Investigation of the acclimatization period: example of the microbial aerobic degradation of volatile organic compounds (VOCs)

Sandrine Bayle Luc Malhautier (corresponding author) Jean-Louis Fanlo Laboratoire Génie de L’Environnement Industriel, Ecole des Mines d’Alès, 6 Avenue de Clavières, Alès Cedex 30319, France E-mail: luc.malhautier@ema.fr Valérie Degrange Laboratoire Ecologie Microbienne du Sol, Université Claude Bernard Lyon 1, 43 boulevard du 11 novembre 1918, Villeurbanne Cedex 69622, France Jean-Jacques Godon Laboratoire de Biotechnologie de l’Environnement, Institut National de la Recherche Agronomique, Avenue des Etangs, Narbonne 11100, France The aim of this study is to better evaluate the occurrence of an acclimatization-enrichment period, defined as a selection period of consortia having the capability to biodegrade pollutants. In order to perform this evaluation, two experimental strategies were carried out and the results were studied carefully. Two laboratory-scale reactors were inoculated with activated sludge from an urban treatment plant. During the experiment, these reactors were supplied with a gaseous effluent containing VOCs. For both reactors, the composition is different. Three parameters were monitored to characterize the microflora: bacterial activities, bacterial densities, and the genetic structure of Bacteria and Eukarya domains (Single Strand Conformation Polymorphism fingerprint). The obtained results showed that the resultant biodegradation functions were equivalent. The bacterial community structure differs even if six co-migrated peaks were observed. These data suggest that the microbial communities in both reactors were altered differently in response to the treatment but developed a similar capacity to remove VOCs at the issue of this period. Furthermore, it is suggested that the experimental strategies developed in this work lead to an enrichment in terms of functionality and microbial diversity almost equivalent.


INTRODUCTION
In ecology, "acclimatization" is defined as the habituation of an organism's physiological response to environmental conditions (usually applied to laboratory environments) (Townsend et al. 2002).Nevertheless, this term has been widely used and, in particular areas, is commonly applied to both the acclimatization and the enrichment steps.The enrichment period corresponds to the selection, from a complex ecosystem, of microorganisms that are already present and are capable of growth and carrying out specific reactions (Madigan et al. 1997).Few authors have studied the acclimatization period, dealing with the degradation of chemicals.Linkfield et al. (1989) characterized the lag period before anaerobic dehalogenation of halobenzoates.
The lengthy acclimatization period appears to represent an enzyme induction phase in which little or no aryl dehalogenation is observed, followed by an exponential increase in activity typical of an enrichment response.Lewis et al. (1986) determined adaptation lag periods for the microbial transformation of p-cresol and showed that adaptation lag periods for the microbial transformation of low concentrations of chemicals may correlate with limiting nutrient concentrations.doi: 10.2166/wst.2009.653For other authors, the distinction between acclimatization and enrichment remains unclear.Pessione et al. (1996) showed that a strain of Acinetobacter radioresistens was able to utilize phenol as the only carbon and energy source, after an acclimatization period of 3 days in which increasing phenol concentrations from 50 to 200 mg L 21 were supplied.Moreno-Andrade & Buitron (2004) studied the variation of microbial activity during acclimatization to 4-chlorophenol (4CP) in an aerobic automated sequencing batch reactor.The results show a reduction in degradation time as the acclimatization process occurred and this reduction is linked to the initial concentration of 4CP.
The authors also showed that, as acclimation took place, the capability of the consortia to biodegrade the toxic increased.Rhine et al. (2003) evaluated microbial community responses to atrazine exposure and nutrient availability.
The authors showed that repeated pesticide exposure may enhance biodegradation through selective enrichment of pesticide-metabolizing microorganisms, particularly when the compound is used as a carbon and energy source.For the treatment of gaseous effluents containing Volatile Organic Compounds (VOCs), an acclimatization period may be carried out before inoculating the bioreactor.It consists in forcing a gaseous effluent containing the VOCs through the suspension of activated sludge.This favors the selection of microbial communities with the ability to remove VOCs (Devinny et al. 1999).Tellez et al. (2002) evaluated the operational performance and petroleumhydrocarbon-removal efficiency of an activated sludge treatment system from oilfield-produced water.The authors initiated an aggressive acclimatization procedure to build the microbial concentration to levels that would achieve effective hydrocarbon removal efficiencies.
It remains difficult to distinguish acclimatization from enrichment and furthermore to know whether a complex mix of acclimatization and enrichment or enrichment only occurs.The aim of this study is therefore to better evaluate the occurrence of this acclimatizationenrichment period.In order to perform this evaluation, two different experimental strategies have been performed and the results have been studied carefully.We considered the biological degradation of chemicals because the results obtained would provide answers to the question.Moreover, microbial biodegradation is one of the most important factors affecting the concentration of chemicals in nature.
In order to check whether acclimatization-enrichment period occurs, the performance of two systems was studied by using two reactors filled with an identical activated sludge suspension but using two different feeding methods.One reactor was supplied with an effluent containing a complex VOC mixture (oxygenated, aromatic and halogenated compounds) for which the concentration of oxygenated compounds varied with incubation time.The other was supplied with an effluent gradually enriched with chlorinated, aromatic and oxygenated compounds.As VOCs concentrations increase, a complex mix of acclimatization to higher concentrations and the selection for bacteria able to grow faster can be suggested.Microbial communities were considered, and more specifically biodegradation activities and genetic structuring.
The mixtures of VOCs were prepared by mixing liquid masses of the different compounds.

Reactor design
Two pilot-scale units, called "C" and "S", were used in this study (Figure 1).The liquid VOC mixture was continuously

Composition of the inlet gas
The gaseous influent composition is given in Table 1.Two periods need to be distinguished.
From the beginning of the experiment to t ¼ 66 days (Period 1), the composition of the polluted gas supplying each reactor was different.The gaseous influent supplying reactor S contained firstly chlorinated compounds only (from the beginning to t ¼ 37 days), then a mixture of chlorinated and aromatic compounds from t ¼ 39 days to t ¼ 66 days (Table 1).
The concentrations of aromatic and chlorinated compounds were 2.1 and 1.4 g m 23 respectively.During period 1, reactor C was supplied with a gaseous influent containing all the chemical groups of VOCs, that is to say oxygenated, aromatic and chlorinated compounds.The concentrations of aromatic and chlorinated compounds were 2.1 and 1.4 g m 23 respectively.The concentration of oxygenated compounds decreased as follows: It was 4.2 g m 23 from the beginning to t ¼ 37 days (0.7 g m 23 for each compound), then 2.35 g m 23 from t ¼ 38 days to t ¼ 66 days (0.4 g m 23 for each compound).The composition of the oxygenated compounds mixture supplied to the reactors through the experiments is the same.
During the second period (period 2), both reactors were supplied with a polluted gas of identical composition (Table 1).The gaseous influent contained a complex mixture of chlorinated, aromatic and oxygenated compounds from t ¼ 67 days to the end of the experiment.The concentrations of aromatic and chlorinated compounds were 2.1 and 1.4 g m 23 respectively.The concentration of oxygenated compounds was 0.84 g m 23 from t ¼ 67 days to t ¼ 82 days (0.14 g m 23 for each compound), then 2.1 g m 23 from t ¼ 83 to t ¼ 96 days (0.35 g m 23 for each compound) and finally 4.2 g m 23 from t ¼ 97 to t ¼ 117 days (0.7 g m 23 for each compound).The composition of the oxygenated compounds mixture supplied to the reactors through the experiments was the same.The composition of gaseous effluent was modified after removal efficiency had been maintained for a minimal period of three days.

Biodegradation monitoring
The biodegradation of VOCs was measured daily by monitoring the concentration of each compound at the sampling inlets and outlets.The reactor sampling ports were directly connected, via Teflon tubes, to a gas chromatograph unit (HP 6890, Hewlett Packard) equipped with a flame ionization detector.A 30-metre HP-1 capillary column was used with a carrier gas (Helium) flow rate of 2.5 mL.min 21 .
The gas sample was injected into the column via an automatic gas sampling valve (250 mL): the gas samples were thus not stored.The temperature of injector and detector were 150 and 2208C respectively.For the detector, the air flow rate was 310 ml min 21 , the hydrogen flow rate was 46.5 ml min 21 and the helium flow rate was 30 ml min 21 .The temperature ramp was as follows.At the beginning, the temperature of the oven was 408C for one minute, then it was increased from 40 to 908C at 158C per min.The temperature was then maintained stable at 908C for 4 min and finally increased to 1508C at 108C per min.The error in the VOC concentration measurement was evaluated using an empty reactor.The pollutant concentration was determined 9 times, enabling the Relative Standard Deviation (RSD) to be estimated at 10%.
A volume of sample (2 ml) was diluted in 9 volumes of 1% sodium hexametaphosphate, and homogenized using a blender (Ultra Turax, T25 basic, IKA) for 2 min at 19,000 rpm.Then, 1mL of this suspension was incubated with 3.7% formaldehyde for 30 min, before counter-staining by DAPI at a final concentration of 20 mg mL 21 for 1 h, in the dark, at an agitation speed of 200 rpm.
Stained bacteria were recovered on a 0.2 mm-polycarbonate membrane filter (Millipore GTBP, Ireland) by microfiltration.The filters were then mounted on microscope slides in Mounting Medium (Sigma, USA) and observed using an epifluorescence microscope (DMLB, Leica, Germany) equipped with a blue excitation filter (BP 340 -380 nm) and an LP 425 barrier filter.For each slide, 30 fields were counted and the Relative Standard Deviation (RSD) was calculated for each count.

Analysis of total DNA by PCR-SSCP
The DNA extraction method has been described in a previous work (Godon et al. 1997).The target DNA amplified was the V3 region of the 16S rDNA gene for the bacteria domain (primers W104 and W049) (Table 2), The SSCP profiles were compared after alignment with the internal weight standard.
The SSCP profiles were first analyzed by visual comparison, then using the SSCP2 software developed in the Environmental Biotechnology Laboratory, Narbonne, France.This software determines and aligns peaks; so co-migrated peaks were recognized.

RESULTS
The microbial aerobic degradation of a complex mixture of VOCs was investigated.Two bioreactors were used and inoculated with activated sludge coming from an urban wastewater treatment plant.Pollutant removal, bacterial densities and the dynamics of microbial diversity were determined for increasing incubation time.The occurrence of an acclimatization-enrichment period was evaluated (from t ¼ 67 to t ¼ 117 day).

Pollutant removal
The removal efficiency for VOCs at t ¼ 66 days is summarized in Table 3.
For reactor S, the removal efficiency of ethylbenzene was 70%, the removal efficiency of toluene and p-xylene was 40%.For reactor C, esters were completely eliminated, the removal efficiency of MEK was 50%, methanol, acetone and MIBK were not eliminated.From t ¼ 70 days to the end of the experiment (Figure 2), it was observed that the VOC removal efficiency became equivalent for the two reactors.
Chlorinated compounds were not eliminated in either reactor.Aromatics were degraded only in reactor S, and their abatement decreased progressively from 50% at t ¼ 66 days to 0% at t ¼ 82 days when the concentration of oxygenated compounds was increased from 0 to 140 mg m 23 .The removal of oxygenated compounds against increasing incubation period is shown in Table 4.     4).

Densities and diversity
The total bacterial densities increased from (3.1 ^0.7) £ 10 8 cells mL 21 at t ¼ 0 day to (1.9 ^0.3) £ 10 9 cells mL 21 for reactor C and (6.2 ^0.5) £ 10 8 cells mL 21 for reactor S at t ¼ 66 days (Figure 3).Between t ¼ 66 and t ¼ 117 days, the bacterial densities were stable in both reactors, with (1.9 ^0.3) £ 10 9 cells mL 21 for reactor C and (7.0 ^0.5) £ 10 8 cells mL 21 for reactor S. The microbial community structure was observed for the Bacteria and Eukarya phylogenetic domains using SSCP analysis.
Figure 3 shows the bacterial community structure against increasing incubation time.The diversity of the total bacterial community coming from activated sludge at the beginning of the experiment (t ¼ 0) shows a wide range.
SSCP profile analysis reveals 25 peaks.Between t ¼ 0 to t ¼ 67 days, the number of peaks decreases.For reactor C, there are some peaks on the left of the pattern (approximately 10 peaks, of which one is dominant) while for reactor S, the peaks are located on the right of the pattern (approximately 7 peaks, of which one is dominant).
For reactor C, between t ¼ 67 and t ¼ 117 days, the patterns do not show any major variations with increasing incubation time.The dominant peaks are located on the left of the pattern throughout the duration of the experiment.
For reactor S, between t ¼ 67 and t ¼ 117 days, the patterns were modified with increasing incubation time.The peaks are firstly largely located on the right of the profile.
Nevertheless, other peaks located on the left are also observed.At t ¼ 114 days, 2 peaks are dominant, and are located at the pattern ends.The average elimination efficiency (%) is indicated; NR: Not removed.The diversity is lower for Eukarya than for Bacteria.
Only six peaks are present at the beginning of the experiment (Figure 5).pollutants, such as cattle manure (Chachkhiani et al. 2004) or aromatic compounds (Stoffels et al. 1998).From t ¼ 67 to t ¼ 114 days, the obtained results (Table 4) show that the resulting biodegradation functions become similar.In both cases, the microbial community could not simultaneously remove all the compounds.The obtained results showed that the operating strategies applied did not lead to the selection of a bacterial community with enhanced biodegradation functions.Moreover, it was shown competitive phenomena between substrates and that oxygenated compounds were used first.Furthermore, the performance of both reactors subsequently became equivalent, whichever the strategy used.While the degradation efficiencies were equivalent for both reactors, the results obtained show that the bacterial community structure differs.Hence, the experimental strategies used lead to the same enrichment in terms of functionality, but different enrichments in terms of microbial diversity.Some authors have reported that injected and vaporized into a filtered air stream operating at 10 L min 21 via a Precidor syringe pump (Infors AG, Switzerland).Some of the polluted air (1 L min 21 ) was forced through a 5 L glass reactor seeded with 4 L of activated sludge.The Empty Bed Residence Time was 4 min.The composition of the activated sludge was 4.5 g of chemical oxygen demand L 21 , 1.1 g of biological oxygen demand L 21 , and 5.2 g of total suspended solids L 21 .Activated sludge came from the domestic sewage treatment plant of Saint Christol lez Ale `s (France) (6,500 populationequivalent).The loading was as follows: 860 mg of chemical oxygen demand L 21 day 21 , 340 mg of biological oxygen demand L 21 day 21 , and 380 mg of total suspended solids L 21 day 21 .The Volatile Suspended Solids (VSS) were around 80%.The sludge age was around 10 days.The volume of sludge was maintained constant by the daily addition of a mineral salt nutrient solution, HCMM3, as developed by Juteau et al. (1999).The experiment was performed at normal room temperature (20 -258C) for 117 days.Neutral pH was daily maintained using a NaOH solution (1 M).
The biodegradation was monitored by calculating the VOC removal efficiency (RE) as follows:RE ¼ Ci 2 Co Ci £100RE: Removal Efficiency (%); Ci: Inlet concentration of VOCs (g m 23 ); Co: Outlet concentration of VOCs (g m 23 ).
by the E. coli sequence.
The removal efficiency measured for ketones, esters and methanol differs.The esters were completely removed.The removal efficiency for other molecules decreased.Acetone and MIBK were poorly eliminated from the 83rd to the 117th day period.The removal efficiency of methanol and MEK respectively ranged from 35 to 49 and 100 to 36% for reactor C and 75 to 57 and 75 to 31% for reactor S. Hence, the evolution of the removal of oxygenated compounds was generally similar for both reactors (Table

Figure 2 |
Figure 2 | VOCs removal efficiency (%) reached for reactors S and C from the 66th to the 117th day.V Reactor C and O Reactor S.

Figure 3 |
Figure 3 | Bacterial community dynamics for reactors S and C: total bacteria densities and genetic structure of total bacterial communities.X-axis: time; Y-axis: peak area.
For reactor C, at t ¼ 40 days, the diversity is modified and the dominant peak is located further on the left than observed at t ¼ 0 day.Between t ¼ 71 and t ¼ 117 days, the patterns are similar.At t ¼ 114 days, six peaks are counted, as on the t ¼ 40 days profile.The new peak B becomes dominant from t ¼ 39 days.For reactor S, the profiles are different for each analysis.The number of peaks varies and increases from two (at t ¼ 40 days) and three (at t ¼ 58 days) to ten peaks at t ¼ 114 days.Peaks X and P are present throughout the duration of the experiment.They dominate the pattern (Figure 5).Superposition of the profiles at the end of the experiment shows distinct bacterial communities in the two reactors.DISCUSSION Two experimental strategies were carefully applied in the laboratory in order to better assess whether acclimatization-enrichment period occurred.The results of these experiments were analyzed to check for the occurrence of acclimatization-enrichment.Firstly, at t ¼ 64 days, it was observed that the VOC mixture led to a decrease in bacterial diversity.This phenomenon has already been observed in soil and sediment studies after the microflora has been contaminated with pollutants such as mercury (Rasmussen & Sørensen 2001), fuel (Stephen et al. 1999) or polycyclic aromatic hydrocarbons (Langworthy et al. 1998), and during the start-up phase of bioreactors supplied with different

Figure 4 |
Figure 4 | Comparison of the bacterial communities at t ¼ 65 and at t ¼ 114 days for reactors S and C : co-migrated peaks.X-axis: time; Y-axis: peak area.

Figure 5 |
Figure 5 | Evolution of the Eukarya domain diversity with increasing incubation time in reactors S and C. ND: no amplification of 16S DNA was obtained for the considered day.X-axis: time; Y-axis: peak area.

Table 1 |
Composition of the polluted gas with increasing incubation time gene) and 1 mL of purified template DNA adjusted to a total volume of 50 mL.

Table 2 |
Primers used to amplify the 16S and 18S rDNA and their characteristics

Table 4 |
Oxygenated compound removal efficiency (%) with incubation time for t ¼ 66 to t ¼ 117 days in both reactors