Minimal NOx emission by Lysinibacillus sphaericus in nutrient poor soil

The aim of this study was to determine whether nitrogen dioxide emissions by Lysinibacillus sphaericus exist in nutrient poor soil. First, we evaluated the presence of two genes involved in denitrification (nosF and nosD) by PCR screening of five strains of L. sphaericus (III (3)7, OT4b.49, OT4b.25, OT4b.31 and CBAM5). We then applied a bacterial consortium made up by L. sphaericus III (3)7 and OT4b.49 into closed microcosms of soil and with minimum salts medium (MSM) supplemented with ammonia to measure the concentration of produced nitrogen dioxide over time. The assays with closed microcosms showed a minimum level of nitrogen dioxide over time. The nosF and nosD primers amplified the expected fragment for the five strains and the sequenced nosF and nosD PCR product showed an ATPase domain and a copper-binding domain respectively, which was consistent with the function of these genes. The basal emission of nitrogen dioxide by L. sphaericus in soil is coupled to its ability to enhance the nitrogen bioavailability for soils deficient in nutrients. Therefore, our results indicate that this microorganism can be considered as a good c and idate to validate the low emission of NOx in field and in the future as an alternative for biofertilization..


Introduction
Nitrogen constitutes one of the most important nutrients for the sustainability of life on Earth. Unfortunately, the balance of this element transformation has been altered by anthropogenic sources, mainly by combustion and chemical fertilization processes (Fowler et al. 2013). One of the fundamental losses of nitrogen to the atmosphere happens through nitrogen oxides ( ). These have adverse effects on the environment due to their implication in the formation of smog, acid rain, global warming and ozone layer depletion (World Health Organization., 2006;Portmann et al., 2012).
The main source of these compounds is anthropogenic as they are produced from fossil fuel combustion. However, bacterial nitrification and denitrification also contribute largely to the emission of nitrogen oxides (Tortoso and Hutchinson., 1990;Kool et al., 2011). It is known that bacterial nitrification is carried out only by autotrophic organisms that belong to β subclasses of Proteobacteria, such as Nitrospira, Nitrobacter and Nitrosococcus spp (Levy-Booth et al., 2014). On the other hand, denitrification by bacteria is performed by a wide range of organisms such as Pseudomonas spp and Paracoccus denitrificans (an aerobic denitrifier), which is the most representative bacteria of the Proteobacteria group (Hayatsu et al., 2008).
However, it has been found that heterotrophic bacilli are also involved in nitrification and denitrification (Verbaendert et al., 2011). (Kim et al., 2005) characterized the presence of Bacillus strains (Bacillus cereus, Bacillus subtilis and Bacillus licheniformis) in these processes for wastewater treatment. Also, there is genomic evidence for genes related to the nitrogen cycle for gram-positive bacilli (Lin et al., 2010). Particularly for Lysinibacillus sphaericus, the presence of genes such as nifU for nitrogen fixation (Hu et al., 2008;Peña-Montenegro and Dussán., 2013), nitrate reductase (Peña-Montenegro et al., 2015) and nosD, nosF, and nosL involved in denitrification (Rey et al., 2016b;Gomez-Garzón et al., 2016) suggests that this microorganism could also be involved in these processes.
It was also found that L. sphaericus can fix nitrogen gas as ammonium and perform in vitro nitrification (Dussán., 2016) where the strains of L. sphaericus III (3)7 and OT4b.49 were the most efficient in this process. Thus, this study aims to evaluate whether L. sphaericus contributes to nitrogen dioxide production when it is added to a nutrient poor soil from the Eastern Llanos basin, Colombia.

Bacterial strains and growth conditions
The five strains of L. sphaericus used in this study are listed in Table 1. They were obtained from the bacteria collection at the Center of Microbiological Research (CIMIC). Nutrient agar was the growth medium used for the microbial consortium (L. sphaericus III (3)7 and OT4b.49), it was incubated at 30°C for 48 hours and used for the closed microcosms assays.
For the partial coding sequence of nosD amplification procedure, the following primers were obtained: nosDf (Forward 92 primer) 5'-TGACACTCAACAGGCAAAGG-3' and nosDr (Reverse primer) 5'-CCATCCATTGACCAAAGCTC-3'.The PCR procedure involved one step of 95°C for one minute, then 30 cycles of 95 °C for 30 s, 49 °C for 30 s, and 72 °C for 30 s, followed by a final step of 72 °C for 5 minutes. The primers used for the amplification of both genes were designed using the sequences annotated from the genomes  of L. sphaericus OT4b.25, III (3)7 and OT4b.49 (Rey et al., 2016a;Gómez-Garzón et al., 2016). For both PCR procedures this strain was used as positive control. The amplification products were visualized in 1.5 % agarose gel at 90V for 70 minutes, purified and sequenced by the DNA Sequencing Laboratory (Universidad de los Andes).

Sequence analysis
Primary sequences were cleaned with CLC Main Workbench and then the nucleotides obtained and their corresponding deduced amino acid sequences were compared against GenBank database using BLASTN from the National Center for Biotechnology Information. We also carried out a conserved domains analysis with the amino acids sequences using Pfam and Interpro (EMBL-EBI).

Closed microcosms assays
The substrates used for these assays were nutrient poor soil (low N and P) from Casanare, Colombia and minimal salts medium (Hartsman et al., 1992) supplemented with 0.5g L -1 of ammonia.
Nitrogen oxides emission measurements were performed using the Williams and Fehsenfeld (1991) chamber method with the following modifications: three treatments of 30 g of mixed soil (1 part remediated: 3 parts of clean soil) with another 3 treatments of 30 mL of minimum salts medium supplemented with ammonia were set up in closed recipients. Following this, a bacterial mix (with 10 9 UFC/g as a final concentration for both strains) or bacterial mix with 0.22 g of fertilizer (NPK) was inoculated into both media. Each treatment was maintained at 30ºC and it was subjected to 2 replicas. Samples of 10 mL of air were taken with a syringe on day 0, 7, 14, and 21. These samples were added into closed recipients with an aqueous solution of sodium hydroxide (pH 9) and then used to indirectly measure the nitric oxide and nitrogen dioxide quantity using SpectroquantNova60A®.
This indirect measurement relies on the aerobic conversion of nitrogen oxide into nitrogen dioxide and oxidation of nitrogen dioxide into nitrite and nitrate according to the following reaction (Kuropka., 2011): Statistical analysis using R project for Statistical Computing was performed using the Kruskal Wallis test (p-value of 0.05).
The amplification results for both denitrification genes showed a band of approximately 300 bp that is consistent with the expected PCR product for all the five strains studied (Figure 1). For nosF and nosD, the sequences obtained showed 99 % identity with the reference genome for L. sphaericus C3-41 (Table 2). The conserved domain analysis for NosF resulted in an ATPase domain, which is consistent with the function that it performs in a NosZ assembly. Furthermore, for NosD, a comparative analysis showed a copper-binding domain for all five sequences according to the domains present for the protein encoded by this gene. Few findings for gene characterization involved in denitrification for Gram-positive bacteria are currently available.  described functional characteristics of Geobacillus thermodenitrificans NG80-2 nosZ cluster revealing that Gram positive and Gram-negative bacteria have conserved the molecular mechanism related to the final step of denitrification. However, further studies on the nosZ sequence for L. sphaericus along with PCR screening and confirmation of gene expression for all the denitrification genes are required. The presence of atypical nosZ cluster does not necessarily indicate denitrification activity as most of bacteria and archaea do not have other denitrification genes (Sanford et al., 2012). It is also necessary to evaluate the activity of these enzymes and their importance related to nitrogen oxides production, as Beaumont et al., (2002) found that the disruption of nitrite reductase in Nitrosomonas europaea is not sufficient to stop nitric and nitrous oxide production. Figure 2 shows that there are significant differences between treatments for day 14, with bigger concentrations for treatments inoculated with the bacterial mix (soil with NPK and MMS) which coincides with the possible biogenic emissions of nitric oxide converted rapidly to nitrogen dioxide by these bacteria (Davidson et al., 2000). As MMS is supplemented with ammonia as nitrogen source, the results obtained indicate that L. sphaericus could perform nitrification and denitrification. Both processes generate nitrogen oxides as intermediates (Hayatsu et al., 2008). The measures of nitrite were not included as the results were similar to the detection level. Although the expected ratio of nitrites and nitrates was equal to the unity (Aoki et al., 1982;Kuropka 2011), the excess nitrogen dioxide led to the oxidation of nitric oxide, and the temperature conditions and volume of the microcosm contributed to the greater concentration of nitrate over nitrite. This is consistent with nitrite formation not being the rate-limiting factor from the oxidation of nitrogen dioxide (Aoki et al., 1982). There are few available studies about nitric oxides emissions in soils and they have found that the ratio between nitric and nitrous oxide production varies from 3 to 10:1 (Smith et al., 1997). Thus, in further studies, we would expect lower emissions of nitrous oxide in this particular soil.
In conclusion, the presence of denitrification genes (nosD and nosF) for five strains of L. sphaericus could be associated with the activity of this microorganism in vitro as our results showed throughout the study. Although the low NOx emissions obtained need to be validated with field trials, they provide important evidence related to the ability of L. sphaericus to perform nitrification and denitrification processes. Thus, this microorganism could be considered as a promising candidate for biofertilization.