A series of promoters for constitutive expression of heterologous genes in fission yeast



Inducible/repressible promoters are useful for the maintenance of toxic genes or timely expression. For ectopic expression of cloned genes in the fission yeast Schizosaccharomyces pombe, the thiamine-regulatable nmt1 promoter has been widely used, since the transcriptional activity of this promoter can be controlled by thiamine. However, this property sometimes limits a certain type of research, since the expression inevitably requires cells to be cultivated under the conditions that induce promoter activation. To allow constitutive expression of heterologous genes, we cloned three promoters of cam1+, tif51+ and ef1a-c+. Construction of a series of vectors comprising these promoters and their introduction into the fission yeast cells demonstrated that the activity was different among these promoters but was not affected by cultured media commonly used in fission yeast. Therefore, a promoter with appropriate strength would be selectable from these promoters, depending on the genes to be expressed. Copyright © 2008 John Wiley & Sons, Ltd.


Expression of a cloned gene of interest is one of the most standard and useful ways to investigate the function of the gene. Although several promoters, including mammalian viral promoters, have been reported in fission yeast, Schizosaccharomyces pombe, as described by Siam et al. (2004) and references therein, the nmt1 promoter that is repressible by addition of thiamine into a medium has been most widely used for expressing heterologous genes (Maundrell, 1990). To induce varying levels of expression of the genes, two versions of the nmt1 promoter have been additively employed (denoted as nmt1* and nmt1**, which show medium and low transcriptional activities, respectively; Basi et al., 1993). The transcriptional activity of each version can also be varied depending upon the absence and presence of thiamine in the medium. Furthermore, temperature-dependent induction activity of the truncated derivative of the nmt1 promoter was also reported recently (Kumar and Singh, 2006). In addition to the nmt1 promoter, several promoters that are also used for induced expression of heterologous genes have been developed (Bellemare et al., 2001; Erler et al., 2006; Fujita et al., 2006; Iacovoni et al., 1999). These types of inducible promoters are very useful for the maintenance of cloned genes that have harmful effects on cell growth, and timely expression to investigate the function of the gene in a specific situation. In particular, they are indispensable for the construction of expression libraries, especially cDNA libraries, since virtually all of them contain several harmful genes. Despite their usefulness, however, it may be a drawback that these promoters should be stimulated under the specific conditions that allow their activation. The nmt1 promoter, for example, is not suitable for overexpression of genes in YE medium containing intrinsic thiamine, which is commonly used for yeast cell culture. Therefore, this property of the nmt1 promoter may limit the area of research in some cases.

In addition to these inducible promoters, the adh1 promoter has been used in a number of expression vectors, and its transcription level was compared with that of the nmt1 promoter in some experiments (reviewed by Siam et al., 2004)). Although its expression is thought to be constitutive, it is not fully known how much its transcriptional activity varies among different media. Furthermore, it has been reported that several viral promoters can function, and therefore can be used for heterologous gene expression in fission yeast (Gmunder and Kohli, 1989; Jones et al., 1988; Toyama et al., 1992; Toyama and Okayama, 1990). However, it is less known whether or not they are constitutive in any medium used in fission yeast.

In this paper, we report on three novel promoters that can be used for constitutive expression of cloned genes. We estimated the transcriptional activity of each promoter in several media used commonly in fission yeast. The different levels of expression directed by these promoters would allow selection of a promoter that has the suitable transcriptional activity for an individual experiment.

Materials and methods

Sz. pombe strains and media

Sz. pombe wild-type strains JY3 (h90 prototroph) and AM2 (h90leu1-32) were used in this study. Complete medium (YE; 0.5% yeast extract, 2% glucose, 5 µg/ml adenine), minimal medium (SD), and minimal medium (MM) were used for the culture of Sz. pombe cells (reviewed by Forsburg and Rhind, 2006). Thiamine and leucine were added to the media at a final concentration of 15 µ M and 50 µg/ml, respectively, when needed.

Genetic methods and transformation of Sz. pombe

General methods to handle fission yeast cells were as described (Forsburg and Rhind, 2006). High-efficiency transformation of Sz. pombe cells was carried out as described previously (Matsuyama et al., 2000). For chromosomal integration of a series of pDUAL-based plasmids, each DNA was digested with NotI in a volume of 10–20 µl, and 3–4 µl of the resultant solution was directly used for transformation of AM2 (Matsuyama et al., 2004). Leu+ transformants were selected on an SD plate lacking leucine.

Cloning of the promoters used in this study

For cloning of the promoters of cam1+, tif51+ and ef1a-c+, PCR primers were designed to amplify about 700 bp of the promoter region of each gene. Pyrobest® DNA polymerase (TaKaRa Bio Inc.) was used for amplification. Primers used for amplification of the promoters were as follows; SphI-Pcam1-F, AACCAGCTGCATGCGCTATAGCATACTACGTTG; NheI-Pcam1-R, AAGCTGACGCTAGCATCAGTAACTCTAAAGTCCTTC; SphI- Ptif51-F, AACCAGCTGCATGCAACCGGACTC ATCTTTGTGAG; NheI-Ptif51-R, AAGCTGAC GCTAGCTCTAAACGATTCTTAAACTTG; SphI-P-EF1a-F, CCAGCTGCATGCATTTCAATCATT TTAGGAGAC; and NheI-P-EF1a-R, GTCGACG CTAGCGTGAGTTCGATGAAACAGACTC. The 5′- and the 3′-primers contain an SphI restriction site and an NheI site, respectively. The resultant PCR products were digested with these restriction endonucleases and inserted into the same sites of the pDUAL-based plasmids, such as pDUAL–FFH1 (Matsuyama et al., 2004), resulting in the replacement of the pre-existing nmt1 promoter. Plasmids constructed in this study are listed in Table 1. Nucleotide sequence data of these plasmids are available in the DDBJ/EMBL/GenBank databases under Accession Nos AB364119–50.

Table 1. Vectors constructed in this study
NamePromoterTagTag positionccdB*Accession No.
  • *

    This cassette makes the vectors Gateway-compatible.


Measurement of the protein levels by fluorescent microscopy

The transcriptional activity of the promoters used in this study was quantified by measuring the fluorescence intensity of GFP expressed under each promoter. pDUAL–GFH1, − 31, − 41, − 51, − 61 and − 81 (DDBJ/EMBL/GenBank Accession Nos AB364055, AB364132, AB364056, AB364133, AB364134 and AB364057, respectively) were integrated at the chromosomal leu1 locus of AM2. The resultant integrants were cultured in YE, SD, MM and MM containing 15 µ M thiamine to mid-log phase at 30 °C. GFP fluorescence in living cells was monitored by the Axiophot2 microscope (Carl Zeiss) with an appropriate set of filters. GFP fluorescence intensity of each integrant was determined by calculating the mean of fluorescence of 30 cells from three independent colonies. To minimize errors in measurement of the fluorescence intensity, the samples were not exposed to the excitation light before data acquisition. Emission light was incorporated for 3 s in every sample.

β-Galactosidase assay

β-Galactosidase activity was measured in a 96-well plate format as described below. The lacZ reporter gene was amplified by two-step PCR from pMC1871 (GenBank Accession No. VB0127) and subcloned as an entry clone using the Gateway system (Invitrogen). PCR primers used for amplification of the lacZ ORF were as follows: LacZ-CF, AAAAAGCAGGCTCTCATATGGTCGTTTTACAACGTCGTGA; and LacZ-CR, AGAAAGCTGGGTCTCGAGTGATTTTTGACA- CCAGACCAACT. After verification by sequencing, the lacZ ORF subcloned into pDONR221 (Invitrogen) was transferred to the Sz. pombe expression vector pDUAL–FFH1c and its promoter derivatives via the LR reaction. The resultant expression plasmids were digested with NotI and introduced into AM2. Each integrant was cultured as in the case of the GFP reporter assay. Cells were harvested in a 96-well plate and resuspended in Y-PER Yeast Protein Extraction Solution (Pierce) containing o-nitrophenyl-β-D-galactopyranoside (ONPG). The reaction was stopped by the addition of 1 M sodium carbonate. After centrifugation, OD420 of the supernatants were measured using SpectraMax M2e (Molecular Devices). β-Galactosidase activity was expressed in Miller units: 1000 × OD420/reaction time (min)/culture density (OD600)/volume of culture used for assay (ml).

Results and discussion

Cloning of promoters with different transcriptional activity

To screen for the promoters suitable for expression of a cloned gene, we first estimated the relative amount of all gene transcripts by analysing total RNAs prepared from the wild-type strain grown in minimal medium MM, using a DNA microarray (our unpublished experiments). Among the promoters with a certain level of expression, we then selected several promoters that were expected to show constitutive expression, irrespective of culture media, according to the previously reported microarray data (Chen et al., 2003; Rustici et al., 2004). Considering the restriction enzyme recognition sites together with these data, we finally selected three promoters derived from cam1+ (encoding calmodulin), tif51+ (eIF5A), and ef1a-c+ (EF-1α). Each promoter was amplified by PCR from the genomic DNA of a wild-type strain and was inserted into a series of pDUAL-based vectors (Matsuyama et al., 2004), thereby replacing the pre-existing nmt1 promoter. Thus, a new series of pDUAL vectors, which can be used for chromosomal integration, were generated (Table 1).

Measurement of the promoter activity

To examine the relative transcriptional activities of these promoters, we used pDUAL–GFH1 (Matsuyama et al., 2004) and its promoter derivatives. These plasmids can express the green fluorescent protein (GFP) fused with the FLAG epitope and a hexahistidine tag (His6). To measure them more precisely, we made a set of strains in which plasmids were integrated at the chromosomal leu1 locus by using pDUAL-based vectors. We measured the expression levels of GFP when integrants are cultured in YE, SD and MM media that are commonly used in fission yeast study. In addition, we examined the effect of thiamine on these promoters (Figure 1). Among the promoters tested, the nmt1 promoter had the pre-eminent activity in MM medium. In addition to the wild-type nmt1 promoter, the weakened derivatives of the nmt1 promoter (nmt1* and nmt1**) were repressed in other media. In contrast, other promoters showed constitutive transcriptional activity in any medium, suggesting that these promoters cannot be repressed by thiamine or by the substances contained in the media tested. The tif51 and the ef1a-c promoters showed similar transcriptional activity, although the ef1a-c promoter activity seemed slightly stronger in any medium. The cam1 promoters showed transcriptional activity comparable to the fully induced activity of the nmt1* promoter.

Figure 1.

The activity of promoters in different media as assessed by GFP fluorescence. The expression level of GFP was calculated using fluorescent microscopy. pDUAL–GFH integrants, in which GFP is expressed under the promoter indicated, were cultured to mid-log phase in the indicated media at 30 °C. Values were calculated as the mean of fluorescence of 30 cells from three independent transformants. Error bars represent standard deviation (SD) from the mean. nmt1*, medium-strength nmt1 promoter; nmt1**, low-strength nmt1 promoter

To further confirm the transcriptional activity of these promoters, we also estimated the expression levels of lacZ using the set of promoters. The lacZ reporter gene was fused to pDUAL–FFH1c (Matsuyama et al., 2004) and its promoter derivatives. These plasmids were then integrated into the leu1 locus of AM2, as was the case for GFP. We carried out β-galactosidase assays using integrants grown in the media that were also used to monitor the GFP expression. Although the signal range was different, a similar tendency was observed as compared with the GFP reporter assay (Figure 2). Whereas three versions of the nmt1 promoter showed a nearly 10- to 20-fold activation in MM medium containing no thiamine, transcriptional activity of three promoters cloned in this study were not largely altered in any medium. The tif51 and the ef1a-c promoters showed similar transcriptional activities, which were slightly stronger than the induced activity of the nmt1* promoter in this assay. The cam1 promoter showed weaker activity as compared with the tif51 and the ef1a-c promoters. Some small differences in the relative transcriptional activity were seen in two reporter assays, especially in comparison of the activities between the nmt1 promoter derivatives and the promoters cloned in this study. For example, the nmt1* promoter showed relatively stronger activity than measured using the GFP reporter. However, these differences are probably mostly due to differences in the systems that were used to quantify the transcriptional activity, as discussed previously (Grünenfelder and Winzeler, 2002).

Figure 2.

The promoter activity assessed by β-galactosidase assay. Transcriptional activity of each promoter was assessed by measuring β-galactosidase activity of pDUAL–FFH integrants cultured as described in Figure 1. Values were calculated using three independent transformants. Error bars represent SD from the mean


As shown in two reporter assays, since all the promoters newly cloned were not affected by the normal medium for cell culture, the use of them enables us to constitutively express cloned genes. By estimating their transcriptional activity using GFP and lacZ as a reporter, we could divide them into two classes: cam1 (low-level expression) and tif51 and ef1a-c (high-level expression). A suitable promoter for heterologous expression of a certain gene of interest should be ideally determined for each experiment. As reported previously, however, since mRNA abundance of a transgene sometimes has a poor correlation with protein abundance when overexpressed under the control of a heterologous promoter (Matsuyama et al., 2006), varying levels of transcription by promoters with different activities does not always result in the desired protein levels. These properties often make it difficult to know how much level of transcriptional activity is suitable for expression of a cloned gene of interest. Therefore, the more promoters whose transcriptional activities are characterized when used for heterologous gene expression are listed, the more variety of analyses would be feasible. In this sense, in addition to the previously reported promoters, the promoters described herein would become a useful option for functional analysis of genes, especially for those having no adverse effect on cells.


This work was supported in part by the CREST Research Project, Japan Science and Technology Agency. A.M. was supported in part by Grant-in-Aid for Young Scientists (A) from the Ministry of Education, Culture, Sports, Science and Technology of Japan.