Biostimulation by methanol enables the methylotrophic yeasts Hansenula polymorpha and Trichosporon sp. to reveal high formaldehyde biodegradation potential as well as to adapt to this toxic pollutant
- First Online:
- Cite this article as:
- Kaszycki, P., Walski, T., Hachicho, N. et al. Appl Microbiol Biotechnol (2013) 97: 5555. doi:10.1007/s00253-013-4796-y
- 308 Views
The methylotrophic yeasts Hansenula polymorpha and Trichosporon sp. revealed enhanced biodegradation capability of exogenously applied formaldehyde (Fd) upon biostimulation achieved by the presence of methanol, as compared to glucose. Upon growth on either of the above substrates, the strains proved to produce the activity of glutathione-dependent formaldehyde dehydrogenase—the enzyme known to control the biooxidative step of Fd detoxification. However, in the absence of methanol, the yeasts’ tolerance to Fd was decreased, and the elevated sensitivity was especially pronounced for Trichosporon sp. Both strains responded to the methanol and/or Fd treatment by increasing their unsaturation index (UI) at xenobiotic levels below minimal inhibitory concentrations. This indicated that the UI changes effected from the de novo synthesis of (poly) unsaturated fatty acids carried out by viable cells. It is concluded that the yeast cell response to Fd intoxication involves stress reaction at the level of membranes. Fluidization of the lipid bilayer as promoted by methanol is suggested as a significant adaptive mechanism increasing the overall fitness enabling to cope with the formaldehyde xenobiotic via biodegradative pathway of C1-compound metabolism.
KeywordsMethylotrophic yeastHansenula polymorphaTrichosporon sp.Membrane adaptationFatty acid unsaturationNonconventional yeastFormaldehyde biodegradation
Among the nonconventional yeasts, methylotrophic species have focused much attention regarding their wide biotechnological use (Hartner and Glieder 2006; Yurimoto et al. 2011). These yeasts are able to metabolize single-carbonic (C1) compounds employing the enzymatic pathway reactions of methanol utilization (Hartner and Glieder 2006; Gleeson and Sudbery 1988; van der Klei et al. 2006) and can serve as convenient systems for heterologous gene expression as well as favorable models for studying cell physiology, mechanisms of gene induction, repression, and cell stress reactions. In particular, some of the methylotrophs, namely Hansenula polymorpha and Trichosporon sp. proved highly tolerant to exogenous methanol and formaldehyde (Fd), both known as environmentally hazardous xenobiotics. For that reason, both strains have great application potential and were proposed as good biotechnological tools capable of bioselective environmetal monitoring and efficient biodegradation of industrial pollutants (Glancer-Soljan et al. 2001; Gonchar et al. 2002; Kaszycki and Koloczek 2000; Kaszycki et al. 2001, 2006; Khlupova et al. 2007; Sigawi et al. 2011).
The methylotrophic pathway reactions, as described for H. polymorpha and several other species, are based on complex enzymatic interactions involving distinct cellular compartments (van der Klei et al. 2006; Yurimoto et al. 2011). Methanol utilization undergoes subtle control which enables to coordinate important mechanisms and processes such as catabolite repression, derepression, and induction as well as other cell regulatory systems like peroxisome proliferation, glutathione and NADH metabolism, membrane transport, and others (Gleeson and Sudbery 1988; Sibirny et al. 1988). Methanol, a substrate for growth, serves as a key inducer of most of the enzymes involved in methylotrophy. Formaldehyde, in turn, is usually regarded as a central and toxic intermediate in methylotrophic conversions and at this point the methanol metabolism is branched into assimilatory and dissimilatory reactions (Yurimoto et al. 2005).
The efficiency of a biotechnological process is often limited by microbial tolerance to stress (Auesukaree et al. 2009). Especially in environmental biotechnological applications, the contact with toxic chemical stressors requires microorganism adaptation which determines successful biodegradation. The cytoplasmic membrane acts as a primary target site for most of organic stress agents and thus bilayer structural damage as well as alteration of its physical state has severe implications for the whole cell physiology and survival (Mykytczuk et al. 2007; Walker 2000). In order to counteract the deleterious effects of toxic chemicals and prevent their accumulation, cells are able to adapt their membranes to unfavorable conditions usually by modulating bilayer fluidity (Heipieper and de Bont 1994; Sikkema et al. 1995; Weber and de Bont 1996). This physiological adaptation process is known as homeoviscous adaptation and determined predominately by the investigation of the lipid fatty acid composition and packing the amount of sterols and lipid–protein interactions (Denich et al. 2003; Mykytczuk et al. 2007; Sinensky 1974; van der Rest et al. 1995).
The response of eukaryotic microorganism to xenobiotics, as compared to prokaryotes, is far less investigated. To date, the reaction of conventional yeasts Saccharomyces cerevisiae and Schizosaccharomyces pombe to the externally applied ethanol was the most extensively studied case (Alexandre et al. 1994; Jones et al. 1987; Koukou et al. 1990; Rodríguez-Vargas et al. 2007) and showed the membrane-fluidizing effect of this alcohol. In turn, for S. cerevisiae supplemented with the herbicide 2,4-Dichlorophenoxyacetic acid desaturase inhibition was found which rendered the membrane less fluid and thus less permeable (Viegas et al. 2005). However, regarding membrane lipid composition, both S. cerevisiae and S. pombe are not real model organisms for eukaryotic cells as their membranes do not contain polyunsaturated fatty acids. Therefore, the so-called nonconventional yeasts which typically contain also linoleic (18:2) and linolenic acids (18:3) are particularly suitable for research. In these organisms, the active modification of the lipid unsaturation index (UI) appears as one of the most pronounced mechanism that controls membrane fluidity. UI is defined as the average ratio of double bonds per fatty acid molecule and its value is optimized by the stress-inducible regulatory system involving saturase/desaturase activities (Heipieper et al. 1996; Martin et al. 2002; Weber and de Bont 1996).
Despite the strong scientific interest, there is no report concerning membrane fatty acid adaptation of methylotrophic yeasts upon growth and/or treatment with C1 compounds. In this study, the two mentioned yeast species were therefore tested for their membrane fatty acid unsaturation level as an adaptive mechanism responsive to treatment with methanol and/or formaldehyde. The attempt was also made to correlate the observed changes in lipid fluidity upon methanol incubation with the formaldehyde biodegradation potential.
Materials and methods
Yeast strains and their taxonomy
Two methylotrophic yeast strains were selected for the study: (1) a prototrophic revertant strain of H. polymorpha (Pichia angusta) NCYC 2309 (Leu−); (2) Trichosporon sp. CBM 84 (microorganism collection of the Centre of Microbiological Research and Autovaccines, ul. Sławkowska 17, 31–016 Kraków, Poland). Both strains were reported to be methanol- and formaldehyde-resistant and proved to be capable of efficient biodegradation of these compounds. The collection yeast H. polymorpha used to serve as a model strain whose methylotrophy was studied in detail (Gleeson and Sudbery 1988; Sibirny et al. 1988; van Zutphen et al. 2010). On the other hand, Trichosporon sp., an extremophilic environmental yeast, was shown to reveal a unique, alcohol oxidase-independent methylotrophy (Kaszycki et al. 2006). This strain was isolated from oil slurry-contaminated soil, and it was classified as a genus Trichosporon employing the appropriate molecular identification procedures based on rDNA subregion amplification reactions (Kaszycki et al. 2006). Taxonomic lineages (given according to the NCBI—Taxonomy Browser): (1) H. polymorpha: Phylum: Ascomycota, Subphylum: Saccharomyctina, Class: Saccharomycetes, Order: Saccharomycetales, Family: Saccharomycetaceae, Genus: Pichia; (2) Trichosporon sp.: Phylum: Basidiomycota, Subphylum: Agaricomycotina, Class: Tremellomycetes, Order: Tremellales, Genus: Trichosporon.
Culture conditions and growth media
Solid YPD media for yeast culture plating contained 20 g/l peptone, 10 g/l yeast extract, 20 g/l glucose, and 20 g/l bacto-agar. The optimal liquid growth medium contained (per liter): 3 g (NH4)2SO4, 0.5 g KH2PO4, 0.3 g MgSO4×7H2O, 0.2 g CaCI2×6H2O, 2 g yeast extract, and 2 g peptone.
Prior to each experiment, the yeasts were inoculated from agar plates into 300-ml Erlenmeyer’s flasks containing 50 ml of optimal liquid medium. The optical density (OD540) was measured in order to dilute the culture suspension to achieve OD540 ≈ 0.03. Then, the source of carbon was added (glucose or methanol at 1 % w/v), and the flasks were agitated at yeasts’ temperature optima (37 °C for H. polymorpha, 25 °C for Trichosporon sp.) for 14–17 hours, to reach a logarithmic growth phase OD540 ≈ 2.0 for H. polymorpha and OD540 ≈ 1.8 for Trichosporon sp. Then, the yeasts were treated with xenobiotics by adding formaldehyde or methanol directly to the culture. Control cultures (not treated) were always examined.
Determination of yeast growth and survivability
Turbidimetric measurement of liquid yeast cultures was performed at a wavelength of λ = 540 nm using a Jasco V-530 UV/VIS Spectrophotometer. The OD540 determination was used to assess yeast growth rates as well as growth inhibition caused by xenobiotics.
Yeast survivability assessment was carried out by the Koch plating method. Appropriate serial dilutions ranging from 10−2 to 10−6 in sterile water were prepared, and 25 μl aliquots of each were plated on Petri dishes with YPD agar. The number of cells per 1 ml of culture (expressed as colony forming units, CFU, per 1 ml) was then calculated based on the number of colonies formed on the solid medium. The error bars in the figures were calculated as average standard deviation for each of the experiments.
Determination of growth inhibition
Where: x1—final OD540 of the culture; x0—OD540 prior to addition of a xenobiotic; t—time duration of the experiment (6 h).
Lipid extraction and transesterification of fatty acids
After 6-h treatment with xenobiotics, the cultures were centrifuged at 1,000×g, cell pellets (150–200 mg of cell wet mass) washed with deionized water, and the lipids were extracted by the method of Bligh and Dyer (1959) with a modification of Schneiter and Daum (2006). In brief, 1 ml of methanol (analytical grade) was added to glass tubes containing wet biomass, and then the tubes were vortexed till the yeast pellet became resuspended completely. Afterwards, 1.75 ml of chloroform was added, and the tubes were vortexed again for 3 min. The chloroform solvent contained 0.01 % of butylated hydroxytoluene to prevent peroxidation of lipids. In the following step, 0.5 ml deionized water was added, the tubes were vigorously shaken and centrifuged at 1,000×g at 4 °C. After centrifugation, the aqueous upper phases were discarded, and the organic bottom ones were transferred into clean glass tubes containing 0.5 ml water or 0.9 % NaCl, then shaken and centrifuged again at the conditions described above. The resultant bottom phase were then transferred to clean glass tubes and dried under the stream of nitrogen.
To enable analysis of the fatty acid composition with GC-FID, the extracted fatty acids need to be converted to fatty acid methyl esters (FAMEs). FAME synthesis was performed following the protocol of Morrison and Smith (1964). According to that, 0.6 ml boron trifluoride (20 % in methanol) was added to the dried fatty acids followed by incubation at 95 °C for 15 min. The reaction was stopped by adding 0.3 ml distilled water and 0.5 ml n-hexane and shaking for 1 min. The organic phase containing the FAMEs was transferred to clean GC autosampler vials and stored at 5 °C.
GC-FID analyses and determination of the UI
FAME samples were analyzed by gas chromatography using a gas chromatograph with flame ionization detector (GC-FID) (6890 N, Agilent Technologies)) equipped with a split/split-less injector (7683B, Agilent Technologies). The column was a CP-Sill 88 capillary column (Chrompack, ID: 0.25, 50 m; 0.2 μm film). Helium was used as a carrier gas. The injection was splitless; the temperature of the injector and detector was 240 °C. The initial oven temperature was 40 °C for 2 min, and the final temperature was 220 °C with an increase of 8 °C/min to 220 °C, followed by 10 min isothermal conditions. Data acquisition was performed using the GC-Chemsation Software (Agilent Technologies). Peak identification was performed by comparison with a commercial qualitative standard of bacterial acid methyl esters (BAME, Supelco).
The error analysis based on replicate determinations yielded the value of analytical-system relative error for UI (standard deviation/mean) as 2–5 %.
Preparation of cellular extracts and zymographic staining
Yeast cell cultures were centrifuged (3,000×g for 5 min) to remove the medium, then to each 1 g of the cellular pellet 1 ml buffer (25 mM Tris/HCl, pH 7.5, containing 50 mM EDTA and 10 % dithiothreitol) was added. The tube with cell suspensions was cooled with ice, and then the cells were disrupted in four 15-min cycles with a laboratory sonicator UP 50H Ultrasonic Processor (dr. Hielscher GmbH) using an MS7 sonotrode (50 % maximum output power, 0.5 s pulses).
Protein determination in cellular extracts was done according to the method of Bradford (1976). Zymographic analyses of the extracts were performed to reveal activities of a glutathione-dependent formaldehyde dehydrogenase, FdD. The reagent solution for the native-PAGE enzymatic staining contained 25 mg of NAD+, 15 mg of p-Nitro blue tetrazolium chloride (NBT), 1 mg of phenazine methosulfate (PMS), 10 ml of 0.2 M Tris/HCl pH 7.5, 1 ml of 50 mM KCl, 35 ml of water, 5 ml of 2 mM glutathione (reduced), and 5 ml of 100 mM formaldehyde as a substrate.
Fd concentrations in biodegradation tests were determined using the method of Nash (1953). Defined volumes (600 μl) of the appropriate sample dilutions were mixed with the same amount of a Nash reagent and incubated at 60 °C for 10 min. Then, the samples were cooled down to conserve the colored reaction. The absorbance was measured spectrophotometrically at a wavelength of 412 nm. A calibration curve was made upon dilutions of the standard formaldehyde solution ranging from 0.01 to 0.3 mM.
Yeast media components: yeast extract, casein peptone, and bacteriologic agar were purchased from BTL, Poland. All other chemical compounds were of analytical grade purchased from Sigma-Aldrich. The solutions and buffers were made using bidistilled water. Whenever required, fully sterile conditions were applied. All experiments were carried out in triplicate. Standard deviation was always lower than 5 %.
Biodegradation of exogenous formaldehyde in the presence of methanol
The above data indicate different mechanisms of biodegradation for each of the strains. In the presence of methanol, H. polymorpha proved to be a highly efficient biodegrader at lower Fd concentrations (close to a 100 % up to 750 mg/l Fd) and then, along increasing Fd, the degradation efficiency decreased to 60 % at 2,000 mg/l). Detailed Fd decay kinetics for different initial concentrations of this xenobiotic had already been studied and published in previous papers (Kaszycki and Koloczek 2000; Kaszycki et al. 2001). In the case of Trichosporon sp., the yeast revealed very high tolerance enabling it to efficiently degrade Fd over the broad range of concentrations reaching extreme levels of 7,000 mg/l which was close to maximum values reported previously (Kaszycki et al. 2006). At the same time, the absolute values of biodegradation yield were significantly lower (60–40 % within 6 h) than for H. polymorpha and showed smaller dependence on the initial concentration of Fd.
Effect of the growth substrate on effective concentration causing 50 % growth inhibition (EC50) and minimal inhibitory concentrations (MIC) of formaldehyde
Detection of formaldehyde dehydrogenase activity
Determination of membrane fluidity at growth-inhibitory concentrations of formaldehyde and methanol
Methanol was of relatively low toxicity for the studied methylotrophs, and the total inhibition of growth was observed at concentrations above 10 %. Formaldehyde treatment inhibited the yeasts’ growth at much lower levels, and its negative impact was especially manifested when the yeasts were grown on glucose. Table 1 shows the EC50 and MIC values as determined based on the growth inhibition analysis. The respective curves representing the inhibition of growth after 6 h incubation with increasing xenobiotic concentrations are shown in Figs. 4, 5 and 6.
The above data were confirmed by the direct plating method of yeast survival analysis. For H. polymorpha as well as for Trichosporon sp. high survival rates were reported for the yeasts grown on methanol, up to formaldehyde levels of 1,500 and 3,000 mg/l, respectively. After cultivation on glucose, Fd concentrations of 1,000 mg/l for H. polymorpha and below 250 mg/l for Trichosporon sp. led to the total population death (data not shown).
Effect of carbon source on the membrane fatty acid unsaturation index (UI)
The membrane UI of H. polymorpha cultured on methanol increased significantly upon intoxication with growth-inhibitory concentrations of formaldehyde (Fig. 4b). The highest UI value (UI = 1.31) was observed when 750 mg/l Fd was added. As the Fd concentration increased and growth inhibition was more pronounced, the response of cells was less intense, which was similar to the earlier case. However, the concentration of Fd that suppressed the UI modulation was considerably higher for the case of yeasts grown on methanol compared to those grown on glucose. For Trichosporon sp. cultured on methanol, unlike for the glucose substrate, this yeast reacted to the increasing inhibitory concentrations of formaldehyde by rising UI values (Fig. 5b). The maximum value of UI (UI = 1.59) was observed upon treatment with 1,500 mg/l of Fd. At greater Fd levels, the reaction was only slightly less efficient. Moreover, even the highest amount of formaldehyde (6,000 mg/l) did not hamper the response at the level of unsaturation index. It is important to notice that the control yeast cultures (not treated with formaldehyde) when grown on methanol revealed higher UI values than the cultures grown on glucose which agrees with the observations presented above (cf. Table 2). For H. polymorpha cultured on methanol, the UI value was approximately 1.0 compared to around 0.6 on glucose, for Trichosporon sp. it was about 1.0 and 0.2, respectively.
In addition, the yeasts’ reactions to phenol treatment were also investigated. This aromatic xenobiotic caused a decrease in UI with a maximum around the EC50 values (data not shown), which was in accordance with the previous data (Heipieper et al. 1992, Heipieper and de Bont 1994; Kim et al. 2001) and gave further evidence of the ability of both strains to adapt their membranes.
The tolerance of methylotrophic yeasts towards formaldehyde and methanol is usually linked to their ability to metabolize these compounds (Kaszycki and Koloczek 2000; Gleeson and Sudbery 1988; Sibirny et al. 1988; Yurimoto et al. 2005, 2011). No attempt has been made so far to evaluate the possible involvement of a membrane-specific mechanism. The presence of such a mechanism should be hypothesized since it was shown for many organic chemicals to induce membrane adaptive changes as a primary cell response to treatment with xenobiotics. In this study, we have therefore combined the experimental data on formaldehyde biodegradation with the methanol-promoted membrane adaptive changes achieved by lipid unsaturation.
The main finding of this work is that the increasing Fd tolerance and enhanced biodegradation potential was accompanied by fluidization of the cell membranes of H. polymorpha and Trichosporon sp. This effect was made possible by methanol’s direct influence on the degree of lipid unsaturation as well as by yeast biostimulation with this alcohol that enabled the yeast membranes to further adapt to the presence of formaldehyde. Aldehydes themselves are known to be extraordinarily toxic to all living systems. This is mainly because of the fact that next to their effect on membrane fluidity, they also show an additional chemical toxicity mainly due to their ability to disrupt proteins by the formation of Schiff’s bases (Löffler et al. 2010). Therefore, a precultivation on methanol induces the cellular machinery for formaldehyde degradation/detoxification to the far less toxic formate. In addition, also the adaptation to methanol on other cellular levels, such as, e.g., the cytoplasmic membrane fluidity, seems to increase the overall fitness to respond to Fd and prevents its accumulation. It is emphasized here that the observed UI changes occurred as a result of metabolic yeast activity, enabling them to de novo synthesize membrane lipids and incorporate double bonds. This was made possible only at xenobiotic concentrations that were subinhibitory for cell growth. Both of the studied strains proved to be able to synthesize the polyunsaturated fatty acids linoleic acid (18:2) and linolenic acid (18:3), which is unlike S. cerevisiae (Lu et al. 2000) and makes these yeasts, next to Kluyveromyces lactis (Heipieper et al. 2000, Cialfi et al. 2011), particularly interesting research objects.
At present, it is difficult to determine the importance of the novel mechanism in terms of the methylotrophic yeast ability to deal with the exogenously applied formaldehyde. Based on the hitherto collected data and on the current understanding of methylotrophy we may infer, however, that the role of any membrane-associated processes have been undervalued, so far.
It is widely accepted that FdD is the enzyme responsible for detoxification of excess formaldehyde (Baerends et al. 2002; 2008; Kato et al. 1982; Lee et al. 2002; Sasnauskas et al. 1992; Sibirny et al. 1988; Yurimoto et al. 2005) which is either generated upon methanol oxidation with alcohol oxidase (AOX) or appears as exogenous pollutant. This detoxification is achieved by triggering biodegradative pathway which finally oxidizes Fd to H2O and CO2. In a novel, AOX-independent methylotrophic pathway exhibited by several environmental yeast isolates including Trichosporon sp. (Kaszycki et al. 2006), an FdD activity was produced too. Therefore, this enzyme might be responsible for the observed high Fd biodegradation potential.
It should be noted that both H. polymorpha and Trichosporon sp. were earlier reported to require methanol in order to reveal maximum exogenous formaldehyde biodegradation effect (Kaszycki and Koloczek 2000; Kaszycki et al. 2001, 2006). Among the possible explanatory mechanisms, one should consider the methanol-based induction of methylotrophy including the FdD activation (Hartner and Glieder 2006; Gleeson and Sudbery 1988; Yurimoto et al. 2011; van der Klei et al. 2006) as well as the presence of independent biodegradative step with the methyl formate synthase as proposed by the Kato group (Murdanoto et al. 1997; Sakai et al. 1995). The presence of the third mechanism, i.e., the methanol-induced membrane adaptation finds considerable support when the presence of FdD activity in both experimental variants (glucose vs. glucose-methanol supplementation) is considered. The results of zymographic analyses as well as the analytical detection of the FdD reaction product proved no glucose repression over FdD, although under such noninducible conditions, the FdD activity was partially lowered. Therefore, an efficient Fd biodegradation could be expected even in the presence of a glucose substrate alone.
For H. polymorpha, the observed decreased Fd tolerance and poorer degradation ability could still be explained by the lack of methanol induction. Furthermore, glucose is known to be a strong repressor of a key enzyme, AOX, and thus the whole pathway might become inhibited. In addition to that, for this strain, but not for Trichosporon sp., the increase of the UI parameter was also induced by formaldehyde present at noninhibitory concentrations under glucose growth. These observations might explain why the methanol-biostimulation effect was less pronounced for H. polymorpha. As regards Trichosporon sp., its extreme sensitivity to formaldehyde and a very low Fd biodegradation yield in glucose medium make the necessity for adaptation at the level of membranes more straightforward and suggest that these characteristics should be directly confronted with the relatively rigid state of the lipid bilayer.
To sum up, it appears that the role of membrane adaptive changes upon C1 compound utilization by methylotrophic yeasts may be different depending on the species involved in methylotrophy as well as on the particular pathway of methanol utilization. The methanol-promoted fluidization might prevent formaldehyde from rapid entering the cytoplasm, thus enabling the yeasts to be screened against toxification. Consequently, the more favorable conditions achieved by membrane adaptation might allow for undisturbed FdD-based biodegradation, efficient enough to keep the cytoplasmic Fd level below the lethality threshold.
The presented data should be confirmed by complementary analyses employing genetic and molecular biochemistry studies and in particular detection of desaturase activities and genes. Recently, in a corresponding study by van Zutphen et al. (2010), a H. polymorpha transcriptome was analyzed upon transfer of this yeast from glucose to methanol medium. The results showed profound rearrangement of the cell architecture, physiology, and enzymatics as inferred from the microarray-based identification and quantification of transcripts. Most of the data are consistent with the earlier-described models of methylotrophic induction. However, the information concerning membrane-acting factors is lacking, although the authors claim that more than 30 % of the transcriptome still awaits detailed characterization.
In conclusion, it seems likely that membrane fluidization achieved by the metabolism-driven extensive fatty acid unsaturation plays a significant role in adaptation to exogenous formaldehyde and is involved in enhancing Fd resistance and biodegradation potential of methylotrophic yeasts. The presence of methanol contributes to this mechanism as it promotes the membrane adaptive reaction towards formaldehyde and increases the lipid bilayer fluidity itself. At least for some methylotrophs, as exemplified by Trichosporon sp., biostimulation with methanol is a crucial condition enabling formaldehyde biooxidation at high concentrations. In order to fully exploit the potential of methylotrophic yeast in environmental biotechnology, a deeper insight into stress adaptation processes needs to be provided. This will help to find out whether the yeasts’ tolerance to certain xenobiotics could be improved by means of controlled membrane fluidity modulation, which in turn will be of great importance in terms of the potential of bioremediation processes.
This work was partially supported by the European Commission within its Seventh Framework Program Project BACSIN (Contract no. 211684). This work contributed to the CITE Research Programme of the Helmholtz Centre for Environmental Research.