Dictyostelium hybrid polyketide synthase, SteelyA, produces 4-methyl-5-pentylbenzene-1,3-diol and induces spore maturation


  • Editor: Pauline Schaap

Correspondence: Tamao Saito, Faculty of Science and Technology, Sophia University, 7-1 Kioi-cho Chiyoda-ku, Tokyo 102-8554, Japan. Tel.: +81 3 3238 3367; fax: +81 3 3238 3361; e-mail: tasaito@sophia.ac.jp


The genome of Dictyostelium contains two novel hybrid-type polyketide synthases (PKSs) known as ‘Steely’; the Steely enzyme is formed by the fusion of type I and type III PKSs. One of these enzymes, SteelyB, is known to be responsible for the production of the stalk cell-inducing factor DIF-1 in vivo. On the other hand, the product(s) and expression pattern of SteelyA are not clearly understood, because there are two different reports associated with the in vitro products of SteelyA and its expression pattern. To solve this problem, we first examined the expression pattern using two different primer sets and found that it was quite similar to that shown in the dictyExpress database. stlA expression peaked at approximately 3 h and declined, but showed a small peak around the end of development. Next, we examined the in vivo product of SteelyA using a stlA null mutant and found that the mutant lacked 4-methyl-5-pentylbenzene-1,3-diol (MPBD). This null mutant showed aberrant, glassy sori, and most of the cells in the sori remained amoeba-like without a cell wall. This defect was restored by adding 200 nM of MPBD to the agar. These results indicate that SteelyA produces MPBD in vivo and induces spore maturation.


Dictyostelium is an excellent model organism to study a developmental system that is regulated by the secreted signal molecules. Starvation triggers Dictyostelium cells to aggregate and form a multicellular mound, eventually forming a fruiting body that contains two types of differentiated cells: stalk and spore cells (Kessin, 2001). In the mound stage, cells begin to differentiate into prespore and prestalk cells at random positions. The differentiated cells sort out and form an anterior prestalk and posterior prespore zones at the slug stage (Kay & Thompson, 2009).

The prestalk cell population is composed of several subtypes of cells (Williams, 1997), whereas the prespore cells are believed to be rather homogeneous. Under the appropriate conditions, the slug transforms into the final morphology, the fruiting body, which consists of a spore mass atop vacuolated and dead stalk cells.

In Dictyostelium, the developmental process is a stress response triggered by starvation and the cell type differentiation process is mainly controlled by extracellular signals. Among them, cAMP is essential for both prespore and prestalk differentiation (Kay, 1982; Meima & Schaap, 1999). Polyketides are also signal compounds that control the differentiation of Dictyostelium. The polyketide synthases (PKSs) inhibitor, cerulenin, inhibits Dictyostelium differentiation and therefore its development (Serafimidis & Kay, 2005).

The diversity of the biological activity of polyketides has rendered these secondary metabolites and the PKS genes that regulate their production the focus of biomedical and biopharmaceutical research.

The completion of the Dictyostelium genome project revealed that the Dictyostelium genome contains more than 40 PKS genes, indicating that it has huge potential for polyketide production. In addition, the Dictyostelium genome contained two novel hybrid-type PKS genes (Eichinger et al., 2005; Zucko et al., 2007). This novel structure was known as ‘Steely’. In Steely PKS proteins, the type III PKS domain was fused to the C-terminus of a multidomain type I PKS (Eichinger et al., 2005; Austin et al., 2006). The two Steely-type PKSs were called SteelyA and SteelyB. SteelyB was reported to be responsible for the production of the stalk-inducing factor DIF-1 and the knockout mutant of stlB lacked DIF-1 (Austin et al., 2006). This stlB mutant aided the elucidation of the functions of DIF-1 in vivo (Saito et al., 2008). However, two different reports associated with SteelyA expression pattern and its products have been identified. According to one report in 2006, an in vitro product was identified as pyrone and the stlA gene was expressed maximally in early development before cell aggregation. Another report in 2008 identified 4-methyl-5-pentylbenzene-1,3-diol (MPBD) as the main in vitro product and the stlA gene was found to be expressed only in late development (Austin et al., 2006; Ghosh et al., 2008).

In this study, we re-examined the expression pattern of stlA using two different primer sets and observed that it was similar to that in the dictyExpress database and our previous report (Austin et al., 2006; Rot et al., 2009). Furthermore, we used an stlA mutant and showed that one of the in vivo products of SteelyA was MPBD, a differentiation-inducing factor that was identified in the conditioned medium for a dmtA mutant (Saito et al., 2006). Finally, we observed that MPBD induced the formation of mature spore cells in the fruiting body.

Materials and methods

Strain and cell culture

The Dictyostelium discoideum Ax2 strain was grown in an axenic medium at 22 °C and was harvested at a density of approximately 5 × 106 cells mL−1. A stlA null strain that we reported previously (Austin et al., 2006) was grown in an axenic medium in the presence of 10 μg mL−1 balsticidin S. The axenically grown cells were washed and were developed at 22 °C on the phosphate buffer (2.7 mM Na2HPO4/10.7 mM K2HPO4 pH 6.2) agar plates at a density of 1–2 × 106 cells cm−2.

For reverse transcription (RT)-PCR analysis, developing cells were harvested every 3 h until t21 (late culmination stage) and used for RNA purification. For developmental analysis, cells were developed on a filter paper placed on a phosphate-buffer agar plate.


RT-PCR primer sets were designed to amplify a unique RNA sequence.

blast similarity searches were used to confirm that each primer sequence amplified the unique RNA sequence. Before using each primer set for RT-PCR, we used genomic DNA as a template to evaluate the quality of the primer set.

Total RNAs obtained at each time point were extracted using the RNeasy kit (Qiagen) according to the manufacturer's instructions.

RT-PCR was performed using a One-step RT-PCR kit and/or RT-PCR (AMV) kit (Takara). The RT-PCR conditions were as follows: one cycle of 30 min at 50 °C for RT and one cycle of 2 min at 94 °C, followed by 22 cycles of 40 s at 94 °C, 40 s at annealing temperature and 30–70 s at 68 °C for extension. The details of the primer sets, annealing temperatures and the size of the products are summarized in Table 1. PCR was performed for the stlA keto-synthase domain using the primers stlA-KSf and stlA-KSr (Table 1). For the stlA type III PKS domain, the primers used were exactly the same as those used by Ghosh et al. (2008).

Table 1.   Summary of the primer sets used in the RT-PCR
Primer nameSequenceAnnealing temperature (°C)CyclesProduct size (bp)

RT-PCR products were subjected to 1% agarose gel electrophoresis to evaluate the expression profile. Ig7 (mitochondrial large rRNA) was used as the RT-PCR control and total RNA was used as the loading control.

Analysis of the factors

Axenically grown cells were harvested in the late log phase and allowed to develop for 3 days on a filter paper supported on a stainless-steel mesh whose under surface was in contact with phosphate buffer and Amberlite XAD-2 resin beads to bind nonpolar compounds. After complete development, the beads were collected and extracted with ethanol. The extracted materials were concentrated by rotary evaporation and taken up in 40% methanol and filtered using a DISMIC 13HP filter (Advantec). Filtered samples were analyzed by reverse-phase HPLC (TSK gel-ODS-120T) eluting at 1 mL min−1 with a gradient of 40–100% methanol containing 2% acetic acid in 1 h. Samples obtained at 32–38 min were collected and analyzed by GC–MS using a Saturn 2000 ion-trap mass spectrometer (Varian Inc., Walnut Creek, CA) connected to a Varian 3800 gas chromatograph equipped with a BPX70 capillary column. The oven was maintained at 170 °C for 3 min, programmed to increase to 260 °C at 20 °C min−1 and then maintained at 260 °C for 7.5 min. Helium gas was used as the carrier gas. GC–MS was operated at an ionization voltage of 70 eV and a trap temperature of 175 °C with a mass range of 40–650 atomic units.

Spore formation assay

To examine spore morphology, the sori were collected from the mature fruiting bodies of each strain, suspended in phosphate buffer and examined under a microscope (Axiovert135, Zeiss). To examine the effect of MPBD on spore maturation in the stlA null strain, cells were developed on the phosphate agar containing MPBD. After 40 h, the sori were collected using an iron loop, suspended in phosphate buffer and examined under a microscope.

The number of viable spores was determined by suspending them in 10 mM EDTA (pH 7.5) and heating for 30 min at 37 °C (Richardson & Loomis, 1992). The number of viable spores present after heating was counted under the microscope.

Results and discussion

Expression of stlA

Two independent developmental expression profiles of stlA obtained by RT-PCR have been reported previously. These two reports used different primers and showed different expression patterns, leading to the re-examination of the expression pattern (Austin et al., 2006; Ghosh et al., 2008). Figure 1 shows the expression profiles obtained with two different primer sets. Primer set 1 included the primers stlA-KSf and stlA-KSr and was designed based on the keto-synthase domain, which has 119 bp of intron between the positions of these primers. Primer set 2 was identical to that used in a previous report (Ghosh et al., 2008). We obtained identical results with the two different primer sets. The expression of stlA peaked around the early stage of development and declined as development progressed. However, in the last stage of development, it showed a weak peak. These results were in accordance with the previously obtained results. Recently, a database of RNA sequences obtained from developmental stages (dictyExpress) was published (Rot et al., 2009), and our expression profile was in accordance with that shown in the dictyExpress database. Two previously reported studies used the same Ax2 strain and allowed the cells to grow in an axenic medium. On the other hand, the dictyExpress database used a different strain Ax4 grown in the association with Klebsiella aerogenes. We found that stlA expression in the vegetative stage was induced by the presence of Klebsiella (Akabane et al., in preparation). Despite these differences, the present expression pattern was in accordance with that shown in the dictyExpress database.

Figure 1.

 Expression profile of the stlA gene was analyzed by RT-PCR. Ig7 (mitochondrial large rRNA) and total RNA served as RT-PCR and loading controls, respectively. The developmental time is indicated above the panels. Two independent primer sets (see Materials and methods) produced the same results and the expression patterns were in accordance with the result in the dictyExpress database.

The stlA null strain lacks MPBD in vivo

Two different gene products have been reported for SteelyA. MPBD was the main in vitro product according to one report, but another report identified pyrone as the gene product (Austin et al., 2006; Ghosh et al., 2008).

Because the structure of MPBD has been examined thoroughly (Saito et al., 2006), we first focused on MPBD. To test whether MPBD is the product of SteelyA, we compared the materials released from mature fruiting bodies of the stlA null strain and Ax2, wild-type strain.

Nonpolar compounds released from the cells were collected using the Amberlite XAD-2 resin. After the elution of bound compounds from the resin, extracted materials were dissolved in 40% methanol and separated by reverse-phase HPLC. This method was used in a previous study in which MPBD was purified and identified as a differentiation-inducing factor (Saito et al., 2006).

We detected the HPLC peak from the Ax2 sample, but not from the stlA null sample. To confirm that the stlA mutant lacked MPBD, we further analyzed the HPLC fractions by GC–MS. Because the MPBD synthesized was eluted at around 35 min in our HPLC experiment, we collected three fractions eluted between 32 and 38 min (2 min for each fraction) and subjected them to GC–MS analysis. Figure 2 shows the representative results of GC–MS total ion current chromatogram. The main peak in Ax2 was the same with MPBD by GC–MS analysis.

Figure 2.

 Representative result of the GC–MS total ion current chromatogram of the compounds released by the stlA null mutant. The stlA null strain lacked the MPBD peak at around 35 min in our HPLC experiment, whereas the Ax2 strain showed the peak. To confirm the HPLC results, fractions obtained between 32 and 38 min (2 min for each fraction) were collected and analyzed by GC–MS: stlA null strain lacked the MPBD peak unlike the Ax2 strain.

Three independent stlA null strains failed to accumulate a detectable level of MPBD, indicating that SteelyA produced MPBD.

MPBD induces spore maturation

To investigate the function of MPBD in the development of Dictyostelium, we examined the phenotype of the stlA mutant. MPBD was identified as a differentiation-inducing factor that stimulated not only stalk cell differentiation but also spore cell differentiation (Saito et al., 2006). The stlA mutant cells developed normally and produced normal fruiting bodies (data not shown). However, the spore mass differed from that of the wild-type strain and had a glassy appearance (Fig. 3a). We then examined the morphology of spores under the microscope and observed that most of them remained in the amoebae-like form and not encapsulated spores. To confirm this observation, we stained sorus with Calcofluor, which fluoresces when in contact with cellulose of mature spore cells. Figure 3b (arrows) indicates that the amoebae-like cells were not encapsulated. We then heated the cells in the sorus with 10 mM EDTA (pH 7.5) at 37 °C for 30 min and counted the number of spores (Richardson & Loomis, 1992). Table 2 shows the result of the spore maturation test. The ratio of encapsulated spores in the stlA mutant was about 20% of that in the wild-type cell (Table 2).

Figure 3.

 Fruiting bodies of the wild-type strain and the stlA null mutant. The stlA mutant fruiting body produced glassy sorus (a). Spores were suspended in phosphate buffer and examined using a microscope (b: top panels). The cells in the sori were stained with Calcofluor and photographed under phase contrast (middle panels) and UV illumination (bottom panels). The wild-type strain contained encapsulated spores in the sori. In case of the stlA null mutant, most of the cells in the sori were amoebae-like (arrows) and encapsulated spores were also detected. This defect of stlA null mutants was restored by the addition of 200 nM MPBD in the agar. Scale bar=200 μm (a), 100 μm (b: top panels), 20 μm (b: middle and bottom panels), respectively.

Table 2.   Effect of MPBD for the spore encapsulation of the stlA null mutant
Concentration of MPBDEncapsulation ratio (%)
(nM)Ax2stlA null
  1. The number of encapsulated spores was determined by suspending them in 10 mM EDTA (pH 7.5) and heating for 30 min at 37°C. The number of viable encapsulated spores was counted under the microscope. Ax2 without MPBD was used as a standard for each experiment. Values are the mean of nine independent experiments.

None10022 ± 17
1098 ± 735 ± 19
10097 ± 1265 ± 20
200102 ± 983 ± 17
500100 ± 786 ± 16

As mentioned above, GC–MS analysis showed that the stlA mutant lacks MPBD. An alternative interpretation of this result is that SteelyA produced a polyketide that was not MPBD, but was essential for normal development and was therefore indirectly involved in MPBD production.

To rule out this possibility, we attempted to compensate the defect of the stlA mutant by adding MPBD in the agar.

As shown in Fig. 3b, the normal spore phenotype was restored in the stlA mutant by supplying 200 nM of MPBD in the agar. MPBD was first identified as a stalk-inducing factor and synthetic MPBD was also shown to stimulate spore cell differentiation (Saito et al., 2006). Our in vivo analysis demonstrated that SteelyA hybrid-type PKS produced MPBD in vivo and regulated the spore maturation. Because the fruiting body of the stlA null strain produced sori, it appeared that MPBD was involved in spore maturation rather than prespore differentiation. To confirm this, we analyzed the expression of the cell-type-specific genes in the stlA null mutant (Fig. 4). The prestalk markers (ecmA and ecmB) and prespore markers (pspA and cotB) expressed normally. Unexpectedly, the expression of spiA specifically in prespore and spore cells during culmination (Richardson & Loomis, 1992) was normal.

Figure 4.

 Effect of stlA on the expression of cell-type-specific genes. Ig7 (mitochondrial large rRNA) and total RNA served as RT-PCR and loading controls, respectively. The developmental time is indicated above the panels. The development of the stlA null mutant was delayed by about 3 h compared with that of Ax2; therefore, we collected the cells until t25. Total RNA was prepared from Ax2 and stlA null mutant cells. The expression of indicated genes was analyzed by RT-PCR.

To the best of our knowledge, the hybrid-type Steely PKS has been found only in slime molds. Two types of Steely PKS occur in D. discoideum: SteelyA and SteelyB. SteelyB has been reported to be responsible for the production of the polyketide backbone of stalk cell-inducing factor DIF-1. In this study, we found that SteelyA was responsible for the production of MPBD, a differentiation-inducing factor identified in the material released by the dmtA mutant and MPBD induced spore maturation.

Extracellular cAMP is essential for prespore differentiation, but is not sufficient to induce the formation of mature spores (Kay, 1982; Schaap & van Driel, 1985). Several (pre)spore-inducing factors have been reported so far (Oohata, 1995; Anjard et al., 1997, 1998; Oohata et al., 1997; Serafimidis & Kay, 2005; Saito et al., 2006) Two active spore-inducing factors were detected in a conditioned medium, one of which was called the psi factor (Oohata et al., 1997). In addition, the peptides SDF-1 and SDF-2 promote the terminal differentiation of spores (Anjard et al., 1998). The present results indicated that MPBD also regulated the terminal differentiation of spores. How these factors regulated spore differentiation and interacted with each other constitutes the next step of our research.


This work was supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan to T.S. (No. 20510196). T.B.N. is grateful for the Sophia type III scholarship.