Introduction

Soil microbiome, a small but highly dynamic living organic matter pool, plays a pivotal role in nutrient cycling and thus it is important in affecting soil fertility (Schloter et al. 2018). Due to its nature, soil microbiome quickly responds to biotic and abiotic factors, thus becoming a sensitive indicator of most disturbances and changes occurring in the soil ecosystem (Laudicina et al. 2012).

Accurate and rapid methods, such as the chloroform (CHCl3) fumigation-incubation (CFI) and the CHCl3 fumigation-extraction (CFE) have been used to measure the size of soil microbiome, i.e. the soil microbial biomass C (SMBC) (Jenkinson and Powlson 1976; Vance et al. 1987). Both methods involve the fumigation of soil for 24 h by CHCl3 vapours, but CHCl3, an alkyl halide, is toxic to both humans and the environment (Lionte 2010). Furthermore, the 24 h CHCl3 fumigation does not lyse all soil microbial cells. Badalucco et al. (1990, 1992) demonstrated that amounts of phenol-reactive C and anthrone-reactive C, i.e. ascribable to sugars, represented higher proportions of total extractable C after than before CHCl3 fumigation, suggesting that some non-biomass sugars were solubilized during the 24 h fumigation. In addition, since the CO2 produced during the 24 h CHCl3 fumigation in 8 different soils ranged from 20 to 85% of that biotically produced (respired) in the absence of CHCl3, Badalucco et al. (1997) questioned that the percentage of microbial biomass surviving chloroform exposure, or partially lysed, was not negligible. The efficiency of CHCl3 in lysing microbial cells depends on soil texture, due to its low diffusion in clayey soils (Badalucco et al. 1997). However, even in sandy loam soils, not all soil microbial cells were lysed, as shown by Toyota et al. (1996) reporting that approximately 10% of bacterial colony forming units survived a 5-day chloroform fumigation. Likely, the exopolysaccharides secreted by bacteria might have protected bacterial cells from the action of chloroform vapours. Later, Haubensak et al. (2002), investigating the reliability of the CFE method in organic and mineral soil horizons of 11 mature forests, demonstrated that different fumigation conditions (CHCl3 exposure from 1 to 10 days at field moist or field capacity water content) resulted in different C/N ratios of the organic compounds made extractable by CHCl3 fumigation from soil. Moreover, Alessi et al. (2011) demonstrated that a significant amount of chloroform was adsorbed by the clay fraction of soil and not removed by increasing the chloroform evacuation time or subjecting the fumigated soil to high N2 pressures. It was suggested that the CFE method should be corrected for the soil clay content. Finally, Rotbart et al. (2020) showed that the residual (i.e. non-evacuated) chloroform concentrations in the fumigated soils correlated positively with soil organic matter content. Therefore, the assessment of soil microbial biomass through CHCl3 fumigation methods, based on prior disruption of microbial cells for the subsequent extraction (CFE) or mineralization (CFI) of intracellular matter still shows some problems.

In this study, we tested a new technique for lysing soil microbial cells by the high gas pressurization of soil. The gas can penetrate and fill soil microbial cells, and when the applied gas pressure is suddenly released, the absorbed gas can rapidly expand inside the cells, causing the mechanical rupture of the cell, like a popped balloon (Fig. S1). Likely, the enhanced concentration of gas might cause an increase in fluidity of the lipid double-layer of the microbial membranes, thus increasing their permeability to gas, which can rapidly diffuse into microbial cells (Spilimbergo and Bertucco 2003). Moreover, an intensive localized cooling due to the Joule–Thomson effect, when pressurized gas is expanded, might also play a role in cell lysis during the decompression (Garcia-Gonzalez et al. 2007). This technique, which was first reported by Fraser (1951), and then implemented, involves the sterilization/inactivation of microbial cultures in liquid medium or heat-sensitive foods, such as milk or jams (Nakamura et al. 1994; Garcia-Gonzalez et al. 2007), but it has never applied in soil.

The aim of this study is to test an alternative approach to CHCl3 fumigation for lysing soil microbial cells by exposing the soil to gases at high pressure, followed by rapid depressurization via the gas release. Different combinations of gas pressure, type of gas used, and time of pressurization were tested because these factors affect the penetration of the gas inside the cells. The lysis of soil microbial cells by high pressurization with CO2 or N2, here called CO2HP or N2HP, respectively, was determined by measuring either the amount of CO2 released during ten days of soil incubation, or the organic C extracted by a KCl solution after soil pressurization, and these values were compared with those obtained by CFI and CFE methods, respectively.

Materials and methods

Soils and experimental setup

Two soils, one agricultural (AGR) and one forest (FOR), with different physico-chemical properties, mainly with contrasting organic matter amounts, were collected to assess the response of CO2HP and N2HP approaches with variable soil types. Both soils were collected in Sicily (Italy) by a spade: the AGR soil was surface (0–20 cm) collected at Camporeale, PA province, (37°54′53.6" N 13°04′47.9" E, Typic Haploxerert, vineyard) after discarding plant residues and stones from the surface, in February 2020; the FOR soil was collected at San Fratello, ME province, (38°02′12.6" N 14°36′27.1" E, Typic Haploxerept, oak forest) at 5–20 cm in order to exclude L/O horizons, in July 2020. About 30 kg were sampled for each soil. Immediately after their collection, soils were sieved (Ø < 2 mm), stored at room temperature and their chemical and physical properties determined (Table 1). A soil aliquot was pre-incubated at 25 °C and at 50% of the water holding capacity (WHC) for 7 days, in order to stabilize the biological activity, before the lysis of soil microbial cells by CHCl3 fumigation, CO2HP or N2HP methods. Three distinct and independent experiments were carried out: Experiment I in November 2020; Experiment II in January 2021; Experiment III in February 2021.

Table 1 Main soil properties
Full size table

CO2 and N2 high pressurization (CO2HP and N2HP)

The pressurization experiments were conducted using the Cells Disruption Bomb equipment (Parr Instrument Company, USA), which is a non-stirred steel pressure vessel with an internal volume of 1850 mL and a manometer, located in the upper part, giving internal pressure values up to 20.7 Mpa (Figure S2).

Four aliquots of 15 g for each pre-incubated soil and for each pressurization treatment (see later) were put into 50 mL plastic Falcon tubes, which were placed without caps in the steel vessel (it can contain up to 12 tubes per pressurization cycle). The vessel was tightly closed and incubated at around 40 °C, as recommended by the manufacturer to allow the spatially homogeneous gas diffusion. In order to evaluate the best cell lysis by the high pressurization method, the gas pressure ranged from 2.76 to 5.51 MPa and the duration from 2 to 32 h (Table 2). At the end of each pressurization treatment, the outlet valve of the vessel was immediately opened. The complete depressurization took from 20 to 40 s, depending on the pressure x duration combination, as verified by the manometer connected to the vessel.

Table 2 Framework of the three experiments performed to compare CHCl3-fumigation based methods (CFI and CFE) with the high pressurization method using N2 or CO2 gases (N2HP and CO2HP)
Full size table

CHCl3 fumigation

After AGR and FOR soils pre-incubation, CHCl3 fumigation was performed according to Jenkinson and Powlson (1976). Four soil aliquots (each 15 g) were put inside glass beakers and fumigated with CHCl3 (stabilized with amylene), in a sealed desiccator (40 cm diameter) for 24 h at 22 °C in the dark. Then, the fumigant was removed by a vacuum pump by ten cycles of suction (2 min each) and air re-introduction (1 min each).

Determination of organic C released before (control) and after CHCl3 fumigation, CO2HP or N2HP

In the Experiment I, soil organic C released following CHCl3 fumigation, CO2HP or N2HP was determined similarly to the CFI method (Jenkinson and Powlson 1976), i.e. by measuring the evolved CO2 during the 10 days incubation. Evolved CO2 was also determined from the untreated soil (control). Then, in the Experiments II and III, which only differed for the combination of CO2 high pressures and durations, the released organic C, which was extracted with 2 M KCl, was determined similarly to the CFE (Vance et al. 1987). This salt was preferred to K2SO4 since the extracted organic C (see later) was determined by elemental analysis (Murage and Voroney 2007).

Soil incubation was carried out in 125 mL glass jars sealed with rubber stoppers holding silicone septa and placed in the dark at 22° C for 10 days. The CO2 accumulated after 1, 3, 7 and 10 days of incubation in the headspace of the glass jars was determined by injecting 1 mL of air from each jar into a gas chromatograph (TraceGC, Thermo Fisher Scientific, Milan, Italy) equipped with a thermal conductivity detector and a Poropak Q column, using He as the carrier (oven 50 °C, injector at 225 °C, column flow 40 mL/min, split mode 10:1). At each CO2 sampling, jars were ventilated with fresh air for 30 min and then sealed again, after adding distilled water to adjust soil moisture, if needed. Soil extraction was carried out with a sample weight:extractant volume (g/mL) ratio of 1:4. Soil suspensions were horizontally shaken for 45 min at 70 rpm. At the end of shaking, the suspensions were filtered with Whatman No. 42 paper and then organic C was determined by a TOC-L elemental analyzer (Shimadzu Italia, Milano).

Statistical analysis

The results reported are the means with standard deviations of 4 pseudo-replicates (n = 4) and are expressed on dry soil weight basis at 105 °C. For each of the three described experiments, data were subjected to one-way ANOVA (with replication), with the only factor being the specific treatments applied time by time, i.e. control, fumigation (CFI or CFE), CO2HP at various durations, N2HP at various durations. Before performing the ANOVA, the normal distribution and homogeneity of variance of the data were checked by Kolmogorov-Smirnoff and Levene's goodness-of-fit tests, respectively. Significant differences at p < 0.05 among various treatments and within each of the three independent experiments were assessed by the Least Significant Difference test (LSD). Statistical analysis was performed using RStudio 2022.02.3 + 492 "Prairie Trillium" (libraries tidyverse and emmeans).

Results and discussion

Comparison between high pressurization (4.13 MPa) and CFI method in lysing soil microbial cells (Experiment I)

The cumulative CO2-C (Cflux) evolved during 10 days of incubation from AGR soil was lower than from FOR soil for the same treatment (Fig. 1), likely due the lower content in KCl-extractable C (Fig. 2), which may be considered as a proxy of available C (Laudicina et al. 2013). Indeed, the increase in CO2 cumulated between 7 and 10 days of incubation of the AGR soil, regardless of treatment, was smaller than the correspondent increase present in the FOR soil, likely because the microbial available C was nearly exhausted in the AGR soil, while it was still present even after 10 days in FOR soil (Fig. 1). In addition, the Cflux cumulated throughout the whole 10-d incubation for all three AGR soil N2HP treatments did not significantly differ from the control treatment, while in N2HP treated FOR soils the Cflux was higher than in control soil by 23.4%, 41.8% and 62.4% after 16, 24 and 32 h of N2 pressurization, respectively (Tables 3, 4, and 5; Fig. 1). These results could depend on: 1) N2 gas pressurization at 4.13 MPa for the duration time from 16 to 32 h was ineffective in the AGR soil or scarcely efficient in the FOR soil in lysing microbial cells because none or little extra-flush of organic C was respired in comparison with control; 2) the N2 gas partly remained inside soil pores even after depressurization, thus avoiding the oxygen penetration into soil pores with the result of slowing down microbial respiration. The nitrogen gas, being a not polar molecule, is poorly soluble in water (its hydrosolubility is 19 mg/L at 20 °C and 0.1 MPa pressure) (Garcia-Gonzalez et al. 2007) and might not have reached living microbial cells inhabiting water films surrounding soil particles. In addition, the lower MBC of AGR than FOR soil (Table 1) may also explain the different effect of N2HP in lysing microbial cells of the two soils.

Fig. 1
figure 1

Cumulative CO2 emitted from AGR and FOR soils during a 10-days incubation at 25 °C and 50% of WHC, subjected to constant high pressure (4.13 MPa) for 16, 24, and 32 h with molecular nitrogen (N2, N2HP) or carbon dioxide (CO2, CO2HP). NF, not fumigated (control); F, CHCl3 fumigated. Values represent means ± SD (n = 4)

Full size image
Fig. 2
figure 2

KCl-extractable C from AGR and FOR soils subjected to various high CO2  pressures (CO2HP, 2.76, 3.44, 4.13, 4.82 and 5.51 MPa) for 4 or 24 h. NF, not fumigated (control); F, CHCl3 fumigated. Values represent means ± SD (n = 4). Different lowercase letters indicate significant differences at p < 0.05

Full size image
Table 3 One-way ANOVA for the three experiments using AGR and FOR soils
Full size table
Table 4 Mean, standard deviation (n = 4) and variation coefficient values of CO2-C cumulated during 10-d incubation (Exp I-CFI) or values of KCl-extractable C (Exp II-CFE and Exp III-CFE) for AGR soil following the framework reported in Table 2
Full size table
Table 5 Mean, standard deviation (n = 4) and variation coefficient values of CO2-C cumulated during 10-d incubation (Exp I-CFI) or values of KCl-extractable C (Exp II-CFE and Exp III-CFE) for FOR soil following the framework reported in Table 2
Full size table

The 10-d Cflux of the AGR and FOR CHCl3 fumigated soils was higher than values of the respective unfumigated controls as it increased by 87.7 mg kg−1 and 403.4 mg CO2-C kg−1 (Tables 3, 4, and 5; Fig. 1), respectively. These increases in Cflux were likely due to the microorganisms surviving the chloroform fumigation, which respired the cytoplasmic materials released from lysed microbial cells (Jenkinson and Powlson 1976).

The cumulative 10-d Cflux from both AGR and FOR soils under CO2HP at 4.13 MPa and any pressurization duration was higher than the values of CHCl3 fumigated soils (p < 0.05). Likely, the CO2HP was more effective than CHCl3 fumigation in lysing microbial cells, as shown by the higher mineralization of cytoplasmatic material by the surviving microorganisms. Moreover, the Cflux did not increase with the duration of CO2 pressurization in AGR soil, whereas the Cflux of the FOR soil was significantly (p < 0.05) linearly related to the duration (Tables 3, 4, and 5; Fig. 1). The CO2HP method, compared to CFI method, increased the Cflux of the AGR soil by 62.4%, as an average, whereas the Cflux of the FOR soil increased by 13.5%, 26.3% and 48.5% with CO2HP durations of 16 h, 24 h and 32 h, respectively. Therefore, these results suggest that pressurized CO2 was likely effective in lysing soil microbial cells. The CO2 can cross microbial membranes because it is a polar molecule (Garcia-Gonzalez et al. 2007). The CO2 solubility in water at 20 °C and 0.1 MPa pressure is 1690 mg/L, i.e. nearly ninety times higher than that of N2.

By subtracting the Cflux of the untreated control from either 10-d C flux of the CHCl3 fumigated soil or CO2HP treated soil, and applying the Jenkinson's relationship (1988) for estimating MBC by the CFI method \(\left[\mathrm{MBC}=\left({\mathrm{CO}}_{2}-{\mathrm{C}}_{\mathrm{flux \ Fumigated}/\mathrm{CO}2\mathrm{HP}}\mathrm { \ minus }{\mathrm{ \ CO}}_{2}-{\mathrm{C}}_{\mathrm{flux \ Control}}\right)/0.45\right]\), MBC value of the CO2HP treated AGR soil at 4.13 MPa and for 16 h was almost twice the value of the CHCl3 fumigation-incubation method, whereas the estimated MBC value of FOR soil at 4.13 MPa CO2 pressure increased with the duration and was up to 75.2% higher than with CFI method (Tables 4 and 5).

It may be speculated that the evolved CO2 flush of the CO2HP treated soils can also have an abiotic origin, instead of totally deriving from the mineralization of lysed microbial cells. The CO2 dissolved in water, the quantity of which depends on its partial pressure, reacts with the water molecule to form carbonic acid, which is a weak acid, according to the reaction

$${\mathrm{CO}}_2\left(\mathrm{aq}\right)+{\mathrm H}_2\mathrm O\rightleftharpoons{\mathrm H}_2{\mathrm{CO}}_3,\mathrm{with}\;\left[{\mathrm H}_2{\mathrm{CO}}_3\right]/{\left[{\mathrm{CO}}_2\right]}_{\mathrm{aq}}=1.7\times10^{-3}\;\mathrm{mol}/\mathrm L\left(25{}^\circ\mathrm C\right),$$

i.e. only one molecule of CO2 dissolved in water every 588 (Garcia-Gonzalez et al. 2007). Moreover, the CO2 dissolved in water was likely lowered during the high pressurization because the vessel was incubated at 40 °C and the gas hydrosolubility decreases by increasing temperature. On the other hand, the CO2 deriving from the following multiple equilibria \({\mathrm{CO}}_{2}\rightleftharpoons {\mathrm{H}}_{2}{\mathrm{CO}}_{3}\rightleftharpoons {{\mathrm{HCO}}_{3}}^{-}\rightleftharpoons \hspace{0.17em}{{\mathrm{CO}}_{3}}^{2-}\) was negligible because: a) both soils had a pH close to neutrality and thus a significant percentage of the formed carbonic acid immediately dissociated into bicarbonate; b) the CO2 derived by the above multiple equilibria was likely dispersed during the depressurization phase and before the soil incubation.

Comparison between the CO2 high pressurization (CO2HP) and the CFE method in lysing soil microbial cells (Experiments II and III)

As mentioned above, the C flux produced during incubation of CO2HP soils could be partially derived from abiotic processes or from the CO2 trapped in the soil pores even after depressurization, and then slowly released during the 10 days of soil incubation. In order to verify this potential drawback of the method, we performed a second experiment (Experiment II), comparing the CO2HP with the CFE method, thus determining the KCl-extracted organic C (Cextr) produced after CHCl3 fumigation or after CO2 pressurization/depressurization (Tables 3, 4, and 5).

The Cextr of the AGR soil pressurized with CO2 at five different high pressure values (from 2.76 to 5.51 MPa) for 4 h was not significantly different compared to the Cextr of the CHCl3 fumigated soil, thus suggesting that at these conditions the amount of lysed microbial cells was similar to the CHCl3 fumigation. On the contrary, the Cextr by the same five high pressurizations but lasting for 24 h, gave an increase of 33.5%, as an average, compared to the fumigated soil (Table 4; Fig. 2). The Cextr fluxes of the FOR soil after pressurization for 4 h were always significantly lower than the values of CHCl3 fumigated soil, although they increased linearly from 2.76 up to 4.82 MPa, whereas the further pressure increases to 5.51 MPa did not enhance the Cextr (Table 5; Fig. 2). Nevertheless, the CO2 pressurization from 2.76 to 5.51 MPa for 24 h gave Cextr values always higher than Cextr by chloroform fumigation. The highest Cextr value occurred again at 4.82 MPa, although it was not significantly different from those observed at 4.13 and 5.51 MPa; the three high pressurizations ranging from 4.13 to 5.51 MPa for 24 h increased the Cextr of 29.6%, as an average, compared to the fumigated soil (Table 5; Fig. 2). Likely, the optimal high pressure x duration combination for the CO2HP method to be used for organic C-rich soils with high MBC should be equal or higher than 4.13 MPa for 24 h.

By subtracting the Cextr of the untreated control from either the Cextr of CHCl3 fumigated soil or the CO2HP treated soil and applying the relationship by Vance et al. (1987) for estimating MBC with CFE method [MBC = (Cextr Fumigated/CO2HP minus Cextr Control) × 2.64], the 24 h pressurization at 4.82 MPa of the AGR soil gave an 80% higher value than that of the CHCl3 fumigation (Table 4). The estimate of MBC of the FOR soil subjected to the CO2HP at 4.82 MPa for 24 h was 43% higher than the value of the CFE (Table 5).

In order to find the optimal combination of CO2 pressure x duration for increasing microbial cell lysis (Tables 3, 4, and 5), we carried out a third experiment (Exp III) with AGR and FOR soils by testing two CO2 pressures (3.44 and 4.13 MPa) and seven durations (from 2 to 32 h, Fig. 3). The AGR soil, with a low MBC estimated by CFE method, showed that the pressurization at 3.44 MPa for 2 h gave a value of MBC comparable to that of CFE method (Tab. 4; Fig. 3). Prolonging the duration from 2 to 8 h slightly increased the Cextr, which was 23.6% higher than that of CFE method. However, the Cextr of the CO2HP at 4.13 MPa for 16 h duration was equal to that of 8 h duration, while the 24 and 32 h durations at 4.13 MPa increased the Cextr by 46% and 92%, compared to that at 16 h duration, respectively (Tab. 4; Fig. 3). Likely, it seems that with low soil MBC the pressurization duration plays a more important role in lysing microbial cells than the value of the CO2 pressure.

Fig. 3
figure 3

KCl-extractable C from AGR and FOR soils subjected to two constant high CO2 pressures (CO2HP, 3.44 or 4.13 MPa) for 2, 4, 6, 8, 16, 24 or 32 h. NF, not fumigated (control); F, CHCl3 fumigated. Values represent means ± SD (n = 4). Different lowercase letters indicate significant differences at p < 0.05

Full size image

Both Experiments II and III using FOR soil, i.e. with the highest MBC, showed that the durations up to 8 h and at 3.44 MPa of CO2 lysed less microbial cells than CFE method; the difference between the two methods decreased by increasing the duration (Table 5; Fig. 3). However, at 4.13 MPa pressure for 16 h duration, the Cextr was significantly higher by 17.8% than that of CFE method, while the increase at 24 and 32 h was higher by 32.7% and 63.9%, respectively. Likely, in soils with high MBC, both CO2 pressure level and its duration affected the microbial cells lysis. It may be speculated that in soils with high MBC there is a broad spectrum of cells differing in their resistance to lysing agents. This may explain the linear progressive increase in Cextr of the FOR soil by increasing both pressurization duration and applied CO2 pressure (Table 5; Fig. 3).

The following two hypotheses are proposed considering our results: 1) the multitude of soil microbial species shows different resistances to lysing agents; likely by increasing the MBC and the biodiversity, the cell resistance also increases; 2) the highest values of CO2HP pressure and duration might have lysed the quiescent or encysted microbial cells and/or made KCl-extractable some additional soil organic C not ascribable to living organic matter.

Conclusions

Our proposed physical approach based on high pressurization by CO2 (CO2HP) was likely more effective than CHCl3 fumigation in lysing soil microbial cells of two soils differing in organic matter content and MBC values. The soil with higher organic matter content and microbial biomass required higher CO2 pressures and longer pressurization periods than the soil with lower organic matter. Among the tested combinations, the most successful was CO2 pressure at 4.13 MPa for 32 h. However, our results do not exclude that the proposed CO2HP method may release some additional soil organic C not ascribable to living organic matter. Therefore, further studies are needed using several soils with different chemical and physical properties. Future research should also analyze the composition of organic C extracted after the CO2HP treatment to ascertain its microbial origin. Finally, the proposed method should be tested with soils added with known abundances of microbial cells labeled with 13C and/or 15N. Microbial diversity of soil before and after the CO2HP treatment should be also tested. These studies can give insights into the C and/or N made extractable by the CO2HP and if it derives from soil microbiome. These studies may also be the basis to set up a new method to determine the soil microbial biomass.