Introduction

Candida parapsilosis ATCC 7330 is an established biocatalyst for the preparation of numerous optically pure aryl/ aliphatic secondary alcohols and aryl amines using different strategies namely, asymmetric reduction1,2,3, deracemisation4,5,6 and oxidative kinetic resolution7. These optically pure secondary alcohols are produced from either prochiral ketones or racemic alcohols in reaction times as short as 5 min or even up to 24 h. These small organic molecules which are biotransformed as reported in this study, possibly enter the cells by passive diffusion i.e. transport of molecules across the membranes from higher to lower concentration without any additional energy requirement8. Once inside the cell, the molecule gets transformed and is excreted from the cell due to which it is possible to extract the material from the reaction mixture and determine its optical purity, spectral analysis and yield. The present study addresses the question about the site of biotransformation inside the cell. A fluorogenic assay for confocal microscopic studies was developed for this purpose. The experimental results were validated using HPLC.

The most widely used methods for localisation studies are subcellular fractionation, immunohistochemical studies and fluoro-probe techniques. It is known that cells consist of different dehydrogenases and other metabolic enzymes even under extremely unfavourable conditions9. In the case of the three reported bacterial dehydrogenases, class I ADHs, are localised in the periplasm; class II ADHs are present as soluble monomers in the periplasm and class III ADHs are present in acetic acid bacteria and are also localised in the periplasmic side of the cytoplasmic membrane10. Localisation of alcohol dehydrogenases (ADHs) in various mammalian tissues mainly use immunohistochemical studies11,12, immuno-gold labeling13, Western blotting14 and fluorimetric assays15. Only a few reports are available till date on the localisation of alcohol dehydrogenases in yeast cells. ADH-I and ADH-II were localised in the cytosol while ADH-III was localised in the mitochondrial matrix based on subcellular fractionation16,17. Using a green fluorescent protein, a commonly used fluorescent tag, obtained from the jelly fish Aequorea victoria, ADH-IV and ADH-V were localised in the mitochondria and cytoplasm respectively18; and the localisation of ADH VI and VII which were obtained from over-expression studies remains ambiguous18,19,20. However, tracking the site of biotransformation of small organic molecules (alcohols/ketones) by resting yeast cells is not yet reported.

Klein et al., developed an enantioselective fluorimetric assay for alcohol dehydrogenases based on umbelliferone. The chiral secondary alcohol is tagged with umbelliferone. On oxidation in the presence of ADH, the corresponding unstable ketone formed undergoes β-elimination at basic pH in the presence of bovine serum albumin to release the fluorescent umbelliferone21. This strategy can be used when the target protein is a single enzyme. In whole cells where a pool of different enzymes is present, multiple reactions can take place which can be misleading. Ideally, the substrate molecule should be detectable without a tag, as the fluorescent tag can alter its properties including that of permeation through membranes. In many studies, the total ADH activity is determined spectrophotometrically using p-nitroso-N,N-dimethyl aniline to the fluorescent p-amino-N,N-dimethyl aniline22. The most commonly used fluorogenic substrates for ADH activity are 4-methoxy-1-naphthaldehyde and 6-methoxy-2-naphthaldehyde, which are more used to determine ADH-1 and ADH-2 activities respectively in mammalian cells23,24,25. Unlike mammalian cells, yeasts have thick cell wall, which has to factored in while considering the entry of bulky fluorescent substrates in and out of the cell.

This paper presents for the first time, experiments using fluorescent organic small molecules to track their biotransformation [redox reaction] in the yeast cells. The biotransformed product then comes out of the cells.

Results

Selection of the substrate

The fluorescent substrate, 6-methoxy-2-naphthaldehyde 1a (Fig. 1) is often used to study the ADH activity for different mammalian cells23,24,25. This molecule was incubated with wet cells of C. parapsilosis ATCC 7330 using ethanol as the cosolvent and glucose as the cosubstrate1. Two possibilities were considered: a. If the substrate aldehyde 1a got reduced by C. parapsilosis ATCC 7330, it would give the product alcohol which is not fluorescent i.e, (6-methoxynaphthalen-2-yl)methanol 2a (Case Ia) and b. If the alcohol 2a was used as substrate which then gets oxidised to the aldehyde 1a by C. parapsilosis ATCC 7330, it would mean going from a non-fluorescent molecule to a fluorescent molecule (Case Ib), which is preferred for tracking. The reaction was initially carried out for 24 h and later extended to 48 h, and the product formation was monitored by TLC. The TLC indicated no trace of product 2a formation even after 48 h. The next experiment was carried out (Case Ib; Fig. 1) where the alcohol 2a was the substrate for biocatalytic oxidation (Tris-HCl buffer, pH 8.5, 10 mM) with acetone as cosubstrate and ethanol as cosolvent26. The progress of the reaction as monitored by TLC did not show aldehyde formation and the substrate remained unreacted even after 48 h.

Figure 1
figure 1

Biocatalytic reduction of 6-methoxy-2-naphthaldehyde 1a (Case Ia) and oxidation of (6-methoxynaphthalen-2-yl)methanol 2a using C. parapsilosis ATCC 7330 (Case Ib).

Full size image

Earlier reports from our lab have shown that C. parapsilosis ATCC 7330 mediated the asymmetric reduction1,2,3,27 and deracemisation4,28,29,30 of aromatic/aliphatic prochiral ketones and secondary alcohols respectively. It seemed logical to use a prochiral ketone for the biocatalytic reduction which shows structural similarity to 1a for the study. Hence, 1-(6-methoxynaphthalen-2-yl)ethanone 1b, a fluorogenic aldol sensor in antibody-catalysed retro-aldol reaction31, was chosen as the next substrate (Case IIa; Fig. 2) and on incubation with cells of C. parapsilosis ATCC 7330 under reaction conditions optimised earlier, showed only a trace of non-fluorescent product alcohol [1-(6-methoxynaphthalen-2-yl)ethanol] 2b′ by TLC. The conversion was only <10% as monitored by HPLC using a reverse phase column. This substrate 1b could not be used for further studies and in any case the product alcohol is non-fluorescent. The racemic alcohol 2b (Case IIb) was used as a substrate (substrate concentration: 2.8 mM) for deracemisation using the earlier optimised conditions26 and the reaction was carried out for 1 h. The reaction was expected to proceed via stereoinversion mechanism like other aryl racemic alcohols carried out using this biocatalyst4,5,6. The progress of the reaction was monitored using TLC and the intermediate formation of the fluorescent ketone 1b was observed along with the alcohol. The time course of the reaction was monitored by HPLC using the reverse phase column and observed that 38% of ketone 1b along with the unreacted alcohol 2b′ were formed (Supplementary Results, Supplementary Fig. 1), which indicated that the reaction which occurred was oxidative kinetic resolution (Fig. 3). The same reaction when carried out in the absence of acetone, gave only 27% of ketone which showed the important role of acetone in cofactor regeneration32. The chiral HPLC profile showed that the (S)-enantiomer of the racemate undergoes enantioselective oxidation, while the (R)-enantiomer (ee 98%) remains unreacted (Fig. 3). The time course of this reaction from the chiral HPLC profile showed that the reaction is completed within 30 minutes and further extending the reaction time up to 1 h did not show any noticeable change in the conversion of the products formed (Supplementary Fig. 2).

Figure 2
figure 2

Asymmetric reduction of 1-(6-Methoxynaphthalen-2-yl)ethanone 1b using C. parapsilosis ATCC 7330 (Case IIa).

Full size image
Figure 3
figure 3

Enantioselective oxidation of rac-1-(6-Methoxynaphthalen-2-yl)ethanol 2b using C. parapsilosis ATCC 7330 (Case IIb).

Full size image

Substrate concentration

Substrate concentrations in the range 0.0162–3.38 mM were studied using constant biomass (1.2 g wet cells/5 mL). The concentration of the substrate 2b was optimised at 0.87 mM for maximum conversion (38%) to 1b (Supplementary Fig. 3) and used for all further experiments.

Effect of culture age on alcohol dehydrogenase activity of the biocatalyst

All the above experiments were carried out with the whole cells harvested at 14th hour (mid log phase) of the growth phase1. In order to understand the effect of enzyme activity with respect to the culture age, the cells were harvested at different time points (8–40 h) and used for the biotransformation. The results showed that cells harvested at the 24th hour of growth, showed the maximum conversion [49%] to ketone (Supplementary Fig. 4). Even though cells harvested at 14 h produced the same ketone at less conversion (38%), these cells were used to visualise the site of biotransformation. For several other substrates reported earlier from our lab, the 14 h harvested cells produced the desired optically pure products with maximum ee (up to >99%)1,2,3,6,7,26,29,30,33,34,35.

Cell imaging studies using confocal microscopy

The reaction was carried out qualitatively on a microscope slide using a higher substrate concentration (4.4 mM) and added over the cell suspension as a mixture in acetone: ethanol (2:1). The substrate concentration was increased in anticipation that the distribution of the substrate to the cells may not be uniform due to the absence of shaking during the reaction and also the optimised concentration (0.87 mM) might be too low to be detected if the distribution is not uniform. From the images, obtained with the live cells, it was observed that the fluorescence was localised in the vacuole of the cell. Another, interesting observation made from the images is that the formation of the fluorescent ketone molecule occurs in the vacuole region of the cell and as soon as the ketone is formed, it gets released out of the cell in less than 5 min (Fig. 4). The control experiments (using heat killed cells) did not show the formation of ketone (absence of fluorescence within the cells). This confirms the formation of the fluorescent molecule in the live cells due to the active alcohol dehydrogenase enzyme present.

Figure 4
figure 4

Microscopic images of localisation of biotransformation in C. parapsilosis ATCC 7330 in pH 8 sodium phosphate buffer using Confocal laser microscopy (a) Cells with no substrate added; inset: enlarged cells without substrate (b) Cells incubated with rac-1-(6-Methoxynaphthalen-2-yl)ethanol 2b with appearance of fluorescence inside the vacuole; (c) Release of the fluorescent product 1-(6-Methoxynaphthalen-2-yl)ethanone 1b outside the cell; (d) Heat killed cells incubated with rac-1-(6-Methoxynaphthalen-2-yl)ethanol 2b with no fluorescent product formation; inset: enlarged heat killed cells with substrate.

Full size image

Rate of formation of the fluorescent ketone

The images show that the formation of ketone 1b from the racemate 2b occurs inside the vacuole and gets released from the cell spontaneously. The kinetics of ketone formation was followed under optimised reaction conditions. Aliquots at time 0 to 5 min were analysed using reverse phase HPLC and the amount of ketone present in the supernatant was estimated using a standard plot of known concentrations (0.05–0.4 mM) of ketone 1b. It was observed that the formation of the fluorescent ketone begins in the 1st minute (at the rate of 136.67 μg/min/gm of biomass) and the maximum conversion to the fluorescent ketone (up to 38%, 0.32 mM) is achieved in the 3rd minute of the reaction time for this substrate concentration (0.87 mM) (Supplementary Fig. 5). The important observation from this study is that the formation of ketone occurs spontaneously and gets released out of the cells, possibly making way for the other set of non-fluorescent alcohol molecules to get converted to the ketone.

Discussion

The oxidative kinetic resolution of racemic 1-(6-methoxynaphthalen-2-yl)ethanol 2b by Candida parapsilosis ATCC 7330 is a good assay to track the biotransformation in the yeast cells as shown in this study.

Designing fluorogenic substrates which are capable of coupling with the chemical redox reaction and exhibiting the switch in emission properties is reported36. Most of the organic fluorophores designed as redox probes are based on the push-pull structural feature, whereby the electron donating and electron withdrawing groups are connected via an extended π-conjugated system37. The carbonyl group becomes part of the push-pull system and when it gets reduced to the alcohol, the electron withdrawing nature of the carbonyl group changes to the electron donating nature of the hydroxyl group. This shift causes changes in the electronic properties between the reactant and product giving different emission characteristics36. Initially, the substrate 6-methoxy-2-naphthaldehyde 1a was chosen for this study as it satisfies the conditions of an ideal organic fluorescent redox probe. However, substrate 1a remained as such up to 48 h. Likewise, alcohol 2a was not oxidised to the fluorescent aldehyde as expected. Both 1a and 2a are not preferred substrates for this biocatalyst, although the asymmetric reduction of α-oxoaldehydes using this biocatalyst is known to produce the respective diols2. In the case of oxidation, primary alcohols are less preferred as compared to secondary alcohols7. From the earlier reports1,3,27 the asymmetric reduction of prochiral ketones using C. parapsilosis ATCC 7330 is well documented. Therefore, the substrate 1b was subjected to asymmetric reduction but it resulted in low conversion (<10%) to the product alcohol, a phenomenon seen for substrates with bulky groups including naphthyl group which show either less or no conversion to the product alcohols4,38,39.

Most yeast cells including C. parapsilosis ATCC 7330 appear as ovoid or ellipsoidal in shape with a cell length of 2–3 μm to 20–50 μm. As expected, the size of the yeast cell differs with its age and growth conditions40. During unfavourable conditions e.g. nutritional deprivation or oxidative stress, the cytoplasmic proteins and the cell organelles get transported to a vacuole by means of non-selective “cytoplasm-to-vacuole” pathway which tends to overlap with the autophagy for degradation9. It is reported that several NAD+ dependent alcohol dehydrogenases are found in the vacuolar compartment of S. cerevisiae41,42. Evidently, vacuoles of yeast cells undergo morphological changes in response to various intracellular and extracellular stimuli. In the log phase, cells consist of multiple medium-sized vacuoles under normal conditions. These multiple vacuoles, fuse into a single large vacuole during the stationary phase or under nutritional deprivation. They can also get fragmented into multiple small vesicles during osmotic stress. In hypo-osmotic conditions, the vacuoles swell into a single large vacuole occupying a majority of the cellular volume. Thus, the uptake and release of water or ions from the vacuole shows a distinct morphological change in the size and number of vacuoles in the yeast cell43.

In the present study, C. parapsilosis ATCC 7330 cells were harvested from the growth medium at 14 h, which corresponds to the mid log phase of the growth period. All the experiments were carried out using the harvested cells or resting cells. Due to nutritional stress under these conditions, the cellular contents including the cytoplasmic proteins are forced to get transported into the vacuole9 inside the yeast cells. Therefore, under these experimental conditions the representative enantioselective oxidation of racemic-1-(6-methoxynaphthalen-2-yl)ethanol to the fluorescent 1-(6-methoxynaphthalen-2-yl)ethanone occurs in the vacuoles inside the cells (Fig. 4).

A novel fluorescence based method for the site of localisation of the oxidation reaction of 1-(6-methoxynaphthalen-2-yl)ethanol to the fluorescent 1-(6-methoxynaphthalen-2-yl)ethanone in C. parapsilosis ATCC 7330 in addition to reporting a biocatalysed method to prepare the optically pure (R)-1-(6-methoxynaphthalen-2-yl)ethanol and the precursor ketone 1-(6-methoxynaphthalen-2-yl)ethanone is reported in this study. The experimental results show that the enzymes like alcohol dehydrogenases which are used to catalyse various redox reactions are localised in the vacuole of the yeast cells under the conditions studied. Racemic-1-(6-methoxynaphthalen-2-yl)ethanol was oxidised using C. parapsilosis ATCC 7330 enantioselectively to the fluorescent ketone (38%), leaving the (R)-enantiomer unreacted with 98% ee and 62% conversion. The study highlights that the fluorescent ketone once formed in the vacuole gets released out of the cell in less time, which was confirmed from the kinetic studies.

Methods

The chemicals, 6-methoxy-2-naphthaldehyde and 1-(6-methoxynaphthalen-2-yl)ethanone were purchased from Sigma-Aldrich and standard racemic alcohols were synthesised using the reported method44. The growth conditions of Candida parapsilosis ATCC 7330 were followed as mentioned earlier1. The fluorescent intensity of the substrates and products were determined using JASCO spectrofluorimeter FP 8000. The conversion and ee of the products formed were analysed using JASCO PU-1580 HPLC equipped with PDA detector. The confocal microscopic studies were carried out using Zeiss LSM 700 Confocal microscopy and image processing was done using Zen LE- 2012 software.

Enantioselective oxidation of 1-(6-methoxynaphthalen-2-yl)ethanol 1b using Candida parapsilosis ATCC 7330

In a 25 mL Erlenmeyer flask, 1.2 g of wet cells of Candida parapsilosis ATCC 7330 was suspended in 5 mL of sodium phophate buffer (pH 8, 10 mM). The substrate 1-(6-methoxynaphthalen-2-yl)ethanol 1b (2.8 mM) dissolved in 100 μL of ethanol and 200 μL of acetone, was added and the reaction was continued up to 30 min. The progress of the reaction was monitored using TLC and after the reaction time the products were extracted using ethyl acetate. The conversion was determined using HPLC with MERCK Lichrospher 100 RP-18e column with acetonitrile: water (85:15) as mobile phase (flow rate 0.5 mL/min) and ee using Chiralcel OD-H column (hexane: 2-propanol; 95:5; flow rate: 1 mL/min)45 respectively. All the experiments were carried out in triplicate and the control experiments were done in parallel using blank (no cells) as well as using heat-killed cells under identical conditions.

Cell imaging studies using confocal microscopy

The experiments were carried out on microscopic slides by adding 20 μL of the cell suspension (OD600 0.7; pH 8 buffer). The number of cells of C. parapsilosis ATCC 7330 present in the cell suspension was found to be 6.2 × 105 cells/20 μL which was counted using haemocytometer46. About 10 μL of substrate 2b (4.4 mM) in acetone:ethanol mixture (2:1) was added to the cell suspension in the slide. The slide was mounted on the confocal microscope and images were captured using the constant instrumental settings (Supplementary Table 1). The experiments were carried out in triplicate and the control experiments were done in parallel using blank cells (without substrate) as well as using heat-killed cells under identical conditions.

Reaction kinetics of the formation of 1-(6-methoxynaphthalen-2-yl)ethanone 1b from racemic 1-(6-methoxynaphthalen-2-yl)ethanol 2b using Candida parapsilosis ATCC 7330

In a 25 mL Erlenmeyer flask, 1.2 g of wet cells of Candida parapsilosis ATCC 7330 was suspended in 5 mL of sodium phophate buffer (pH 8, 10 mM). The substrate 1-(6-methoxynaphthalen-2-yl)ethanol 2b (0.87 mM) dissolved in 100 μL of ethanol and 200 μL of acetone, was added and the reaction was continued up to 5 min. The aliquots were taken at 1 min intervals and the products were extracted using ethyl acetate. The conversion was determined using HPLC using the conditions mentioned above.

A series of known concentrations (0.05–0.4 mM) of 1-(6-methoxynaphthalen-2-yl)ethanone 1b were suspended in 5 mL of sodium phophate buffer (pH 8, 10 mM) and were extracted using ethyl acetate. The area under the curve was determined using HPLC using the conditions mentioned above. The standard plot was drawn earlier to correlate the area under the curve with respect to the concentration of the ketone 1b.

Preparative scale synthesis of 1-(6-methoxynaphthalen-2-yl)ethanone 1b by enantioselective oxidation of racemic-1-(6-methoxynaphthalen-2-yl)ethanol 2b using the whole cells of Candida parapsilosis ATCC 7330

In a 150 mL Erlenmeyer flask, 15 g of wet cells of Candida parapsilosis ATCC 7330 was suspended in 28.2 mL of sodium phosphate buffer (pH 8, 10 mM), 1.2 mL of acetone was added to it and incubated at 25 °C and 150 rpm. The substrate, 1-(6-methoxynaphthalen-2-yl)ethanol 2b (80 mg, 0.4 mmol) which was dissolved in 600 μL of ethanol was added and the reaction was continued for 30 min. After the reaction time, the crude product was extracted thrice with ethyl acetate, the organic layer was dried over anhydrous sodium sulphate and the solvent was removed by rotary evaporator. The optically pure (R)-1-(6-methoxynaphthalen-2-yl)ethanol 2b′ (ee: 98%, yield: 47%) and 1-(6-methoxynaphthalen-2-yl)ethanone 1b (yield: 41%) were obtained as colourless solids after purification with silica gel column chromatography using hexane:ethyl acetate (90:10) as eluent. The purified products were characterised using spectroscopic (IR, 1H and 13C NMR) techniques and were consistent with the literature reported values45,47 (Supplementary Figs 6 to 9).

Spectral Characterisation of biotransformed products. 1-(6-Methoxynaphthalen-2-yl)ethanone 1b

Colourless solid, 1H NMR (500 MHz; CDCl3; δ in ppm)47: 8.38 (d, J = 1 Hz, 1H), 7.99 (dd, J = 8.5, 2 Hz, 1H), 7.83 (d, J = 9 Hz, 1H), 7.75 (d, J = 9 Hz, 1H), 7.19 (dd, J = 9, 2.5 Hz, 1H), 7.14 (d, J = 2 Hz, 1H), 3.93 (s, 3H), 2.68 (s, 3H); 13C NMR (CDCl3; 125 MHz; δ in ppm): 197.83, 159.71, 137.23, 132.56, 131.06, 130.01, 127.76, 127.04, 124.61, 119.67, 105.70, 55.37, 26.50; IR (neat; cm−1): 3060, 3000, 2969, 2938, 1733.

(R)- 1-(6-Methoxynaphthalen-2-yl)ethanol 2b'

Colourless solid, 1H NMR (500 MHz; CDCl3; δ in ppm): 7.71–7.74 (m, 3H), 7.47 (dd, J = 8.5, 1.5 Hz, 1H), 7.12–7.16 (m, 2H), 5.04 (q, J = 6.5 Hz, 1H), 3.92 (s, 3H), 1.57 (d, J = 6.5 Hz, 3H); 13C NMR (CDCl3; 125 MHz; δ in ppm): 157.65, 140.89, 134.04, 129.39, 128.73, 127.16, 124.35, 123.76, 118.95, 105.68, 70.53, 55.30, 25.05; IR (cm−1): 3343, 3008, 2966, 2935, 1074; Specific rotation: [α]D26 + 28.4 (c 0.8, CHCl3). The compound was resolved by HPLC using the Chiralcel OD-H column using hexane: 2-propanol (95:5) as mobile phase with flow rate 1.0 mL/min with retention times (min): 14.92 (S, minor); 21.13 (R, major)45.

Additional Information

How to cite this article: Venkataraman, S. et al. Direct observation of redox reactions in Candida parapsilosis ATCC 7330 by Confocal microscopic studies. Sci. Rep. 6, 34344; doi: 10.1038/srep34344 (2016).