The Reporter System for GPCR Assay with the Fission Yeast Schizosaccharomyces pombe

G protein-coupled receptors (GPCRs) are associated with a great variety of biological activities. Yeasts are often utilized as a host for heterologous GPCR assay. We engineered the intense reporter plasmids for fission yeast to produce green fluorescent protein (GFP) through its endogenous GPCR pathway. As a control region of GFP expression on the reporter plasmid, we focused on seven endogenous genes specifically activated through the pathway. When upstream regions of these genes were used as an inducible promoter in combination with LPI terminator, the mam2 upstream region produced GFP most rapidly and intensely despite the high background. Subsequently, LPI terminator was replaced with the corresponding downstream regions. The SPBC4.01 downstream region enhanced the response with the low background. Furthermore, combining SPBC4.01 downstream region with the sxa2 upstream region, the signal to noise ratio was obviously better than those of other regions. We also evaluated the time- and dose-dependent GFP productions of the strains transformed with the reporter plasmids. Finally, we exhibited a model of simplified GPCR assay with the reporter plasmid by expressing endogenous GPCR under the control of the foreign promoter.


Introduction
In mammalians, G protein-coupled receptors (GPCRs) constitute the largest and most divergent protein families. GPCRs are activated by hormones, odorants, peptides, neurotransmitters, and so on [1][2][3]. Not surprisingly, it is oen the case that GPCRs are associated with various diseases; GPCRs are one of the most potential drug targets, along with enzymes. However, it is difficult to reproduce appropriate and functional GPCR expression in heterologous cells due to the distinctive conformation and unknown mechanism of its trafficking and folding. erefore, a large number of GPCRs still do not determine the corresponding ligands. Today, a wide variety of hosts and their transformants have been developed to resolve these problems [4][5][6][7][8][9][10].
Fission yeast, Schizosaccharomyces pombe, is a unicellular eukaryote and the best model organism especially in cell cycle. Fission yeast shares more similarities to mammals in terms of mRNA splicing, posttranslational modi�cation, and so on than other yeasts including budding yeast, Saccharomyces cerevisiae [11]. Indeed, many genes of �ssion yeast can be complemented by the mammalian homologs, and many mammalian proteins are successfully expressed in �ssion yeast cells [12][13][14][15][16][17]. Although �ssion yeast has so many advantages in expressing heterologous protein, it is preferable not to select �ssion yeast in the GPCR study.
Fission yeast endogenously follows two alternative GPCR pathways. One is the nutrient (glucose) signaling pathway [18] and the other is the mating pathway. is mating pathway closely resembles the mitogen-activated protein kinase (MAPK) pathway in mammalian cells [19,20]. A �ssion yeast cell usually divides by mitosis in rich medium. However, exposed to nutrient starvation (particularly nitrogen starvation), the cell converts mitosis to meiosis by which the cell develops into robust spores with opposite mating type via pheromone communication for resistance to environmental stress [20,21]. One mating type, plus cell (h+ or P cell), secretes diffusible peptide pheromone, which is P-factor encoded by map2 gene. Another mating type, minus cell (h− or M cell), receives P-factor by the endogenous GPCR, which is Mam2 protein encoded by mam2 gene [22,23]. e pheromone reception triggers sequential and synergistical activation of mating speci�c genes, which results in morphological changes for later conjugation to opposite mating type [23,24].
In this study, we proposed the screening system with �ssion yeast through the endogenous mating pathway to �nd the ligand of a number of orphan GPCRs. As a reporter gene, the intense genes such as LacZ or luciferase are oen utilized in GPCR assay with yeasts [8][9][10]25], but we used green �uorescent protein (GFP) with a focus on easy and inexpensive detection in spite of a very weak signal compared to that of intense genes. In order to overcome this disadvantage, we constructed the tractable reporter plasmids. Escherichia coli strain DH5 was used for the subcloning of the plasmid preparation.

Materials and Methods
Peptide and oligonucleotide synthesis were performed by Operon Co. Ltd. (Tokyo, Japan). Each peptide was prepared as a stock solution of 1 mM in Milli-Q water and stored at −80 ∘ C.

Plasmid Constructions and Transformations
2.2.1. Gene Disruption. e gene disruption was performed by standard homologous recombination method. e detailed construction of the plasmids for gene disruption was described previously [26]. Brie�y, about 1000 bp of 5 ′ and 3 ′ �anking sequences of a target gene were used as the chromosomal integration regions. To delete the ura4 selection marker, the ura4 gene was sandwiched with about 200 bp of 3 ′ �anking sequence of the target gene. For negative selection of the ura4-cells, the cells were plated onto the YES-FOA plates (0.5% Yeast Extract, 3% glucose and SP Supplements, 2% Bacto agar, 0.1% 5-�uoroorotic acid). e resultant ura4-cells were available for the subsequent gene recombination.

Reporter
Plasmids Based on pAL7. e reporter plasmids were engineered based on pAL7 which was a high copy plasmid for �ssion yeast [27]. e schematic illustration was described in Figure 1. e main fragment including the replication origin and selection marker was ampli�ed from pAL7 using primers pAL7invforward and pAL7invreverse. e fragment was ligated with a series of a certain promoter, GFP, and Lipocortin I (LPI) terminator. To replace  Table 1 were constructed in this way.
the promoter region with each upstream region of the pheromone-dependent gene, inverse PCR was performed from the resultant plasmid using primers pAL7invreverse and GFPORFforward. To replace the LPI terminator, inverse PCR was performed using primers pAL7invforward and GFPORFreverse. Open reading frame (ORF) of GFP used in this study was ampli�ed from a Monster Green Fluorescent Protein phMGFP Vector (Promega Japan, Tokyo, Japan). LPI terminator was ampli�ed from a pSU1Z vector (Asahi Glass Co., Ltd, Tokyo, Japan). Upstream and downstream regions were ampli�ed from a genomic DNA, and the primers were listed in Table 3. To express mam2 gene under the control of other promoters, we replaced hCMV promoter with nmt1 or urg1 promoter. e region of nmt1 promoter was referred to pREP1 vector and that of urg1 was referred to the region reported by Watt et al. [28]. To repress the nmt1 promoter, more than 15 M of thiamine were added to the medium.
To induce the nmt1 promoter, the cells were incubated in fresh EMM without thiamine for 20 hours. During the preincubation, urg1 promoter was consistently repressed by the absence of uracil. On the contrary, urg1 promoter was constitutively activated during the assay by the nitrogen starvation.
All PCR products used for the plasmid construction were prepared using the KOD-plus-Neo (Toyobo, Osaka, Japan) in accordance with the supplier's instructions. All fragments without the replication origin of Escherichia coli were phosphorylated with T4 Polynucleotide Kinase (Toyobo) and the ligation reactions were performed with a Ligation-Convenience Kit (Nippon gene, Tokyo, Japan). e sequence of each plasmid was veri�ed by the nucleotide sequence analysis.

2.2.4.
Transformation. e �ssion yeast was transformed using a lithium acetate method [27,29]. Transformed cells were plated onto MMA plates or MMA plates supplemented with leucine. e plates were incubated at 32 ∘ C for 2-3 days, and positive colonies were selected. To check for correct integration, PCR was performed on the extracted DNA using SapphireAmp Fast PCR Master Mix (Takara Bio Inc., Otsu, Japan). All the parental strains in this study lacked the function of both leu1+ and ura4+. Before assay, leu1-32 was complemented with LEU2 derived from pAL7 (or its derivatives) and ura4-D18 was complemented with ura4 gene derived from pSU1Z vector (or its derivatives). Each value of �uorescence intensity was mean (±SD) from more than three independent experiments. a (−): incubation without P-factor. b (+): incubation with P-factor. c SNR: signal to noise ratio� SNR was obtained by dividing the �uorescence intensity at 24 h (+) by that at 24 h (−).

T 3: Primer information.
Primer name Target gene +945 AAGTTTAATATCGGAAAATTTAA CGGAAGTTAGGCTTGTGTGC a Region represents how distant from ATG or stop codon of its target gene is ORF. Aliquots of culture medium (2 L) were mounted on slides and visuali�ed by a �uorescence microscope. Aer the addition of P-factor, the morphology was gradually changed and the production of GFP increased. Scale bar = 10 m, (c) dose-dependent response. e cells were exposed to the various concentrations of P-factor and were incubated for 24 hours aer the addition of P-factor. Filled circle depicts OSP210-2 (sxa2Δ, pAL7-Umam2-GFP-LPI) and �lled square depicts OSP210-17 (sxa2Δ, pAL7-Usxa2-GFP-DSPBC4.01). -axis denotes a percentage of the maximum response. e values were the means of triplicate determinations from a typical experiment.

Time-Dependent
Assay. e cells were grown in EMM at 32 ∘ C for 24-36 hours and were inoculated into 5 mL of the fresh EMM. en the cells were grown at 30 ∘ C for 20 hours and harvested. Aer washing twice with sterile water, the cells were transferred to assay medium to give an initial optical density of 1.5 at 600 nm. Immediately, P-factors were added at a �nal concentration of 1 or 0 M. e cells in 500 L were taken every hour (0-6 h), two hours (6-12 h), or twelve hours (12-24 h). Aliquots were resuspended in 500 L of Milli-Q water aer being washed once. Fluorescence intensity of GFP produced in the cells was measured by a Hitachi F-2700 Fluorescence Spectrophotometer (Hitachi, Tokyo Japan). e relative �uorescence intensity was averaged and calculated as the ratio of the sample to the control strain harboring pAL7 instead of the reporter plasmid (OSP210-0 or OSP220-0). e images were taken on a Zeiss Axiovert 200M Inverted Microscope with the AxioCam MRm charge-coupled camera controlled by Axiovision. Assays were performed using independent transformants.
is the Hill coefficient.

Tractable Reporter Plasmid Developed from pAL7.
To prepare the transformants for reporter assay, we constructed the new reporter plasmids with both tractability and sensitivity. e reporter plasmids made those transformants produce green �uorescent protein (GFP) through the Mam2-P-factor signal transduction pathway (mating pathway). To determine the most appropriate pheromone-dependent reporter region on the reporter plasmid, we focused on the 7 genes, which were previously reported to have been sharply activated by the addition of P-factor [24,30]. e upstream regions of those genes were used as an inducible promoter for GFP. As the downstream region of GFP, we employed LPI terminator, which correctly worked as a gene terminator in the �ssion yeast cell. Each of the resultant plasmids (listed in Table 2) was transformed into the sxa2Δ strain, OSP210 (h−, leu1-32, ura4-D18, sxa2Δ). e reason why sxa2 gene was deleted was because Sxa2 protein was the speci�c peptidase of Pfactor. e response to the 1 or 0 M of P-factor under nitrogen starvation was shown in Table 2. All upstream regions other than the dhc1 upstream region expressed GFP. e mam2 upstream region responded much more intensely than others. Nevertheless, the signal to noise ratio (SNR) of mam3 and sxa2 upstream regions was better than that of the mam2 upstream region. Next, we determined whether some downstream regions were more suitable for 3 ′ UTR of GFP than the LPI terminator since it was reported that 3 ′ UTR of mRNA regulated its stability and had an effect on the expression level [31]. LPI terminator in the reporter region of each plasmid was replaced with the downstream region corresponding to its upstream region to reproduce the native genomic context. e combination of dhc1 regions did not respond, suggesting that the important region for transcription might locate on ORF or the outside of the regions used in this study. e spk1 downstream region decreased the background under the absence of P-factor. e mam2 and mam3 downstream regions obviously decreased the response to P-factor, while that of sxa2 showed little difference from LPI terminator. e SPBC4.01 and rgs1 downstream regions raised the response dramatically, but it was clear that the rgs1 downstream region had raised the expression level without relying on P-factor. In an effort to prepare the appropriate reporter region, the mam2, mam3, and sxa2 upstream regionS presumably including the intense pheromone-dependent upstream activation sequence were combined with SPBC4.01 downstream region. As we had expected, the combination of the sxa2 upstream and SPBC4.01 downstream regions had raised the response with little increase of background. ereby, the best SNR was obtained with this combination. We circumstantially investigated the two reporter plasmids; one is pAL7-Umam2-GFP-LPI including the most intense reporter region and the other is pAL7-Usxa2-GFP-DSPBC4.01 whose reporter region exhibited the best SNR. We explored the time-dependent GFP production of the strains harboring pAL7-Umam2-GFP-LPI (OSP210-2) or pAL7-Usxa2-GFP-DSPBC4.01 (OSP210-17). e strains were exposed to 1 or 0 M of P-factor under nitrogen starvation (Figure 2(a)). e strain OSP210-2 exhibited the  response with not only the intensity but also the rapidity, which enabled it to discriminate the positive reaction within 3 h aer the addition of P-factor. However, the cells had produced measurable GFP protein without the addition of Pfactor. On the contrary, pAL7-Usxa2-GFP-DSPBC4.01 had caused the cells to produce little GFP protein in the absence of P-factor. e maximum value was relatively high, and the positive signal could be discriminated within 5 h aer the addition of P-factor. ese results corresponded to the observation with the �uorescent microscope (Figure 2(b)). In both strains, the luminous cells began to appear at the time when the GFP signal was detected by the �uorescent spectrometer. And the elongation of the cell body to form shmoos was also observed, which was the typical response induced by the reception of P-factor [23]. Curiously, the strain OSP210-2 had begun to elongate about two hours earlier than the strain OSP210-17. We also explored the dose-dependent GFP production (Figure 2(c)). e response formed sigmoid curve clearly indicated that the Mam2-P-factor interaction was quantitatively measurable by these strains. eir EC50 values did not show a signi�cant difference between the strains. It was suggested that the efficiency of the subsequent signal transduction pathway had to remain hardly affected by increasing the copy number of plasmids.

e cyr1Δ Strain with Reporter
Plasmid. e adenylyl cyclase encoded by cyr1 gene is oen deleted from the assay in the mating pathway. e cyr1Δ strain exhibits constitutive starvation regardless of the existence of nitrogen and carbon [32]. Being ready to receive the ligands at any time has an advantage in the mating pathway of the assay system, so we examined the effect of cyr1-deletion with reporter plasmids. e cells were exposed to 1 or 0 M of P-factor under the presence or absence of nitrogen, and the timedependent responses were measured (Figure 3). Under the presence of nitrogen, the cyr1Δ strain harboring pAL7-Umam2-GFP-LPI (OSP220-2) responded most intensely at 12 h. Compared to the strain without P-factor, the difference in GFP production began to appear at 3 h and was completely distinguishable at 4 h and beyond. However, these strains had already exhibited high background at 0 h. On the other hand, the cyr1Δ strain harboring pAL7-Usxa2-GFP-DSPBC4.01 (OSP220-17) became distinguishable at 4 h. In addition, the strain without P-factor did not produce GFP protein at all. Unsurprisingly, the cyr1+ strain harboring pAL7-Usxa2-GFP-DSPBC4.01 (OSP210-17) with or without P-factor did not show the response in the presence of nitrogen. Interestingly, the cyr1+ strain harboring pAL7-Umam2-GFP-LPI (OSP210-2) could respond to P-factor by 12 h having passed even under the presence of nitrogen. Under the absence of nitrogen, the cyr1Δ strain OSP220-2 was with high background and began to produce GFP at about 3 h as is the case in the presence of nitrogen. However, this strain did not raise the response as we had expected. e cyr1Δ strain OSP220-17 became distinguishable about one hour earlier than cyr1+ strain OSP210-17, and the intensity of the response was also increased. e cyr1-deletion was quite effective in the strain OSP220-17 regardless of the existence of the nitrogen.

e Ectopic mam2 Expression in the Strain Harboring
Reporter Plasmid. As a model for heterologous GPCR assay with the reporter plasmid, the mam2 gene was ectopically expressed under the control of other promoters in the endogenous mam2Δ strain, OSP230 (h−, leu1-32, ura4-D18, sxa2Δ, mam2Δ). ree promoters were used to express the mam2 gene; the �rst is no message in thiamine 1 (nmt1) promoter, which is the most common inducible promoter and exhibits the strongest expression in the �ssion yeast [33,34]; the second is uracil regulatable 1 (urg1) promoter, which is an inducible promoter and expressed moderately. While urg1 promoter is repressed by the removal of uracil from the medium, it was maximally induced by either the addition of the uracil or the removal of the nitrogen [28]; the third is human cytomegalovirus (hCMV) promoter which is the strong promoter activated constitutively [35]. e mam2 gene was driven on either of the chromosomal integration vectors for �ssion yeast, pSU1Z vector (or its derivatives) or the reporter plasmid (Figure 4(a)). e transformation to introduce the receptor and the reporter was performed at the same time. When the parental strain OSP230 was cotransformed by 150 ng of the reporter plasmid and 150 ng of the linearized integration vector, 37 (SD ± 14) colonies appeared. Exposed to 1 M of P-factor, the response could be adequately detected in 5 mam2 expression patterns out of 6, which fell slightly below that in the native expression pattern (Figure 4(b)). Perhaps too much accumulation of Mam2 protein might hinder the GFP production in the strain OSP230-17h1 whose mam2 was constitutively expressed under the control of hCMV promoter on the high copy plasmid. On the other hand, the GFP production was not inhibited by the constitutive chromosomal expression whose mam2 expression was smaller than episomal expression. e induced nmt1 promoter shows little difference between the episomal and chromosomal expression. Under this experimental condition in which the cells were pre-incubated for 20 h, the mam2 gene was strongly expressed by induced nmt1 promoter for 2-5 h aer the intracellular thiamine was completely carried out of the cell body. Regulating the pre-incubation time might improve the response. In spite of the addition of thiamine, the nmt1 promoter was not completely repressed. is result coincided with the previous data [28]. e urg1 promoter did not show a signi�cant difference between the episomal and chromosomal expression. In both strains, it was very easy to operate urg1 promoter since that was automatically repressed and induced under this experimental condition. Although we followed the same procedure with pAL7-Umam2-GFP-LPI as we had with pAL7-Usxa2-GFP-DSPBC4.01, the assay had gone wrong. e reason why we could not assay was that the cells formed aggregates that did not respond to P-factor at all during the preculture (Figure 4(c)).

New Reporter
System. e reporter system constructed in this study was considerably tractable. ese reporter plasmids were easily transformed with high efficiency even at the same time as the chromosomal integration vector was introduced. In addition, the plasmid is so �exible that it is easy to reconstruct; for example, the selection marker could be changed depending on the genotype of laboratory stock strains. Moreover, as enough GFP for detection was produced in this assay system, the reporter gene did not have to be an intense one such as lacZ or luciferase, which required the substrate degradation at the detection for sensitive readout. As a reporter region of the plasmid, we concluded that a combination of sxa2 upstream region and the SPBC4.01 downstream region was the most suitable due to the best SNR. e cells responding to P-factor began to appear at 5 h and became obviously luminous at 6 h, while few luminous cells appeared for 24 h in the absence of P-factor. e combination of mam2 upstream region and LPI terminator responded most rapidly and intensely of all, and began to respond at 3 h and became absolutely distinguishable at 4 h. However, we were not able to recommend using the reporter region for two reasons: �rstly because of the excessively high background (many luminous cells appeared within 8 h and SNR was not so good compared to others), and secondly because nonspeci�c aggregation was readily formed.

e Combination of the Reporter Plasmid and cyr1-Deletion.
In the combination of pAL7-Usxa2-GFP-DSPBC4.01, the cyr1-deletion had both positive and negative aspects. e positive aspects were that the assay medium was not limited and that the cells responded a few hours earlier. e negative aspects were that the cyr1Δ strains became more difficult to deal with; for example, it took a longer time to divide (maximum doubling time of OSP210-17 → OSP220-17: 3.20 h → 4.76 h at 30 ∘ C in EMM) and the transformation efficiency became signi�cantly lower (about one hundredth). In addition to those demerits, what was worse was what occurred in the cyr1Δ strain harboring pAL7-Umam2-GFP-LPI. Under nitrogen starvation, the cyr1Δ strain responded less intensely than the cyr1+ strain. One explanation could be that there was a difference in the optimal pre-incubation time to maximally respond to Pfactor between the strain OSP220-2 and other strains. It was implied that in order to retain pAL7-Umam2-GFP-LPI, the intracellular condition had to be slightly affected (described below).

Ectopic mam2
Expression. As a model for the heterologous GPCRs, we expressed mam2 gene under the control of three kinds of promoters in the mam2Δ strain. It was fundamentally convenient for the associated experiments to express the receptor gene on the reporter plasmid, although the huge receptor expression could cause problems in the response. e expression by the nmt1 and urg1 promoter on the high copy plasmid had little impact on the reception of Pfactor, but the reduction of response was seen in the strain whose mam2 gene was constitutively expressed by hCMV promoter on the high copy plasmid. It was suggested that too much Mam2 production presumably caused the unfolded protein response, which had a harmful effect on the cell [36]. us, it would be indispensable to repress the heterologous GPCR expression during mitosis. In fact, nmt1 promoter could reproduce the response well in spite of demonstrating stronger expression than hCMV promoter. When the assay with urg1 promoter was performed, the operations were very easy because the cells were automatically repressed and induced in this system. However, note that the induction of urg1 promoter can depend on the intracellular condition [28], so the cyr1Δ strain might be unable to repress the expression of urg1 promoter as well as the cyr1+ strain.

Intracellular Effect of pAL7-Umam2-GFP-LPI. e Cyr1
protein is required through the nutrient pathway to perceive not nitrogen but rather a carbon resource [37]. In fact, the cyr1-deletion makes the cell unable to perceive the carbon resource. However, carbon starvation does not result in the autophagy for meiosis, which is induced by nitrogen starvation [38]. e reason why the cyr1-deletion (carbon starvation mimic) affects the mating pathway is presumably because it facilitates expression of the ste11 gene which commonly plays a key role in several intracellular signal transduction pathways including not only mating pathway but also nutrition signaling and stress signaling pathways [30,37]. In this study, there was some circumstantial evidence suggesting that such the ste11-induction was seen in the cyr1+ strain harboring pAL7-Umam2-GFP-LPI (OSP210-2). e strain OSP210-2 could respond to P-factor even under the existence of nitrogen ( Figure 3). Furthermore, the morphological change made progress about two hours earlier than that of the strain OSP210-17 ( Figure 2(b)). e promoter and/or activating regions must be related to the GFP production, but the regions are unlikely to have an effect on the morphological change. It was suggested that DNA and/or RNA derived from pAL7-Umam2-GFP-LPI made ste11 gene more likely to be induced as is oen the case with stress response and nutrient starvation.