular	circular	circular	irregular	oval	oval
Margin	entire	entire	entire	entire	wavy	spreading	uniformly edged	entire	regular	entire	entire	entire	regular	regular	regular
Elevation	slightly convex	convex	convex	slightly convex	convex	convex	slightly convex	slightly convex	convex	convex	convex	convex	undulate	convex	convex
Texture	smooth	wet	smooth	smooth	smooth	smooth	smooth	smooth	smooth	smooth	smooth	smooth	smooth	smooth	dull
Color with special characters (Pigmentation)	red	grey colored colonies	slightly gummy colonies	white or creamish white	creamish white	slighly brown	white or creamish white	shiny	white	white	white	white	white	creamish white	white
Opacity	translucent	translucent	opaque	opaque	translucent	opaque	translucent	opaque	translucent	opaque	opaque	opaque	translucent	opaque	translucent
Gram’s Staining	(-ve) with 0.5-0.8 μm in length; with rounded ends	(-ve) with 1.0 mm dia; coccoid rods	(-ve) with 0.3 -1.0 μm; straight rods arranged in short chains	(+ve) with 1–2 mm in dia; large cocci	(-ve) with 0.3-1.5 μm in length; slightly curved rods	(-ve) with 0.2-0.7 x 1.0-5.0 μm; rods straight to slightly curved	(+ve) with 1.0–1.5 mm in dia; coccoid occurring in pairs and in tetrads	(+ve) with 1 mm in dia; small slunder rods	(+ve) with 0.7 – 2.5 μm in size; rods	(+ve) with 1–2 mm in dia; rods with short chains ended	(+ve) with 1 mm in dia; rods with short ends	(+ve) rod with short in chains with 1–2 mm in diameter	(+ve) 0.5 -1.0 μm; straight rods arranged in short chains	(+ve) with 1–2 μm; rods with short rounded ended	(+ve) with 1 to 2 μm; short rods

Identification of culturable bacteria isolated from compost

Marked changes in the profiling patterns of bacteria between the initial, mid and final stages of the composting process were observed. The changes in the structure of bacterial community were analyzed on the basis of 16S rRNA gene sequence chronometer from day one to end of composting. The amplified PCR products of bacterial 16S rRNA genes were sequenced partially.

All sequences were compared with 16S rRNA gene sequences present in the Genebank using BLAST and their percentage similarity was also compared and recorded in Table 4. The majority of the bacterial isolates (78.8%) were affiliated with Firmicutes (especially the genera Bacillus sp., Terribacillus sp. and Lysinibacillus sp. etc.), whereas only 9.1, 6.1 and 6.1% of bacterial isolates were affiliated to the members of γ-proteobacteria, β-proteobacteria and actinobacteria, respectively (Figure 3). Apart from spore forming Bacilli other genera in the compost were Staphylococcus, Serratia, Klebsiella, Enterobacter, Microbactrium, Kocuria, Acidovorax and Comamonas.

Interestingly, genera like Kocuria, Microbacterium, Acidovorax and Teribacillus have been reported for the first time from the compost population from agricultural by-products. The heat generated during composting destroyed all pathogenic bacteria in the final mature compost and was found to be free from Staphylococcus, Klebsiella, Enterobacter and Serratia. The phylogenetic affiliation of compost isolates with their accession numbers and their nearest neighbors of the GenBank database are shown in (Figure 4 and Table 4).

Nutrient status of mature compost

The results showed a significant increase in minerals (w w^-1) in agricultural byproducts composting (Table 1) and no gradual fluctuations were observed after 40^th day. Janakiram and Sridevi [28] attempted the composting of Kattamanakku (Jatropha curcas) waste with slurries of cow dung by an aerobic composting method; the percentages of N, P, K, Na, Ca and Mg increased after 30 and 60 days of composting. The findings correlated with the present study. Similarly Felton et al. [29] reported that total P increased during the compost process. The metal concentrations like Cu, Mn and Zn increased rapidly during first 49 days of composting from swine (Susdo mesticus) feces [30]. The stability and solubility of various compounds in compost is influenced by the pH of the compost [31, 32].

Microbial population

Kell et al. [33] studied that at the simplest level, bacteria may be classified into two physiological groups: those that can, and those that cannot readily be grown to detectable levels in vitro. The viable count usually refers to the number of individual organisms in compost that can be grown to a detectable level, in vitro by forming colonies on an agar-based medium. However, the number of viable cells approximates to the number of colony forming units [34]. Changes in bacterial population were analyzed by cultivation-based method (cfu g^-1) to reveal changes in the number of mesophilic and thermophilic bacteria during the composting process.

Hargerty et al. [35] reported that there was maximum increase in microbial population in the early stages of composting which was dependent on initial substrate used and environmental conditions of the composting. High content of degradable organic compound in the initial mixture might have stimulated microbial growth involved in self-heating during initial stage of composting [36]. An equivalent tendency does not occur with regard to mesophilic and thermophilic bacteria in the present study when the population density decreased from 10⁹ to 10⁷ cfu g^-1. However from thermophilic to cooling and maturation phase, the gradual decrease in 10⁷ to 10⁵ cfu g^-1 could be due to the unavailability of nutrients during maturation phase. During peak heating the bacterial populations declined by approximately 10-fold at 40°C and 100-fold at 50°C, followed by population growth at cooling phase, which decreased by 1000 fold as compared to the mesophilic (starting) phase of composting [7]. The Gram-positive bacteria dominated the composting process as they accounted for 84.8% of total population and the remaining 15.2% were Gram-negative as illustrated in Figure 2.

For bacteria, 16S rRNA gene sequence analysis is a widely accepted tool for molecular identification [37, 38]. Franke-Whittle et al. [39] also investigated the microbial communities in compost by using a microarray consisting of oligonucleotide probes targeting variable regions of the 16S rRNA gene. During the present investigation, thirty three bacterial isolates were cultured, out of which twenty six isolates (78.8%) belonged to class firmicutes; two isolates (6.1 %) belonged to actinobacteria; three isolates (9.0 %) belonged to class γ-proteobacteria and the remaining two isolates (6.1%) showed sequence similarity to class β-proteobacteria (Figure 3). Table 4 and Figure 4 summarizes all the bacterial taxa reported in agricultural byproduct compost based on sequence similarity, which were categorized in four main classes: Firmicutes, β-proteobacteria, γ-proteobacteria and actinobacteria in concurrence with the findings of Ntougias et al. [40] and Chandna et al. [41].

The present study determined the microbial succession of the dominating taxa and functional groups of microorganisms, as well as the total bacterial activity during composting of agricultural byproducts, using incubation, isolation, and enumeration techniques. The bacterial population showed differences between mesophilic, thermophilic and maturing stages of compost. Ryckeboer et al. [7] analyzed the bacterial diversity and found that both Gram-positive and Gram-negative bacteria increased during the cooling and maturation phases of biowaste composting in compost bin. In the present study, the level of firmicutes increased markably during mesophilic phase, and then decreased during the next phase upto cooling and maturation. The number of actinobacteria count remained stable during mesophilic and thermophilic phase of composting. Population of β-proteobacteria remained insignificant in thermophilic phase whereas, the level of γ-proteobacteria increased slightly during mesophilic phase and then decreased markably during thermophilic phase. Similarly, Fracchia et al. [6] observed the prevalence of Gram-positive organisms belonging to the firmicutes and actinobacteria.

In the present study a few Serratia, Enterobacter, Klebsiella and Staphylococcus sp. were also isolated during early phase of composting. Silva et al. [42] also found Serratia sp. in bagasse and coast-cross straw during the first stage of composting. Enterobacter sp. was predominantly present at an early stage of composting process and died off at increased temperature [43] in accordance with the present study. Moreover, Enterobacter sp. is common in soil, water and even in compost too and mainly survives as saprophytes [44]. Strauch [45] found that the Klebsiella sp. was present at the beginning of thermophilic phase till the temperature was below 60°C. Similarly, Ahlawat and Vijay [46] also isolated Staphylococcus sp. from mushroom research farm compost at a wider temperature range (43–55°C). Importantly no pathogen could be detected during the curing phase of compost produced from agricultural byproducts. Thus our composting process also resulted in the eradication of pathogens, as has been reported by Danon et al. [47].

Heating is essential to enable the development of a thermophilic population of microorganisms, which is capable of degrading the more recalcitrant compounds, to kill pathogens and weed seeds [48]. Bacillus sp. was able to survive in the compost pile due to their property to form endospores during thermophillic stage. Various researchers investigated that Bacillus sp. was a predominant genera present throughout the composting process [25, 49], and the most dominant bacterial taxon recovered from compost feedstock [50]. Additonally, Kocuria sp. was one of the isolates, cultured from present studied compost. Similarly, Vaz-Moreira et al. [51] also isolated Kocuria palustris from vermicompost from food wastes.

BLAST analysis (http://blast.ncbi.nlm.nih.gov/Blast.cgi) of 16S rRNA gene sequence revealed similarity to sequences of the species Comamonas kerstersii, a β-Proteobacterium of the Comamonadaceae family, as published in GenBank. Young et al. [52] isolated Comamonas sp. from food waste compost. It had the ability to metabolize complex organic compounds as energy sources for growth [53]. Moreover, Comamonadaceae, a new family encompassing the Acidovorans[54], was also recovered from agricultural byproduct compost. Pinel et al. [55] isolated β-proteobacterial Acidovorax sp. symbionts from the nephridia of four different species of earthworms. Pizl and Novokova [56] also showed the establishment of different kinds of relationship between earthworms and microbes. The nephridial symbionts form their own monophyletic group closely related to the genus Acidovorax[57]. The bacteria reduced the biodegradable organic content and help in mineralization of solid waste [58].

Site selection, raw material for composting

During the composting process, the temperature in the pile (5 to 30 cm from the top) was measured daily using a dry bulb thermometer. Similarly, the environment temperature was also recorded during composting near the pile. The samples were collected at every 10^th day for microbial and physicochemical analysis. The composting was terminated after 50 days. The duplicate samples were used to assess the consistency or reproducibility in the method.

Physiochemical analysis of compost

Compost pH and electrical conductivity (EC) were measured by preparing a (1:5 w v^-1 compost: water) mixture as described by Rhoades [59] and Blakemore et al. [60] respectively. The percent organic carbon (C) in the compost was determined by the wet digestion method outlined by Walkley and Black [61]. Total nitrogen (N) was estimated by Kjeldahl method [62] and total sulfur according to the method of Steinbergs [63]. The potassium was estimated by ammonium-acetate method [64]. The samples were analyzed for micronutrient by atomic absorption spectrophotometer (Model 3030, Perkin-Elmer, USA). Macronutrients like calcium (Ca), magnesium (Mg) were determined following the methodology of Moral et al. [65] and sodium (Na) by using the method of Thompson and Wood [66]. The trace metals; copper (Cu), zinc (Zn), iron (Fe) and manganese (Mn) were estimated by ICP-MS (Induced coupled plasma Mass Spectrometer) as per methodology of Koplık et al. [67]; Fingerová and Koplık [68]; Jenn-Hung and Shang-Lien [30], respectively.

Isolation and enumeration of bacteria during composting

Bacteria were isolated from compost by serial dilution method by plating 100 μl of diluted suspension from each phase {the mesophile (30 and 35°C), thermophile (40 and 50°C), maturation and cooling phase (35 and 30°C) samples} were spread plated on nutrient agar (NA) plates. The plates were incubated at 30°C, 35°C, 40°C and 50°C for 24 h. Colonies were counted and populations were expressed in term of cfu g^-1. Morphologically different colonies were purified on NA plates. All isolates and were preserved on slants at 4°C and glycerol stock at -20°C in 20% (v v^-1). All chemicals and media were of molecular grade and procured from either Merck Pvt. Ltd or Himedia, India.

Morphological, biochemical and molecular characterization

Presumptive identification was carried out by colony morphology and use of the first stage diagnostic biochemical tests for Gram-positive and Gram-negative bacteria. Further identification was carried out by standard biochemical tests by using Himedia tests kits (Hi motility™ and Assorted™ Biochemical kit, Hi Carbohydrate™ kit, Hi IMViC™ Biochemical test kit).

Genomic DNA extraction, purification and quantification

Loopful of selected bacterial isolates were streaked and grown on NA plates at their relevant temperature and freshly grown isolates were used to inoculate in 50 ml of Luria-Bertani (LB) broth (Himedia, India). The broth cultures were grown at their respective temperature of the isolates with shaking at 200 rpm till the cultures reached OD₆₀₀ of 0.4-0.5. Thereafter, cells were pelleted by centrifugation at 9167 × g for 10 min at 4°C and washed with TE buffer [10 mM Tris–HCl pH 8.0, 1 mM ethylenediaminetetraacetic acid (EDTA)] and pellets were either frozen (-20°C) for storage or used immediately for genomic DNA extraction by using the method of Sambrook & Russell [69].

DNA samples were quantified by running on agarose gel electrophoresis using 0.8% agarose gel in 1 × tris-boric acid EDTA (TBE) (89 mM tris pH 7.6, 89 mM boric acid, 2 mM EDTA) and visualized by ethidium bromide (0.5 μg ml^-1) staining, to determine DNA size and to assess RNA contamination.

PCR Amplification and sequencing

Amplifications were performed in 50 μl reaction mixture containing 75 ng of template DNA, 1-unit of i-Taq™ polymerase (NEB, UK), 2 mM MgCl₂ (NEB, UK) , 2 μl of 10X PCR buffer, 0.1 mM dNTP (NEB, UK), 100 ng of each forward (8f’:5’-AGAGTTTGATCCTGGCTCAG-3’ [70]), and reverse (1542r’:5′-AAGGAGGTGATCCAGCCGCA-3’ [71]) primers. The amplification was carried out using G-strom thermal cycler (Labtech, UK). Amplification programme consisted of initial cycle of denaturation at 94°C for 5 min, 30 cycles of denaturation at 94°C for 1 min, annealing at 58°C for 1 min, initial extension at 72°C for 1 min 30 sec and final extension at 72°C for 7 min. Amplified products were electrophoresed at 5 Vcm^-1 through 1.5% agarose gel containing 0.5 μg ml^-1 ethidium bromide in 1xTBE electrophoresis buffer with 50 bp DNA Ladder (NEB, UK). The gels were visualized under UV illumination in Gel Documentation system 2000 (Biorad, Hercules CA, USA) and stored as TIFF file format. Sizes of the amplicons were estimated in comparison with 50bp DNA ladder (NEB, UK).

Sequencing of 16S rRNA gene and phylogenetic analysis

The expected DNA band of 1.5 kb was excised from gel and purified using the gel elution kit (Sigma-Aldrich, USA) as per the manufacturer’s protocol. Sequencing reactions were carried out with a BigDye Terminator cycle sequencing kit (Applied Biosystems, USA), standard universal primer forward (8f’) and reverse (1542r’) primer and sequenced by using ABI Prism 3100 genetic analyzer (Applied Biosystems, USA). The sequences thus obtained were assembled and edited using Clone Manager Version 5 (http://www.scied.com/pr_cmbas.htm).

Database search was carried out for similar nucleotide sequences with the BLAST search of Non-reductant (NR) database (http://blast.ncbi.nlm.nih.gov/Blast.cgi) [72]). Partial length 16S rRNA gene sequences of strains closely related to the isolate were retrieved from NCBI for further analysis. For describing their phylogentic relationship, the partial 16S rRNA gene sequences were aligned by using Clustal_X [73]. A phylogenetic tree was constructed by means of neighbor-joining method using MEGA version 5 programme [74].

Nucleotide sequences accession number

The nucleotide sequences of 16S rRNA were obtained and deposited in the GenBank database (EMBL, U.K.) and the accession numbers; AM778178-AM778192, AM884572-AM884579 and FR865468-FR865475 were assigned to their respective sequences.