Separation of active laccases from Pleurotus sapidus culture supernatant using aqueous two-phase systems in centrifugal partition chromatography

Journal of Chromatography B, 1002 (2015) 1–7

Contents lists available at ScienceDirect

Journal of Chromatography B journal homepage: www.elsevier.com/locate/chromb

Separation of active laccases from Pleurotus sapidus culture supernatant using aqueous two-phase systems in centrifugal partition chromatography C. Schwienheer a , A. Prinz b , T. Zeiner b , J. Merz a,∗ a b

Laboratory of Plant and Process Design, Department of Biochemical and Chemical Engineering, TU Dortmund University, D-44227 Dortmund, Germany Laboratory of Fluid Separations, Department of Biochemical and Chemical Engineering, TU Dortmund University, D-44227 Dortmund, Germany

a r t i c l e

i n f o

Article history: Received 10 February 2015 Received in revised form 21 July 2015 Accepted 26 July 2015 Available online 31 July 2015 Keywords: Centrifugal partition chromatography (CPC) Aqueous two-phase systems (ATPS) Downstream processing Laccase from Pleurotus sapidus Activity preservation

a b s t r a c t For the production of bio active compounds, e.g., active enzymes or antibodies, a conserved purification process with a minimum loss of active compounds is necessary. In centrifugal partition chromatography (CPC), the separation effect is based on the different distribution of the components to be separated between two immiscible liquid phases. Thereby, one liquid phase is kept stationary in chambers by a centrifugal field and the mobile phase is pumped through via connecting ducts. Aqueous two phase systems (ATPS) are known to provide benign conditions for biochemical products and seem to be promising when used in CPC for purification tasks. However, it is not known if active biochemical compounds can “survive” the conditions in a CPC where strong shear forces can occur due to the two-phasic flow under centrifugal forces. Therefore, this aspect has been faced within this study by the separation of active laccases from a fermentation broth of Pleurotus sapidus. After selecting a suitable ATPS and operating conditions, the activity yield was calculated and the preservation of the active enzymes could be observed. Therefore, CPC could be shown as potentially suitable for the purification of bio-active compounds. © 2015 Elsevier B.V. All rights reserved.

1. Introduction In course of the development process in biotechnology, the purification of bio-active compounds becomes more meaningful. Thereby, the active function of the biochemical compound has to be preserved [1,2]. Therefore, the examination of existing and development of new purification techniques has to be taken into account, with regards to a gentle separation environment for the products. Centrifugal partition chromatography (CPC) is mainly known for purification of natural products in preparative scale, while typically aqueous–organic systems are used [3,4]. CPC is a kind of solid or support free liquid–liquid chromatography in which the separation mechanism is based on the principle of different partitioning of components between two immiscible liquids. Using a centrifugal force, one phase is kept stationary in chambers which are connected by ducts to a chamber-cascade, while the other phase is pumped through as mobile phase (principle shown in Fig. S1 in the Supplementary material). Due to the shear forces between both phases, the mobile phase is dispersed in the stationary phase and thus,

∗ Corresponding author. Fax: +49 231 755 2341. E-mail address: [email protected] (J. Merz). http://dx.doi.org/10.1016/j.jchromb.2015.07.050 1570-0232/© 2015 Elsevier B.V. All rights reserved.

provides an interfacial area for mass transfer. According to their partition coefficients (Kd), components of a mixture to be separated will elute at different times, e.g., components which are higher partitioned to the mobile phase will elute earlier. According to [5], partition coefficient should be in the range of the “sweet spot” of 0.5 < Kd < 2. Lower values result in a poor peak resolution and higher values in a high dilution of the eluted compounds. Suitable Kd’s for CPC separations therefore are generally different than for classical continuous extraction, were highest or lowest possible Kd’s are desired for a maximized extraction effect. In CPC, which is a discontinuous extraction, a Kd of 1 (or at least in the sweet spot) for the target compound is favored and components are separated discontinuously in form of component peaks as known from classical chromatography. As a major advantage compared to other chromatographic devices, no supporting solid phase is used in CPC. Thus, CPC can provide gentle separation conditions, like no irreversible adsorption, and a high amount of stationary phase (60–80% of the total column volume) is accessible to the sample solutes resulting in a high capacity [6]. Due to these advantages the CPC is recently investigated by several research groups for the use as separation technology for biochemical products [7–10]. Several successful separations of

2

C. Schwienheer et al. / J. Chromatogr. B 1002 (2015) 1–7

biochemical products using CPC have been reported in the last years [8–13]. Thereby, aqueous two-phase systems (ATPS) were preferred as systems. With ATPS, both liquid phases are mainly composed of water and thus provide gentle, conditions for the separation of biotechnological compounds [14]. The biphasic character of ATPS is achieved by adding phase forming components like two incompatible polymers, one polymer and one salt or two different salts above a critical amount and a certain temperature. Classical ATPS could be proven as gentle and activity preserving extraction systems for biotechnological products [14–16]. The combination of using ATPS as phase system in CPC offers new operating windows for e.g., separating fragile or large biochemical compounds which might not be processable by classical solid liquid chromatography (e.g., due to size exclusion limitations due to diffusion or molecular damage by adsorption to the stationary phase’s surface). ATPS are well known to provide gentle conditions, however it has to be evaluated if biochemical compounds can “survive” the influences in the CPC apparatus itself. Molecular structures might get damaged and thus, the activity of the component can decrease, due to shear or interfacial forces resulting from the liquid two phase in CPC. Some research groups therefore measured the enzyme activities after a separation run using a Counter Current Chromatography (CCC) device [17,18]. Similar in its application but significantly different in working principle, the stationary phase is thereby immobilized in a simple tube wound to a coil that is rotating around two axes (see [19,20]). The mechanical principle of CCC and CPC are generally different and so are the shear forces between the phases in both apparatus. It can be assumed, that the shear forces in a CPC chamber are considerable higher. Additionally, enzyme activity has yet not been accounted before and after a separation process using CCC or CPC devices and thus, the activity yield could not been calculated yet. Therefore, a “real” culture supernatant containing active laccases was used for separation experiments with a CPC device. The focus of work was not laid on separating any specific component and thus, separation was not optimized. The main objective was to evaluate the ability of the CPC to preserve the activity of the biochemical components and therefore, activities before and after being processed through the CPC were balanced and the activity yield was calculated.

every supernatant was filtered prior to experimental application using filter paper 413 (pore size: 5–13 ␮m) from VWR. 2.3. Activity measurement of laccases As described by Majcherczyk et al. [22] the activity of laccases can be measured by the time dependent oxidation of 2,2 -azinobis-(3-ethylbenzothiazoline-6-sulphonic acid) and the product is detected spectrophotometrically (ABTS assay). The activity measurements were performed according to the modifications by Prinz et al. [23]. All measurements were performed in triplicates and the averaged values were calculated. For all activity measurements the average absolute deviation was below 1.22 U L−1 . 2.4. Preparation of ATPS For this study, a system composed of PEG-3000 with an amount of 13% (w/w) and phosphate salt with 6% (w/w) related to the amount of phosphate ions (PO4 3− ) was used, as it was previously investigated for the extractive separation of laccases from the culture supernatant of Pleurotus sapidus [23]. All necessary data about the system’s phases composition and influence of NaCl on laccases activity partition coefficient is given in [23]. As described in the introduction, a partition coefficient around one is desired for CPC separations. By addition of NaCl the activity partition coefficient Kd,act of the laccases can be influenced, whereby increasing the mass fraction of NaCl in the ATPS will decrease the activity partition coefficient of laccase from P. sapidus as measured previously by Prinz et al. [23]. Thereby, Kd,act is defined according to equation 1 as the total activity actTP of laccases measured in the top phase (TP) divided by the total activity actBT in the bottom phase (BT). However, P. sapidus is known to produce different types of laccases [21] and thus, the activity partition coefficient represents a mixed value of all laccases in the supernatant. An amount of 2.2% (w/w) of NaCl was chosen to achieve a Kd of one. Additionally, a system with 2.5% (w/w) NaCl was chosen, resulting in an activity partition coefficient lower than one. Higher amounts of NaCl may results in denaturation of laccases as shown by [23] and thus, were not considered. Kd,act =

2. Materials and methods 2.1. Chemicals PEG-3000 was purchased from Merck KGaA (Darmstadt, Germany). Phosphate salts were used in the hydrated forms, namely K2 HPO4 ·3H2 O and NaH2 PO4 ·2H2 O, from Merck KGaA (Darmstadt, Germany). NaCl was purchased from Carl Roth GmbH + Co., KG (Karlsruhe, Germany). The dye Cibacron brilliant blue, used in the hydrodynamic experiments for selectively coloring of the PEG rich upper phase, was purchased from Kremer Pigmente GmbH & Co., KG (Aichstetten, Germany). Millipore water was used for the analytical solutions in the ABTS assay. The following chemicals, used for the ABTS assay, were purchased from Sigma–Aldrich (St. Louis, MO, USA): 2,2-azino-bis-(3ethylbenzothiazoline-6-sulphonic acid) (ABTS, ≥98.0%, Order-Nr. A1888), sodium acetate (≥99.7%, Order-Nr. S-2889) and acetic acid (≥99.8%, Order-Nr. 33209).

2.2. Laccases from Pleurotus sapidus The fermentation was performed as described by Linke et al. [21]. The culture supernatant was frozen at −20 ◦ C. After thawing,

actTP actBT

(1)

For the preparation of ATPS, stock solutions with deionized water were used. A stock solution was prepared with 50% (w/w) of PEG-3000. The phosphate stock solution consists of the hydrated phosphate salts K2 HPO4 ·3H2 O and NaH2 PO4 ·2H2 O in a mass ratio (mK2HPO4·3H2O /mNaH2PO4·2H2O ) of 0.47 to buffer the solution at a pH value of 7 and was prepared with 25% (w/w) related to the amount of phosphate ions (PO4 3− ). An NaCl stock solution was prepared with 25% (w/w) of the salt. The necessary amount of stock solutions and pure water were calculated while scaled to the total amount of ATPS needed for each separation experiment in CPC (about two liters) and weighted afterwards. All compounds were combined and mixed for several hours. The ATPS were equilibrated overnight and tempered at 25 ◦ C. The settled two phases were not divided in separate flasks for the use in CPC, to keep them in equilibrium state for the whole experimental time. 2.5. CPC experiments For the CPC separation experiments a FCPC® (Fast Centrifugal partition Chromatography) by Kromaton (Annonay, France) was used. The rotor is built up of 20 discs, containing 1320 chambers and 200 mL internal volume in total. The peripheral setup is as described in further studies [24,25]: solvent supply is done by two

C. Schwienheer et al. / J. Chromatogr. B 1002 (2015) 1–7

SD-1 pumps (Agilent, Santa Clara (CA), USA), one for each phase. Six-port valves (Knauer, Berlin, Germany; type: Smartline S6, 3channel) were used to select the phase to be pumped to the CPC and for sample injection. The flow out of the CPC was cooled down by a steel tube in iced water before UV-detection to decrease the noise in the detector baseline as described in a former study [24]. For UVdetection a ProStar 340 detector (Agilent, Santa Clara (CA), USA) at 210 nm and a fraction collector L-7655 (Merck-Hitachi, Darmstadt, Germany) was used. The phase systems used for the CPC experiments were tempered at 25 ◦ C (Lauda; Lauda-Königshofen, Germany; type: Alpha RA8). It has been shown in earlier studies [24] that the retention of stationary phase for ATPS composed of PEG and phosphate salts was better when using the PEG rich upper phase as stationary phase. Thus, the descending mode of operation was chosen where the lower salt rich phase of the ATPS is the mobile phase. The rotational speed was set to 1000 rpm for all experiments within this study. Hydrodynamic measurements in the CPC chamber were performed to determine the volume flow of mobile phase for the separation experiments. Detailed information about hydrodynamic measurement and the experimental setup of a single disc CPC is given in a former publication [24]. 2.5.1. Preparing the injection sample solution Since ATPS are mainly composed of water with the phase forming components enriched above a critical concentration, it is not useful to inject the aqueous culture supernatant directly. This will lead to a dilution of the phase system and thus to a different ATPS or the system will become one-phasic. In each case the equilibrium between mobile and stationary phase in the CPC will be disturbed. To avoid this, the water of 15 mL of the equilibrated lower phase was evaporated using a rotary evaporator (Rotavapor r-205, Büchi AG, Uster, Switzerland) at 40 ◦ C and 10 mbar until the dissolved phase forming components (and supplementary components, e.g., NaCl) remained in the flask. Afterwards, 15 mL of the filtrated culture supernatant was used to re-dissolve these components. This solution was then used as injection sample to the CPC. Alternatively, the culture supernatant for injection could be enriched with the necessary amount of phase forming components and additives like NaCl of the phase to be injected. However, this requires the precise knowledge of the composition of the resulting phases after preparation of the ATPS at the corresponding mixing point. The phase’s composition was not measured during this study. However, for industrial scale application, it might be more economical to measure this once and to prepare the solutions used for injection by adding the corresponding amount of components. 2.5.2. CPC separation experiments of laccases from fermentation broth Prior the separation experiments the rotor had to be filled with stationary phase, i.e., the PEG rich upper phase of the ATPS. To remove the storage solution (ethanol and water in equal volumetric amounts) the lower phosphate rich phase was pumped in ascending mode (5 mL min−1 , 1000 rpm) until at least twice the internal rotor volume was used. Then the upper phase was pumped in descending mode (5 mL min−1 , 1000 rpm) to push the salt phase out. The liquid flow out of the CPC was collected in a volumetric flask (±1 mL) to determine the volume of discharged phase. This procedure was repeated until in each case the internal rotor volume was pushed out. Finally, the upper PEG rich phase as stationary phase remained in the rotor. For the separation experiments the mode of operation and the rotational speed was kept constant and the flow of mobile phase started at a chosen volume flow. The flow was maintained until no further stationary phase was discharged from the rotor and thus the UV-detector did show a straight base line. The eluted liquid

3

volume from the rotor containing discharged stationary phase was collected in a volumetric flask (±1 mL) for calculation of the stationary phase retention (Sf ) according to Eq. (2), where VC is the internal volume of the rotor, Vs,dis the discharged volume of stationary phase and Vdead the dead volume of the system without the rotor (Vdead = 14.3 mL including the 10 mL sample loop). Sf =

VC − Vs,dis + Vdead VC

(2)

After reaching the equilibrium between stationary and mobile phase in the rotor the rotational speed and mobile phase flow were maintained at set values. A sample of the injection solution was taken and maintained at 25 ◦ C for the whole separation time as blank for the later calculation of laccase activity yields. Afterwards, the sample loop was loaded with the injection solution. The injection was started simultaneously with recording the UVchromatogram. 10 mL of the sample was injected corresponding to 5% (v/v) of the internal rotor volume as described by Sutherland and co-workers [7,8]. The fraction collector was started with a dead time of 14–15 min and a sample interval of 20 s for the first 43 samples and 90 s until the end of the separation experiment. These sample intervals had been found to be suitable in preliminary experiments. 2.5.3. Calculation of separation characteristics For the calculation of peak characteristic data, the peaks of the UV-diagram were fitted by exponentially modified Gauss equations (EMG) to consider overlapping peaks [26]. Using the EMG peak fitting the mobile phase retention volume VR,i and peak width wi , as characteristic data for an individual component peak i, were calculated according to Eqs. (3) and (4). Thereby, t,i is the first and  i the second momentum of the peak, xj is the UV-detector signal at the time tj with the sampling interval t and V˙ mob the mobile phase volume flow [27]. nj=1 tj × xj × t VR,i  =t,i ≈ nj=1 xj × t V˙ mob



wi 4 × V˙ mob

2

≈ i2 ≈

nj=1 (tj − t,i )2 × xj × t nj=1 xj × t

(3)

(4)

Based on these values, the separation performance could be calculated. The resolution Ra,b (Eq. (5)) between two peaks indicates the separation performance between the corresponding compounds [27]. Ra,b =

2(VR,1 − VR,2 ) withVR,a > VR,b w1 + w2

(5)

To calculate the laccase activity acti of a selected peak i, Eq. (6) was used. Thereby actj is the activity measured as described in Section 2.3 in a CPC fraction j collected by the fraction collector and Vj is the corresponding fraction volume. The activity peak starts at fraction j = m and ends at fraction j = n. acti =

nj=m actj × Vj

(6)

nj=m Vj

The Laccase’s activity recovery Yi for a selected activity peak can be calculated according to Eq. (7), where Vinj is the injected sample volume and actinj the activity measured in the blank taken before injection. Yi [%] =

acti × nj=m Vj actinj × Vj

× 100%

(7)

The recoveries of several activity peaks can be summed up to calculate the total recovery of laccase activity or the activity loss during CPC separation, respectively.

4

C. Schwienheer et al. / J. Chromatogr. B 1002 (2015) 1–7

Fig. 1. UV-detector signal and activity trend for separation experiments 1 and 2 with different ATPS (2.2 and 2.5% (w/w) of NaCl). The corresponding data for the three main peaks is given in Table 1.

3. Results and discussion 3.1. Selection of operating parameters The retention of stationary phase was measured at varying volume flow of mobile phase as described in earlier studies [24]. The flow of mobile phase and the corresponding retention of stationary phase (calculated by retention videos) are shown in Fig. S2 in the Supplementary material. Compared to other ATPS that were commonly used (e.g., an ATPS composed of PEG-1000/K2 HPO4 /H2 0 with 12.5/12.5/75% w/w [8]), the selected system for this study is relatively stable in CPC. That means, that the phase separation in the chambers is fast, resulting in high stationary phase retention even at comparably high flow rates of mobile phase. For an efficient separation by CPC, this is of crucial importance, as a higher flow rate reduces the time needed for a separation run. As a general phenomenon observable for CPC, mass transfer and therewith separation efficiency normally does not decrease with increasing mobile phase volume flow [28]. This is due to a better dispersion and thus increased interfacial area between the phases when increasing the volume flow. Based on the observations about ATPS selection for use in CPC in [24], the good stationary phase retention might be caused by the PEG of higher molecular mass (PEG-3000). A good retention of more than 70% (v/v) is possible, even at a mobile phase volume flow of 7 mL min−1 . However, at a certain flow rate, the stationary phase retention drastically decreased, which was observable for the selected ATPS between 9 and 11 mL min−1 . This behavior could be noticed for several other ATPS and some aqueous–organic systems [24,25]. It is known from earlier studies, that the stationary phase retention may be slightly lower using the separation rotor (20 discs, 1320 chambers) in contrast to the single disc rotor (66 chambers) used for the hydrodynamic measurements. That is why the mobile phase volume flow was set to 4 mL min−1 for all following experiments. 3.2. Separation experiments An ATPS with 2.2% (w/w) of NaCl was chosen for experiment 1, resulting in equal partitioning of the total laccase activity between mobile and stationary phase (i.e., Kd,act = 1) [23]. With this phase system a separation run was performed and the activity was measured in the collected fractions. The corresponding UV-chromatogram and activity chromatogram is shown in Fig. 1. In the UV-chromatogram three major peaks were obtained and the corresponding peak data is listed in Table 1. Peak 1, a small and narrow peak was obtained at a retention volume of 79 mL,

shortly after the mobile phase dead volume in the rotor (calculated to 58 mL according to the measured stationary phase retention of 71% (v/v)). Peak 2 at 105 mL retention volume is the highest peak and overlaps peak 1. Peak 3 at a retention volume of 247 mL is the broadest peak (compare peak width given in Table 1). The activity chromatogram of the collected fractions shows two major peaks. A first major activity peak obviously belongs to peak 1 and the second activity peak is included in peak 3 of the UV-detector signal. It can be said, that there are at least two different laccases with a different partitioning behavior. Thereby, the partition coefficient of the first peak is very low and thus, the corresponding components mostly partition to the mobile phase and elute close to the mobile phase dead volume peak. Components included in the third peak prefer to partition to the stationary phase and elute in a very broad peak. Hence, the components including the active laccases in peak 3 are strongly diluted by mobile phase. To lower this dilution the mass fraction of NaCl was increased to 2.5% (w/w) (experiment 2), the compromise between a low partition coefficient and deactivation of laccases by NaCl. It was assumed, that all laccases are influenced in the same manner, so that both activity peaks observed will show a lower activity partition coefficient, resulting in a lower dilution especially of the second laccase activity peak. Results of this second separation experiment are also shown in Fig. 1 and Table 1. The small change in amount of NaCl seems not to influence the ATPS and its hydrodynamic behavior strongly, as the stationary phase retention and pressure drop in equilibrium do not change significantly. The elution volume of peaks 1 and 2 in Fig. 1 is almost unchanged, while peak 3 elutes slightly earlier. Since peak 2 is slightly higher and peak 3 lower, it can be assumed that some components are shifted in the direction of peak 2 after increasing the amount of NaCl in the phase system. The maximum of the second activity peak (peak 3) is also shifted to a lower elution volume. Hence, the effect of the amount of NaCl is as suggested. However, the activity peak width of the second activity peak is almost unchanged. A possible explanation is that this peak consists of several laccases, each with a slightly different partitioning behavior. As reported by Linke et al. [29] P. sapidus is able to produce several different laccases. Due to the lower dilution of the laccases, the following experiments were performed with an amount of 2.5% (w/w) in the ATPS. However, as indicated in Table 1, this slightly lowers the resolution between peaks 1 and 3, containing the main laccase activity. Two other experiments, namely experiments 3 and 4 were performed and the UV and activity chromatograms were measured as described above. Both experiments were performed using an ATPS containing 2.5% (w/w) of NaCl, while experiment 3 was performed in the same manner as experiment 1 and 2. To increase the productivity of the CPC separation, the culture supernatant used in experiment 4 was concentrated before injection. Therefore, 30 mL of the filtrated culture supernatant was concentrated using a centrifugal filter Vivaspin 20 by Sartorius (Göttingen, Germany) with a 10 kDa cut-off on a Eppendorf centrifuge 5804 R (Hamburg, Germany) until 1 mL was retained. The retentate in addition with 14 mL of ultrapure water was used to prepare the injection solution as described above. This was done to increase the laccase activity in the injected solution and to remove small compounds (below 10 kDa). The concentrated culture supernatant (experiment 4) was injected after experiment 3 without displacing the stationary phase. By this, the separation conditions were kept constant, which allows a better comparison of the results of both experiments. The plan was to double the injected activity by the concentrating the injection solution. However, according to the values given in Table 2 the activity was only increased by a factor of 1.3 comparing the injection solutions. This might have

C. Schwienheer et al. / J. Chromatogr. B 1002 (2015) 1–7

5

Table 1 Separation characteristics and corresponding UV-peak data for separation experiments with two different ATPS. Phase system consists of PEG-3000 (13% w/w), PO4 3− (6% w/w) and a variation in mass fraction of NaCl. No.

Mass fraction of NaCl (% (w/w))

Sf (%)

p (bar)

VR,Peak1 (wPeak1 ) (mL)

VR,Peak2 (wPeak2 ) (mL)

VR,Peak3 (wPeak3 ) (mL)

R3,1 (−)

1 2

2.2 2.5

71 72

19 19

79.2 (13.1) 81.3 (12.8)

104.8 (17.1) 104.2 (16.7)

246.7 (57) 232.9 (55.3)

4.8 4.11

Fig. 2. UV-detector signal and activity trend for two separation experiments with and without a concentrated injection solution. The phase system consists of PEG3000 (13% w/w), PO4 3− (6% w/w) and NaCl (2.5% w/w).

been caused by the concentration process itself, whereby some laccases might have been deactivated or were adsorbed in the filter. In Fig. 2 the UV- and activity-chromatogram is shown. Comparing both separation runs, experiment 4 with concentration and experiment 3 without concentration of the injected sample, the activity in both activity peaks was in the same order higher as the activity measured in the injected sample (see Table 2). Comparing the UV-detector signals, the major part of the signal of peaks 2 and 3 is missing after the concentration process. Thus, most components that absorb at the detector wavelength have to be of small molecular size (≤10 kDa) like phenolic compounds that may be present in the culture supernatant of P. sapidus according to [29] and were removed during the concentration process. Peak 3 did show a very flat UV-signal after concentration. However, due to the remaining activity, at least the laccases with a larger molecular size are included in peak 3. This indicates that the total amount of laccases is low in the culture supernatant. In peak 2 a small amount of the detected components by the UV detector is left. This is probably caused by the components in the retained volume of 1 mL during the concentration process from which the small components have not been removed. In contrast peak 1 changed only slightly in size, indicating that the UV-signal is produced by a significant amount of higher molecular compounds like laccases. For all separation experiments performed the fractions were analyzed for Lacasse activity, allowing the calculation of activity distribution and yields. The separation results and activity yields are listed in Table 2 for all four experiments. Comparing the activity in the culture supernatant (actferm ) with the activity in the lower phase of the ATPS (actinj ) for experiment 1 to 3, it is obvious that there is a decrease of activity due to the dissolved phase forming components and NaCl. The injected activity is about 80% for experiments 1 and 2 and 73% for experiment 3 compared to the original activity in the filtrated culture supernatant. This is most probably caused by the higher amount of phosphate salts and NaCl, when mixed in the ATPS’s lower phase. For the calculation of the activity yields, the collected fractions 1–43, each containing a specific mobile phase flow out from the CPC were assigned to the first activity peak. The fractions 44 until the end of sampling were assigned to the second activity peak. The corresponding activities, volumes and yields for both activity peaks

are given in Table 2, as well as the overall laccase activity yield after separation in the CPC. The activity (Y1 and Y2 ) of both activity peaks is almost evenly distributed, with respect to the total laccase activity. In the third experiment a slightly higher activity yield can be calculated for the first activity peak. During the CPC separation, the sample is diluted with mobile phase. For peak 1 the components are diluted by a factor of about 6, with respect to the injected volume of 10 mL. If only the fractions 6–23 would be considered in the calculations the dilution can be lowered to a factor of 4.4 with a small loss in activity yield to Y1 = 54.5%. Due to the high affinity to the stationary phase the activity in the second peak was diluted by a factor of approximately 30. This high dilution of the laccases in the second peak is unfavorable for further downstream processing. If the purification of these enzymes is of interest, the ascending mode should be used. This means, that the upper PEG rich phase would be used as mobile phase and thus, components with a high partition coefficient elute faster, resulting in a lower peak broadening. However, as described earlier [24] the separation performance of the CPC will be lowered when using the more viscous PEG rich phase as mobile phase. When evaluating the total yield of laccase activity Ylaccases , no further of activity under the influence of CPC conditions during the separation process could be observed. The activity actually increases slightly as can be seen by the yields of more than 100%. Similar results were observed in earlier studies using a single mixer settler apparatus [23]. Recoveries of more than 100% are possible when an activation of the enzymes takes place. This can be affected by a change of the physiological environment. For the calculation of the activity yields a blank from the injected solution was taken and maintained at 25 ◦ C for the same time that was required for the separation experiment. The blank consisted of the culture supernatant with the phase forming components of the lower phase, i.e., a low concentration of PEG. As measured earlier for the investigated system, PEG seems to stabilize the laccase and increase their activity. In the CPC device a PEG rich upper phase is present as stationary phase with which the proteins are interacting according to their partitioning behavior. This can result in a yield of more than 100%. The total activity of the laccases investigated could be preserved during the separation in the CPC. Conditions like shear forces during dispersing and settling of the mobile phase seem not to influence the laccase activity significantly. Thus, the separation conditions in the CPC could be regarded as gentle and in combination with the gentle media properties of ATPS a suitable method for the purification of biochemically produced laccases was demonstrated. The purification of biochemical products by CPC represents a complementary method e.g., to classical extraction using ATPS (i.e., ATPE) with the advantages of a higher loading capacity and the flexibility of stationary phase (concerning selection, renewing and replacement). A combination of ATPE for product clarification and capturing (e.g., as described by [30,31]) followed by CPC using ATPS for further purification might be of particular interest, as both techniques can be used with the same or similar ATPS, respectively. For example, the product after back-extraction with ATPE is concentrated in one phase of an ATPS. After adjusting the partition coefficient to a suitable value for CPC separation, e.g., by adding NaCl or other displacement salts, this concentrate could then be injected to the CPC apparatus for further and more selective purification.

57.4 54.7 60.0 59.1

10.2 10.0 6.3 10.0

308.0(4) 312.0 308.7(5) 308.7(5)

57.1 57.9 48.6 57.4

114.5 112.6 108.6 116.6

4. Conclusion Within this study, the potential applicability of the separation of active enzymes by CPC was faced. Therefore, active laccases from a P. sapidus supernatant were separated by CPC using an ATPS as phase system. As a separation result, two main laccase activity peaks could be detected, as well as a third peak that mainly consists of low molecular weight compounds (≤ 10 kDa). The total amount of laccases and higher molecular compounds (e.g., other proteins) in the culture supernatant was low. The activity of the laccases was measured for fractions of different separation experiments, allowing the calculation of an overall activity yield. The activity yield was always higher than 100%, related to the injected activity in each experiment. This indicates that a major part of the laccases seems to remain active during the separation process. Due to the environmental activity dependency of enzymes, it cannot be said, whether some laccases were deactivated or even destroyed. In general, it can be said, that the CPC in combination with ATPS as phase system provides suitable conditions for a gentle separation of active compounds. The shear and interfacial forces due to the equipment conditions did not lower the activity of the laccases significantly. However, this observation can not necessarily be transferred to all biologically active compounds. Hence, the activity loss during a separation process should be checked in each individual case, but CPC so far seems to be a promising unit operation for the purification of biochemical products. Acknowledgement

58.7 58.7 58.7 58.7

The authors gratefully thank Prof. Dr. Holger Zorn (Justus-Liebig Universität Gießen, Institut für Lebensmittelchemie & Lebensmittelbiotechnologie) for providing the Pleurotus sapidus cell supernatant used for the separation experiments. Appendix A. Supplementary data

71 72 67 67

691.8 671.2 544.2 542.8

549.8 538.1 401.1 536.3

53.8 50.2 41.0 54.1

Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jchromb.2015. 07.050

1 2.2 2 2.5 3 2.5 4(3)2.5

No.Mass fraction of NaCl (% (w/w))Sf (%)actferm (1) [U L−1 ]actinj (2) [U L−1 ]act1 [U L−1 ] (m = 1, n = 43)V1 (mL) (m = 1, n = 43)

Y1 (%) (m = 1, n = 43)act2 (U L−1 ) (m = 44, n = end)V2 (mL) (m = 44, n=end)Y2 (%) (m = 44, n = end)Ylaccases (%)

C. Schwienheer et al. / J. Chromatogr. B 1002 (2015) 1–7 Table 2 Separation results and activity yields of laccases for different separation experiments. (1) Experiments 1 and 2 and experiments 3 and 4 with same frozen sample from culture supernatant, separately filtrated. (2) 10 mL injection volume. (3) Filtrated cell supernatant was concentrated before injection. (4) Sampling interval was 1 min. (5) Wait time of fraction collector was raised to 15 min due to the lower stationary phase retention.

6

References [1] P.A.J. Rosa, I.F. Ferreira, A.M. Azevedo, M.R. Aires-Barros, Aqueous two-phase systems: a viable platform in the manufacturing of biopharmaceuticals, J. Chromatogr. A 1217 (2010) 2296–2305. [2] U. Gottschalk, Bioseparation in antibody manufacturing: the good, the bad and the ugly, Biotechnol. Prog. 24 (2008) 496–503. [3] K.D. Yoon, Y.-W. Chin, J. Kim, Centrifugal partition chromatography: application to natural products in 1994–2009, J. Liq. Chromatogr. Relat. Technol. 33 (2010) 1208–1254. [4] R.C. Bruening, F. Derguini, K. Nakanishi, Preparative scale isolation of 11-cis-retinal from isomeric retinal mixture by centrifugal partition chromatography, J. Chromatogr. A 357 (1986) 340–343. [5] J.B. Friesen, G.F. Pauli, G.U.E.S.S.—a generally useful estimate of solvent systems for CCC, J. Liq. Chromatogr. Relat. Technol. 28 (2005) 2777–2806. [6] W. Murayama, T. Kobayashi, Y. Kosuge, H. Yano, Y. Nunogaki, K. Nunogaki, A new centrifugal counter-current chromatograph and its application, J. Chromatogr. A 239 (1982) 643–649. [7] I.A. Sutherland, G. Audo, E. Bourton, F. Couillard, D. Fisher, I. Garrard, P. Hewitson, O. Intes, Rapid linear scale-up of a protein separation by centrifugal partition chromatography, J. Chromatogr. A 1190 (2008) 57–62. [8] I. Sutherland, P. Hewitson, R. Siebers, R. van den Heuvel, L. Arbenz, J. Kinkel, D. Fisher, Scale-up of protein purifications using aqueous two-phase systems: comparing multilayer toroidal coil chromatography with centrifugal partition chromatography, J. Chromatogr. A 1218 (2011) 5527–5530. [9] N. Mekaoui, K. Faure, A. Berthod, Advances in countercurrent chromatography for protein separations, Bioanalysis 4 (2012) 833–844. [10] S.A. Oelmeier, C. Ladd Effio, J. Hubbuch, High throughput screening based selection of phases for aqueous two-phase system-centrifugal partitioning

C. Schwienheer et al. / J. Chromatogr. B 1002 (2015) 1–7

[11]

[12]

[13]

[14] [15] [16]

[17]

[18]

[19] [20] [21]

chromatography of monoclonal antibodies, J. Chromatogr. A 1252 (2012) 104–114. L. Boudesocque, R. Kapel, C. Paris, P. Dhulster, I. Marc, J.-H. Renault, Concentration and selective fractionation of an antihypertensive peptide from an alfalfa white proteins hydrolysate by mixed ion-exchange centrifugal partition chromatography, J. Chromatogr. B 905 (2012) 23–30. S. Morel, A. Landreau, V.H. Nguyen, S. Derbré, P. Grellier, P. Le Pape, F. Pagniez, M. Litaudon, P. Richomme, Preparative Isolation, fast centrifugal partition chromatography purification and biological activity of cajaflavanone from Derris ferruginea stems, Phytochem. Anal. 23 (2012) 152–158. O. Couillerot, S. Loqman, A. Toribio, J. Hubert, L. Gandner, J.-M. Nuzillard, Y. Ouhdouch, C. Clément, E.A. Barka, J.-H. Renault, Purification of antibiotics from the biocontrol agent Streptomyces anulatus S37 by centrifugal partition chromatography, J. Chromatogr. B 944 (2014) 30–34. P.-A. Albertson, Chromatography and partition of cells and cell fragments, Nature 177 (1956) 771–774. Kula, Trends and future prospects of aqueous two-phase extraction, Bioseparation 1 (1990) 181–189. K. Berggren, H.-O. Johansson, F. Tjerneld, Effects of salts and the surface hydrophobicity of proteins on partitioning in aqueous two-phase systems containing thermoseparating ethylene oxide–propylene oxide copolymers, J. Chromatogr. A 718 (1995) 67–79. Y. Shibusawa, Y. Eriguchi, Y. Ito, Purification of lactic acid dehydrogenase from bovine heart crude extract by counter-current chromatography, J. Chromatogr. B 696 (1997) 25–31. K.L. Wade, Y. Ito, A. Ramarathnam, W.D. Holtzclaw, J.W. Fahey, Purification of active myrosinase from plants by aqueous two-phase counter-current chromatography, Phytochem. Anal. 26 (2015) 47–53. A. Berthod, Countercurrent Chromatography, Elsevier, 2002. Y. Ito, M. Weinstein, I. Aoki, R. Harada, E. Kimura, K. Nunogaki, The coil planet centrifuge, Nature 212 (1966) 985–987. D. Linke, H. Bouws, T. Peters, M. Nimtz, R.G. Berger, H. Zorn, Laccases of Pleurotus sapidus: characterization and cloning, J. Agric. Food Chem. 53 (2005) 9498–9505.

7

[22] A. Majcherczyk, C. Johannes, A. Hüttermann, Oxidation of aromatic alcohols by laccase from Trametes versicolor mediated by the 2,2-azino-bis-(3-ethylbenzothiazoline-6-sulphonic acid) cation radical and dication, Appl. Microbiol. Biotechnol. 51 (1999) 267–276. [23] A. Prinz, J. Hönig, I. Schüttmann, H. Zorn, T. Zeiner, Separation and purification of laccases from two different fungi using aqueous two-phase extraction, Process Biochem. 49 (2014) 335–346. [24] C. Schwienheer, J. Merz, G. Schembecker, Selection and use of poly ethylene glycol and phosphate based aqueous two-phase systems for the separation of proteins by centrifugal partition chromatography, J. Liq. Chromatogr. Relat. Technol. 38 (2014) 929–941. [25] S. Adelmann, T. Baldhoff, B. Koepcke, G. Schembecker, Selection of operating parameters on the basis of hydrodynamics in centrifugal partition chromatography for the purification of nybomycin derivatives, J. Chromatogr. A 1274 (2013) 54–64. [26] P. Naish, S. Hartwell, Exponentially modified Gaussian functions—a good model for chromatographic peaks in isocratic HPLC? Chromatographia 26 (1988) 285–296. [27] H. Schmidt-Traub (Ed.), Preparative Chromatography, Wiley-VCH Verlag GmbH & Co., KGaA, 2005. [28] D. Armstrong, G. Bertrand, A. Berthod, Study of the origin and mechanism of band broadening and pressure drop in centrifugal countercurrent chromatography, Anal. Chem. 60 (1988) 2513–2519. [29] D. Linke, H. Zorn, B. Gerken, H. Parlar, R.G. Berger, Laccase isolation by foam fractionation—new prospects of an old process, Enzyme Microb. Technol. 40 (2007) 273–277. [30] R.N. D’Souza, A.M. Azevedo, M.R. Aires-Barros, N.L. Krajnc, P. Kramberger, M.L. Carbajal, M. Grasselli, R. Meyer, M. Fernandez-Lahore, Emerging technologies for the integration and intensification of downstream bioprocesses, Pharm. Bioprocess 1 (2013) 423–440. [31] A. Prinz, K. Koch, A. Górak, Z. Tim, einer, Multi-stage laccase extraction and separation using aqueous two-phase systems: experiment and model, Process Biochem. 49 (2014) 1020–1031.