Skip to main content
PMC home page
Plant Physiology logoLink to Plant Physiology
. 2006 Oct;142(2):797–806. doi: 10.1104/pp.106.085944

A Heteromeric RNA-Binding Protein Is Involved in Maintaining Acrophase and Period of the Circadian Clock1,[W]

Dobromir Iliev 1,2, Olga Voytsekh 1,2, Eva-Maria Schmidt 1, Monika Fiedler 1, Alla Nykytenko 1, Maria Mittag 1,*
PMCID: PMC1586056  PMID: 16920878

Abstract

The RNA-binding protein CHLAMY1 from the green alga Chlamydomonas reinhardtii consists of two subunits. One (named C1) contains three lysine homology motifs and the other (named C3) has three RNA recognition motifs. CHLAMY1 binds specifically to uridine-guanine-repeat sequences and its circadian-binding activity is controlled at the posttranslational level, presumably by time-dependent formation of protein complexes consisting of C1 and C3 or C1 alone. Here we have characterized the role of the two subunits within the circadian system by measurements of a circadian rhythm of phototaxis in strains where C1 or C3 are either up- or down-regulated. Further, we have measured the rhythm of nitrite reductase activity in strains with reduced levels of C1 or C3. In case of changes in the C3 level (both increases and decreases), the acrophase of the phototaxis rhythm and of the nitrite reductase rhythm (C3 decrease) was shifted by several hours from subjective day (maximum in wild-type cells) back towards the night. In contrast, both silencing and overexpression of C1 resulted in disturbed circadian rhythms and arrhythmicity. Interestingly, the expression of C1 is interconnected with that of C3. Our data suggest that CHLAMY1 is involved in the control of the phase angle and period of the circadian clock in C. reinhardtii.

Circadian rhythms are biological rhythms that persist with a period of about 24 h under constant conditions of light and temperature. They have been described in many organisms where they regulate a wide variety of physiological and cellular processes. While the physiological properties of circadian rhythms are well conserved among different organisms including bacteria (Kondo et al., 1993), only some of the molecular components are conserved across evolution. But, in all model systems studied so far a similar mechanism consisting of positive and negative feedback loops was found to be a key feature of the endogenous pacemaker (Harmer et al., 2001; Reppert and Weaver, 2002; Dunlap and Loros, 2004; Johnson, 2004). Thus, transcription factors and their protein-protein interactions, as well as temporal phosphorylation events, have been found to play important regulatory roles within the circadian oscillator.

In the green alga Chlamydomonas reinhardtii several processes are under control of the circadian clock such as phototaxis, chemotaxis, stickiness to glass surfaces, UV sensitivity, and the cell cycle (for review, see Mittag et al., 2005). Phototaxis, also known as photoaccumulation, was already demonstrated more than 30 years ago by Bruce (1970). As biflagellate unicell, C. reinhardtii moves to regions of brighter light during day phase, allowing the cells to optimize their light absorption for photosynthesis. Automated computerized measurement devices have been developed (Mergenhagen, 1984; Johnson et al., 1991; Kondo et al., 1991) to study this circadian rhythm under free-running conditions over several days.

Based on the availability of the entire genome sequence of C. reinhardtii (Grossman et al., 2003; Grossman, 2005), homology searches have been undertaken to check if known clock genes of other model organisms are conserved in this green alga. Thereby, it was found that two cryptochrome (CRY)-like proteins are present in C. reinhardtii (Mittag et al., 2005). One of them is more closely related to higher plants, while the other is closer to animal CRY. Even though both still have to be functionally characterized, the occurrence of both in C. reinhardtii is of high interest, since mammalian CRY is part of the endogenous oscillator while plant CRY acts as photoreceptor. Thus, it has been hypothesized that C. reinhardtii might have an amalgam clock (Breton and Kay, 2006).

In addition, kinases and phosphatases that are involved in posttranslational modifications of key components of the oscillatory system in other eukaryotic model systems are well conserved in C. reinhardtii (Mittag et al., 2005). First approaches have been undertaken to characterize one of these proteins functionally. Silencing of casein kinase1 (CK1) by RNAi caused a period shortening for several days and finally arrhythmicity of circadian phototaxis (Schmidt et al., 2006). Thus, at least one of the well-conserved kinases also plays an important role in the circadian system of C. reinhardtii. However, the target proteins of these kinases such as frequency (FRQ) in Neurospora crassa or period in Drosophila melanogaster have not been found through homology searches (Mittag et al., 2005). Thus, it was concluded that C. reinhardtii may also have its own clock proteins.

Based on functional homology with the circadian controlled translational regulator (CCTR) from the dinoflagellate Gonyaulax polyedra, an RNA-binding protein named CHLAMY1 has been identified in C. reinhardtii (Mittag et al., 1994; Mittag, 1996). The CCTR binds to a uridine-guanine (UG)-repeat region situated in the mRNA of the luciferin-binding protein (LBP) in a circadian manner (Mittag et al., 1994). LBP was indeed, to our knowledge, the first example where it was shown that its circadian expression is regulated at the translational level (Morse et al., 1989). While LBP reaches a peak in the night phase, its mRNA is constantly present. The CCTR was considered to inhibit translation of lbp mRNA during day phase when it is bound to it. CHLAMY1 also recognizes specifically UG≥7-repeat sequences that are present in the 3′ untranslated regions (UTRs) of several mRNAs of C. reinhardtii and its binding activity is controlled by the circadian clock (Mittag, 1996; Waltenberger et al., 2001). The proteins encoded by these mRNAs include members of nitrogen and CO2 metabolism such as chloroplastic Gln synthetase2, nitrite reductase (NII), or one of the small subunits of Rubisco. Further investigations showed that CHLAMY1 represents a novel type of heteromeric RNA-binding protein that consists of two subunits comprising three RNA recognition motifs (RRMs; C3 subunit) and three Lys homology motifs (KH; C1 subunit), respectively (Zhao et al., 2004). The c1 cDNA encodes a protein of 51,706 D, and the c3 cDNA encodes a protein of 44,979 D. While the C3 subunit migrates according to its molecular mass in SDS-PAGE, the C1 subunit migrates at a higher apparent mass of approximately 60 kD (Zhao et al., 2004). This was observed not only in C. reinhardtii extracts, but also in the recombinant protein overexpressed in Escherichia coli. Interaction of the C1 and C3 subunits was confirmed by immunoprecipitation assays as well as by supershift mobility assays along with a labeled UG-repeat transcript and anti-C1 or anti-C3 antibodies (Zhao et al., 2004). Both C1 and C3 bear domains that may be involved in their protein-protein interaction. C1 has a WW domain at its C terminus that is known to communicate protein-protein interactions (Sudol and Hunter, 2000). C3 that is also conserved in other organisms including mammals belongs to the CELF family of RNA-binding proteins. The spacer region between the second and third RRM domains has been postulated to be involved in protein-protein interactions (Ladd et al., 2001).

Surprisingly, the amounts of both subunits were found to be rather constant over the circadian cycle, indicating that posttranslational events control the circadian-binding activity of CHLAMY1. At the same time, it was found that a ≥670 kD protein complex consisting of the C1 subunit occurs only during subjective day, while an approximately 158 kD complex consisting of C1 and C3 is present both during day and night phase (Zhao et al., 2004). Thus, circadian-binding activity of CHLAMY1 appears to be controlled at the posttranslational level by time-dependent formation of different protein complexes involving C1 and C3. Thereby, it cannot be ruled out that the ≥670 kD protein complex may consist not only of C1, but may have some novel interaction partners.

Here we have characterized the role of CHLAMY1 within the circadian system by independent up- or down-regulation of its two subunits and measurements of the phototaxis rhythms in the transgenic strains. Further, we measured the rhythms of NII activities in the silenced strains. We can show that changes in the level of the C1 or the C3 subunit result in severe disturbances of the circadian clock either affecting its acrophase or causing arrhythmicity. Further, changes in the expression of the C1 subunit (slight overexpression or silencing below a critical level) resulted in large changes in the level of the C3 subunit, suggesting that C3 regulation is strongly dependent on C1. In contrast, C3 up- and down-regulation influence the C1 level only slightly (C3 up-regulation) or not significantly (C3 down-regulation).

RESULTS

Overexpression and Silencing of the C3 Subunit of CHLAMY1 Phase Shift Circadian Phototaxis by Several Hours

To obtain functional information about the role of the C3 subunit in the circadian system of C. reinhardtii, it was either silenced or overexpressed, and circadian phototaxis of the modified strains was measured. For overexpression of C3, the major part of its open reading frame (ORF) was put under the control of the strong truncated hsp70A/rbcS2 tandem promoter (Fig. 1A; for details, see “Materials and Methods” and Supplemental Protocol S1). Transformed strains growing under selection of paromomycin were used for further analysis. Cells were grown to a cell density of about 1 to 5 × 106 cells/mL and crude extracts were prepared. For comparison, a crude extract from nontransformed wild-type cells was used. C3 overexpression was analyzed in westerns with anti-C3 antibodies (Fig. 1B). Different amounts of proteins from wild type (90, 60, and 30 μg per lane) were separated on SDS-PAGE and quantitatively compared to proteins from transformed strains (30 μg per lane) after immunoblotting with the anti-C3 antibody. Equal loading was checked by Ponceau staining. Two strains where C3 was overexpressed at least 3-fold were used for further analysis (C3-ox90 and C3-ox100).

Figure 1.

Figure 1.

Open in a new tab

Overexpression and silencing of C3 and the effects on C1 expression. A, The overexpression construct used for C3 is shown. Promoter indicates the truncated hsp70A/rbcS2 tandem promoter along with the first intron of rbcS2 that bears an enhancer (see “Materials and Methods”). The tagged c3 cDNA bears the major part of the c3 ORF (the first 17 amino acids are missing) and the entire 3′ UTR (see Supplemental Protocol S1). B and C, Different amounts of proteins from a crude extract (90, 60, and 30 μg per lanes) of wild-type (WT) cells were separated on SDS-PAGE and used for western analysis with anti-C3 (B) or anti-C1 (C) antibodies along with proteins of crude extracts (30 μg per lane) from two different C3-overexpressing strains (C3-ox90 and C3-ox100). The positions of C3 and C1 are indicated by arrowheads. D, The RNAi construct used for silencing of C3 is shown. The numbers represent the involved exons. Introns are indicated by lines. The light-gray area represents the predicted 5′ UTR, and c3 prom the potential promoter region of c3 (see Supplemental Protocol S2). E and F, Different amounts of proteins from a crude extract (100, 50, and 25 μg per lanes) of wild-type (WT) cells were separated on SDS-PAGE and used for western analysis with the anti-C3 (E) and anti-C1 (F) antibodies along with proteins of crude extracts (100 μg per lane) from two different C3-silenced strains (C3-sil6 and C3-sil7). The positions of C3 and C1 are indicated by arrowheads.

In all studied model organisms the oscillatory system is driven by positive and negative feedback loops. Thereby, variations in protein level of any clock component can affect the normal expression of other components. For example, in N. crassa, a knockout of FRQ leads to decreased levels of white collar-1 (WC-1; Lee et al., 2000). Also, coregulation was found in some multiprotein complexes of C. reinhardtii, such as components of the photosystems (Wostrikoff et al., 2004; Göhre et al., 2006). Therefore, we were interested to understand if up- or down-regulation of any of the two subunits of CHLAMY1 could cause an effect on the normal expression of the other subunit. Using the C3-overexpressing strains, we examined if the C1 level was affected in parallel. In comparison to wild type, the level of C1 was slightly decreased in the C3-ox strains (Fig. 1C), indicating that there may be a weak coregulation.

For silencing of C3, an RNAi construct was created (Fig. 1D; for details, see “Materials and Methods” and Supplemental Protocol S2). C3 silencing was checked by western analysis with the anti-C3 antibodies (Fig. 1E). Different amounts of proteins from wild type (100, 50, and 25 μg per lane) were separated on SDS-PAGE and quantitatively compared to proteins from transformed strains (100 μg per lane) after immunoblotting with the anti-C3 antibody. Again, equal loading was checked by Ponceau staining. In the two strains C3-sil6 and C3-sil7, the C3 level was reduced to about 25% (C3-sil7) or clearly below 25% (C3-sil6; Fig. 1E) in comparison to the wild-type protein level. In this case, no significant coregulation with regard to the level of C1 was observed in the C3-silenced strains (Fig. 1F). Thus, the C1 levels in C3-sil6 and C3-sil7 were very close to that of the wild type.

The strains where C3 is up- or down-regulated were used for checking automated circadian phototaxis. As a control, the wild-type strain SAG 73.72 was used. Its period ranges between 24.3 and 24.7 h as determined in several independent experiments. A representative circadian phototaxis rhythm of the wild-type strain (period of 24.7 h) is shown (Fig. 2A). Phototaxis reaches its maximum during the middle of the subjective day and a minimum at the end of the subjective night. Both strains where C3 was overexpressed showed a period that is slightly increased in comparison to wild type (25.2 and 25.5 h, respectively; Fig. 2, B and C). With both C3-ox strains significant, phase angle (acrophase) shifts could be observed in comparison to wild type. This can be best visualized when the phototaxis data are amplified and compared to wild type, which has been done for the C3-ox100 strain (Fig. 2D). Thus, the maximum of circadian phototaxis was shifted by a few hours to the very early subjective day and no longer occurred during the middle of the subjective day.

Figure 2.

Figure 2.

Open in a new tab

The peak time of circadian phototaxis of cells maintained in constant darkness is significantly shifted toward the night phase in C3-ox and C3-sil strains as compared to the wild-type strain SAG 73.72. Ordinate, Cell density documented as extinction E in mV; abscissa, time in days. The free-running period of the strains is indicated. Phototaxis curves of wild-type (A), C3-ox strains (C3-ox90 [B] and C3-ox100 [C]) and C3-sil strains (C3-sil6 [E] and C3-sil7 [F]) are shown. Amplification and comparison of phototaxis curves of C3-ox100 and WT (D) and of C3-sil6 and WT (G) over the first 3 d under constant conditions are also shown.

Two C3-silenced strains (C3-sil6 and C3-sil7) showed a period of 24.3 and 24.2 h, respectively (Fig. 2, E and F), that is either in the range found for wild-type cells or very close to it. Again, acrophase shifts could be observed that were even more pronounced as with the C3-ox strains. This is again best highlighted by amplification of the phototaxis data as done with one of the two strains and comparison to wild type (Fig. 2G). Thus, the maximum of phototaxis was shifted by about 6 h to the very end of the subjective night and the minimum to the middle of the subjective day.

Since both up- and down-regulation of the C3 subunit significantly shift the acrophase of the circadian clock, it can be postulated that the C3 subunit represents a component of the circadian clock in C. reinhardtii that is involved in phase maintenance, as will be discussed later.

Up-Regulation of the C1 Subunit of CHLAMY1 Causes Coregulation of the C3 Subunit and Arrhythmic Phototaxis

For overexpression of C1, the same strategy was used as for C3 (Fig. 3A; Supplemental Protocol S1). An overexpression of C1 of about 2-fold was observed in the transgenic strains (C1-ox18, C1-ox93; Fig. 3B). Also in this case, it was determined if the other subunit C3 was influenced in its expression level. Clearly, the level of C3 was up-regulated in the C1-ox18 and C1-ox93 strains (Fig. 3C) with an even higher amplitude (3- to 4-fold) as observed for C1. Thus, in the C1-ox strains the entire C1 and C3 comprising CHLAMY1 complex should be overexpressed. Analysis of phototaxis with the two C1-ox strains revealed that their circadian clock is severely disturbed when examined over a 7 d time period under constant conditions of darkness (Fig. 4). C1-ox18 showed a normal rhythmic behavior over the first 2 d in constant conditions as could be seen in several measurements, but then always became altogether arrhythmic (Fig. 4A). C1-ox93 had a tendency to arrhythmicity from the very beginning under constant conditions including double peaks during day and night phase (Fig. 4B) or showing rather complete arrhythmicity in some measurements (data not shown).

Figure 3.

Figure 3.

Open in a new tab

Overexpression of C1 and coregulation of C3. A, The overexpression construct used for C1 is shown. Promoter indicates the truncated hsp70A/rbcS2 tandem promoter along with the first intron of rbcS2 that bears an enhancer (see “Materials and Methods”). The tagged c1 cDNA bears the major part of the c1 ORF (the first 23 amino acids are missing) and the entire 3′ UTR (see Supplemental Protocol S1). B and C, Different amounts of proteins from a crude extract (90, 60, and 30 μg per lanes) of wild-type (WT) cells were separated on SDS-PAGE and used for western analysis with anti-C1 (B) and anti-C3 (C) antibodies along with proteins of crude extracts (30 μg per lane) from two different C1-overexpressing strains (C1-ox18 and C3-ox93). The positions of C1 and C3 are indicated by arrowheads.

Figure 4.

Figure 4.

Open in a new tab

Overexpression of C1 causes arrhythmic behavior. A, Circadian phototaxis of C1-ox18 was measured using the automated phototaxis-measuring unit developed by Mergenhagen (1984). Ordinate, Cell density documented as extinction E in mV; abscissa, time in days. White circles over layered by an arrow indicate that the scale for mV measurement was exceeded. The free-running period of 25 h that is observed during the first 2 d is indicated, then the cells became completely arrhythmic. B, Phototaxis of C1-ox93 under constant conditions showing disturbances in the circadian clock.

These data show that a wild-type-like level of the C1 subunit is essential for an intact circadian clock and that its overexpression of about 2-fold can already cause arrhythmicity for the endogenous clock of C. reinhardtii.

Down-Regulation of the C1 Subunit of CHLAMY1 Can Cause Arrhythmicity

For silencing of C1, the same RNAi strategy as for C3 was applied (Fig. 5A; Supplemental Protocol S2). Silencing down to a level of about 25% to 30% was observed in some transgenic strains (e.g. C1-sil35), but some strains also showed only silencing down to 40% to 70% (e.g. C1-sil72; Fig. 5B). Again, we examined in the C1-sil strains if the level of C3 was changed in parallel. In strains where C1 was silenced to 25% to 30%, a strong coregulation of the C3 level was observed (e.g. C1-sil35; Fig. 5C), whereby C3 was even more silenced than C1 (below 25%). In contrast, a less pronounced or no coregulation of the C3 subunit was found in the transgenic strains that had shown a smaller degree of C1 silencing (e.g. C1-sil72; Fig. 5C). These data indicate that strong coregulation of C3 in case of C1 silencing depends on a critical low level of C1 that has to be reached in the cell.

Figure 5.

Figure 5.

Open in a new tab

Silencing of C1 below a critical level causes strong coregulation of C3. A, The RNAi construct used for silencing of C1 is shown. The numbers represent the involved exons. Introns are indicated by lines. The light-gray area represents the 5′ UTR, and c1 prom the potential promoter region of c1 (Supplemental Protocol S2). B and C, Different amounts of proteins from a crude extract (100, 50, and 25 μg per lanes) of wild-type (WT) cells were separated on SDS-PAGE and used for western analysis with the anti-C1 (B) and anti-C3 (C) antibodies along with proteins of crude extracts (100 μg per lane) from different C1-silenced strains (C1-sil35 and C1-sil72). The positions of C1 and C3 are indicated by arrowheads.

A challenging task was the analysis of the strong C1-silenced strains since the levels of C1 and of C3 reverted within a few weeks back to wild-type level. After transformation and selection on paramomycin plates, colonies were always grown up and checked by western analysis for their degree of silencing. If silencing of C1 was occurring, the same colony was grown up and checked again by western analysis to verify the result. In most but not all cases, silencing was still maintained within this time range. But in case of a further repeat 1 or 2 weeks later, reversion back to wild-type level was already completed in most strains analyzed, including C1-sil35. As will be discussed later, such a quick reversion of RNAi strains is unusual and also did not occur with the C3-silenced strains that were prepared by the same strategy. Phototaxis measurements with several strains, where C1 was silenced below 40% and C3 was strongly cosilenced, were done directly after the second verification of C1 silencing and in all cases they showed wild-type-like behavior of phototaxis. For example, phototaxis of C1-sil35 is shown (Fig. 6A). However, the wild-type-like behavior could be due to the fact that the C1 level had already been reverted to wild type since every analysis of the silencing level of C1 after the phototaxis assay revealed complete reversion of the C1 level back to wild type in these strains (data not shown).

Figure 6.

Figure 6.

Open in a new tab

Silencing of C1 can cause arrhythmic behavior. A, Circadian phototaxis of C1-sil35 was measured using the automated phototaxis-measuring unit developed by Mergenhagen (1984). Ordinate, Cell density documented as extinction E in mV; abscissa, time in days. The free-running period of 24.7 h is indicated. It should be noted that the C1 as well as the C3 levels were fully reverted back to wild-type level when C1-sil35 cells were checked after the phototaxis experiment. B, Phototaxis of C1-sil72 under constant conditions showing complete arrhythmicity. In this case, silencing was still observed after the phototaxis experiment.

Only in the case of strains where silencing was not as pronounced, reduced levels of C1 could still be found by western analysis after the phototaxis assay showing that the cells were still silenced in C1 during the assay. Notably, these transgenic lines showed arrhythmic behavior from the very beginning under constant darkness (C1-sil72; Fig. 6B) or after 3 d under constant conditions (data not shown). These data, in combination with the data from the C1-ox strains, suggest that C1 is an essential component of the circadian clock in C. reinhardtii.

The Circadian Rhythm of NII Activity Is Also Affected in Strains Where the C1 or C3 Levels Are Reduced

Automated measurement of the phototaxis rhythm over 7 d is well suited to obtain comprehensive information about the effects on phase and period in the transgenic strains. However, it was also of interest to find out if another circadian output rhythm was disturbed in the same way. Therefore, we have manually measured the rhythm of NII activity in strains where either the C3 or C1 levels are reduced in comparison to wild type. NII represents one of the key enzymes of nitrogen metabolism. Its activity was shown before to be diurnally regulated with a maximum during the middle of the light period (Pajuelo et al., 1995). Nii mRNA bears an UG repeat in its 3′ UTR that is recognized by CHLAMY1 (Waltenberger et al., 2001). Thus, it can be seen if CHLAMY1 may regulate one of its direct output targets in the same way as phototaxis. For this purpose, we have released cells that were grown under a light-dark (LD) cycle to constant dim light (LL) and measured NII activity starting from the second day during subjective night (LL 38) until the end of the next subjective day (LL 58) in a 4 h interval. In wild-type cells, NII showed a circadian rhythm in its activity, however, with a rather modest amplitude. Maximal activity occurred during early day phase (Fig. 7A). In the C3-sil6 strain, the acrophase of the peak of NII activity was shifted toward the night phase (Fig. 7B). Thus, maximal NII activity appeared at the end of the subjective night and decreased already during early subjective day. Thereby, the amplitude was enhanced in comparison to wild type.

Figure 7.

Figure 7.

Open in a new tab

NII activities over the circadian cycle in wild-type cells compared to cells where C3 or C1 are silenced. Cells were grown under a LD (12:12) cycle and then transferred to constant conditions of dim light (LL 0). After 38 h, cells were harvested every 4 h at the indicated times. Crude extracts were prepared and enzyme activities were determined as described in “Materials and Methods.” Subjective night, gray background; subjective day, white background. A, Wild-type cells SAG 73.72. B, Strain C3-sil6. C, Strain C1-sil81. Repetition of the NII measurements in the different strains resulted in similar results as shown. D, Different amounts of proteins from a crude extract (100, 50, and 25 μg per lanes) of wild-type (WT) cells were separated on SDS-PAGE and used for western analysis with the anti-C1 antibodies along with proteins of a crude extract (100 μg per lane) from the C1-sil81 strain.

For analyzing NII activity in a C1-silenced strain, we chose a most recent characterized strain, which is not very strongly silenced to avoid potential reversion during the experiments, as mentioned above. In C1-sil81 (Fig. 7D) the C1 level was reduced to about 53%. Immediately after the western analysis, the strain was used for the measurements of NII activity and it was verified after the experiment that the C1 level was still reduced. In this case, arrhythmic NII activity was observed (Fig. 7C). Thus, reduced levels of C3 or C1 cause the same effects (shifts in acrophase or arrhythmicity) on the rhythm of NII activity as on the phototaxis rhythm.

DISCUSSION

Strategies Used to Silence and Overexpress the C1 and C3 Subunits, Respectively

In this study, we have analyzed the role of the two subunits of CHLAMY1 within the circadian system by either silencing them via RNAi or overexpressing them under the control of a strong promoter. For RNAi, the method that was developed by Fuhrmann et al. (2001) for silencing of chlamyopsin was chosen that involved usage of the endogenous promoter of the gene. Thus, the potential promoter regions of the c1 and c3 genes together with the first exons and introns were fused to an inverted cDNA part so that a double-stranded RNA will be formed in the cell that triggers silencing. Such an approach had already been successfully applied for silencing of CK1 in our lab (Schmidt et al., 2006). Maximal silencing down to about 25% (with C1) or even below 25% (with C3) of wild-type level was observed, which is in the range of other RNAi silencing approaches conducted in C. reinhardtii (for review, see Schroda, 2006). While silencing of CK1 and C3 was relatively stable for a few months when the cells were kept on paromomycin plates, and only little reversion was observed during this time frame, strong silencing of C1 was not stable over time and could only be maintained for a few weeks. Reversion of RNAi strains back to wild-type level over a few months has been observed with RNAi and antisense constructs of C. reinhardtii (Schroda, 2006). However, the relatively quick reversion of C1 seems to be a particular case of C1 especially when it is silenced below a critical level of about 40% and may be triggered by a specific mechanism within the cell.

For overexpression, the strong truncated hsp70A/rbcS2 tandem promoter together with the first rbcS2 intron that bears an enhancer (Lumbreras et al., 1998; Sizova et al., 2001) was used. The truncated rbcS2 promoter was shown to be independent of light (Fuhrmann et al., 2004). The truncated tandem promoter was checked recently with regard to a potential circadian expression and found to mediate rather constant expression over a circadian cycle (S. Kiaulehn, O. Voytsekh, M. Fuhrmann, and M. Mittag, unpublished data). While C3 could be overexpressed more than 3-fold using this promoter, C1 overexpression ranged about 2-fold at most. Also, in the case of C1 overexpression, reversion back to wild type was observed, however, not in such a short time range as was found with its silencing. These data give some indications that high up- or down-regulation of C1 might have some toxic effects for the cell.

Changes in the Level of C3 Phase Shift the Peak of Circadian Phototaxis and of NII Activity

The major purpose of the up- and down-regulation of the C1 and C3 subunits was to study the function of the heteromeric RNA-binding protein within the circadian system. For this purpose, measurement of circadian phototaxis is well suited as an indicator for potential disturbances within the circadian clock since it can be automatically measured under constant conditions over a 7 d time range. Thus, sound conclusions can be drawn about effects on the phase and period of the circadian clock.

To examine the influence of up- and down-regulation of the C3 subunit, the phototaxis rhythm was measured in the transgenic C3-ox and C3-sil strains. Increasing amounts of the C3 level caused a slight period lengthening (about 0.5 h) in comparison to wild type. Decreases in the C3 amounts did not change the period in a significant way. However, both overexpression and silencing of C3 shifted the acrophase of the rhythm significantly. Thus, the maximum of circadian phototaxis, which occurs during the middle of the day phase in wild type, appeared at either the end of the night phase in C3-sil strains or the very beginning of the day phase in C3-ox strains. Thereby, the acrophase was shifted up to 6 h towards the night and this shift remained stable over the 7 d of measurement (Fig. 2). The shift in acrophase toward the night was also observed when another output rhythm, namely NII activity, was measured in C3-sil6. While the rhythm of NII activity has its maximum during early day phase in wild-type cells, maximal activity of NII in the C3-sil strain occurred during late night phase. These data suggest that the C3 subunit is involved in maintaining the correct phase angle of output rhythms such as phototaxis or NII activity.

Physiologically, phase shifts can be caused, for example, by pulses of light to which the cells are exposed at specific times of a circadian cycle as was first demonstrated by Hastings and Sweeney (1958) with a marine dinoflagellate. In C. reinhardtii action spectra have been determined for shifting the phase of its circadian phototaxis rhythm by light (Johnson et al., 1991; Kondo et al., 1991). These experiments suggested the involvement of blue- and red-light photoreceptors.

A disturbance in the acrophase of a circadian rhythm caused by up- or down-regulation of specific proteins had been already shown, for example, in case of the PAS/LOV protein VIVID (VVD) of N. crassa (Heintzen et al., 2001; Elvin et al., 2006). In those experiments, knockout of vvd resulted in a phase shift of about 4 h with regard to the conidiation rhythm in wild-type cells. Also, it was shown that temporal overexpression of the cyanobacterial clock gene kaiC can reset the phase of the rhythm (Ishiura et al., 1998). In Arabidopsis thaliana, phase-specific circadian regulation of transcription was recently found (Harmer and Kay, 2005). Further, it was shown in N. crassa that WC-1 possesses a promoter element that can modulate phase (Káldi et al., 2006). Thus, molecular components that are involved in phase adjustment are under investigation. The finding that an RNA-binding protein such as C3 can be involved in phase determination is, however, novel.

C3 does not have a PAS/LOV domain such as VVD or WC-1 that could perceive light information. We therefore hypothesize that the C3 subunit functions as a transducing component that can receive light information from a circadian photoreceptor and then alter the phase angle of circadian output rhythms.

The Duo of C1 and C3 Is Essential for an Intact Circadian Clock in C. reinhardtii

Previous experiments have demonstrated that CHLAMY1 consists of the C1 and C3 subunits and that their interaction is necessary to bind to their RNA targets (Zhao et al., 2004). Interestingly, overexpression of C1 caused a parallel increase in C3. Thus, the level of the CHLAMY1 complex as a whole should be up-regulated in the C1-ox strains. In these strains the circadian clock, as judged by measuring the phototaxis rhythm, is severely disturbed and arrhythmicity occurs. While C1-ox18 still shows circadian behavior for the first 2 d under constant conditions before it becomes arrhythmic, C1-ox93 has disturbances in its circadian rhythm from the very beginning. This is especially impressive when one considers that the level of C1 is altered very little (increase of only about 2-fold) in the two C1-ox strains. In the case of C1-ox18, where a wild-type-like circadian behavior was still visible during the first 2 d under constant conditions, there was no phase shift observed over this time range even though C3 was strongly co-up-regulated. Thus, one has to assume that the influence of increased C3 on acrophase shifting can be counteracted by the increased level of C1.

In the case of C1 silencing, arrhythmic behavior of phototaxis could be found in strains where C1 was only slightly silenced (e.g. C1-sil72). In such strains silencing of C1 was still observed after the phototaxis assay was finished when analyzed by westerns. C1-sil strains that had a more pronounced silencing below 40% (e.g. C1-sil35) were not stable over time. Reversion of C1 and in parallel of C3 back to its wild-type level occurred within a relatively short period of time and in each case examined the reversion was already completed when cells were analyzed after the phototaxis experiment. Thus, the wild-type-like circadian rhythm of C1-sil35, for example, cannot be interpreted in an unambiguous way. We assume that the cells were already reverted before the phototaxis measurement and thus a wild-type-like behavior occurred. It would be rather surprising if slight overexpression (about 2-fold) as well as slight silencing of C1 (e.g. down to about 70%) can cause arrhythmicity, while stronger silencing of C1 would not. However, we cannot rule out this possibility for sure.

Also in the case of C1 silencing, we checked the rhythm of NII activity as a second circadian output process using C1-sil81. In this strain the C1 level was reduced to about 53% and the reduced level could be still confirmed after the NII assay. Clearly, arrhythmicity of NII activity was observed. These data show that changes in the C1 level can cause arrhythmicity with regard to both analyzed output rhythms and corroborate its important role in the circadian clock of C. reinhardtii.

Coregulation of the C1 and C3 Subunits

In all model organisms studied so far the mechanism of circadian oscillation is triggered by positive and negative feedback loops (Harmer et al., 2001; Reppert and Weaver, 2002; Dunlap and Loros, 2004; Johnson, 2004) whereby transcription factors are involved. Here, we show that changes in a heteromeric duo of RNA-binding proteins, the KH-domain bearing C1, and the RRM-spanning C3 proteins, which can interact with each other, are able to modulate acrophase, alter slightly period, and result in arrhythmicity. Thus, key features of a circadian clock are affected in strains where C1 and/or C3 levels have been altered. It is of interest that the two subunits are interconnected in their expression level. An increase in C1 causes up-regulation of C3 and a decrease in C1 below a critical point results in strong down-regulation of C3. In both cases, the expression of C3 is altered with an even higher amplitude than C1 in comparison to wild type. In contrast, changes in the C3 level have little (C3 overexpression) or no significant (C3 silencing) effect on C1 expression. Therefore, C1 seems to be the core unit in the CHLAMY1 complex. A slight increase in C1 triggers an increase of C3, as observed with both C1-ox strains. In case of C1 silencing, there seems to be a critical low level of C1 that has to be reached to cause strong cosilencing of C3. Thus, the well-silenced C1-sil strains (e.g. C1-sil35) showed strong coregulation of C3, while the less-silenced C1 strains (e.g. C1-sil72) showed only weak coregulation.

Currently, we do not know by which mechanism this coregulation occurs, but this matter will be of great interest in future studies. In the case of the assembly governed regulation of the complex biogenesis of subunits of PSI and PSII, it was found that translation is involved in this process and that the 5′ UTRs communicate this regulation (Wostrikoff et al., 2004; Minai et al., 2006). If a comparable mechanism occurs with C1 and C3, there exists the possibility that C1 itself as an RNA-binding protein might regulate the translation of C3, for example. However, C1 might also activate some yet unknown factor that influences the expression of C3.

MATERIALS AND METHODS

Cell Culture

Chlamydomonas reinhardtii cells (wild-type strain SAG 73.72) were grown in Tris-acetate phosphate (TAP) medium (Harris, 1989) under a 12-h light/12-h dark cycle (LD 12:12) with a light intensity of 71 μE m−2 s−1 (1 E = 1 mol of photons) at 24°C. Cells were grown under LD cycle unless otherwise indicated. The beginning of the light period is defined as time zero (LD 0) and the beginning of the dark period is LD 12.

Preparation of Plasmid Constructs for Overexpressing of C1 and C3

For overexpression of C1 or C3, major parts of their cDNAs were put under control of the strong truncated hsp70A/rbcS2-tandem promoter of pRbcBRL that is fused with a two-amino acid spacer to the first intron of rbcS2 that bears an enhancer (Lumbreras et al., 1998; Sizova et al., 2001; Fuhrmann et al., 2004). The detailed cloning procedures for the overexpressing vectors can be found in Supplemental Protocol S1.

Preparation of Plasmid Constructs for Silencing of C1 and C3 via RNAi

Silencing of C1 and C3 by RNAi was done according to Fuhrmann et al. (2001). The detailed cloning procedures for the RNAi vectors can be found in Supplemental Protocol S2.

Transformation of C. reinhardtii with the RNAi and Overexpressing Vectors

C. reinhardtii wild-type strain SAG 73.72 was grown in TAP medium under a LD 12:12 cycle with a light intensity of 71 μmol·m−2·s−1 at 24°C up to a cell density of 2 to 5 × 106 cells/mL for transformation. Twenty micrograms of DNA of each plasmid DNA that was linearized with ScaI were used for transformation according to Kindle (1990) with the following modification. Vortexing of the cells with glass beads was carried out for 15 s. Cells were then transferred in liquid TAP medium in light overnight according to Davies et al. (1992) and plated with 0.5% cover agar on TAP-paromomycin plates (selection medium) following the protocol of Sizova et al. (2001).

Preparation of Crude Extracts of C. reinhardtii and Western-Blot Analysis

Cells were grown to a cell density of 1 to 5 × 106 cells per mL, harvested by centrifugation at LD6, and stored at −80°C after being frozen in liquid nitrogen. Aqueous extracts were prepared according to Zhao et al. (2004). All extracts were prepared in the presence of Complete Proteinase Inhibitor Cocktail (Roche-Applied-Science) according to the user's manual.

Western-blot analysis was done as described in Zhao et al. (2004) with chemiluminescence detection.

Automated Measurement of Circadian Phototaxis (Photoaccumulation) with Wild-Type, RNAi, and Overexpressing Strains

Phototaxis measurements were done with a custom-made machine, developed and described by Mergenhagen (1984). Briefly, a light beam that passed through the culture in a cuvette (30 mL flat Falcon tube) was measured by a photocell, such that the amount of light transmitted depended on the number of cells in the light path. The accumulation of the cells in the light path was determined by the cell density (documented as extinction in mV) with the minimum value (few or no cells in the light path) being arbitrarily set at zero and was measured every 2 h for a period of 20 min. The recording system was installed in a temperature-controlled dark room of 22.5°C.

Assay for NII Activity

C. reinhardtii cells (wild-type strain SAG 73.72) were grown in TAP medium with NO3− as nitrogen source under a LD 12:12 cycle and then put under constant conditions of dim light (LL; 15 μE m−2 s−1) before further use. The beginning of the dim light period is defined as LL 0. Hours under which cells have been kept under dim light are given.

For the NII assay, cell pellets were washed with 50 mm Tris/5 mm EDTA, pH 8, two times before their storage at −80°C. For extracts, cells were resuspended in a buffer of 50 mm Tris/5 mm EDTA pH 8/14 mm dithiothreitol, and lysed by vortexing (highest speed) with glass beads (diameter: 0.25–0.30 mm) for 5 × 1 min. After each min of vortexing, they were placed on ice for 2 min. Cell debris was removed as described in Zhao et al. (2004) and the resulting crude extract was immediately further used. NII activity was determined by a dithionite assay as described by Vega et al. (1980) along with 15 μg of total protein from the crude extract. The reaction was carried out for 20 min at 40°C. The assay involved sodium dithionite as reductant and methyl viologen as electron carrier. Enzymatic activity was followed by measuring colorimetrically (540 nm) the rate of disappearance of nitrite. One unit of activity was defined as the amount of enzyme that catalyzed the reduction of 1 μmol of nitrite per minute.

Sequence data from this article can be found in the GenBank/EMBL data libraries under accession numbers AAR90343 (C1) and AAR90344 (C3).

Supplemental Data

The following materials are available in the online version of this article.

  • Supplemental Protocol S1. Plasmid constructs for overexpressing C1 and C3.

  • Supplemental Protocol S2. Plasmid constructs for silencing of C1 and C3.

Supplementary Material

[Supplemental Data]
plntphys_pp.106.085944_index.html (860B, html)

Acknowledgments

We thank Markus Fuhrmann for providing pSI103 and pRbcBRL, Arntraut Götsch for providing pAG2, and Woody Hastings for helpful suggestions concerning the manuscript. Special thanks is given to Dieter Mergenhagen for the donation of the automated phototaxis measuring unit. We appreciate the free delivery of information by the U.S. (Department of Energy) genome project of C. reinhardtii very much.

1

This work was supported by the Deutsche Forschungsgemeinschaft (grant nos. Mi373/6–2 to 3).

The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (www.plantphysiol.org) is: Maria Mittag (m.mittag@uni-jena.de).

[W]

The online version of this article contains Web-only data.

References

  1. Breton G, Kay SA (2006) Circadian rhythms lit up in Chlamydomonas. Genome Biol 7: 215–219 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Bruce VG (1970) The biological clock in Chlamydomonas reinhardtii. J Protozool 17: 328–334 [Google Scholar]
  3. Davies JP, Weeks DP, Grossman AR (1992) Expression of the arylsulfatase gene from the beta 2-tubulin promoter in Chlamydomonas reinhardtii. Nucleic Acids Res 20: 2959–2965 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Dunlap JC, Loros JJ (2004) The Neurospora circadian system. J Biol Rhythms 19: 414–424 [DOI] [PubMed] [Google Scholar]
  5. Elvin M, Loros JJ, Dunlap JC, Heintzen C (2006) The PAS/LOV protein VIVID supports a rapidly dampened daytime oscillator that facilitates entrainment of the Neurospora circadian clock. Genes Dev 19: 2593–2605 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Fuhrmann M, Hausherr A, Ferbitz L, Schodl T, Heitzer M, Hegemann P (2004) Monitoring dynamic expression of nuclear genes in Chlamydomonas reinhardtii by using a synthetic luciferase reporter gene. Plant Mol Biol 55: 869–881 [DOI] [PubMed] [Google Scholar]
  7. Fuhrmann M, Stahlberg A, Govorunova E, Rank S, Hegemann P (2001) The abundant retinal protein of the Chlamydomonas eye is not the photoreceptor for phototaxis and photophobic responses. J Cell Sci 114: 3857–3863 [DOI] [PubMed] [Google Scholar]
  8. Göhre V, Ossenbühl F, Crèvecoeur M, Eichacker LA, Rochaix J-D (2006) One of two Alb3 proteins is essential for the assembly of the photosystems and for cell survival in Chlamydomonas. Plant Cell 18: 1454–1466 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Grossman AR (2005) Paths toward algal genomics. Plant Physiol 137: 410–427 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Grossman AR, Harris EE, Hauser C, Lefebvre PA, Martinez D, Rokhsar D, Shrager J, Silflow CD, Stern D, Vallon O, et al (2003) Chlamydomonas reinhardtii at the crossroads of genomics. Eukaryot Cell 2: 1137–1150 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Harmer SL, Kay SA (2005) Positive and negative factors confer phase-specific circadian regulation of transcription in Arabidopsis. Plant Cell 17: 1926–1940 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Harmer SL, Satchidananda P, Kay SA (2001) Molecular bases of circadian rhythms. Annu Rev Cell Dev Biol 17: 215–253 [DOI] [PubMed] [Google Scholar]
  13. Harris EH (1989) The Chlamydomonas Sourcebook: A Comprehensive Guide to Biology and Laboratory Use. Academic Press, San Diego [DOI] [PubMed]
  14. Hastings JW, Sweeney BM (1958) A persistent diurnal rhythm of luminescence in Gonyaulax polyedra. Biol Bull 115: 440–458 [Google Scholar]
  15. Heintzen C, Loros JJ, Dunlap JC (2001) The PAS protein VIVID defines a clock-associated feedback loop that represses light input, modulates gating, and regulates clock resetting. Cell 104: 453–464 [DOI] [PubMed] [Google Scholar]
  16. Ishiura M, Kutsuna S, Aoki S, Iwasaki H, Andersson CR, Tanabe A, Golden SS, Johnson CH, Kondo T (1998) Expression of a gene cluster kaiABC as a circadian feedback process in cyanobacteria. Science 281: 1519–1523 [DOI] [PubMed] [Google Scholar]
  17. Johnson CH (2004) Precise circadian clocks in prokaryotic cyanobacteria. Curr Issues Mol Biol 6: 103–110 [PubMed] [Google Scholar]
  18. Johnson CH, Kondo T, Hastings JW (1991) Action spectrum for resetting the circadian phototaxis rhythm in the CW15 strain of Chlamydomonas. II. Illuminated cells. Plant Physiol 97: 1122–1129 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Káldi K, Gonzáles BH, Brunner M (2006) Transcriptional regulation of the Neurospora circadian clock gene wc-1 affects the phase of circadian output. EMBO Rep 7: 199–204 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Kindle KL (1990) High-frequency nuclear transformation of Chlamydomonas reinhardtii. Proc Natl Acad Sci USA 87: 1228–1232 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Kondo T, Johnson CH, Hastings JW (1991) Action spectrum for resetting the circadian phototaxis rhythm in the CW15 strain of Chlamydomonas. I. Cells in darkness. Plant Physiol 95: 197–205 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Kondo T, Strayer CA, Kulkarni RD, Taylor W, Ishiura M, Golden SS, Johnson CH (1993) Circadian rhythms in prokaryotes: luciferase as a reporter of circadian gene expression in cyanobacteria. Proc Natl Acad Sci USA 90: 5672–5676 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Ladd AN, Charlet-B N, Cooper TA (2001) The CELF family of RNA binding proteins is implicated in cell-specific and developmentally regulated alternative splicing. Mol Cell Biol 21: 1285–1296 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Lee K, Loros JJ, Dunlap JC (2000) Interconnected feedback loops in the Neurospora circadian system. Science 289: 107–110 [DOI] [PubMed] [Google Scholar]
  25. Lumbreras V, Stevens DR, Purton S (1998) Efficient foreign gene expression in Chlamydomonas reinhardtii mediated by an endogenous intron. Plant J 14: 441–447 [Google Scholar]
  26. Mergenhagen D (1984) Circadian clock: genetic characterization of a short period mutant of Chlamydomonas reinhardtii. Eur J Cell Biol 33: 13–18 [PubMed] [Google Scholar]
  27. Minai L, Wostrikoff K, Wollman FA, Choquet Y (2006) Chloroplast biogenesis of photosystem II cores involves a series of assembly-controlled steps that regulate translation. Plant Cell 18: 159–175 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Mittag M (1996) Conserved circadian elements in phylogenetically diverse algae. Proc Natl Acad Sci USA 93: 14401–14404 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Mittag M, Kiaulehn S, Johnson CH (2005) The circadian clock in Chlamydomonas reinhardtii. What is it for? What is it similar to? Plant Physiol 137: 399–409 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Mittag M, Lee DH, Hastings JW (1994) Circadian expression of the luciferin-binding protein correlates with the binding of a protein to the 3′ untranslated region of its mRNA. Proc Natl Acad Sci USA 91: 5257–5261 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Morse D, Milos PM, Roux E, Hastings JW (1989) Circadian regulation of the synthesis of substrate binding protein in the Gonyaulax bioluminescent system involves translational control. Proc Natl Acad Sci USA 86: 172–176 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Pajuelo E, Pajuelo P, Clemente MT, Marquez J (1995) Regulation of the expression of ferredoxin-nitrite reductase in synchronous cultures of Chlamydomonas reinhardtii. Biochim Biophys Acta 1249: 72–78 [DOI] [PubMed] [Google Scholar]
  33. Reppert SM, Weaver DR (2002) Coordination of circadian timing in mammals. Nature 418: 935–941 [DOI] [PubMed] [Google Scholar]
  34. Schmidt M, Geßner G, Luff M, Heiland I, Wagner V, Kaminski M, Geimer S, Eitzinger N, Reißenweber T, Voytsekh O, et al (2006) Proteomic analysis of the eyespot of Chlamydomonas reinhardtii provides novel insights into its components and tactic movements. Plant Cell 18: 1908–1930 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Schroda M (2006) RNA silencing in Chlamydomonas: mechanisms and tools. Curr Genet 49: 69–84 [DOI] [PubMed] [Google Scholar]
  36. Sizova I, Fuhrmann M, Hegemann P (2001) A Streptomyces rimosus aphVIII gene coding for a new type phosphotransferase provides stable antibiotic resistance to Chlamydomonas reinhardtii. Gene 277: 221–229 [DOI] [PubMed] [Google Scholar]
  37. Sudol M, Hunter T (2000) New wrinkles for an old domain. Cell 103: 1001–1004 [DOI] [PubMed] [Google Scholar]
  38. Vega JM, Cardenas J, Losada M (1980) Ferredoxin-nitrite reductase. Methods Enzymol 69: 255–270 [Google Scholar]
  39. Waltenberger H, Schneid C, Grosch JO, Bareiss A, Mittag M (2001) Identification of target mRNAs from C. reinhardtii for the clock-controlled RNA-binding protein Chlamy 1. Mol Genet Genomics 265: 180–188 [DOI] [PubMed] [Google Scholar]
  40. Wostrikoff K, Girard-Bascou J, Wollman F-A, Choquet Y (2004) Biogenesis of PSI involves a cascade of translational autoregulation in the chloroplast of Chlamydomonas. EMBO J 23: 2696–2705 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Zhao B, Schneid C, Iliev D, Schmidt EM, Wagner V, Wollnik F, Mittag M (2004) The circadian RNA-binding protein CHLAMY 1 represents a novel type heteromer of RNA recognition motif and lysine homology domain-containing subunits. Eukaryot Cell 3: 815–825 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

[Supplemental Data]
plntphys_pp.106.085944_index.html (860B, html)
plntphys_pp.106.085944_1.pdf (24KB, pdf)

Articles from Plant Physiology are provided here courtesy of Oxford University Press

RESOURCES