LAMMER Kinase Lkh1 Is an Upstream Regulator of Prk1-Mediated Non-Sexual Flocculation in Fission Yeast

Abstract

The cation-dependent galactose-specific flocculation activity of the Schizosaccharomyces pombe null mutant of lkh1⁺, the gene encoding LAMMER kinase homolog, has previously been reported by our group. Here, we show that disruption of prk1⁺, another flocculation associated regulatory kinase encoding gene, also resulted in cation-dependent galactose-specific flocculation. Deletion of prk1 increased the flocculation phenotype of the lkh1⁺ null mutant and its overexpression reversed the flocculation of cells caused by lkh1 deletion. Transcript levels of prk1⁺ were also decreased by lkh1⁺ deletion. Cumulatively, these results indicate that Lkh1 is one of the negative regulators acting upstream of Prk1, regulating non-sexual flocculation in fission yeast.

1. Introduction

Flocculation is a well-known natural phenomenon of active aggregation, and is defined as the reversible aggregation of cells into flocs [1,2]. This process is important in industrial processes related to fermentation technology such as brewing, wine-making, and bioconversion; however, it is also a nuisance in many industrial processes, clinical settings, and in the laboratory [3]. Flocculation is due to divalent-cation-dependent bonding between the cell-surface proteins (lectins) of flocculent cells and specific sugar residues of the mannans embedded in the yeast cell wall [4].

Shankar and Umesh-Kumar [5] purified a cell-surface lectin from a flocculent strain of Saccharomyces cerevisiae, which, in the presence of Ca²⁺, binds specifically to the mannose and mannan residues isolated from yeast cell walls belonging to intact cells. Several dominant flocculation genes have been identified in Sc. cerevisiae: FLO1, FLO5, FLO8, FLO9, FLO10, FLO11, and FLO2 [6–10]. These genes are now considered a part of the ‘flocculin family’ and encode highly homologous proteins [8,11]. Different FLO genes confer variable degrees of flocculation, or encode proteins that have different responses to proteases and heat treatment or sugar binding capabilities [12].

Heterothallic haploid strains of the fission yeast Schizosaccharomyces pombe are usually non-flocculent when cultured independently [11]. However, the deletion or mutation of some negative regulators of flocculin expression, such as Ume5 kinase homolog Prk1 [13], LAMMER kinase homolog Lkh1 [14], transcriptional repressor Tup12 [15], and ribosomal protein Rpl32 [16,17], and the overexpression of some adhesins [18] induced non-sexual flocculation in Sch. pombe. Earlier, we reported that Lkh1-mediated flocculation is a cation-dependent, galactose-specific process [14], and that Lkh1 phosphorylates global transcription repressor Tup12, deletion of which showed asexual flocculation [15]. As of now, no report on the physicochemical properties of Prk1-dependent flocculation, or genetic interaction between the two protein kinases, Lkh1 and Prk1, in the asexual flocculation of Sch. pombe has been published. In order to explore the possible association of these two kinases with flocculating ability, we investigated ion-dependence and sugar-specificity of flocculation in the Prk1 null mutant, and genetic interaction between prk1⁺ and lkh1⁺.

In this study, we found that disruption of prk1⁺ produced cation-dependent galactose-specific flocculation similar to that of lkh1⁺. Also, the overexpression of Prk1 reversed the flocculation phenotype of the lkh1⁺ null mutant. Moreover, we found a significant decrease in the transcriptional level of prk1⁺ in the lkh1⁺ null mutant.

2. Materials and methods

2.1. Strains, media, and vectors

The strains used in this study are listed in Table 1. The rich medium was YE medium, and the selective medium was Edinburgh synthetic minimal medium with appropriate supplements. Yeasts were grown at 30 °C. Standard techniques for fission yeast molecular genetics were used following the methods described in the books written by Moreno et al. [19] and Alfa [20]. Escherichia coli and Sch. pombe cells were transformed by the ultra-competent and the lithium acetate method, respectively [20,21]. pREP2 (100X), a Sch. pombe expression vector containing the thiamine-repressible nmt1⁺ promoter was used for overexpression of prk1⁺ in the lkh1⁺ null mutant. Thiamine (2 μM) was added to the medium to repress transcription from the nmt1⁺ promoter [22].

Table 1.

Schizosaccharomyces pombe strains used in this study.

Strain	Genotype	Source
ED665	h- ade6-M201 leu1-32 ura4-D18	Lab. Collection
PHM5L	h- ade6-M201 leu1-32 ura4-D18 lkh1::leu2	Kim et al. [14]
SP107	h- ade6-M201 leu1-32 ura4-D18 prk1::ura4	This study
SP109	h- ade6-M201 leu1-32 ura4-D18 lkh1::leu2 prk1::ura4	This study
SP202	h- ade6-M201 leu1-32 ura4-D18 lkh1::leu2 pREP2-prk1:ura4	This study

2.2. Disruption and cloning of prk1+

The prk1⁺ gene was replaced with the 1.8-kb Sch. pombe ura4⁺ cassette [23]. To construct the prk1⁺ disruption cassette, the upstream non-coding region was amplified by PCR using primers Prk-uf and Prk-ur. The downstream non-coding region was amplified using primers Prk-df and Prk-dr (Table 2). Transformation was performed using the lithium acetate method [20,24]. Stable Ura⁺ transformants were initially screened by PCR, and the replacement of the prk1⁺ loci was confirmed by Southern blot analysis. To amplify the prk1⁺ gene, the sense primer Prk-F, and the antisense primer Prk-R were used. The pREP2-Prk1 vector was constructed by cloning a SalI–BamHI fragment containing the full-length prk1⁺ coding sequence into pREP2.

Table 2.

Primers used in this study.

Name	Sequence (5' to 3')*	Mutagenic information
<Deletion & Cloning>
Prk-uf	TAGAATTCTGGCATACTTCTAAAGTGCTATTG	EcoRI site
Prk-ur	TAGGATCCTTTCATGCTGAAATGACCCTA	BamHI site
Prk-df	TAGTCGACAACTCACTTTTTAGCCCATTTTT	SalI site
Prk-dr	TAAAGCTTTCGCTAGAGTTATCAGCCTCA	Hind III site
Prk-F	CTGTCGACTATGAAAGACGGTTATAAAATTATTG	SalI site
Prk-R	CTGGATTCTTATTAAAAATGGGCTAAAAAGTG	BamHI site
<qRT-PCR>
Act1-RT-F	CACCCTTGCTTGTTGACTGA
Act1-RT-R	ATTTACGCTCAGGAGGAGCA
Prk1-RT-F	GGAAAGGTTTATAAAGCGG
Prk1-RT-R	TTCAAAGCTGAGTCAAAGC

*The initiator ATG and the stop anticodon TTA are underlined and restriction enzyme sites included are emboldened.

2.3. Flocculation assay

The effects of metal ions and sugars on flocculation were tested by employing the previously described methods [13,14,19]. Briefly, cells grown to early stationary phase (∼10⁸ cells/ml) in YE medium were washed with 10 mM EDTA, followed by excess deionized water to repress flocculation activity, and resuspended in deionized water. Ions or sugars were added to the cell suspension, and then CaCl₂ was added to a final concentration of 10 mM to initiate the flocculation reaction. Percentage flocculation was estimated as described previously [25]. Briefly, the yeast suspension (20 ml) was vigorously shaken in 100-ml flasks and the optical density (OD₆₀₀) (A1) was measured subsequently. The suspension was left for 30 min at room temperature and the optical density of the upper layer of the cell suspension (A2) was measured again. Percentage flocculation was calculated according to an equation: %Floc = [(A1 – A2)/A1] × 100.

2.4. RNA preparation and qRT-PCR

Cells at early stationary phase were harvested and total RNAs were prepared with TRIzol reagent (Invitrogen, Carlsbad, CA, USA) as recommended by the manufacturer. Relative mRNA levels were determined by real-time quantitative reverse transcription (qRT) PCR using the specific primer sets (Table 2). Briefly, total RNA was reverse transcribed into cDNA using the M-MLV Reverse Transcriptase (Elpis Biotech, Daejeon, Korea) according to the manufacturer’s instructions. Each gene-specific primer was optimized for expression on a CFX96 Real-Time PCR (Bio-Rad Laboratories, USA) using TOPreal^TM qPCR 2X PreMIX (Enzynomics, Daejeon, Korea). The PCR cycling parameters were 5 min at 95 °C, 45 cycles of 10 sec at 95 °C, 15 sec at 60 °C, and 30 sec at 72 °C, and 10 sec at 95 °C. A melting curve analysis to confirm specificity was performed for each primer set. β -actin gene was used for normalization and the relative expression was calculated according to a previously reported method [26].

3. Results and discussion

3.1. Deletion of prk1⁺ slightly increases the flocculating ability of the lkh1⁺ null mutant

We previously reported that Lkh1, a LAMMER kinase homolog, negatively regulates asexual flocculation in fission yeast by showing flocculation of the lkh1⁺ null mutant, which is cation-dependent and galactose-specific in liquid culture [14]. Next, to determine the physicochemical properties of non-sexual flocculation caused by the prk1⁺ deletion, and the genetic interaction between lkh1⁺ and prk1⁺ during flocculation, we constructed Δlkh1, Δprk1, and Δlkh1Δprk1 double mutants, and then assayed them for flocculation.

As reported [13,14], ED665 (wild-type) cells remained in suspension during the course of the assay, whereas PHM5L (Δlkh1) and SP107 (Δprk1) cells flocculated and settled rapidly. Interestingly, while the disparity in flocculation was not much, the additional deletion of the prk1⁺ gene (SP109, Δlkh1Δprk1) nonetheless increased the flocculation of the lkh1⁺ null mutant (PHM5L) (Figure 1), clearly suggesting that Prk1 is functionally linked to the Lkh1-mediated asexual flocculation in fission yeast.

Figure 1.

Effect of prk1⁺ deletion on the flocculation phenotype of the lkh1⁺ null mutant. Cells grown to early stationary phase (∼10⁸ cells/ml) in YE medium were washed with 10 mM EDTA and excess water. CaCl₂ was added to the cell suspension to initiate the flocculation reaction. Each culture tube was photographed 30 min after initiation of the flocculation reaction. ED665 (wild-type), PHM5L (Δlkh1), SP107 (Δprk1), and SP109 (Δlkh1Δprk1) cells were used.

3.2. prk1⁺ also acts as a negative regulator of divalent-cation-dependent asexual flocculation in fission yeast

Watson and Davey [13] reported the involvement of the Prk1 protein kinase in Sch. pombe flocculation. However, a substantial number of features of prk1⁺-mediated flocculation are unknown. Therefore, we tested the consequences of using metal ions and sugars on the flocculation activity of prk1⁺ null mutants. The flocculation activities of SP107 (Δprk1) and SP109 (Δlkh1Δprk1) cells were completely repressed upon washing them with 10 mM EDTA (data not shown), but restored by the subsequent addition of divalent cations. As in the lkh1⁺ null mutant PHM5L [14], the flocculation activity of SP107 (Δprk1) cells was restored by the addition of Ca²⁺, Mg²⁺, Li²⁺, Zn²⁺, Mn²⁺, or Co²⁺, but not by Cu²⁺ or Ni²⁺ (Figure 2(A)). However, Zn²⁺, Mn²⁺, and Co²⁺ were more effective in restoring the flocculation activity in SP107 rather than in PHM5L. It is also noteworthy that SP109 cells were similar to PHM5L cells in terms of their divalent-cation requirements for flocculation activity. On the other hand, the effect of metal ions on the flocculation of SP107 cells differed slightly relative to the Sch. pombe AAD-1 mutant, for which the Ca²⁺, Mn²⁺, Li²⁺, Cu²⁺, and Zn²⁺ ions proved to be effective, unlike Mg²⁺ and Co²⁺ ions [27]. The requirement for different cations might be due to the involvement of different flocculins. Interestingly, several cations, including Li²⁺, were almost equally efficient in the flocculation of kinase gene disruptants, in contrast to Ca²⁺, which was most effective in the flocculation of Sc. cerevisiae and Sch. pombe AAD-1 mutants. Our results also indicate that flocculation of the prk1⁺ null mutants is divalent-cation-dependent, rather than Ca²⁺-specific, which is the case with the lkh1⁺ null mutant.

Figure 2.

Effect of divalent cations and sugars on flocculation. Stimulatory effect of divalent cations (A) and inhibitory effect of sugars (B) on flocculation. Cells in suspension represent the percentage of cells remaining in the upper layer 30 min after initiation of the flocculation reaction. The number of cells in suspension was determined by measuring optical density (OD 600 _nm). The bars indicate the standard deviation of four independent measurements for ED665 (wild-type, black bar), PHM5L (Δlkh1, gray bar), SP107 (Δprk1, oblique line bar), SP109 (Δlkh1Δprk1, dotted bar) cells. Each of the measurement was made in triplicates.

3.3. Prk1-dependent asexual flocculation is galactose-specific

Sugar-specificity for the asexual flocculation of SP107 (Δprk1) and SP109 (Δlkh1Δprk1) cells was determined. After suppressing the flocculation activity of cells by washing with 10 mM EDTA, sugar moieties were added to saturate the sugar-specific lectins. Then, CaCl₂ was added to initiate flocculation. The repression of flocculation, when a specific sugar is added, indicates that this sugar has saturated the lectins. These lectins interact with the same kinds of sugars on other cell surfaces, implying that this sugar-specific interaction is involved in the flocculation process. Galactose and lactose significantly inhibited the flocculation activity of SP107 and SP109 cells. On the other hand, glucose and maltose had no effect (Figure 2(B)). These results accord well with previous results for AAD-1 [27] and PHM5L [14], indicating that galactose residues, the major components of the cell-wall galactomannoproteins of Sch. pombe, may act as receptors for the divalent-cation-dependent aggregation of cells into floccules.

3.4. Lkh1 acts at the upstream of the Prk1 for non-sexual flocculation in fission yeast

Because prk1⁺ and lkh1⁺ are involved in the asexual flocculation of fission yeast [13,14], we constructed a lkh1⁺ null mutant transformed with a Prk1 overexpression vector, and tested whether Prk1 reverses the flocculation phenotype of the lkh1⁺ null mutant. The overexpression vector was constructed by inserting the ORF of prk1⁺ (1.2 kb) into the multicloning site of pREP2, under the control of the thiamine-repressible nmt1⁺ promoter. Cells were cultured in defined minimal medium in the absence or presence of 2 μM thiamine. ED665 (wild-type) cells showed no significant flocculation (Figure 3(A), tube 1) but PHM5L (Δlkh1) cells flocculated and settled rapidly (tube 2). Overexpression of prk1⁺ partially reversed the flocculation phenotype of the lkh1⁺ null mutant (tube 3), but repression of prk1⁺ by thiamine had no effect on the flocculation phenotype of PHM5L cells transformed with the Prk1 overexpression vector (tube 4).

Figure 3.

Reversal of the lkh1⁺ null mutant phenotype by prk1⁺. Cells were grown to early stationary phase (∼10⁸ cells/ml) in YE medium. Culture tubes containing ED665 (wild-type, tube 1), PHM5L (Δlkh1, tube 2), SP202 (pREP2-prk1⁺, tube 3), and SP202 with 2 μM thiamine (tube 4) were photographed 30 min after initiation of the flocculation reaction (A). Absorbance of culture supernatants of ED665 (closed circle), PHM5L (open circle), SP202 (closed triangle), and SP202 with 2 μM thiamine (open triangle) was measured at 10-min intervals (B).

Absorbance values (OD₆₀₀) for the supernatants of ED665, PHM5L, and SP202 (Δlkh1-pREP2-prk1) cell cultures were 2.5, 0.28, and 0.87, respectively, after 10 min at rest. This tendency was maintained for 30 min, after which the wild-type cells started to settle slowly. In the presence of thiamine, the absorbance of SP202 cells was similar to that of PHM5L cells. These results indicate that the overexpression of prk1⁺ partially reversed the flocculation phenotype of null mutant. In the previous section 3.3, it was shown that the flocculation activity of the prk1⁺ null mutant differed slightly from that of the lkh1⁺ null mutant in terms of its requirement for divalent cations, and Δlkh1Δprk1 double mutant cells also flocculated and settled rapidly (see Figure 1).

Taken together, these results indicate that Lkh1 acts upstream of Prk1 for non-sexual flocculation. To explore the connection between lkh1⁺ and prk1⁺ further, effect of lkh1⁺ on transcription of prk1⁺was tested via qRT-PCR. The expression levels of prk1⁺ were observed to be significantly diminished (41%, p-value = .013) in the lkh1⁺-null mutant compared to those in wild type (Figure 4). Although it remains to be demonstrated conclusively, it can be inferred from our results that Prk1 is functionally correlated with Lkh1 through unidentified signaling pathways involved in the non-sexual flocculation of fission yeast.

Figure 4.

Relative expression level of prk1⁺. The expression level of prk1⁺ in cells of ED665 (wild-type) and PHM5L (Δlkh1) was analyzed by qRT-PCR. Three biological samples were taken for the first cDNA synthesis, and subsequent quantitative real-time PCR analysis was performed using specific primer set for prk1⁺. The levels of prk1⁺expression in both strains were standardized to that of beta-actin gene and calculated by the 2 − ΔΔCt method. The asterisk indicates statistically significant change due to the deletion of LAMMER kinase (Δlkh1) as calculated with the Student’s t-test for p ≤ .02.

Although more experimental evidence is required, these results allow us to better understand the molecular mechanisms of flocculation in Sch. pombe.

Funding Statement

This work was financially supported by a research fund from Chungnam National University (2016) to H.M.Park.

Disclosure statement

No potential conflict of interest was reported by the author.