The sulfur acclimation SAC3 kinase is requiredfor chloroplast transcriptional repression undersulfur limitation in Chlamydomonas reinhardtiiVered Irihimovitch and David B. Stern*

Boyce Thompson Institute for Plant Research, Cornell University, Tower Road, Ithaca, NY 14853

Edited by Maarten J. Chrispeels, University of California at San Diego, La Jolla, CA, and approved March 17, 2006 (received for review December 21, 2005)

Sulfur (S) deprivation responses have been studied extensively inalgae and land plants; however, little is known of the signals thatlink perception of S status to chloroplast gene expression. Here, wehave compared the chloroplast S limitation response in WT vs. sac1and sac3 sulfur acclimation mutants of the green alga Chlamydo-monas reinhardtii. We provide evidence that in the WT, chloroplasttranscriptional activity rapidly decreases after removal of S fromthe medium, leading to reduced transcript accumulation. Thisdecrease correlates with reduced abundance of a �70-like factor,Sig1, which is most likely the unique chloroplast transcriptionspecificity factor. We further show that reduced transcriptionactivity and diminished Sig1 accumulation are mediated by theSAC3 gene product, a putative Snf1-type Ser�Thr kinase previouslyshown to have both positive and negative effects on nuclear geneexpression. Inclusion of the protein kinase inhibitor 6-dimethyl-aminopurine during S limitation yielded a pattern of expressionthat was largely similar to that seen in the sac3 mutant, lendingsupport to the hypothesis that Sac3 kinase activation leads totranscriptional repression and Sig1 proteolysis. The finding thatSac3 regulates chloroplast gene expression suggests that it has apreviously unknown role in integrating the S limitation response inmultiple subcellular compartments.

sigma factor � dimethylaminopurine � RNA polymerase

Sulfur (S) is a required macronutrient; it is a constituent ofproteins, lipids, electron transport components, and many

cell metabolites (1, 2). Most photosynthetic organisms have thecapacity to assimilate S as a sulfate anion (SO4

2�) and translocateit to the plastid, where primary S metabolism takes place. Inchloroplasts, S is first activated by ATP sulfurylase to form APS(5�-adenylyl sulfate), which is a branch-point intermediate. APScan be phosphorylated by APS kinase, and the product can beused in sulfation reactions or to synthesize cysteine and methi-onine (1). Chloroplasts are thus crucial in the assimilation of S,because ATP and photosynthetic reductant are needed for thefirst activation and reduction steps.

S insufficiency constitutes a serious stress situation for plantsand algae, which respond with a series of metabolic adjustments(3, 4). Studies of S deprivation in the green alga Chlamydomonasreinhardtii have shown that limiting S leads to altered photosyn-thetic performance, reduced levels of photosynthetic proteins,and induction of genes encoding high-affinity S transporters (5,6). S deprivation also leads to increased accumulation of aryl-sulfatase (ARS), which releases sulfate anion from esterifiedorganic sulfates, allowing cells to access new sulfur stores (7).These acclimation responses are thought to prevent photodam-age under conditions where the cell has limited capacity formetabolism of S-containing compounds, particularly proteins.

Several Chlamydomonas mutants that do not respond cor-rectly to S deprivation have been identified based on theirinability to regulate ARS1 expression. Among these mutants aresac1 and sac3 (in which ‘‘sac’’ indicates sulfur acclimation). Thesac1 mutant is aberrant in most of the normal responses to Sdeprivation and dies rapidly under �S conditions. It is unable to

synthesize ARS, exhibits abnormal S uptake, and does notdown-regulate photosynthesis (8). The SAC1 gene encodes aprotein with hydrophobic domains similar to Na��SO4

2� iontransporters found in cell membranes of several organisms (8),but it has not been established whether it functions as an Stransporter in Chlamydomonas. Unlike sac1, sac3 exhibits con-stitutive low-level expression of ARS under normal growthconditions, but, like WT cells, it increases ARS transcriptionunder �S conditions. However, sac3 is unable to fully induce thehigh-affinity S transport system (9). Thus, SAC3 has bothpositive and negative effects on nuclear gene expression. It hasbeen shown that SAC3 encodes a putative Ser�Thr kinase relatedto Snf1p of Saccharomyces cerevisiae (9); however, the substratesand�or activators of Sac3 are undefined.

Although S limitation has been correlated with altered chlo-roplast functions, little is known of the signals that link percep-tion of S status to the chloroplast. Here, we have studied thechloroplast S stress response in Chlamydomonas, comparing theWT to sac1 and sac3. We provide evidence that chloroplasttranscriptional activity decreases rapidly upon S limitation andthat this response requires SAC3. Our data further suggest thatSac3 represses chloroplast transcription, at least in part througha proteolytic mechanism requiring phosphorylation activity. Weconclude that Sac3 is a response regulator that integratesmetabolism across cellular compartments.

ResultsAbundance of Several Chloroplast mRNAs Decreases Rapidly Under SStarvation. Studies in Chlamydomonas have shown that �Sconditions induce a reduction in the rate of oxygen evolution,which correlates with a decrease in photosystem (PS) II activecenters and reduced abundance of proteins of the photosyntheticapparatus (6, 8, 10). S deprivation was also linked to reducedaccumulation of nuclear transcripts encoding proteins involvedin PSI, PSII, light-harvesting complex, and photosynthetic elec-tron transport machinery (11). Given the above observations, itwas of interest to investigate whether under S limitation, chlo-roplast mRNA levels are adjusted as part of a global cellularresponse.

As a first step, the accumulation of several chloroplast mRNAswas measured under normal and �S conditions. To do so, WTcells were grown in S-replete medium, and then the culture wasdivided into two portions, one with S and one without. Sampleswere removed at various intervals, and, after 24 h, the starvedculture was transferred back to S-containing medium for 2 h. Fig.1 shows a representative RNA gel blot of the chloroplast genes

Conflict of interest statement: No conflicts declared.

This paper was submitted directly (Track II) to the PNAS office.

Abbreviations: ARS, arylsulfatase; cpRNA, chloroplast RNA; DMAP, 6-dimethylamino-purine; PEP, plastid-encoded polymerase; PTK, plastid transcription kinase; SnRK2, SNF1-related kinase 2; TAP, Tris�acetate�phosphate; TAP-S, sulfate-deficient TAP.

*To whom correspondence should be addressed. E-mail: ds28@cornell.edu.

© 2006 by The National Academy of Sciences of the USA

www.pnas.org�cgi�doi�10.1073�pnas.0511042103 PNAS � May 16, 2006 � vol. 103 � no. 20 � 7911–7916

PLA

NT

BIO

LOG

Y

Dow

nloa

ded

by g

uest

on

Janu

ary

11, 2

022

atpB (encoding the �-subunit of ATP synthase), petD and petB(encoding subunit IV and cytochrome b6 of the cytochrome b6�fcomplex, respectively), and tufA (encoding elongation factorTu). The results showed that although levels of chloroplastRNAs (cpRNAs) remained mainly unchanged under normalconditions, removal of S from the medium resulted in a decreaseof 20–90% within 4–8 h and a decrease of 40–60% after 24 h.This decrease was reversible, and transcript abundance recov-

ered to at or above the prestarvation levels after 2 h of Sreplenishment. This result contrasts with the accumulation of thenuclear ARS1 transcript, which is known to be induced under �Sconditions (12). As shown in Fig. 1, ARS1 mRNA did notaccumulate under �S conditions but increased rapidly under �Sconditions, subsequently declining upon the replenishment of S.Thus, under conditions where the accumulation of some tran-scripts is increasing, several cpRNAs decrease in response to Sstress, correlating with reduced photosynthesis.

Sac3 Is Involved in Rapid Repression of Chloroplast TranscriptionActivity upon S Starvation. The RNA gel blot data presented aboverevealed that cpRNA accumulation decreased under �S condi-tions. One way of assessing whether this phenomenon is aspecific response to S limitation is to determine whether it isunder genetic control. Therefore, cpRNA accumulation wasexplored after S deprivation in the sac1 and sac3 mutants, whoseabnormal responses to S limitation were described above. Be-cause cpRNA abundance decreases in WT cells after 4–8 h (Fig.1), we focused on a relatively early part of the starvationresponse (i.e., the first 12 h). To confirm the S limitationcondition, ARS1 mRNA was monitored. Consistent with previ-ous reports, ARS1 was induced upon S deprivation in the WT andsac3 cells but not in sac1 cells (Fig. 2). In addition, low expressionof ARS1 was detected in sac3 cells when grown in the presenceof S (0-h time point), as previously reported.

When cpRNA accumulation was compared between strains,both the WT and sac1 cells exhibited decreases under �Sconditions. The decrease appeared to be more rapid in sac1 cells,with 50–90% decreases in transcript levels at the earliest timepoint taken (2 h), contrasting with only minor changes in theWT. More striking was the constant or increasing cpRNAabundance in sac3 cells up to 12 h after starvation, suggestingthat SAC3 function is required for depletion of cpRNA. To verifythis observation, cpRNA levels were monitored in the samemutant that had been complemented (9) with the WT SAC3gene (sac3::SAC3). This strain showed a WT pattern of ARS1induction and decreased cpRNA accumulation under �S con-ditions, confirming the role of SAC3. Thus, chloroplast transcript

Fig. 1. cpRNA and ARS1 accumulation under normal and �S conditions. WTcells were grown to midlogarithmic phase in TAP medium, collected, resus-pended in �S or �S medium, and incubated for 0, 4, 8, 12, or 24 h beforeharvesting. At the end, the S-starved culture was transferred back to S-containing medium for 2 h (SR). Five micrograms of total RNA isolated fromthe indicated samples was separated by electrophoresis, blotted onto a mem-brane, and hybridized initially with the probes shown on the left. The signalscorresponding to cpRNAs were normalized to 25S rRNA, and the initial timepoint was set to 1.0. The numbers beneath each lane show the ratio of RNAaccumulations for each track. Two filters were used, and the 25S reprobing ofeach is shown. The use of each of these filters is denoted by either one or twoasterisks to the right of the experimental blots.

Fig. 2. Chloroplast and ARS1 RNA accumulation in WT cells, sac mutants, and complemented sac3 cells after S deprivation. Cells were grown to midlogarithmicphase in TAP medium, collected, resuspended in �S medium, and incubated for 0, 2, 4, 8, or 12 h before harvesting. RNA was isolated and analyzed as describedin the legend of Fig. 1. Data were also analyzed collectively (see Fig. 7, which is published as supporting information on the PNAS web site).

7912 � www.pnas.org�cgi�doi�10.1073�pnas.0511042103 Irihimovitch and Stern

Dow

nloa

ded

by g

uest

on

Janu

ary

11, 2

022

reduction is SAC3-dependent and perhaps modulated directly orindirectly by SAC1.

Decreased cpRNA levels under �S conditions could resultfrom changes in transcription rates and�or RNA stability. Toassess the contribution of transcriptional changes, run-on assayswere performed. In these experiments, chloroplast transcriptionactivity was compared at time 0 (�S medium) and 8 h afterstarvation by using WT and sac3 cells permeabilized by freeze�thaw cycles, followed by a short incubation in the presence of[�-32P]UTP to extend previously initiated transcripts (13). RNAswere labeled and extracted identically from both strains induplicate and then used as probes on DNA dot blots to estimatetranscription rates of nine chloroplast genes encoding compo-nents of the photosynthesis and gene expression machineries.

Fig. 3 shows that after S starvation, radioactive incorporationwas reduced into all cpRNAs examined in WT cells, indicatingthat decreased cpRNA abundance during S limitation results atleast in part from reduced transcription activity. In contrast andconsistent with the lack of cpRNA reduction under �S condi-tions, no decrease was observed in chloroplast transcription ratesfor sac3. This finding suggests that the putative Sac3 kinasemediates repression of chloroplast transcription under �S con-ditions. It should be noted, however, that our results do not ruleout a contributory role of decreased transcript stability under �Sconditions, which we have not measured directly. We were ableto discount the possibility that reduced cpDNA copy numberlimits transcription under �S conditions, because cpDNA abun-dance does not change under �S conditions in WT cells (S.Yehudai-Resheff and D.B.S., unpublished data).

The Chloroplast S Limitation Response Is Regulated by Phosphoryla-tion. Previous analysis of the sac3 mutant demonstrated thatSAC3 has both positive and negative effects on nuclear geneexpression. Here, our results indicate that SAC3 can regulatechloroplast transcription. Because SAC3 encodes a putativekinase, it can be envisaged that this regulation takes place bymeans of activation of the Sac3 kinase under �S conditions. Asone approach to test this hypothesis, we carried out experiments

in the presence of a general kinase inhibitor, 6-dimethylami-nopurine (DMAP) (14), previously used in Chlamydomonas(15).

When cpRNA accumulation was examined under �S conditions,the addition of DMAP was found to prevent or mitigate thedecrease in cpRNA levels during S starvation (Fig. 4A), consistentwith the notion that Sac3 kinase activity represses transcription. Onthe contrary, cpRNA levels decreased 70–90% over 12 h in thecontrol experiment, in which the solvent DMSO was used. To seewhether DMAP affected chloroplast transcription activity, run-onassays were performed, and, as shown in Fig. 4B, �S conditionsresulted in reduced transcription in the DMSO control but not inthe presence of DMAP. Together, these results provide in vivoevidence that protein kinase activity is necessary to reduce cpRNAaccumulation under �S conditions. It was also noted that in thepresence of DMAP, ARS1 mRNA levels increased only after 12 h,in contrast to the increase seen after 4 h in the control experiment(Fig. 4A), suggesting that phosphorylation, although not necessarilymediated by Sac3, also regulates ARS1 derepression under Slimitation.

Sac3 May Deactivate and�or Destabilize the Chloroplast Sigma Factor.The above data support the concept that Sac3-mediated chlo-roplast transcriptional repression is stimulated under �S con-ditions. In land plants, chloroplast transcription is carried out by

Fig. 3. Changes in transcription rates for chloroplast-encoded genes in WTcells and sac3 mutants after S deprivation. Cells were grown to midlogarithmicphase in TAP medium, collected, and resuspended in �S medium for 8 h. Cells(5 � 107) were removed at 0 and 8 h after starvation and permeabilized forrun-on transcription. All of the 32P-labeled transcription products were hy-bridized to DNA dot blots (in duplicate) containing 0.2 �g (0.5�) or 0.4 �g (1�)of PCR products as shown. pBluescript plasmid DNA was used as a negativecontrol. Dot blots were hybridized, washed, and exposed under identicalconditions and visualized by using the Storm scanner.

Fig. 4. Effects of DMAP on transcription and RNA accumulation. (A) Chlo-roplast and ARS1 RNA accumulation after S deprivation in the presence ofDMAP. Cells were grown to midlogarithmic phase in TAP medium and thenresuspended in TAP-S with 3 mM DMAP or an equal volume of DMSO as acontrol. Samples were taken at the time the kinase inhibitor was added (0) and4, 8, and 12 h after starvation. Isolated RNA was analyzed as described in thelegend of Fig. 1. Data were also analyzed collectively (Fig. 7). (B) Run-ontranscription was performed as described in the legend of Fig. 3, except DMSOor 3 mM DMAP was added as shown.

Irihimovitch and Stern PNAS � May 16, 2006 � vol. 103 � no. 20 � 7913

PLA

NT

BIO

LOG

Y

Dow

nloa

ded

by g

uest

on

Janu

ary

11, 2

022

two distinct types of RNA polymerase (16). One is a bacterial-type enzyme whose catalytic subunits are encoded in the chlo-roplast (plastid-encoded polymerase, PEP), and the other is aphage-type enzyme encoded in the nucleus (nuclear-encodedpolymerase, NEP). Chlamydomonas chloroplasts appear to pos-sess only PEP (17), whose catalytic core consists of �-, �-, ��-,and ��-subunits. PEP additionally requires a sigma factor forpromoter recognition, which is encoded by a gene family in landplants (18) but by a single gene in Chlamydomonas. This gene,RPOD, encodes the protein termed Sig1 or CrRpoD (19, 20).The fact that Chlamydomonas chloroplasts possess PEP but notNEP and that Sig1 apparently acts as the only specificity factorwould appear to facilitate global control of chloroplast tran-scription activity.

Given the above observations, it was reasonable to hypothe-size that under �S conditions, Sac3 kinase might alter RPODexpression, Sig1 stability, and�or Sig1 activity. To test the firstpossibility, RT-PCR analysis was performed to examine theaccumulation of mRNAs encoding PEP subunits, including Sig1,after S deprivation. Fig. 5 shows that RPOD mRNA abundanceexhibited only small variations both in WT and sac3 cells underthese conditions. RT-PCR analysis was also performed for thechloroplast rpoA and rpoB1 genes, which encode the �- and�-subunits of PEP, respectively. This analysis revealed decreas-ing abundance of both transcripts in WT but not sac3 cells, anobservation that is consistent with the notion that Sac3 mediatesgeneral chloroplast transcriptional repression under �S conditions.

Because chloroplast transcription appeared to be regulatedindependently of RPOD transcript abundance, it was of interestto determine whether Sig1 might be a target of regulation. To doso, immunoblot analysis was performed by using an anti-peptideantibody that recognizes a single polypeptide in total protein andchloroplast isolates of �85 kDa, corresponding in size to Sig1(20). Fig. 6A shows immunoblot analysis of total protein col-lected at various intervals after S deprivation, revealing that at12 h after starvation Sig1 abundance declines to between 20%and 50% of its initial level in WT cells with �-tubulin as areference. The sac3 mutant, however, retained, on average, theinitial amount of Sig1 even after 12 h of starvation. Furthermore,as shown in Fig. 6B, addition of DMAP to WT cells, whentransferred to �S, prevented the decline in Sig1 as comparedwith the tubulin control. In contrast, Sig1 was reduced upon Slimitation in the control experiment, in which DMAP wasomitted. These results indicate that increased proteolysis and�orreduced synthesis of Sig1 occurs after S limitation and that Sac3may be involved in a signaling cascade targeting Sig1 fordegradation.

DiscussionPhotosynthetic organisms use different strategies to manage Slimitation. Although several aspects of Chlamydomonas S dep-rivation have been studied (10, 21), little has been learned of themolecular or genetic links between perception of S limitationand chloroplast responses. The results shown here indicate rapid

repression of plastid transcription activity after removal of Sfrom the culture medium and provide evidence that a SAC3-encoded protein kinase is involved in this process. Reducedchloroplast transcription correlated with reduced abundance ofSig1, the single Chlamydomonas PEP sigma factor, a phenom-enon that also requires kinase activity. In sum, the present studyassigns an unanticipated role for the Sac3 Ser�Thr kinase inregulating S limitation responses.

Upon S deprivation, the abundance of at least six cpRNAsdecreases, with a 40–90% reduction observed, depending on thegene and experiment. Gel blot analysis also showed that thedecreased abundance of cpRNA during S limitation requiresSAC3 function but that SAC1 has, at most, a minor role. cpRNAlevels declined more abruptly for some genes in S-deprived sac1cells, mirroring what occurs for nucleus-encoded transcriptsencoding photosynthesis components (11). It should be takeninto consideration, however, that sac1 is very sensitive to Slimitation (8), failing to survive when grown under �S condi-tions in the light. Thus, any conclusions about a specific role ofSAC1 in regulating chloroplast transcription should be madewith caution. Double mutant analysis suggests that SAC1 andSAC3 act in parallel (22), which would argue against oneregulating the other directly.

Analysis of chloroplast transcription rates under �S condi-tions revealed reduced cpRNA synthesis in WT cells but not insac3 cells. Because all cpRNAs that were examined showed thesame trend, S deprivation appears to globally affect transcrip-

Fig. 5. RT-PCR analysis for genes encoding PEP subunits under �S conditions.rpoA and rpoB1 cDNAs were amplified with 30 PCR cycles, and RPOD cDNA wasamplified with 35 cycles. Actin mRNA was used as an internal standard, using30 PCR cycles.

Fig. 6. Immunoblot analysis of Sig1 under �S conditions. Total proteins (7.5�g) from times 0, 4, 8, and 12 h after starvation from WT and sac3 cells (A) andWT cells treated with DMAP or DMSO (B) were separated in SDS-polyacryl-amide gels and transferred to nitrocellulose. Blots were probed first withanti-Sig1 and then with anti-�-tubulin. Densitometric quantification of Sig1and tubulin levels is shown. The relative values (from time 0) represent themean � SD of two independent experiments.

7914 � www.pnas.org�cgi�doi�10.1073�pnas.0511042103 Irihimovitch and Stern

Dow

nloa

ded

by g

uest

on

Janu

ary

11, 2

022

tion, consistent with the single type of RNA polymerase foundin Chlamydomonas chloroplasts. Plastid transcription activitycan also be globally regulated in plants, generally correlatingwith the developmental state (23–25). Although the mechanismscontrolling general plastid transcriptional regulation are not wellunderstood, in the case studied here, we have hypothesized thatSac3 kinase activity triggers transcriptional repression. Thisconclusion is supported by the fact that addition of a generalkinase inhibitor prevented repression under �S conditions (Fig. 4).

Sac3-mediated transcriptional repression may have some par-allels with PEP regulation in mustard, where in vitro studiesidentified two forms of PEP, A and B, which could be intercon-verted by (de)phosphorylation (26). PEP-B occurs in etioplastsand is associated with phosphorylated sigma factor, which ispostulated to inhibit its release from the catalytic core. Lighttriggers dephosphorylation, activating transcription as chloro-plasts mature (27). It was subsequently shown that sigma phos-phorylation is carried out by a plastid transcription kinase (PTK)related to the �-subunit of nucleocytoplasmic casein kinase 2(28). Another example of PEP regulation by phosphorylation isin barley, where in vivo experiments showed that protein kinaseand extraplastidic Ser�Thr phosphatases regulate plastid tran-scription in response to light (29).

SAC3 encodes a putative Ser�Thr kinase of the SnRK2(SNF1-related kinase 2) subfamily featuring a canonical kinasedomain at the N terminus and a C-terminal acidic domain (30),which may confer specificity. The emerging concept from yeastand plant studies is that members of the SNF1 family protect cellsagainst nutritional and environmental stress (31, 32). SeveralSnRK2 family members from Arabidopsis and rice have beenidentified as stress-activated kinases involved in response todrought and humidity (33, 34). Evidence from wheat showed thatthe SnRK2 member PKABA1 is activated by abscisic acid(ABA) and phosphorylates a downstream transcription factor(35). Similarly, the ABA-responsive transcription factor TRAB1is activated by SAPK10 (stress-activated protein kinase 10) inrice protoplasts (36). Another example is AAPK (ABA-activated protein kinase) from Vicia faba, which regulates sto-matal closure by targeting an RNA-binding protein (37). Thus,SnRK2s appear to act both on the transcriptional and posttran-scriptional levels. However, so far, no relationship had beenestablished between SnRK2 proteins and chloroplast transcrip-tional activity.

Targeting prediction programs do not predict a chloroplastlocalization for Sac3 (data not shown). This indication leads toa model in which Sac3 is activated in the cytosol under �Sconditions, initiating a cascade of events leading to chloroplasttranscriptional repression. How this cascade is transduced to thechloroplast, and whether it involves PTK, is an open question.The Chlamydomonas nuclear genome encodes two PTK homo-logues (gene models C�670029 and C�200187), and one couldenvisage transduction of the Sac3 signal to PTK with subsequentphosphorylation of Sig1. This modification would lead to tran-scriptional arrest, as in the mustard PEP-based mechanism,and�or to degradation of Sig1, perhaps stimulated by a limitationof PEP core subunits. Although PTK has not been tested withSig1, recombinant Sig1 can be phosphorylated in vitro by humancasein kinase 2 (CK2) (data not shown), consistent with Sig1containing multiple consensus CK2 recognition sites (as noted inref. 19). We also note that alternative models for Sac3 functionexist, including one where it is phosphorylated under �S conditionsand dephosphorylated when activated (38). However, our data aremost consistent with activation through phosphorylation.

It was recently proposed that in Chlamydomonas, circadianrhythms of chloroplast transcription are regulated by transcrip-tional oscillations of RPOD (19). In contrast, the data presentedhere show a constant level of RPOD mRNA (Fig. 5) but a60–80% decrease in the Sig1 protein level under �S conditions

(Fig. 6). Further analysis showed that Sig1 degradation occurredin a SAC3-dependent manner and required phosphorylation,with the latter conclusion based on prevention of Sig1 disap-pearance by the addition of DMAP. Regulation by degradationor stabilization of sigma factors is known and perhaps beststudied for Escherichia coli RpoS, which is stress-induced. Thelevel of RpoS remains low under normal conditions because ofdegradation by the ClpXP protease (reviewed in ref. 39). Thedegradation pathway depends on RssB, a response regulator thatdelivers RpoS to ClpXP through a still-undefined mechanism (40).

The regulatory model proposed here presumes that transcrip-tional repression under �S conditions is at least partly due todestabilization and�or reduced synthesis of Sig1 triggered byactivated Sac3. Whether Sig1 is degraded inside or outside thechloroplast, and whether Sig1 phosphorylation initially repressestranscription initiation and then stimulates decay, remains to beestablished. The decrease in Sig1 activity and�or abundancemight represent an adjustment of the chloroplast transcriptionapparatus under �S conditions, coincident with reduced expres-sion of catalytic component rpo genes, and a general reductionof photosynthetic activity. This regulation highlights the impor-tance of integrating the chloroplast genetically into nutrientstress response.

Materials and MethodsStrains and Culture Conditions. C. reinhardtii WT strain CC-125,the mutant strains sac1 and sac3, and the complemented sac3strain (are10-12-C1), were grown in nutrient-replete Tris�acetate�phosphate (TAP) medium (41). For starvation experi-ments, cells were grown under continuous light at 25°C on arotary shaker (125 rpm) to midlogarithmic phase (1–5 � 106 cellsper ml). Cultures were then harvested by centrifugation, washedin sulfate-deficient TAP (TAP-S) (22), and resuspended inTAP-S. Repletion was achieved by harvesting cells, washing withTAP, and resuspending in TAP medium for 2 h. In some cases,DMAP (dissolved in DMSO) was added to cells transferred toTAP-S to a final concentration of 3 mM, or the same volume ofDMSO was added as a control.

RNA Isolation and Gel Blot Analysis. For RNA gel blot analysis, totalRNA was isolated by using TRI Reagent (Molecular ResearchCenter, Cincinnati). Five micrograms of RNA was resolved in1% agarose�6% formaldehyde gels. Electrophoresis, blotting,and hybridization were performed as described in ref. 42. TheDNA probes were PCR products for atpB, petD, petB, tufA,ARS1, and 25S ribosomal DNA labeled with [�-32P]dCTP. Theradioactive bands were visualized by using a Storm scanner(Amersham Pharmacia Biosciences). Quantification of bandswas performed by using IMAGEQUANT software (AmershamPharmacia Biotech).

Run-On Transcription in Permeabilized Cells. For run-on transcrip-tion assays, WT and sac3 cells were grown in TAP medium,harvested, adjusted to 1 � 106 cells per ml, and transferred toTAP-S. Permeabilized cells for the assay were prepared by thefreeze�thaw method (13), and aliquots were stored at �80°Cuntil use. Of the pelleted permeabilized cells, 30 �l (�5 � 107

cells) were mixed with 15 �l of 4� run-on transcription buffer(1 M sucrose�60 mM MgCl2�15 mM DTT�50 mM NaF�50 mMHepes, pH 7.5), 2 �l of rRNasin RNase inhibitor (Promega), 0.25mM GTP, 0.5 mM ATP, 0.25 CTP, and 125 �Ci (1 Ci � 37 GBq)of [�-32P]UTP (43). The mixture was incubated on ice for 1 minand then at room temperature for 15 min. The reactions wereterminated by addition of 10 �l of 20% SDS, and radiolabeledRNAs were extracted by using TRI Reagent. Pellets weredissolved in 100 �l of H2O and used for DNA dot blot hybrid-ization. The DNA probes were PCR products for the chloroplast

Irihimovitch and Stern PNAS � May 16, 2006 � vol. 103 � no. 20 � 7915

PLA

NT

BIO

LOG

Y

Dow

nloa

ded

by g

uest

on

Janu

ary

11, 2

022

genes as shown, which were blotted on GeneScreen nylon filtersby using the Bio-Dot microfiltration apparatus (Bio-Rad).

RT-PCR Analysis. For RT-PCR, 2.5 �g of total RNA was treatedwith 1 unit of RQI RNase-free DNase I (Promega) for 30 minat 37°C in a total volume of 10 �l. Then DNase was inactivatedby heating to 65°C for 10 min, and 2-�l aliquots were used for50-�l single-tube RT-PCR (Access RT-PCR System; Promega),according to the manufacturer’s instructions. The primers usedwere as follows: for RPOD, 5�-ACTACAAGCTGGTCAT-GACGGTGT-3� and 5�-TTACGTTCAGCGCCTCTCCA-ATCT-3�; for rpoA, 5�-ATGACAATTTATCCAAATTTA-AAAAAAATC-3� and 5�-TTAAAATTTTCTTGCATA-TTCAAAAGTAAG-3�; for rpoB1, 5�-GATAACGTTAATG-TATCCCAAATTGAT-3� and 5�-TCAACGGGACAAATAC-GACCAT-3�; and for actin, 5�-AATCGTGCGCGACAT-CAAGGAGAA-3� and 5�-TTGGCGATCCACATTTGCT-GGAAGGT-3�. The first step was reverse transcription at 48°Cfor 45 min, followed by 30 or 35 PCR cycles. PCR products weresequenced to confirm their identities.

Immunoblot Analysis. For protein extraction, cells were grown asdescribed above. Ten-milliliter samples were pelleted and

washed with water, and total proteins were extracted by acetoneprecipitation. Protein concentration was determined by usingBradford reagent (Bio-Rad), and equivalent masses of proteinwere resolved in 12% SDS-polyacrylamide gels and transferredto nitrocellulose membranes with a TransBlot semidry cell(Bio-Rad). Immunoblotting was performed by using a peptideantibody directed against Sig1 (kindly provided by David Herrin,University of Texas, Austin) and a monoclonal anti-sea urchin�-tubulin antibody (Sigma). The Sig1 antibody was against thepeptide LADELERLLGPTTSC, to which a nonencoded termi-nal cysteine was added for coupling. Secondary antibodies werehorseradish peroxidase (HRP)-conjugated anti-rabbit IgG (forSig1 detection) or HRP-conjugated anti-mouse (for tubulindetection), and enhanced chemiluminescence (Amersham Phar-macia Biotech) was used. Quantification of bands was performedby using IMAGEQUANT software after scanning of x-ray film andimportation as TIFF files.

We thank Arthur Grossman (Carnegie Institution, Stanford, CA) forhelpful suggestions and for providing strains carrying sac mutations,David Herrin for the anti-Sig1 antibody, and Katia Wostrikoff for acritical reading of the manuscript. This work was supported by NationalScience Foundation Award DBI-0211935.

1. Leustek, T. & Saito, K. (1999) Plant Physiol. 120, 637–644.2. Leustek, T., Martin, M. N., Bick, J. A. & Davies, J. P. (2000) Annu. Rev. Plant

Physiol. Plant Mol. Biol. 51, 141–165.3. Hirai, M. Y., Fujiwara, T., Awazuhara, M., Kimura, T., Noji, M. & Saito, K.

(2003) Plant J. 33, 651–663.4. Nikiforova, V., Freitag, J., Kempa, S., Adamik, M., Hesse, H. & Hoefgen, R.

(2003) Plant J. 33, 633–650.5. Yildiz, F. H., Davies, J. P. & Grossman, A. R. (1994) Plant Physiol. 104,

981–987.6. Zhang, L., Happe, T. & Melis, A. (2002) Planta 214, 552–561.7. de Hostos, E. L., Schilling, J. & Grossman, A. R. (1989) Mol. Gen. Genet. 218,

229–239.8. Davies, J. P., Yildiz, F. H. & Grossman, A. (1996) EMBO J. 15, 2150–2159.9. Davies, J. P., Yildiz, F. H. & Grossman, A. R. (1999) Plant Cell 11, 1179–1190.

10. Wykoff, D. D., Davies, J. P., Melis, A. & Grossman, A. R. (1998) Plant Physiol.117, 129–139.

11. Zhang, Z., Shrager, J., Jain, M., Chang, C. W., Vallon, O. & Grossman, A. R.(2004) Eukaryotic Cell 3, 1331–1348.

12. de Hostos, E. L., Togasaki, R. K. & Grossman, A. (1988) J. Cell Biol. 106, 29–37.13. Gagne, G. & Guertin, M. (1992) Plant Mol. Biol. 18, 429–445.14. Neant, I. & Guerrier, P. (1988) Exp. Cell Res. 176, 68–79.15. Reisdorph, N. A. & Small, G. D. (2004) Plant Physiol. 134, 1546–1554.16. Cahoon, A. B. & Stern, D. B. (2001) Trends Plant Sci. 6, 45–46.17. Eberhard, S., Drapier, D. & Wollman, F. A. (2002) Plant J. 31, 149–160.18. Allison, L. A. (2000) Biochimie 82, 537–548.19. Carter, M. L., Smith, A. C., Kobayashi, H., Purton, S. & Herrin, D. L. (2004)

Photosynth. Res. 82, 339–349.20. Bohne, A. V., Irihimovitch, V., Weihe, A. & Stern, D. B. (February 2, 2006)

Curr. Genet., 10.1007�s00294-006-0060-7.21. Grossman, A. (2000) Protist 151, 201–224.22. Davies, J. P., Yildiz, F. & Grossman, A. R. (1994) Plant Cell 6, 53–63.23. Deng, X. W. & Gruissem, W. (1987) Cell 49, 379–387.24. Mullet, J. E. & Klein, R. R. (1987) EMBO J. 6, 1571–1579.25. Cahoon, A. B., Harris, F. M. & Stern, D. B. (2004) EMBO Rep. 5, 801–806.

26. Pfannschmidt, T., Ogrzewalla, K., Baginsky, S., Sickmann, A., Meyer, H. E. &Link, G. (2000) Eur. J. Biochem. 267, 253–261.

27. Tiller, K. & Link, G. (1993) EMBO J. 12, 1745–1753.28. Ogrzewalla, K., Piotrowski, M., Reinbothe, S. & Link, G. (2002) Eur. J. Bio-

chem. 269, 3329–3337.29. Christopher, D. A., Li, X., Kim, M. & Mullet, J. E. (1997) Plant Physiol. 113,

1273–1282.30. Hardie, D. G., Carling, D. & Carlson, M. (1998) Annu. Rev. Biochem. 67,

821–855.31. Halford, N. G. & Hardie, D. G. (1998) Plant Mol. Biol. 37, 735–748.32. Sanz, P. (2003) Biochem. Soc. Trans. 31, 178–181.33. Umezawa, T., Yoshida, R., Maruyama, K., Yamaguchi-Shinozaki, K. & Shi-

nozaki, K. (2004) Proc. Natl. Acad. Sci. USA 101, 17306–17311.34. Kobayashi, Y., Yamamoto, S., Minami, H., Kagaya, Y. & Hattori, T. (2004)

Plant Cell 16, 1163–1177.35. Johnson, R. R., Wagner, R. L., Verhey, S. D. & Walker-Simmons, M. K. (2002)

Plant Physiol. 130, 837–846.36. Kobayashi, Y., Murata, M., Minami, H., Yamamoto, S., Kagaya, Y., Hobo, T.,

Yamamoto, A. & Hattori, T. (2005) Plant J. 44, 939–949.37. Li, J., Kinoshita, T., Pandey, S., Ng, C. K., Gygi, S. P., Shimazaki, K. &

Assmann, S. M. (2002) Nature 418, 793–797.38. Pollock, S., Pootakham, W., Shibagaki, N., Moseley, J. & Grossman, A. (2005)

Photosynth. Res. 86, 475–489.39. Peterson, C. N., Mandel, M. J. & Silhavy, T. J. (2005) J. Bacteriol. 187,

7549–7553.40. Peterson, C. N., Ruiz, N. & Silhavy, T. J. (2004) J. Bacteriol. 186, 7403–7410.41. Harris, E. H. (1989) The Chlamydomonas Sourcebook: A Comprehensive Guide

to Biology and Laboratory Use (Academic, San Diego).42. Drager, R. G. & Stern, D. B. (1998) in The Molecular Biology of Chloroplasts

and Mitochondria in Chlamydomonas, eds. Rochaix, J.-D., Goldschmidt-Clermont, M. & Merchant, S. (Kluwer, Dordrecht, The Netherlands), Vol. 7,pp. 165–181.

43. Sakamoto, W., Kindle, K. L. & Stern, D. B. (1993) Proc. Natl. Acad. Sci. USA90, 497–501.

7916 � www.pnas.org�cgi�doi�10.1073�pnas.0511042103 Irihimovitch and Stern

Dow

nloa

ded

by g

uest

on

Janu

ary

11, 2

022