Quick Delivery of Aldehyde Dehydrogenase into Yeast Vacuoles by QRPL Peptide Sequence on Carboxypeptidase Y

Background The signal peptide sequence is known to increase transport eciency to organelles in eukaryotic cells. In this study, we focus on the signal peptide of the vacuolar protein for vacuolar targeting. Results The signal peptide sequence QRPL of carboxypeptidase Y (CPY), a vacuolar protein, was inserted inside the green uorescent protein (GFP) that does not locate in vacuole for vacuolar targeting. The protein location was then conrmed by confocal microscopy. Fascinatingly, the green uorescent protein that contains QRPL inside the sequence could be expressed faster than its natural form (within 1 hour after induction). In addition, aldehyde dehydrogenase 6 (ALD6), a cytosolic protein has engineered the sequence with QRPL to be transported to the vacuole. The aldehyde removal activity of ALD6 protein in the recombinant yeast was then analyzed by measuring the luminescent intensity in Vibrio scheri . Conclusions In summary, the signal peptide QRPL could be used not only to transport target proteins accurately to vacuole but also to improve the protein activity, as well as to shorten the induction time.


Abstract
Background The signal peptide sequence is known to increase transport e ciency to organelles in eukaryotic cells. In this study, we focus on the signal peptide of the vacuolar protein for vacuolar targeting. Results The signal peptide sequence QRPL of carboxypeptidase Y (CPY), a vacuolar protein, was inserted inside the green uorescent protein (GFP) that does not locate in vacuole for vacuolar targeting. The protein location was then con rmed by confocal microscopy. Fascinatingly, the green uorescent protein that contains QRPL inside the sequence could be expressed faster than its natural form (within 1 hour after induction). In addition, aldehyde dehydrogenase 6 (ALD6), a cytosolic protein has engineered the sequence with QRPL to be transported to the vacuole. The aldehyde removal activity of ALD6 protein in the recombinant yeast was then analyzed by measuring the luminescent intensity in Vibrio scheri . Conclusions In summary, the signal peptide QRPL could be used not only to transport target proteins accurately to vacuole but also to improve the protein activity, as well as to shorten the induction time.

Background
The vacuole is known as a cellular organelle in Saccharomyces cerevisiae. Many researchers have reported that vacuoles participate in vesicular tra cking processes 1,2 . The vacuole of S. cerevisiae contains about (50-60) acid hydrolases that play essential roles in several biological processes. Some special vacuolar enzymes can secrete cytotoxic molecules outside the cells. The degradable function of molecules typically takes place by endocytosis, after being activated as a response to stressful conditions 3,4 . In previous studies, we have investigated the antimicrobial activity of vacuoles from eukaryotes 5,6 . Here, we focus on the tra cking of vacuolar proteins by signal peptide sequence. Various signal peptide sequences that have been reported in most vacuolar enzymes are related to tra cking in cellular mechanisms [7][8][9][10] . Most vacuolar enzymes are delivered from the endoplasmic reticulum and tra cked to the Golgi complex, the pre-vacuole, or the late endosome. They are involved in the CPY pathway, in which proteins are transported from the Golgi complex to the vacuole 9 . Other known mechanisms of protein tra cking include the autophagy pathway, in which unnecessary vacuole or proteins can be converted into double-membraned autophagosomes 10 . In addition, the late endosome/multivehicle body pathway (MVB pathway) is related to delivery protein to the vacuole by endosomes. In a similar fashion, all proteins are transported in the cytosol by signaling mechanism. In 1980, signal peptide sequences were intensively studied [11][12][13] . Speci cally, vacuolar proteins contain various signal peptide sequences for protein tra cking in S. cerevisiae. Signal peptide sequences related to the transportation of protein or cargos to the vacuole and sorting mechanisms have been reported 12,13 . In our study, we focus on the QRPL signal peptide sequence of Carboxypeptidase Y (CPY), which is a typical vacuolar protein in S. cerevisiae [14][15][16][17][18] . Transfer vectors were designed by using QRPL to deliver the protein of interest to the vacuole that is not expressed in vacuole. QRPL has inserted in the middle of the CPY sequence at the 24th to 27 th residues for the development of transfer vector.
Furthermore, we tried to transport aldehyde dehydrogenase 6 (ALD6), which has been reported as an aldehyde removal protein through NAD-dependent reaction in the cytosol, to the vacuole [19][20][21][22] . Aldehydes are known to cause problems to the environment and animals including humans, by creating exogenous and endogenous precursors during numerous physiological processes 22 .
In this study, MBTL-Q-DJ1 and MBTL-Q-DJ2 were developed using the QRPL signal peptide sequence with target proteins as transfer vectors. Those protein expression levels were evaluated by uorescence microscopy. MBTL-Q-DJ1 is a model of the transfer vector for the transport of GFP to the vacuole, and MBTL-Q-DJ2 contained ALD6 for the transport of a foreign protein to the vacuole by signal peptide sequence. Furthermore, the aldehyde removal e ciency of ALD6 was then evaluated. Therefore, this development of the transfer vector using a signal peptide sequence is useful for delivering proteins or peptide to the vacuole. In addition, protein that contains QRPL sequence could be expressed in vacuole after only 1 hour of induction. This investigation might be a potential approach for the therapy of vacuolar diseases.

Results
Construction of pYES2::QRPL::GFP and pYES2::QRPL::ALD6::GFP for vacuolar targeting Carboxypeptidase Y (CPY) is known to be transported from ER to vacuole, including a signal peptide sequence near the N-terminal of the DNA fragment (QRPL sequence). In this study, we considered the possibility of the signal peptide delivering foreign proteins to the vacuole, after generation by the nucleus.
Here, the yeast vacuole was stained with LysoTracker-Red; the uorescent signals of vacuole and GFP were observed by uorescence microscopy, to con rm the location of GFP in the vacuole. The merged images show a yellow color indicating the combination of green and red color, whereas the mock strain did not show any green or yellow color. The result of the co-localization in vacuole means that GFP is successfully expressed in the vacuole ( In the same way, we inserted the QRPL sequence inside aldehyde dehydrogenase 6 (ALD6), followed by GFP to determine the localization of ALD6.

Effect of QRPL sequence on the induction time for protein expression
The induction with galactose is essential for protein expression using GAL promoter in pYES2 vector. A regular induction of GAL promoter in S. cerevisiae that was used in this study requires (10-12) h 23 ; thus we wanted to con rm whether the QRPL could provoke any effect on the induction process. Figure 3 shows different expression levels of green uorescent protein after various induction times of (0.5, 1, 3, 6, and 12) h of the strains MBTL-Q-DJ1 and MBTL-Q-DJ2. Interestingly, the yellow color in the merged images of recombinant yeasts provided that the GFP of MBTL-Q-DJ1 and ALD6 protein of MBTL-Q-DJ2 were successfully transported and expressed in the vacuole after 1 h of induction. These strains occupy QRPL signal peptide sequence in the middle of the DNA sequence that leads the proteins to the vacuole after only 1 h, whereas the mock strain, which does not carry QRPL, could not show the co-localization of vacuole with the proteins of interest during induction.
Aldehyde-reducing ability of the strain MBTL-Q-DJ2 To evaluate the aldehyde dehydrogenase activity of ALD6 in the recombinant yeast, formaldehyde was exposed with the vacuolar enzymes that included ALD6 for 1 h at 62.5, 125, 250, 500, and 1,000 ppm.
The mixture was then centrifuged at 20,000× g at 4 °C for 10 min to remove pellet, and the supernatant collected for luminescent reaction with Vibrio scheri. V. scheri is known as having luminescence that could be decreased by being stressed. In this study, the remnant of aldehyde would react with the cell to reduce the luminescence. Table 1 shows the formaldehyde removal e ciency of MBTL-Q-DJ2 strain compared with the mock and control sample (no treatment), by measuring the luminescent intensity in V. scheri. A high value of luminescence means V. scheri was alive by reducing formaldehyde. The vacuolar enzyme that included ALD6 showed a high e ciency of decrease in formaldehyde at 62.5 and 125 ppm, whereas the cytoplasmic enzyme did not show any in uence. Table 1 The formaldehyde removal ability of ALD6. The value in the table presents the luminescent intensity of all samples; "control" means sample without treatment. The decreased e ciency of aldehyde was compared with those of mock and recombinant strain by measuring the luminescent intensity in V. scheri. The high value of luminescence means V. scheri was alive by reducing formaldehyde. ALD 6 showed excellent e ciency to decrease formaldehyde at (62. 5 and 125) ppm, while the cytoplasmic enzyme did not show any reduction of formaldehyde effect. Furthermore, it is necessary to determine the optimized concentration of vacuolar enzymes for the highest reduction effect. Figure 4 presented the luminescent level after exposing formaldehyde to different concentrations of vacuolar enzymes of 300, 500 and 1,000 μg/μL. The highest gap of luminescence was observed at 500 μg/μL of the vacuolar enzyme from MBTL-Q-DJ2.

Discussion
Many proteins contain signal peptide sequences, which enable them to be delivered accurately for cellular mechanisms. The delivery of proteins from the late Golgi in yeast depends on which protein tra cking pathway that targets protein is involved. In this work, the signal peptide sequence of vacuolar protein Carboxypeptidase Y (CPY) of S. cerevisiae, QRPL, was used. Even though QRPL was investigated a long time ago, this sequence was not studied in the protein targeting eld. Here, the transfer vector that was designed with QRPL signal peptide sequence demonstrated the capability of leading GFP and ALD6, a cytosolic protein, to the vacuole. Therefore, this transfer vector could be used to deliver the protein of interest or peptide to the yeast vacuole.
Interestingly, QRPL was found to accelerate the induction process of the protein of interest in the vacuole.
The recombinant yeast S. cerevisiae s2805 that contains the plasmid pYES2 with GAL promoter regularly requires (10-12) h of galactose induction for the protein expression 23 ; however, the presence of QRPL in the recombinant yeast quickened the protein expression, which was observed 1 h after adding galactose.
Further investigation with a shorter induction time should be considered for clearer understanding of this phenomenon.
In conclusion, we designed a useful transfer vector that could deliver a target protein to the vacuole using the signal peptide sequence of the vacuolar protein. The signal peptide sequence of CPY, QRPL was inserted at the 24th to 27th residues with the target protein, and it was possible to transport it to the vacuole. Therefore, our investigation provided a promising approach for protein targeting that could be applied in varied elds.

Materials And Methods
Culture of S. cerevisiae, chemicals, and materials

DNA isolation, manipulation, and transformation
Genomic DNA was prepared from S. cerevisiae s2805 using Wizard SV genomic DNA kit (Promega, Madison, USA), according to the manufacturer's instructions. Plasmid DNA, such as pYES2.0 (Invitrogen, USA) and pEGFP-C1 (Clontech Laboratories, Inc., USA), was prepared from recombinant E. coli cells using an alkaline lysis technique with a QIA spin Miniprep kit (Qiagen, Germany). All of the DNA modi cations were analyzed by agarose gel electrophoresis, and ligation was performed using standard procedures (Sambrook et al., 2001). The PCR experiments were carried out using a T Gradient Thermocycler (Biometra, Germany), Ex TaqTM DNA polymerase (Takara Bio Inc., Japan), and genomic DNA as the template.
The PCR products were puri ed using a QIA quick PCR puri cation kit (Qiagen, Germany). The oligonucleotides used for PCR ampli cation were purchased from AccuOligo (Bioneer Co., Korea). Figure   1 describes two diagrams of the transfer vector with QRPL signal peptide sequence (QRPL). The transformation to E. coli was performed by electroporation with an Electro cell Manipulator (BTX Technologies Inc., USA); and yeast transformation was performed by lithium acetate method 24 .
Design of transfer vector (pYES2::QRPL::GFP) Plasmids used in this study were constructed for the expression of QRPL signal peptide sequence with green uorescence protein (GFP) (Fig. 1 a). To amplify the QRPL gene (5'-CAA CGG TCT TTG -3'), the primer was designed to include QRPL signal peptide sequence from CPY in the pYES2 plasmid. PCR products were prepared of the two parts of the gene, then primer R1 and F2 included QRPL with EarI restriction enzyme site. PCR via primer: Primer-F1 ( Plasmids used in this study were constructed for the expression of QRPL signal peptide sequence with Aldehyde dehydrogenase 6 (ALD6). The DNA containing QRPL with ALD6 was synthesized by Bioneer Co., Korea. The PCR product was inserted into the BamHI and NotI site of pYES2.0, to make pYES2::QRPL::ALD6 plasmid. After that, the GFP gene was ampli ed from pEGFP using PCR via primers: Primer-GFP-F (5'-ATT GCG GCC GCA GTG AGC AAG GGC GAG G -3') and Primer-GFP-R (5'-GGT GCA TGC CTT GTA CAG CTC ATC CAT -3'). The DNA fragment of GFP was then cloned into the NotI and SphI site to make pYES2::QRPL::ALD6::GFP. This vector was transformed into S. cerevisiae to make recombinant strain MBTL-Q-DJ2 (Fig. 1 b).

Fluorescence microscopy
Recombinant and mock strains were cultured in SD and SG medium. The cells were washed with PBS buffer and stained with 100 nM Lyso-Tracker Red DND-99 (Molecular Probes, Leiden, The Netherlands) in PBS for 30 °C for 10 min. All strains were washed by PBS twice, and the intensity of protein expression observed by uorescence microscopy (Zeiss, Germany). Fluorescent images were analyzed using the Zeiss image Browser.

Isolation of vacuole and vacuolar enzyme from recombinant yeast
Recombinant and mock strains were both cultured in SD medium. To induce pYES2.0 plasmid, cells were cultured on SG medium for 12 h. All strains were harvested after growing at OD 600 0.8-1.0, then the vacuole was isolated from mock and recombinant strains.
The 0.1 M Tris-SO 4 buffer (100 mM Tris-SO 4 (pH 9.4), 10 mM dithiothreitol (DTT)) was mixed with harvesting cells. This mixture was incubated at 30 °C at 90 rpm for 10 min and centrifuged at 3,000 rpm for 5 min. The supernatant was discarded, and the mixed pellet exposed to Lyticase (2,000 U/mL) (Sigma, St. Louis, USA) for 1 h, to make it exible in cell membranes. This was followed by centrifugation for 5 min, and the pellet was washed twice with Sorbitol K + phosphate buffer (20 mM K + phosphate buffer, 1.2 M Sorbitol). In the next step, breaking buffer (20 mM Tris-Cl (pH 7.4), 0.6 mM Sorbitol, and 1 mM phenyl methane sulfonyl uoride (PMSF)) was mixed with pellet, and ultra-sonication applied at 40 W for 30 min (20 s on / 10 s off pulse) in the ice. The mixture was centrifuged for 10 min, and the supernatant was discarded. The pellet was mixed again with breaking buffer, and a second ultra-sonication was applied for 20 min (10 s on / 10 s off pulse). Lastly, the rst centrifuge (500× g, 5 min) was made to collect supernatant, and the second centrifuge (20,000× g, was made for 30 min at 4 °C). The pellet contains vacuoles from cells 5,25,26 .
The vacuolar enzyme was extracted by 0.1 % NP40 buffer containing 1 mM DTT at a ratio of 1:1. The mixture was vortexed for 10 min, followed by the reaction in ice for 30 min. After incubation, the mixture was centrifuged at 13,000 rpm at 4 °C for 10 min. The supernatant contained vacuolar enzymes that would be used in the aldehyde reduction assay.

Luminescence assay
The vacuolar enzyme was exposed to formaldehyde to evaluate the e ciency of aldehyde reduction. Formaldehyde was prepared at 62.5, 125, 250, 500, and 1,000 ppm, and then the vacuolar enzyme was exposed to the formaldehyde at room temperature for 1 h. After that, the mixture was centrifuged at 20,000× g at 4 °C for 10 min, the supernatant was collected for luminescent reaction with Vibrio scheri at a ratio of 1:1 at 30 °C for 30 min, and the luminescence was measured by GloMax ® Explorer system (Promega Co., Ltd., USA) 27,28 .

Data analysis
All the data were obtained from three independent samples, run simultaneously for error analysis. The data were analyzed using Sigma Plot (SPS, Chicago, IL. USA), and the results were reported with the standard deviations. In addition, the correlation between cell viability and several experimental conditions was reported. A p-value of <0.05 was considered signi cant. The designed recombinant plasmids that contain QRPL signal peptide sequence. (a) QRPL is inserted to become the 24th and 27th amino acids of GFP. This transfer vector is called MBTL-Q-DJ1. (b) QRPL is inserted to become the 24th and 27th amino acids of Aldehyde dehydrogenase 6 (ALD6). This transfer vector is called MBTL-Q-DJ2. The protein expression at different induction time by galactose of MBTL-M-DJ, MBTL-Q-DJ1, and MBTL-Q-DJ2. The "LysoTracker" column presents the red signal of yeast vacuoles after staining with LysoTracker-Red; the "GFP" column shows the green uorescent signals, and the "Merged" column means the combination of red and green signals, which re ects the expression of GFP in vacuoles.