The Chlamydomonas reinhardtii PSR1 gene is required for proper acclimation of the cells to phosphorus (P) deficiency. P-starved psr1 mutants show signs of secondary sulfur (S) starvation, exemplified by the synthesis of extracellular arylsulfatase and the accumulation of transcripts encoding proteins involved in S scavenging and assimilation. Epistasis analysis reveals that induction of the S-starvation responses in P-limited psr1 cells requires the regulatory protein kinase SNRK2.1, but bypasses the membrane-targeted activator, SAC1. The inhibitory kinase SNRK2.2 is necessary for repression of S-starvation responses during both nutrient-replete growth and P limitation; arylsulfatase activity and S deficiency-responsive genes are partially induced in the P-deficient snrk2.2 mutants and become fully activated in the P-deficient psr1snrk2.2 double mutant. During P starvation, the sac1snrk2.2 double mutants or the psr1sac1snrk2.2 triple mutants exhibit reduced arylsulfatase activity compared to snrk2.2 or psr1snrk2.2, respectively, but the sac1 mutation has little effect on the abundance of S deficiency-responsive transcripts in these strains, suggesting a post-transcriptional role for SAC1 in elicitation of S-starvation responses. Interestingly, P-starved psr1snrk2.2 cells bleach and die more rapidly than wild-type or psr1 strains, suggesting that activation of S-starvation responses during P deprivation is deleterious to the cell. From these results we infer that (i) P-deficient growth causes some internal S limitation, but the S-deficiency responses are normally inhibited during acclimation to P deprivation; (ii) the S-deficiency responses are not completely suppressed in P-deficient psr1 cells and consequently these cells synthesize some arylsulfatase and exhibit elevated levels of transcripts for S-deprivation genes; and (iii) this increased expression is controlled by regulators that modulate transcription of S-responsive genes during S-deprivation conditions. Overall, the work strongly suggests integration of the different circuits that control nutrient-deprivation responses in Chlamydomonas.

---

THE elements phosphorus (P) and sulfur (S) are essential macronutrients for sustaining life. P is a structural component of nucleic acids and phospholipids, and is a ubiquitous modifier of carbohydrates and proteins, while S is incorporated into sulfolipids, polysaccharides, proteins, cofactors, and a wide variety of important metabolites including S-adenosyl-methionine, glutathione, and phytochelatins. The preferred forms of P and S that are assimilated by plants and microbes are the orthophosphate ion, PO₄^3– (Pi), and the sulfate ion, SO₄^2–. The available pools of these anions can vary significantly as environmental conditions change. Most organisms have a limited capacity to store S, and thus require a continual supply of S-containing nutrients for survival. In contrast, cells often have considerable reserves of P, which are bound in polymers of DNA, RNA, and polyphosphate. The ability of microbes to acclimate to periods of nutrient insufficiency is essential to their survival in the natural environment (reviewed by GROSSMAN and TAKAHASHI 2001).

The green, unicellular alga Chlamydomonas reinhardtii (Chlamydomonas throughout) exhibits both specific and general responses when experiencing P or S deprivation. The general responses are common to a number of different stress conditions, while the specific responses enable processes that are advantageous during particular nutrient deficiencies, often allowing for better scavenging of the limiting nutrient from internal and external stores. P and S limitation elicit qualitatively similar effects on growth and photosynthesis, differing only in that the responses to S starvation occur more quickly following exposure of cells to medium devoid of S. General responses to nutrient limitation that have been analyzed include the cessation of growth at low cell densities (SHIMOGAWARA et al. 1999; ZHANG et al. 2002) and a reduction in photosynthetic O₂ evolution, which is mostly a consequence of reduced photosystem II (PS II) activity (WYKOFF et al. 1998). The specific S-deprivation responses include an elevated rate of extracellular SO₄^2– uptake (YILDIZ et al. 1994), secretion of extracellular arylsulfatases (LEIN and SCHREINER 1975; DE HOSTOS et al. 1988), and an increased cellular capacity to assimilate SO₄^2– by increasing levels of enzymes required for cysteine biosynthesis (RAVINA et al. 2002). Mechanisms for conserving S during limiting conditions include the rapid turnover of sulfolipids and their replacement with phospholipids (SUGIMOTO et al. 2007, 2008), the synthesis of putative cell wall proteins with a very low abundance of S-containing amino acids (TAKAHASHI et al. 2001), and a potential change in the polypeptide composition of light harvesting complexes, favoring the synthesis of complexes with polypeptides containing low levels of sulfur amino acid (NGUYEN et al. 2008). Many S starvation-elicited responses appear to be controlled, at least in part, at the level of gene expression. Genes encoding the arylsulfatases (DE HOSTOS et al. 1989; RAVINA et al. 2002), extracellular proteins (TAKAHASHI et al. 2001), ATP sulfurylase (YILDIZ et al. 1996), adenylylphosphosulfate reductase, serine O-acetyl transferase, OAS (thiol)-lyase (RAVINA et al. 2002), sulfite reductases, and putative SO₄^2– transporters (ZHANG et al. 2004), are upregulated in cells deprived of S.

A suite of specific responses also enables Chlamydomonas to acclimate to P limitation. P-starved cells induce high-affinity Pi uptake (SHIMOGAWARA et al. 1999) and synthesize extracellular phosphatases (QUISEL et al. 1996) that enhance Pi scavenging from the environment. Polyphosphate stores (HEBELER et al. 1992; WERNER et al. 2007) and chloroplast DNA (YEHUDAI-RESHEFF et al. 2007) are also mobilized, and phospholipids are replaced by galactolipids and sulfolipids (RIEKHOF et al. 2003) as cells redistribute their internal P resources. P-starved cells also conserve Pi by inhibiting the turnover of chloroplast transcripts by the phosphorylytic chloroplast polynucleotide phosphorylase (YEHUDAI-RESHEFF et al. 2007). Like S-deprivation responses, P deficiency-specific responses are often regulated at the level of gene expression, as transcripts encoding extracellular phosphatases and Pi transporters increase markedly in abundance following elimination of P from the growth medium (KOBAYASHI et al. 2003; CHANG et al. 2005; MOSELEY et al. 2006).

Screens for mutants that fail to acclimate properly to S limitation have enabled identification of three key regulators of the S-deficiency responses (DAVIES et al. 1994; POLLOCK et al. 2005). The Sulfur Acclimation 1 (SAC1) gene encodes an integral membrane protein that is similar to the SLC13 family of transporters (DAVIES et al. 1996). SAC1 is a positive regulator critical for the activation of genes involved in scavenging and assimilating S from the environment (DAVIES et al. 1996; RAVINA et al. 2002; ZHANG et al. 2004; GONZALEZ-BALLESTER et al. 2008). The light sensitivity of S-deficient sac1 strains probably reflects their inability to decrease photosynthetic electron transport (DAVIES et al. 1994; WYKOFF et al. 1998), although the role of SAC1 in the regulation of photosynthesis is not well understood. The SNF1-related protein kinase 2.2 (SNRK2.2) gene, previously known as SAC3 (for Sulfur Acclimation 3), encodes a serine/threonine kinase that acts as a negative regulator of S deficiency-responsive gene expression (DAVIES et al. 1999). The snrk2.2 mutants display low, constitutive arylsulfatase activity and express elevated basal levels of S deficiency-responsive genes in S-replete medium (DAVIES et al. 1994; DAVIES et al. 1999; RAVINA et al. 2002; GONZALEZ-BALLESTER et al. 2008). SNRK2.2 is also required for the proper downregulation of chloroplast transcription in S-starved cells (IRIHIMOVITCH and STERN 2006). There is no clear epistasis relationship between SAC1 and SNRK2.2. The sac1snrk2.2 double mutants have lower constitutive arylsulfatase activity than snrk2.2 cells when grown under nutrient-replete conditions, but in contrast to the sac1 single mutants, the double mutants can still partially upregulate ARS gene expression in response to S deficiency (DAVIES et al. 1994). A related serine-threonine kinase, SNRK2.1, has recently been shown to perform a central regulatory function in controlling S-starvation responses (GONZALEZ-BALLESTER et al. 2008). The snrk2.1 mutants are unable to activate expression of most S starvation-responsive genes and cannot maintain normal basal levels of expression for some genes during S-replete growth. A snrk2.1 null allele is even more light sensitive during S starvation than sac1, suggesting that SNRK2.1 also plays a role in modulating photosynthetic activity during S starvation (GONZALEZ-BALLESTER et al. 2008). In contrast to the nonepistatic relationship between SAC1 and SNRK2.2, the phenotype of a snrk2.1snrk2.2 double mutant is indistinguishable from that of the snrk2.1 single mutant, indicating that SNRK2.1 is epistatic to SNRK2.2 (GONZALEZ-BALLESTER et al. 2008). Models based on genetic data propose that SAC1, SNRK2.1, and SNRK2.2 interact to form a signaling cascade that senses the cellular S status, maintains basal levels of the transcripts encoded by the S-responsive genes in S-replete cells, and activates (or represses) the expression of target genes during S limitation (POLLOCK et al. 2005; GONZALEZ-BALLESTER et al. 2008).

To date, a single regulator of P-deficiency responsive gene expression, encoded by the Phosphate Starvation Response 1 (PSR1) gene, has been identified (WYKOFF et al. 1999). Two independent genetic screens resulted in the generation of multiple psr1 alleles (SHIMOGAWARA et al. 1999). The psr1 strains fail to accumulate extracellular phosphatases or activate high-affinity Pi uptake during P starvation, and consequently grow poorly on a hydrolysable P source such as glucose-1-phosphate (SHIMOGAWARA et al. 1999). In contrast to the phenotype of S-deprived sac1 strains, P-starved psr1 cells downregulate photosynthesis more rapidly than wild-type cells (SHIMOGAWARA et al. 1999). Nevertheless, when psr1 cells are deprived of P they become more sensitive to high light intensities than wild-type cells (MOSELEY et al. 2006).

PSR1 is a complex protein with characteristics of transcriptional regulators, including a DNA-binding MYB domain (reviewed by LIPSICK 1996), a coiled-coil domain that may be involved in protein–protein interactions (reviewed by BURKHARD et al. 2001) and a helix-loop-helix motif (reviewed by JONES 2004). The C-terminal region of PSR1 is rich in glutamine, a common feature of transcriptional activators (reviewed by PABO and SAUER 1992). More than 95% of PSR1 is located in the nucleus, irrespective of cellular P status (WYKOFF et al. 1999). Although sequence-specific DNA binding has not been demonstrated for PSR1, the Arabidopsis thaliana homolog, PHR1, has been shown to bind regulatory motifs in the promoters of P-starvation inducible genes (RUBIO et al. 2001). The vascular plant homologs of PSR1 contain similar DNA binding and protein–protein interaction domains, but lack the glutamine-rich transcriptional activation domains of the algal protein (WYKOFF et al. 1999; RUBIO et al. 2001). Candidate gene targets under PSR1 control have been identified on the basis of altered transcript accumulation in P-deficient vs. P-replete conditions, and the loss of differential gene expression in psr1 mutants (MOSELEY et al. 2006). PSR1 controls expression of genes encoding enzymes with roles in P scavenging, including the major secreted phosphatase, PHOX, putative high-affinity Pi transporters, and proteins potentially involved in mobilizing polyphosphate from internal stores. It has also been implicated in the activation of genes whose products may reduce the accumulation of excitation energy in the photosynthetic complexes during P starvation (MOSELEY et al. 2006).

Studies of fungal and plant model systems have demonstrated that various regulatory pathways linked to nutrient-deprivation responses have a set of common target genes (reviewed by GASCH and WERNER-WASHBURNE 2002). Many genes that allow cells to cope with stress (e.g., a decreased potential for growth because of suboptimal conditions) are regulated similarly in Saccharomyces cerevisiae during P and S deprivation (SALDANHA et al. 2004). A more specific intersection between the regulation of P and S metabolism in S. cerevisiae was demonstrated with the finding that Pho4p, a P deprivation-responsive transcription factor, could functionally substitute for the sequence-specific DNA-binding protein Cp1p, which is required for activation of methionine biosynthetic genes. Activation of Pho4p by P deprivation, overexpression of Pho4p, and the use of constitutive alleles, all resulted in suppression of methionine auxotrophy in the S. cerevisiae cep1 mutant, which lacks the Cp1p factor (O'CONNELL and BAKER 1992). In this study we explore genetic interactions between the regulators of P- and S-starvation responses in Chlamydomonas, and provide evidence supporting the integration of these pathways.

Media and growth conditions:

Tris-acetate-phosphate (TAP) medium was made according to the standard recipe (GORMAN and LEVINE 1966). To make Pi-free (TA) medium, potassium phosphate (1.8 mM potassium, 1.1 mM Pi) was replaced with 1.8 mM potassium chloride. SO₄^2–-free (TAP –S) medium was prepared by substituting chloride salts for the SO₄^2– salts. Solid medium contained 0.75% agarose; TA plates were always supplemented with 100 µM glucose-1-phosphate while TAP –S plates always contained 2 mM sodium or potassium thiocyanate (cells grew identically with either cation). TA liquid medium was not supplemented with glucose-1-phosphate, and TAP –S was not supplemented with thiocyanate. Liquid cultures were grown in Erlenmeyer flasks at 25°, with agitation at 200 rpm, and under fluorescent bulbs at intensities of 30–40 µE m^–2 s^–1. Dark-grown liquid cultures were grown similarly except that the flasks were securely wrapped in aluminum foil and were only exposed to light for 5 min during sampling. Cells on plates were grown at 25° at 40–70 µE m^–2 s^–1.

Chlorophyll measurements and viability staining:

To determine the chlorophyll content of liquid cultures, 0.75 ml of cell culture were transferred to a 1.5-ml eppendorf tube and centrifuged for 5 min at 13,800 x g to pellet the cells. The supernatant was discarded and the chlorophyll in the pellet was extracted with 0.75 ml of a mixture of 80% acetone, 20% methanol. The chlorophyll concentration per ml of cultured cells was estimated from the absorbance at 652 nm divided by 34.5 (ARNON 1949). Viablity staining with Evans blue dye was performed as described previously (YEHUDAI-RESHEFF et al. 2007).

Phosphatase and sulfatase assays:

Colorimetric assays for alkaline phosphatase (SHIMOGAWARA et al. 1999) and arylsulfatase (DAVIES et al. 1994) activities were performed. Cells were grown on solid medium for 4–5 days and the color reactions were allowed to develop for 20–24 hr before the cells were washed from the surface of the medium with deionized water to allow examination of the colored precipitate embedded in the medium. Liquid assays for alkaline phosphatase and arylsulfatase activities were performed as described previously (LEIN and SCHREINER 1975; DE HOSTOS et al. 1988; QUISEL et al. 1996).

Strains and genetic analysis:

A comprehensive list of strains used in these studies is presented in Table 1. Wild-type strains were CC-125 (nit– mt+) and CC-1690 (nit+ mt+). The psr1 strains were psr1-1nit– mt+ which had been backcrossed five to six times with CC-124 and CC-125, and psr1-2 (psr1-2::ARG7 cw15nit–) (SHIMOGAWARA et al. 1999). The psr1-1 mutant was complemented with plasmid pKS1, containing a genomic PSR1 sequence (WYKOFF et al. 1999), to generate the psr1-complemented strain. A psr1-1nit– mt– strain was backcrossed three times to CC-1690 to generate the psr1-1nit+ mt– strain, which was used in subsequent experiments and for a number of crosses. psr1 progeny were scored as growing poorly on TA + glucose-1-phosphate plates and failing to express abundant alkaline phosphatase on the basis of the colorimetric assay for this activity (SHIMOGAWARA et al. 1999). The sac1 mutant (ars5-3nit– mt–) (DAVIES et al. 1994) was backcrossed three times to CC-1690 to produce a sac1(ars5-3) nit+ mt+ strain, while a snrk2.2 strain [sac3(are10)nit– mt–] (DAVIES et al. 1994) was crossed once to CC-1690 to produce a snrk2.2(are10)nit+ strain. The psr1-1nit+ mt– strain was crossed to sac1(ars5-3)nit+ mt+, sac1(ars4-15)nit– mt+ and sac1(ars4-55)nit– mt+. The sac1 progeny were identified by their inability to synthesize arylsulfatase on TAP –S + 2 mM thiocyanate plates. A representative sac1(ars5-3) mutant and a psr1sac1(ars5-3) double mutant from a tetratype tetrad were used for gene expression studies and enzyme activity assays. The snrk2.2(are10)nit+ mt+ and snrk2.2(are16)cw15mt+ strains (DAVIES et al. 1999) were crossed to psr1-1nit+ mt–. The snrk2.2 progeny were recognized on the basis of their constitutive arylsulfatase activity on TAP medium. Representative snrk2.2(are10) and psr1snrk2.2(are10) strains were chosen from a tetratype tetrad for further analyses. Representative snrk2.1(ars11) and psr1snrk2.1(ars11) strains were selected by crossing psr1-1nit+ mt– to the snrk2.1::aphVIII(ars11)nit– cw15mt+ strain (POLLOCK et al. 2005; GONZALEZ-BALLESTER et al. 2008). The snrk2.1 mutants, which are paromomycin resistant due to integration of the aphVIII gene, grew poorly and did not induce arylsulfatase on TAP –S + thiocyanate (POLLOCK et al. 2005; GONZALEZ-BALLESTER et al. 2008). Triple mutants were generated by crossing the appropriate double-mutant strains, and a representative sac1snrk2.2(are10) strain was selected from the progeny of a backcross of the psr1sac1snrk2.2(are10) strain to CC-1690. The presence of the sac1(ars5-3) and snrk2.2(are10) alleles in the respective triple-mutant strains was verified by amplification of the pJD76 and pJD67 insertional markers from isolated genomic DNA (see DAVIES et al. 1994 and supplemental data). cw+ nit+ representative mutant strains were selected for phenotypic and epistasis analysis to eliminate any potential contributions to the observed phenotypes by the cw15, nit1, or nit2 mutations. Although for some genotypes the phenotypic expressivity varied significantly between individual isolates, all of the representative strains used in the experiments described in the RESULTS section were selected because they displayed average responses within an observed phenotypic range.

View this table: In this window In a new window   TABLE 1 Chlamydomonas strains used in this study

 

RNA isolation:

Total RNA was isolated using a protocol modified from SCHLOSS et al. (1984). Briefly, cells from 50 ml of culture were pelleted by centrifugation, frozen in liquid N₂, and stored at –80°. The pelleted cells were kept frozen until they were lysed in 3 ml of a solution containing 50 mM Tris-HCL pH 7.5, 150 mM sodium chloride, 15 mM EDTA pH 8, 2% SDS and 40 µg/ml proteinase K (Invitrogen, Carlsbad, CA). The aqueous emulsion of lysed cells was extracted four times with 3 ml of a 50:50 mixture of acidic (pH 4.5) phenol (United States Biochemicals, Cleveland) and chloroform (Fisher, Pittsburgh), followed by two extractions with 3 ml chloroform (Fisher). Nucleic acids were precipitated by adding 3 ml (volume equal to the original cell suspension volume) of 8 M lithium chloride to the extracted solution, incubating the mixture at 4° for at least 4 hr and then collecting the nucleic acids by centrifugation at 14,500 x g for 30 min. Nucleic acid pellets were washed with 70% ethanol, vacuum-dried, and dissolved in water. Trace DNA contamination was removed by digestion with TurboDNase (Ambion, Austin, TX) according to the manufacturer's instructions.

Real-time quantitative RT–PCR:

Real-time quantitative RT–PCR was performed using the iScript One-Step RT–PCR kit with SYBR green and the Chromo4 Real-Time System (Bio-Rad Laboratories, Hercules, CA) according to the protocol described previously by MOSELEY et al. (2006). The primers for amplification were: CBLP, forward (5'-CTTCTCGCCCATGACCAC-3'), reverse (5'-CCCACCAGGTTGTTCTTCAG-3'); PHOX, forward (5'-TTCCGTTTCCGTTCTCTGAC-3'), reverse (5'-CCCTGCATCTTGTTCTCCAG-3'); ARS1, forward (5'-CGCGCCGTCACTTGTTTGTTG-3'), reverse (5'- GCCCACTTCTTTACCCAGCACCTC-3'); ARS2, forward (5'-GTACTGCCGCGTGCGTGATTCC-3'), reverse (5'-CCTAAACATTTGGCTCCGCAGTCC-3'); SLT1, forward (5'- ACGGGTCTTCGAGCGAATTGC -3'), reverse (5'-CGACTGCTTACGCAACAATCTTGG-3'); SLT2, forward (5'-GTACGGAGTTCC TTACGCGC-3'), reverse (5'-TTCTTCGCCACCGATGAGC-3'); SULTR2, forward (5'- ACGTGGCATGCAGCTCAT-3'), reverse (5'-CTTGCCACTTTGCCAGGT-3'); SAT1, forward (5'-TGCGGCACCCTTTCCCAGAG-3'), reverse (5'-CACCCGATCACACGCAAACTTCA-3'); OASTL4, forward (5'-GGCCCTCCTGGAAGCTGAATCA-3'), reverse (5'-CCCCACCCCACCCATTAGTAGTC-3'); ECP76, forward (5'-CCTCGCTCTCCTCGCTGCTG-3'), reverse (5'-CGGCCGACTTGGGTAATTGC-3'); SBDP, forward (5'-GGACGGCAGCATCATGGTGAGC-3'), reverse (5'-TCCACACGCCCTTGACCTTGAG-3'). The sizes of the amplification products range from 100 to 300 bp. The purity of PCR amplification products was assessed by melting curve analysis and by gel electrophoresis.

Arylsulfatase is induced in P-starved psr1 cells:

The psr1 mutants, represented by psr1-1 in this study, are unable to acclimate properly to P starvation and consequently bleach and exhibit much less growth than wild-type strains on TA medium supplemented with glucose-1-phosphate as the sole P source (SHIMOGAWARA et al. 1999) (Figure 1A, growth, –P +S). The P-starved psr1 strain also displays very low phosphatase activity compared with the wild-type strain CC-125 or the psr1 mutant that was complemented with the PSR1 gene (SHIMOGAWARA et al. 1999; WYKOFF et al. 1999) [Figure 1, A (phosphatase activity, –P +S) and B]. However, we noted that psr1 cells exhibit significant arylsulfatase activity in Pi-free medium, and that this activity is not observed in the wild-type or the psr1-complemented strains [Figure 1, A (sulfatase activity, –P +S) and C]. Similar induction of arylsulfatase is observed when the psr1-2 strain is starved for Pi (see supplemental Figure S1). The activation of arylsulfatase in P-deficient psr1 cells is not a consequence of secondary light stress since similar arylsulfatase activity is produced by psr1 cells grown without Pi in the light or in the dark (compare Figure 1, C and D). The wild-type, psr1, and psr1-complemented strains all show copious arylsulfatase activity after growth in SO₄^2–-free medium [Figure 1, A (sulfatase activity, +P –S) and E], indicating that the psr1 mutation shows no apparent alteration in normal acclimation responses to S deprivation. The arylsulfatase activity measured for psr1 cells after 7 days of P starvation is 7% of the peak activity measured after 24 hr of S starvation (compare Figure 1, C and E).

View larger version (34K): In this window In a new window Download PPT slide   FIGURE 1.— psr1 growth and hydrolytic activity. (A) Growth, phosphatase, and sulfatase activities of wild-type (CC-125), psr1, and psr1-complemented strains on solid agar media. +P +S plates were made with TAP medium, which contains 1.1 mM Pi and 0.4 mM SO42–. –P +S plates were made with TA medium, containing no added Pi, but supplemented with 100 µM glucose-1-phosphate. +P –S plates contained 2 mM potassium thiocyanate, but no added SO42–. Cells were grown for 5 days and then assayed for phosphatase and sulfatase activity using the colorimetric assays described in MATERIALS AND METHODS. The colorimetic assays were developed over an 24-hr period. (B) Quantitative analysis of phosphatase activity over 7 days of growth in TA liquid medium (no added Pi, no glucose-1-phosphate) for wild-type, psr1, and the psr1-complemented strain. Alkaline phosphatase activity is represented as absorbance at 410 nm of p-nitrophenol produced per milligram of chlorophyll. Data points are the average of duplicate measurements, ± SD. (C) Arylsulfatase activity in the same cultures was measured using a similar method to that described in B and as reported by DAVIES et al. (1994). (D) Alkaline phosphatase (AP) and arylsulfatase activity (ARS) in wild-type and psr1 cultures grown for 3 days in the dark in TAP (+P) vs. TA (–P) medium. Enzyme activities were normalized to equal cell number (A410 per 4 x 108 cells) rather than equal chlorophyll in this experiment, since dark-grown cells accumulate less chlorophyll than cells grown in the light (FORD et al. 1981). However, for these strains, this number of cells (in the light) is approximately equivalent to 1 mg of chlorophyll (MOSELEY et al. 2002), so the enzyme activites are roughly comparable between B, C, D, and E. (E) Arylsulfatase activity in TAP and after 24 hr of growth in TAP –S medium.

 

Activation of SO₄^2–-assimilation pathway genes in P-deficient psr1 cells:

A previous study analyzed global patterns of P deficiency-responsive gene expression in the parental wild-type strain and the psr1 mutant, demonstrating that a number of genes encoding proteins associated with S assimilation were upregulated in P-starved psr1 cells (MOSELEY et al. 2006). To validate these results we used quantitative real-time PCR to compare the relative abundance of transcripts from a cross section of genes involved in S scavenging and assimilation. After 24 hr of P deprivation in wild-type cells, the transcript abundance of PHOX, which encodes the major, derepressible secreted phosphatase, increased by 5000-fold, whereas expression of the S-assimilation genes either remained constant or declined slightly (Figure 2). In contrast, psr1 cells grown for 24 hr in P-free medium failed to upregulate PHOX, but exhibited 30- to 200-fold increases in the levels of the ARS1 and ARS2 transcripts, which encode the major, secreted arylsulfatase enzymes (Figure 2). Although arylsulfatase activity steadily increased in P-deficient psr1 cultures over a 7-day period (Figure 1C), the peak abundance of ARS2 transcripts was observed at 24 hr after the cells were transferred to medium devoid of Pi (data not shown). The SLT1 and SLT2 transcripts, encoding putative Na⁺/SO₄^2– transporters, increase in abundance by 2000-fold and 50-fold, respectively, compared to the transcript levels in P-replete cells (Figure 2). The amount of the SULTR2 transcript, which encodes a putative H⁺/SO₄^2– transporter, increased in the psr1 strain by 40-fold after 24 hr of P starvation. Smaller increases of between 2.5- and 20-fold were observed for SAT1, OASTL4, ECP76, and SBDP mRNAs, encoding, respectively, a serine acetyl-transferase, an O-acetyl-serine(thiol)lyase (cysteine synthase), an extracellular protein potentially associated with the cell wall and a putative selenium-binding protein. It should be noted that the maximum levels of all of these transcripts in psr1 cells starved for P were one to two orders of magnitude below peak levels attained for wild-type cells starved for S (GONZALEZ-BALLESTER et al. 2008; D. GONZALEZ-BALLESTER and W. POOTAKHAM, unpublished results). The results shown in Figures 1 and 2 clearly demonstrate that the regulatory pathway that controls S deficiency-responsive gene expression is partially activated in P-starved psr1 cells.

View larger version (8K): In this window In a new window Download PPT slide   FIGURE 2.— Expression of S deficiency-responsive genes in P-starved wild-type and psr1 mutant cells. Relative abundance of P deficiency and S deficiency-responsive transcripts compared to the the CBLP control transcript in wild-type (CC-1690) and psr1 cells grown in TAP, and after 24 hr of growth in TA. Relative transcript abundance is calculated from the equation 2–C(t) x 105. PHOX encodes alkaline phosphatase, ARS1 encodes arylsulfatase 1, ARS2 encodes arylsulfatase 2, SLT1 encodes SAC1-like SO42–-transporter 1, SLT2 encodes SAC1-like SO42–-transporter 2, SULTR2 encodes a putative H+/SO42– transporter, SAT1 encodes serine acetyl transferase, OASTL4 encodes O-acetyl serine (thiol)lyase 4, ECP76 encodes a 76-kDa extracellular protein, and SBDP encodes a putative selenium-binding protein. Results from two independent experiments are shown.

 

View this table: In this window In a new window   TABLE 2 Summary of arylsulfatase (ARS) activity and S deficiency-responsive gene expression phenotypes in psr1 and S-deficiency acclimation regulatory mutants

 

Epistasis analysis of PSR1 and regulators of the S-deficiency response:

Are components of the signal transduction pathway that elicit changes in gene expression during S starvation also involved in the aberrant activation of S-assimilation genes in P-starved psr1 cells? To investigate the relationship between PSR1 and genes that encode regulators of S-deficiency acclimation responses, the psr1-1 mutant was crossed to sac1, snrk2.2, and snrk2.1 strains. As mentioned in the introduction, SAC1 and SNRK2.1 encode a putative sensor and serine/threonine kinase, respectively, that are required for activation of S deficiency-responsive gene expression (DAVIES et al. 1996; GONZALEZ-BALLESTER et al. 2008). SNRK2.2 encodes a serine/threonine kinase that acts as a negative regulator of S deficiency-responsive gene expression (DAVIES et al. 1999; RAVINA et al. 2002). Assays for growth, alkaline phosphatase activity, and arylsulfatase activity were used to compare the acclimation response phenotypes of wild-type, psr1, sac1, snrk2.2, snrk2.1, and combinations of double and triple mutants during P and S starvation (Figure 3 and Table 2). In control experiments, strains with wild-type PSR1 exhibited normal induction of phosphatase activity in –P medium, but only low, basal levels of activity were observed for strains carrying the psr1-1 allele [Figure 3, A (column 7) and B]. Conversely, after 24 hr of S starvation arylsulfatase expression was blocked in all strains containing the snrk2.1 allele and in sac1 strains that did not also contain the snrk2.2 mutation [Figure 3, A (column 6) and D]. As has been reported previously, snrk2.2 strains have low, constitutive arylsulfatase activity in S-replete medium, and accumulate approximately one-third as much arylsulfatase activity in S-deficient medium as wild-type cells (DAVIES et al. 1994, 1999) [Figure 3, A (columns 4 and 6) and D]. The epistasis relationship between SAC1 and SNRK2.2 is unclear. Strains containing both the sac1 and snrk2.2 alleles have slightly lower constitutive arylsulfatase activity in S-replete medium and express less arylsulfatase in S-deficient medium than the snrk2.2 single mutant (DAVIES et al. 1994) [Figure 3, A (columns 4 and 6) and D].

View larger version (25K): In this window In a new window Download PPT slide   FIGURE 3.— Epistasis analysis of PSR1, SAC1, SNRK2.2, and SNRK2.1. (A) Phenotypic analysis of growth, alkaline phosphatase, and arylsulfatase activities for single, double, and triple mutants on TAP (+P +S), TA + glucose-1-phosphate (–P +S) and TAP –S + thiocyanate (+P –S) solid medium. All strains were spotted onto plates from TAP medium, grown for 4 days, and then assayed for phosphatase or arylsulfatase activities. (B) Liquid culture assays for alkaline phosphatase activites in TAP and after 24 and 72 hr of growth in TA medium for wild-type cells (CC-1690) and various mutants, including psr1, sac1, snrk2.2, snrk2.1, and double- and triple-mutant strains, described in the text and in Table 1. Data points are the average of duplicate measurements (± SD). (C) Arylsulfatase activity assay of cells from the same strains as in B. (D) Comparison of arylsulfatase activity in TAP vs. after 24 hr of growth in TAP –S medium. Data for A, B, C, and D are from one representative experiment; comparable results were obtained from experimental replicates.

 
The level of arylsulfatase activity in P-starved psr1 and psr1sac1 cells was similar, suggesting a hypostatic relationship between SAC1 and PSR1 with respect to this phenotype [Figure 3, A (column 5) and C]. However, we cannot eliminate the possibility that the sac1(ars5-3) allele may produce some partially functional SAC1 protein since in this allele the ARG7 marker is integrated into the 13th intron of the SAC1 gene without generating a deletion (see supplemental Figure S3) and SAC1 transcript levels are normal (data not shown). However, double mutants of psr1 with either of two additional sac1 alleles (ars4-15 and ars4-55) also produced arylsulfatase during P starvation (data not shown). The sac1(ars4-55) allele is likely to be null since the SAC1 transcript could not be detected by quantitative real-time PCR (data not shown), supporting the conclusion that SAC1 is not required for the accumulation of arylsulfatase activity observed in the psr1 mutant during P starvation.

In contrast to the results discussed for the sac1 alleles, SNRK2.1 is epistatic to PSR1; no P deficiency-inducible arylsulfatase is expressed in any of the psr1 strains that also harbor a snrk2.1 allele [Figure 3, A (column 5) and C]. The SAC1 and SNRK2.1 genes are both essential for normal activation of arylsulfatase when cells are exposed to –S conditions, but only the wild-type SNRK2.1 allele is required for low-level arylsulfatase activity in P-starved psr1 cells. Conversely, a synergistic epistasis relationship is observed between PSR1 and SNRK2.2 since the arylsulfatase activity produced by P-starved psr1snrk2.2(are10) double mutants is far greater than the sum of the activities of either single mutant [compare Figure 3, A (column 5) and C]. Similar results were obtained with psr1snrk2.2(are16) double mutants (see supplemental Figure S1). The level of arylsulfatase activity associated with psr1snrk2.2 cells deprived of P for 72 hr is comparable to the level attained by wild-type cells after 24 hr of S starvation (compare Figure 3, C and D). This result implies that SNRK2.2 inhibits full activation of arylsulfatase in P-starved psr1 cells.

As expected, the triple mutants psr1sac1snrk2.1 and psr1snrk2.2snrk2.1 do not exhibit arylsulfatase on –P medium [Figure 3, A (column 5) and C], demonstrating that SNRK2.1 is essential for arylsulfatase induction during P starvation. Interestingly, the psr1sac1snrk2.2 triple mutant does not display the high level of ARS in P-deficient cells that is observed in the psr1snrk2.2 double mutant; rather the level of ARS activity resembles that of the psr1 single mutant. Consequently we conclude that (1) SNRK2.1 is absolutely essential for all arylsulfatase activity observed during P deprivation of the psr1 mutant, (2) neither SAC1 nor SNRK2.2 is essential for the low level of arylsulfatase activity that occurs when psr1 cells are starved for P, and (3) SAC1 is required for the high arylsulfatase activity measured in P-deprived psr1snrk2.2 mutant cells.

S deficiency-responsive transcripts in psr1 and S-assimilation regulatory mutants:

Comparison of ARS1 transcript abundance in P-replete vs. P-starved wild-type, psr1, sac1, snrk2.2, snrk2.1, and the respective double and triple mutants (Figure 4A) is mostly consistent with the level of ARS activity observed in these strains (Figure 3), with the exception of psr1snrk2.2 relative to psr1sac1snrk2.2 (see below). As shown in Figure 2, wild-type cells display no significant activation of ARS1 after 24 hr of P deficiency, but ARS1 is upregulated 100- to 200-fold in P-deprived psr1 cells. Like the situation in wild-type cells, very limited ARS1 induction occurs in the sac1 and snrk2.1 strains (although there are differences in the basal level of ARS1 transcripts between the tested strains). An average 200-fold increase in ARS1 mRNA is observed in P-deficient psr1sac1 cells, consistent with the hypostatic relationship of SAC1 to PSR1 reported above; on the other hand, ARS1 mRNA does not accumulate in P-starved psr1snrk2.1, psr1sac1snrk2.1, or psr1snrk2.2snrk2.1 mutant cells, confirming that SNRK2.1 is absolutely required for activation of S deficiency-responsive gene expression in psr1 cells that are starved for P.

View larger version (11K): In this window In a new window Download PPT slide   FIGURE 4.— ARS1 and PHOX gene expression in psr1 and the various S-deficiency response mutants. Relative abundance of (A) ARS1 transcripts and (B) PHOX transcripts in wild-type (CC-1690), psr1, sac1, snrk2.2, snrk2.1, and double- and triple-mutant strains grown in TAP (+P), and after 24 hr of growth in TA (–P). The data for the wild-type and psr1 strains is reproduced from Figure 2 to facilitate comparison with the other strains. (C) Comparison of the relative abundance of the indicated transcripts in the psr1snrk2.2 and psr1sac1snrk2.2 strains. Results from one to four experimental replicates are shown for each strain and each condition.

 
The snrk2.2 mutant expresses ARS1 at a high basal level (Figure 4A) (DAVIES et al. 1999), consistent with the constitutive arylsulfatase activity observed in P-replete medium [Figure 3, A (column 4) and C] (DAVIES et al. 1994). However, we were surprised to observe that after 24 hr of P starvation, ARS1 transcript abundance in the snrk2.2 single mutant increased by 30-fold and reached an even higher level than that observed in psr1 (Figure 4A). This increase in ARS1 mRNA does not yield a proportional increase in arylsulfatase activity (Figure 3C), suggesting that post-transcriptional mechanisms may regulate arylsulfatase synthesis. In fact, as with P-deficient psr1 cells, a general derepression of S deficiency-responsive genes is observed in P-starved snrk2.2 cells (data not shown). Like the snrk2.2 single mutant, P-replete psr1snrk2.2 cells have high basal levels of ARS1 and other –S transcripts (Figure 4, A and C), but in this case the transcript levels that are produced in P-deficient cells are comparable with wild-type induction in –S (GONZALEZ-BALLESTER et al. 2008). This high level of expression is consistent with the high arylsulfatase activity observed in P-deficient psr1snrk2.2 cells (Figure 3, A and C). Taken together, these results suggest that activators of the S-assimilation pathway are stimulated both in wild-type and in psr1 under P-starvation conditions, but the negative regulatory effect of SNRK2.2 normally dominates, suppressing full expression of S deficiency-responsive genes in cells starved for P.

Surprisingly, SAC1 appears to have only a limited effect on the abundance of –S transcripts in P-deficient snrk2.2 or psr1snrk2.2 cells. Although P-starved sac1snrk2.2 cells have slightly less arylsulfatase activity compared to snrk2.2 cells, and psr1sac1snrk2.2 cells have significantly less arylsulfatase activity than psr1snrk2.2 strains (Figure 3, A and C), the abundance of ARS1 and other S deficiency-responsive transcripts is minimally affected in any of the strains containing the sac1 mutation (Figure 4, A and C). The discrepancy between the abundance of ARS transcripts and arylsulfatase activity in the psr1snrk2.2 and psr1snrk2.2sac1 strains suggests that SAC1 in essential for post-transcriptional regulation of arylsulfatase synthesis, activity, or both.

psr1snrk2.2 cells bleach during P limitation:

We observed, on a per-volume basis, that psr1snrk2.2 cultures from two independent psr1snrk2.2(are10) isolates began bleaching significantly after 3 days of growth in medium with no added Pi, whereas P-deficient wild-type and psr1 cultures exhibited no decline or less of a decline, respectively, during the same time period (Figure 5A). The cell density increased over the 7 days time course for all of the cultures with the exception of psr1snrk2.1snrk2.2 (wild-type cells divided two to three times, psr1 and psr1snrk2.2 cells divided one to two times) (Figure 5B), indicating that chlorophyll declined on a per-cell basis in the psr1 and psr1snrk2.2 strains. Examination of cell staining with Evan's Blue dye during growth in P-deficient medium revealed that psr1snrk2.2 cells began to lose membrane integrity following the third day in medium devoid of P, in parallel with the bleaching (Figure 5C). Essentially no loss of viability was observed for wild-type cells over the same time period (Figure 5C). Viability of psr1 cells also declined after 3 days of P starvation, but at a slower rate than in psr1snrk2.2 cultures. Interestingly, P-deficient psr1snrk2.1snrk2.2 cells did not bleach and maintained significantly higher cell viability than psr1 or psr1snrk2.2 cultures (Figure 5, A and C). Furthermore, the bleaching of P-deficient psr1 and psr1snrk2.2 cells is partially rescued by growth in the dark (data not shown), suggesting that to some extent the loss of cell viability may be caused by photodamage. These results indicate that the high level of activation of the S-deficiency responses is deleterious to the survival of P-deficient psr1snrk2.2 cells, and that the cells can be rescued to a significant extent by the snrk2.1 lesion, which blocks expression of the S-responsive genes.

View larger version (14K): In this window In a new window Download PPT slide   FIGURE 5.— Chlorophyll content and cell viability in –P cultures. (A) Time course of chlorophyll concentration per culture volume in wild-type cells and various mutants grown in TA (Pi-free) medium for 7 days. Chlorophyll concentrations are the average of duplicate measurements, ± SD. The data are from one representative experiment; similar patterns were observed in two additional, independent experiments (see supplemental Figure S2). (B) Change in cell number for wild-type cells and the different mutant strains at various times following the transfer of cells to TA (Pi-free) medium. The changes in cell number are the average of two to four technical replicates, ± SD. (C) Time course of viability of strains grown in TA (Pi-free) medium. Nonviable cells were distinguished from viable cells by staining with 0.25% Evans Blue dye. Percentage viability is calculated from the average of two to four measurements, ± SD. The data are from one representative experiment; similar results were obtained in an additional, independent experiment (see supplemental Figure S2). For all experiments the wild-type strain used was CC-1690 while the mutants were psr1, two independent psr1snrk2.2 isolates (see Table 1), and psr1snrk2.1snrk2.2.

 

We have demonstrated that the psr1 mutant activates S deficiency-responsive genes during growth in P-deficient medium, and have investigated the role of the regulatory elements SAC1, SNRK2.1, and SNRK2.2, which normally control S-deprivation responses, in this activation response. Epistasis analysis indicates that activation of S deficiency-responsive genes in P-starved psr1 cells does not require a significant positive regulatory input from the SAC1 gene product. On the other hand, the central regulatory kinase, SNRK2.1, is absolutely required for S-assimilation gene activation in P-deficient psr1 cells. The synergistically epistatic relationship between PSR1 and SNRK2.2 shows that the SNRK2.2 kinase has an inhibitory effect on the expression of S-deficiency response target genes in psr1 cells deprived of P, similar to its role in S-replete wild-type cells. However, unlike the situation for nutrient-replete cells, where S deficiency-responsive gene expression is only partially induced, the absence of SNRK2.2 in the psr1snrk2.2 double mutant enables close to full activation of the –S target genes during P deprivation. The phenotype of P-starved psr1snrk2.2 strains points to the existence of an internal sensor of cellular S status that is capable of almost fully activating S deficiency-responsive gene expression even when high levels of SO₄^2– are present in the medium. The psr1sac1 and the psr1sac1snrk2.2 mutants reveal that although SAC1 is not required for the induction of S-deficiency responsive genes in P-deficient psr1 cells, it is required for the high activity of arylsulfatase observed in P-starved psr1snrk2.2 cells, suggesting a role for SAC1 in post-transcriptional regulation of some S deficiency-inducible proteins.

Regulation of S deficiency-response genes:

What is the mechanism of –S gene activation in P-starved psr1 cells? In Figure 6 we present a speculative model to describe the interactions between the regulatory factors that control expression of the S-assimilation genes and activation of the S-responsive genes in the psr1 mutant. Under nutrient-replete conditions where the external concentration of SO₄^2– ion is high, the SNRK2.2 kinase represses S deficiency-responsive gene expression (Figure 6A). Since the cell is both internally and externally replete for S, the positive regulators SAC1 and SNRK2.1 are not activated and only basal levels of transcripts for the S-assimilation genes accumulate. However, exclusion of SO₄^2– from the medium leads to activation of SAC1 and SNRK2.1 (Figure 6B). From the phenotype of sac1 mutants, in which the induction of S deficiency-responsive transcripts is attenuated (DAVIES et al. 1994; ZHANG et al. 2004; GONZALEZ-BALLESTER et al. 2008), we infer that a major function of SAC1 is to inactivate SNRK2.2 in S-starved cells, allowing for full, high-level induction of S deficiency-responsive gene targets by the activator SNRK2.1 (Figure 6B). In –S sac1 mutant strains, repression by SNRK2.2 is never released and consequently there is only low-level activation by SNRK2.1. We have noted that many of the sac1 progeny from genetic crosses have low and variable levels of arylsulfatase activity when deprived of S (see supplemental Figure S1). This may reflect strain-to-strain variation in the balance between repression and activation of S deficiency-responsive gene expression by SNRK2.2 and SNRK2.1, respectively.

View larger version (28K): In this window In a new window Download PPT slide   FIGURE 6.— Models of S deficiency-responsive gene regulation. (A) Wild-type, nutrient-replete conditions. (i) SO42– ion is bound to SAC1 which is in an inactive state. (ii) SNRK2.2 makes the central kinase, SNRK2.1, ineffective in activation of gene expression. (iii) The cell is internally replete for S, so (iv) SNRK2.1 is maintained in a basal activation state. (v) S deficiency-responsive genes are expressed at basal levels. (B) Regulation of S-starvation responses in wild-type cells under S-starvation conditions. (vi) SAC1 no longer binds SO42– ion and becomes activated. SAC1 causes inactivation of (vii) the SNRK2.2 repressor, which in turn allows SNRK2.1 to function as a transcriptional activator. (viii) Internal S deficiency further activates (ix) SNRK2.1, causing (x) full expression of S deficiency-responsive genes. (xi) SAC1 and SNRK2.2 also function in post-translation regulation of arylsulfatase and other sulfur-responsive proteins. (C) Regulation of S-starvation responses in wild-type cells under P-starvation conditions. (xii) The SAC1 protein is bound to SO42– and is unable to block repression of gene expression by SNRK2.2. Sulfolipid biosynthesis increases and SO42– uptake and assimilation may be limited during P stress, potentially causing (xiii) moderate internal S deficiency. Nevertheless, (xiv) SNRK2.2 suppresses activation of the S-starvation responsive genes by (xv) SNRK2.1, so (xvi) expression of S-starvation responsive genes is maintained at the low basal level. (xvii) PSR1 becomes active and elicits (xviii) full expression of the P-starvation responsive genes. The cells fully acclimate to P-starvation conditions. (D) Regulation of S-starvation genes in psr1 mutant cells under P-starvation conditions. Since SO42– is plentiful in the medium, (xix) SAC1 is in the inactive state, and (xx) SNRK2.2 suppresses activation of the S-starvation responsive genes by SNRK2.1. psr1 cells may experience (xxi) severe internal S deprivation as a result of (xxii) their inability to acclimate properly to P starvation. An equilibrium between repression by SNRK2.2 and (xxiii) activation of SNRK2.1 by the internal signal is achieved, resulting in (xiv) a low level of induction of S deficiency-responsive genes.

 
The phenotypes of snrk2.2 strains are very informative with respect to our understanding of the roles of the S-deficiency response regulators. Nutrient-replete snrk2.2 mutants display elevated basal expression of most S-assimilation genes compared to wild-type cells (Figure 4A) (DAVIES et al. 1994; RAVINA et al. 2002; GONZALEZ-BALLESTER et al. 2008), but the loss of SNRK2.2 is not sufficient by itself to cause expression of S-assimilation genes at the levels observed in S-starved cells. S deficiency-responsive transcripts are induced similarly in S-starved wild-type and snrk2.2 cells (DAVIES et al. 1994; RAVINA et al. 2002; GONZALEZ-BALLESTER et al. 2008), suggesting that activation of the –S acclimation response is not achieved simply through repression of SNRK2.2; rather, the SAC1-dependent inactivation of SNRK2.2 is coordinated with the S deprivation-dependent activation of the SNRK2.1 kinase (Figure 6B). Although the absence of SNRK2.2 does not significantly affect induction of S deficiency-responsive genes in S-starved cells, like SAC1 the SNRK2.2 kinase also may play a role in post-transcriptional regulatory processes that affect protein expression and activity. S-starved snrk2.2 strains have only 40–50% of the wild-type level of arylsulfatase activity (Figure 3D) (DAVIES et al. 1994) and fail to accumulate the normal complement of extracellular proteins that are expressed in –S cells (DAVIES et al. 1994).

The partial derepression of S deficiency-responsive genes that is observed in nutrient-replete and P-deficient snrk2.2 mutants also occurs in sac1snrk2.2 double mutants (Figure 4A) (DAVIES et al. 1994). Also, ARS transcript abundance increases in S-starved sac1snrk2.2 cells, albeit to a somewhat lesser extent than in wild-type or in snrk2.2 single mutants (DAVIES et al. 1994), indicating that activation of the S deficiency-responsive genes does not absolutely require SAC1 (DAVIES et al. 1994).

In contrast to the situation in wild-type cells, SNRK2.2 does not completely inhibit expression of S deficiency-responsive genes in P-starved psr1 cells (Figure 6D). This result is consistent with the hypothesis that the metabolic state of P-starved cells leads to a secondary, internal S deficiency (discussed below) that results in partial gene activation. Since the psr1sac1 double and psr1 single mutants under –P conditions have similar levels of transcripts associated with the S-deficiency response, the stimulatory signal that overrides SNRK2.2 repression is not transmitted via SAC1. Therefore, accumulation of S deprivation-associated transcripts in P-starved, psr1 cells must depend either on a positive regulator that interacts with SNRK2.1 or on promiscuous modification/activation of SNRK2.1. Unlike the situation during nutrient sufficiency, elimination of SNRK2.2 repression in P-deficient psr1snrk2.2 mutant cells leads to full activation of the signaling cascade and high S deficiency-responsive gene expression (Figure 4, A and C). Therefore, while low internal SO₄^2– can lead to partial activation of SNRK2.1 in the psr1 strain, additional relief of SNRK2.2 repression can cause full activation. Furthermore, under these conditions the activation of –S responsive genes is uncoupled from SAC1 since similar transcript levels are attained in P-starved psr1snrk2.2 and psr1sac1snrk2.2 cells (Figure 4, A and C).

While SAC1 does not influence the level of transcripts from S derpivation-responsive genes in P-starved psr1snrk2.2 cells, it does have a significant stimulatory effect on the level of arylsulfatase activity in this strain (Figure 3, A and C). SAC1 is an integral membrane protein similar to SO₄^2– transporters of the SLC13 family, and it has been hypothesized that the protein functions as a sensor of external SO₄^2– (DAVIES et al. 1996). In situations where the external SO₄^2– concentration is high and the abundance of ARS transcripts is comparable, such as nutrient-replete snrk2.2 and sac1snrk2.2 cells, or P-starved psr1snrk2.2 vs. psr1sac1snrk2.2, the presence of SAC1 affects the amount of active arylsulfatase. In addition, preliminary data indicates that SAC1 is also required for the accumulation of SO₄^2– transporter polypeptides in psr1snrk2.2 cells starved for P (W. POOTAKHAM, unpublished results), supporting a role for SAC1 in post-transcriptional regulatory processes.

The partial induction of arylsulfatase and high-level expression of the S deficiency-responsive genes in S-starved sac1snrk2.2 strains and P-starved psr1sac1snrk2.2 strains (Figures 3 and 4) confirm that gene activation during S deficiency can be largely uncoupled from SAC1 and SNRK2.2, providing us with new insight into the regulation of the S-responsive pathway. A simple hypothesis is that the activation state of the SNRK2.1 kinase can be directly affected by the intracellular S status (Figure 6). Alternatively, an unknown regulator may interact with SNRK2.1 and stimulate its activity when intracellular S is low. In all cases tested, SNRK2.1 is absolutely required for induction of S-assimilation genes, confirming the central regulatory role of this kinase (Figure 6) (GONZALEZ-BALLESTER et al. 2008).

P-limitation affects intracellular S metabolism:

What leads to the activation of SNRK2.1 in P-starved psr1 cells? A hypothesis that is consistent with the experimental evidence is that P limitation causes a secondary, internal S deficiency, despite the abundance of SO₄^2– in the –P medium (Figure 6C). P-deficient cells could experience internal S limitation as a consequence of at least two acclimation responses: (i) cellular S is redirected toward sulfolipid synthesis for replacement of thylakoid membrane phospholipids (RIEKHOF et al. 2003), and (ii) SO₄^2– assimilation is inhibited, possibly the consequence of a Pi conservation regime in which cells redirect ATP away from SO₄^2– assimilation and toward other essential processes. Recently, it was shown that P-starved cells accumulate 25 times more cysteine than nutrient-replete cells (BOLLING and FIEHN 2005), supporting the idea that P deficiency causes significant changes in the intracellular S status. In psr1 cells starved for P, the degree of internal P deficiency may be even more extreme than in wild-type cells. While psr1 mutants cannot scavenge Pi from the medium (SHIMOGAWARA et al. 1999) and may be defective in mobilization of internal P stores (MOSELEY et al. 2006), they arrest growth and downregulate photosynthesis more quickly than wild-type cells, indicating that they are to some extent able to adjust to their abnormal physiological state (SHIMOGAWARA et al. 1999). In this respect the phenotype of P-deficent psr1 mutants contrasts with the phenotypes of S-starved sac1 and snrk2.1 strains, which fail both to upregulate S scavenging and assimilation and to downregulate photosynthetic electron transport, resulting in photodamage and cell death at moderate light intensities (WYKOFF et al. 1998; SHIMOGAWARA et al. 1999; GONZALEZ-BALLESTER et al. 2008). A possible consequence of the extreme P deficiency experienced by P-starved psr1 cells may be that the level of secondary S stress reaches a threshold that activates SNRK2.1 beyond the point at which it can be fully repressed by SNRK2.2, resulting in low-level induction of the S-assimilation genes (Figure 6D). A constitutively inhibitory allele of SNRK2.2 might produce a similar effect in S-starved cells. These hypotheses could be tested by comparing the levels of S-containing metabolites in P-starved wild-type and psr1 cells and by analyzing the phosphorylation state of SNRK2.1.

Alternative models:

While a preponderence of circumstantial evidence suggests that S metabolism is affected by P deficiency, alternative models can be invoked to explain the mutant phenotypes. For example, there are six additional members of the SNRK2 family of protein kinases besides SNRK2.1 and SNRK2.2 in Chlamydomonas (GONZALEZ-BALLESTER et al. 2008). If one or more of these members functions in a regulatory cascade that activates PSR1 during P starvation, in psr1 mutants where the regulatory pathway is blocked the kinase(s) might interact with SNRK2.1 and cause aberrant induction of S deficiency-responsive gene expression. Other stress signals could also contribute to the activation of S-starvation responses during P limitation. In Arabidopsis, cysteine biosynthesis is enhanced during light and oxidative stress through the activity of the cyclophilin CYP20-3, which assists in the folding of chloroplast serine acetyltransferase (DOMINGUEZ-SOLIS et al. 2008). We have ruled out light stress as the causative agent in the activation of –S responses in the psr1 mutant since similar arylsulfatase activity was observed in P-deficient cultures in the light or in the dark (Figure 1D). Nevertheless, it is possible that other "general" stress signals are responsible for the aberrant –S responses in P-starved psr1 cells.

Proximal vs. secondary phenotypes of psr1:

Genetic analysis of the interactions between regulators of the S deficiency-responsive genes and PSR1 provides a cautionary tale that illustrates the difficulty of deciphering the wild-type function of a gene from its mutant phenotype. Some aspects of a mutant phenotype may be caused by acclimation to a novel physiological state that is caused by the lesion, rather than resulting directly from the loss-of-function of a particular gene. For example, a preponderance of evidence indicates that PSR1 encodes a nuclear-localized transcriptional activator (WYKOFF et al. 1999; RUBIO et al. 2001). A possible model that would account for the expression of the S deficiency-responsive genes in P-starved psr1 cells is that PSR1 itself negatively regulates their expression, either by participating directly in the S deprivation-responsive signaling cascade, interacting with the promoters of the target genes of this cascade, or by inducing expression of a gene encoding a repressor of S-responsive genes during P deficiency. While we cannot fully rule out a role for PSR1 in repressing the S-deficiency responses, this hypothesis does not explain why cells that acclimate properly to P starvation would find it necessary to repress the S deficiency-responsive genes when plenty of SO₄^2– is present in –P growth medium, or why expression of S-assimilation genes should increase in P-starved snrk2.2 cells (Figure 4), even though PSR1-dependent responses occur normally in this strain (Figures 3 and 4). Given the evidence that P starvation affects the internal S status of the cell, regulation of the S-deficiency responses can be understood without invoking any direct involvement of PSR1 in the signal transduction pathway. Instead, the metabolic state of P-starved psr1 cells provides an "artificial" intracellular environment in which interactions between the P- and S-stress responses are unmasked.

Similar care should be taken in ascribing control of aspects of a wild-type acclimation response to a particular regulatory protein. For example, part of the response of wild-type cells to P limitation is to stabilize chloroplast RNA transcripts (cpRNAs), and this stabilization correlates with downregulation of the chloroplast polynucleotide phosphorylase (PNPase) (YEHUDAI-RESHEFF et al. 2007). PSR1 has been implicated in this process, since cpRNA abundance declines drastically in psr1 cells that are starved for P and the expression of the PNP1 gene is not downregulated as it is in P-starved wild-type cells (YEHUDAI-RESHEFF et al. 2007). However, reduced cpRNA abundance is also characteristic of S-starved cells (IRIHIMOVITCH and STERN 2006), making it unclear whether the decline in cpRNAs in P-starved psr1 cells is a consequence of their inability to reduce the level of PNPase activity, the partial activation of the S-deficiency response, or a combination of the two. It should be possible to distinguish between these possibilities by analyzing cpRNA stability and transcription rates in P-deficient psr1 cells.

Why would it be important for wild-type cells to repress S deficiency-responsive genes during P limitation? This would be a logical strategy for a cell that perceives internal P and S deprivation but that can sense that SO₄^2– is not limiting in the external environment. Upregulation of S-deficiency responses would be both energetically costly and futile in this situation since, because of P insufficiency, the cell would be unable to grow even if it acquired more S. In this context, it is unclear whether partial activation of S-deficiency responses provides any benefit to P-starved psr1 cells. Along this vein, we wondered whether the more rapid downregulation of photosynthesis in P-deficient psr1 cells compared to wild-type cells (SHIMOGAWARA et al. 1999) was a byproduct of the partial activation of the S-deficiency response, since photosynthesis is downregulated more quickly during S starvation than in P starvation (WYKOFF et al. 1999). Comparison of PAR curves for P-deficient wild-type, psr1, and psr1snrk2.1 cells revealed that the S-deficiency responses were not likely to be involved in the downregulation of photosynthesis; both the psr1 and psr1snrk2.1 strains showed a similar, rapid reduction in photosynthetic electron transfer rates after 24 hr of P starvation, while wild-type cells maintained relatively high rates of photosynthesis (data not shown). Cell viability during P starvation is actually improved in a strain that harbors both the psr1 and the snrk2.1 mutations, compared to psr1 (Figure 5C). Furthermore, the rapid loss of chlorophyll and cell viability in psr1snrk2.2 cultures during P deprivation (Figure 5) indicates that full activation of the S-deficiency response is harmful rather than beneficial to the cell.

While P and S limitation have qualitatively similar effects on growth and photosynthesis (WYKOFF et al. 1998), some specific responses to these two limitations are diametrically opposed. For example, while P-starved cells turn over phospholipids and increase synthesis of sulfolipids (RIEKHOF et al. 2003), the opposite occurs in S-starved cells; sulfolipids are degraded and phospholipid abundance increases (SUGIMOTO et al. 2007, 2008). It is conceivable that opposing responses prevent maintenance of proper phospho- to sulfolipid ratios in the thylakoid membrane, contributing to the bleaching of –P psr1snrk2.2 cells. Examination of the lipid profiles of P-deficient psr1snrk2.2 cells should be informative in this regard.

Conclusion:

It has been somewhat surprising to find that analysis of the phenotype of the psr1 mutant, which does not acclimate normally to P deprivation, provides insight into regulatory mechanisms controlling S-starvation responses. However, there are still many questions about the SAC1, SNRK2.1, and SNRK2.2 regulatory proteins that remain unanswered. What are the internal signals that trigger S-starvation responses? What are the targets of the kinases? Do the phosphorylation states of the regulatory proteins change during S starvation? Do the regulatory proteins physically interact, and does S starvation change these interactions? Detailed biochemical analyses will be required to achieve a more comprehensive understanding of how cells acclimate to P- and S-limiting conditions and the integration between the two responses.

We thank Steve Pollock, Rachel Muir, and all members of the Grossman and Bhaya labs for helpful discussions. This work was supported by National Science Foundation grants MCB0235878 and MCB-0824469 awarded to A.R.G.

² These authors contributed equally to this study.

ARNON, D. I., 1949 Copper enzymes in isolated chloroplasts, polyphenoloxidase in Beta vulgaris. Plant Physiol. 24: 1–15.[Free Full Text]

BOLLING, C., and O. FIEHN, 2005 Metabolite profiling of Chlamydomonas reinhardtii under nutrient deprivation. Plant Physiol. 139: 1995–2005.[Abstract/Free Full Text]

BURKHARD, P., J. STETEFELD and S. V. STRELKOV, 2001 Coiled coils: a highly versatile protein folding motif. Trends Cell Biol. 11: 82–88.[CrossRef][Medline]

CHANG, C.-W., J. L. MOSELEY, D. WYKOFF and A. R. GROSSMAN, 2005 The LPB1 gene is important for acclimation of Chlamydomonas reinhardtii to phosphorus and sulfur deprivation. Plant Physiol. 138: 319–329.[Abstract/Free Full Text]

DAVIES, J., F. YILDIZ and A. R. GROSSMAN, 1996 Sac1, a putative regulator that is critical of survival of Chlamydomonas reinhardtii during sulfur deprivation. EMBO J. 15: 2150–2159.[Medline]

DAVIES, J. P., F. YILDIZ and A. R. GROSSMAN, 1994 Mutants of Chlamydomonas reinhardtii with aberrant responses to sulfur deprivation. Plant Cell 6: 53–63.[Abstract]

DAVIES, J. P., F. H. YILDIZ and A. R. GROSSMAN, 1999 Sac3, an Snf1-like serine/threonine kinase that positively and negatively regulates the responses of Chlamydomonas to sulfur limitation. Plant Cell 11: 1179–1190.[Abstract/Free Full Text]

DE HOSTOS, E. L., R. K. TOGASAKI and A. R. GROSSMAN, 1988 Purification and biosynthesis of a derepressible periplasmic arylsulfatase from Chlamydomonas reinhardtii. J. Cell. Biol. 106: 29–37.[Abstract/Free Full Text]

DE HOSTOS, E. L., J. SCHILLING and A. R. GROSSMAN, 1989 Structure and expression of the gene encoding the periplasmic arylsulfatase of Chlamydomonas reinhardtii. Mol. Gen. Genet. 218: 229–239.[CrossRef][Medline]

DOMINGUEZ-SOLIS, J. R., Z. HE, A. LIMA, J. TING, B. B. BUCHANAN et al., 2008 A cyclophilin links redox and light signals to cysteine biosynthesis and stress responses in chloroplasts. Proc. Natl. Acad. Sci. USA 105: 16386–16391.[Abstract/Free Full Text]

FORD, C., S. MITCHELL and W.-Y. WANG, 1981 Protochlorophyllide photoconversion mutants of Chlamydomonas reinhardtii. Mol. Gen. Genet. 184: 460–464.[CrossRef]

GASCH, A., and M. WERNER-WASHBURNE, 2002 The genomics of yeast responses to environmental stress and starvation. Funct. Integr. Genomics 2: 181–192.[CrossRef][Medline]

GONZALEZ-BALLESTER, D., S. V. POLLOCK, W. POOTAKHAM and A. R. GROSSMAN, 2008 The central role of a SNRK2 kinase in sulfur deprivation responses. Plant Physiol. 147: 216–227.[Abstract/Free Full Text]

GORMAN, D. S., and R. P. LEVINE, 1966 Cytochrome f and plastocyanin: their sequence in the photosynthetic electron transport chain of Chlamydomonas reinhardtii. Proc. Natl. Acad. Sci. USA 54: 1665–1669.[CrossRef]

GROSSMAN, A., and H. TAKAHASHI, 2001 Macronutrient utilization by photosynthetic eukaryotes and the fabric of interactions. Annu. Rev. Plant Physiol. Plant Mol. Biol. 52: 163–210.[CrossRef][Medline]

HEBELER, M., S. HENTRICH, A. MAYER, D. LEIBFRITZ and L. H. GRIMME, 1992 Phosphate regulation and compartmentation in Chlamydomonas reinhardtii studied by in vivo ³¹p-nmr, pp. 717–720 in Research in Photosynthesis, edited by N. MURATA. Kluwer Academic Publishers, Dordrecht, The Netherlands.

IRIHIMOVITCH, V., and D. B. STERN, 2006 The sulfur acclimation SAC3 kinase is required for chloroplast transcriptional repression under sulfur limitation in Chlamydomonas reinhardtii. Proc. Natl. Acad. Sci. USA 103: 7911–7916.

JONES, S., 2004 An overview of the basic helix-loop-helix proteins. Genome Biol. 5: 226.[CrossRef][Medline]

KOBAYASHI, I., S. FUJIWARA, K. SHIMOGAWARA, T. KAISE, H. USUDA et al., 2003 Insertional mutagenesis in a homologue of a Pi transporter gene confers arsenate resistance on Chlamydomonas. Plant Cell Physiol. 44: 597–606.[Abstract/Free Full Text]

LEIN, T., and O. SCHREINER, 1975 Purification of a derepressible arlysulfatase from Chlamydomonas reinhardtii. Biochimica et Biophysica Acta 384: 1147–1157.

LIPSICK, J. S., 1996 One billion years of Myb. Oncogene 13: 223–235.[Medline]

MOSELEY, J. L., T. ALLINGER, S. HERZOG, P. HOERTH, E. WEHINGER et al., 2002 Adaptation to Fe-deficiency requires remodeling of the photosynthetic apparatus. EMBO J. 21: 6709–6720.[CrossRef][Medline]

MOSELEY, J. L., C. W. CHANG and A. R. GROSSMAN, 2006 Genome-based approaches to understanding phosphorus deprivation responses and PSR1 control in Chlamydomonas reinhardtii. Eukaryot. Cell 5: 26–44.[Abstract/Free Full Text]

NGUYEN, A. V., S. R. THOMAS-HALL, A. MALNOE, M. TIMMINS, J. H. MUSSGNUG et al., 2008 Transcriptome for photobiological hydrogen production induced by sulfur deprivation in the green alga Chlamydomonas reinhardtii. Eukaryot. Cell 7: 1965–1979.[Abstract/Free Full Text]

O'CONNELL, K. F., and R. E. BAKER, 1992 Possible cross-regulation of phosphate and sulfate metabolism in Saccharomyces cerevisiae. Genetics 132: 63–73.[Abstract]

PABO, C. O., and R. T. SAUER, 1992 Transcription factors: structural families and principles of DNA recognition. Annu. Rev. Biochem. 61: 1053–1095.[CrossRef][Medline]

POLLOCK, S., W. POOTAKHAM, N. SHIBAGAKI, J. MOSELEY and A. GROSSMAN, 2005 Insights into the acclimation of Chlamydomonas reinhardtii to sulfur deprivation. Photosynthesis Res. 86: 475–489.[CrossRef][Medline]

QUISEL, J., D. WYKOFF and A. R. GROSSMAN, 1996 Biochemical characterization of the extracellular phosphatases produced by phosphorus-deprived Chlamydomonas reinhardtii. Plant Physiol. 111: 839–848.[Abstract]

RAVINA, C. G., C.-I. CHANG, G. P. TSAKRAKLIDES, J. P. MCDERMOTT, J. M. VEGA et al., 2002 The sac mutants of Chlamydomonas reinhardtii reveal transcriptional and posttranscriptional control of cysteine biosynthesis. Plant Physiol. 130: 2076–2084.[Abstract/Free Full Text]

RIEKHOF, W. R., M. E. RUCKLE, T. A. LYDIC, B. B. SEARS and C. BENNING, 2003 The sulfolipids 2'-O-Acyl-Sulfoquinovosyldiacylglycerol and Sulfoquinovosyldiacylglycerol are absent from a Chlamydomonas reinhardtii mutant deleted in SQD1. Plant Physiol. 133: 864–874.[Abstract/Free Full Text]

RUBIO, V., F. LINHARES, R. SOLANO, A. C. MARTIN, J. IGLESIAS et al., 2001 A conserved MYB transcription factor involved in phospate starvation signaling both in vascular plants and in unicellular algae. Genes Dev. 15: 2122–2133.[Abstract/Free Full Text]

SALDANHA, A. J., M. J. BRAUER and D. BOTSTEIN, 2004 Nutritional homeostasis in batch and steady-state culture of yeast. Mol. Biol. Cell 15: 4089–4104.[Abstract/Free Full Text]

SCHLOSS, J. A., C. D. SILFLOW and J. L. ROSENBAUM, 1984 mRNA abundance changes during flagellar regeneration in Chlamydomonas reinhardtii. Mol. Cell. Biol. 4: 424–434.[Abstract/Free Full Text]

SHIMOGAWARA, K., D. D. WYKOFF, H. USUDA and A. R. GROSSMAN, 1999 Chlamydomonas reinhardtii mutants abnormal in their responses to phosphorus deprivation. Plant Physiol. 120: 1–10.[Free Full Text]

SUGIMOTO, K., N. SATO and M. TSUZUKI, 2007 Utilization of a chloroplast membrane sulfolipid as a major internal sulfur source for protein synthesis in the early phase of sulfur starvation in Chlamydomonas reinhardtii. FEBS Lett. 581: 4519–4522.[CrossRef][Medline]

SUGIMOTO, K., T. MIDORIKAWA, M. TSUZUKI and N. SATO, 2008 Upregulation of PG synthesis on sulfur-starvation for PS I in Chlamydomonas. Biochem. Biophys. Res. Commun. 369: 660–665.[CrossRef][Medline]

TAKAHASHI, H., C. E. BRABY and A. R. GROSSMAN, 2001 Sulfur economy and cell wall biosynthesis during sulfur limitation of Chlamydomonas reinhardtii. Plant Physiol. 127: 665–673.

WERNER, T., N. AMRHEIN and F. FREIMOSER, 2007 Inorganic polyphosphate occurs in the cell wall of Chlamydomonas reinhardtii and accumulates during cytokinesis. BMC Plant Biol. 7: 51.[CrossRef][Medline]

WYKOFF, D., J. DAVIES and A. GROSSMAN, 1998 The regulation of photosynthetic electron transport during nutrient deprivation in Chlamydomonas reinhardtii. Plant Physiol. 117: 129–139.[Abstract/Free Full Text]

WYKOFF, D., A. GROSSMAN, D. P. WEEKS, H. USUDA and K. SHIMOGAWARA, 1999 Psr1, a nuclear localized protein that regulates phosphorus metabolism in Chlamydomonas. Proc. Natl. Acad. Sci. USA 96: 15336–15341.[Abstract/Free Full Text]

YEHUDAI-RESHEFF, S., S. L. ZIMMER, Y. KOMINE and D. B. STERN, 2007 Integration of chloroplast nucleic acid metabolism into the phosphate deprivation response in Chlamydomonas reinhardtii. Plant Cell 19: 1023–1038.[Abstract/Free Full Text]

YILDIZ, F., J. P. DAVIES and A. R. GROSSMAN, 1994 Characterization of sulfate transport in Chlamydomonas reinhardtii during sulfur-limited and sulfur-sufficient growth. Plant Physiol. 104: 981–987.[Abstract]

YILDIZ, F. H., J. P. DAVIES and A. R. GROSSMAN, 1996 Sulfur availability and the SAC1 gene control adenosine triphosphate sulfurylase gene expression in Chlamydomonas reinhardtii. Plant Physiol. 112: 669–675.[Abstract]

ZHANG, L., T. HAPPE and A. MELIS, 2002 Biochemical and morphological characterization of sulfur-deprived and H₂-producing Chlamydomonas reinhardtii (green alga). Planta 214: 552–561.[CrossRef][Medline]

ZHANG, Z., J. SHRAGER, M. JAIN, C.-W. CHANG, O. VALLON et al., 2004 Insights into the survival of Chlamydomonas reinhardtii during sulfur starvation based on microarray analysis of gene expression. Eukaryot. Cell 3: 1331–1348.[Abstract/Free Full Text]

Communicating editor: A. P. MITCHELL

International Access Link

Online ISSN: 1943-2361  Print ISSN: 0016-6731
Copyright 2009 by the Genetics Society of America
phone: 412-268-1812   fax: 412-268-1813   email: genetics-gsa{at}andrew.cmu.edu