Academic onefile – document – a central integrator of transcription networks in plant stress and energy signalling brain anoxia

Photosynthetic plants are the principal solar energy converter sustaining

Life on earth. Despite its fundamental importance, little is known about how

Plants sense and adapt to darkness in the daily light-dark cycle, or how they

Adapt to unpredictable environmental stresses that compromise photosynthesis

And respiration and deplete energy supplies. Current models emphasize diverse

Stress perception and signalling mechanisms [1, 2]. Using a combination of

Cellular and systems screens, we show here that the evolutionarily conserved

Arabidopsis thaliana

Protein kinases, KIN10 and KIN11 (also known as AKIN10/at3g01090 and

AKIN11/at3g29160, respectively), control convergent reprogramming of

Transcription in response to seemingly unrelated darkness, sugar and stress

brain anoxia

Conditions. Sensing and signalling deprivation of sugar and energy, KIN10

Targets a remarkably broad array of genes that orchestrate transcription

Networks, promote catabolism and suppress anabolism. Specific bzip

Transcription factors partially mediate primary KIN10 signalling. Transgenic

KIN10 overexpression confers enhanced starvation tolerance and lifespan

Extension, and alters architecture and developmental transitions.

Significantly, double kin10 kin11

Deficiency abrogates the transcriptional switch in darkness and stress

Signalling, and impairs starch mobilization at night and growth. These

Studies uncover surprisingly pivotal roles of KIN10/11 in linking stress,

Sugar and developmental signals to globally regulate plant metabolism, energy

brain anoxia

Balance, growth and survival. In contrast to the prevailing view that sucrose

Activates plant snrk1s (snf1-related protein kinases) [3, 4, 5, 6], our

Functional analyses of arabidopsis

KIN10/11 provide compelling evidence that snrk1s are inactivated by sugars

And share central roles with the orthologous yeast snf1 and mammalian AMPK in

Energy signalling.

We have previously shown that the arabidopsis

Glucose sensor hexokinase 1 (HXK1) is essential for the glucose repression

Of photosynthesis genes [9]. However, DIN

Gene repression by glucose in the dark was still active in the


-null mutant gin2

( glucose-insensitive2

) (fig. 1c), indicating the involvement of an HXK1-independent signalling

Pathway [10].Brain anoxia to identify the key regulators governing this stress/sugar

Signalling pathway, we developed a cell model system using a sensitive

Reporter, generated by fusing the putative promoter of

Arabidopsis DIN6

(encoding the glutamine-dependent asparagine synthetase, ASN1

) to the luciferase ( LUC

) gene (methods summary). In transfected mesophyll protoplasts,


Was activated by darkness, DCMU, hypoxia/submergence and other stresses

Within 3-6 h (fig. 1d, and J.S., unpublished), in a similar way to the

Endogenous DIN6

Gene [8, 11, 12, 13] (fig. 1a-c; and supplementary fig. 2a). This activation

Was repressed by sucrose or glucose at physiological concentrations as low as

5 mm (supplementary fig. 2b). Interestingly, endogenous DIN6

brain anoxia

Activation by diverse signals was abolished by the protein kinase inhibitor

K252a (supplementary fig. 2c), suggesting the requirement of protein kinases

In the integration of stress and sugar signals.

We first investigated the involvement of members of the snf1-related kinase

(snrk) gene families [3] (supplementary fig. 3a), implicated in metabolic

Regulation owing to their sequence homology to the yeast snf1 and mammalian

AMPKs [3, 14, 15]. Cell-based functional screens revealed the specific

Activation of DIN6-LUC

By the two ubiquitously expressed members of the snrk1 group, KIN10 and

KIN11 (fig. 1e, and supplementary fig. 3b, c). Representative snrk2 and snrk3

Members lacked the same function (fig. 1e) despite their established protein

brain anoxia

Kinase activities in osmotic and salt stress responses (methods summary).

Expression of KIN10 and KIN11 in protoplasts conferred significant kinase

Activity, which required the conserved ATP binding site (K48 and K49) and

T175 (ref. 14; supplementary fig. 4, and fig. 1f). However, unlike T172

Phosphorylation in AMPK, T175 phosphorylation in KIN10 was not correlated

With activation by dark, DCMU or hypoxia treatment (data not shown). Thus,


Served as a sensitive and quantitative transcription reporter to monitor

Endogenous KIN10/11 activity in vivo


To substantiate this finding further, we screened arabidopsis

BZIP transcription factors [16] that have been characterized as G-box

Binding factors (gbfs), and found that GBF5/bzip2 (at2g18160) specifically

brain anoxia

Activated DIN6-LUC

(fig. 2e). Importantly, co-expression of KIN10 with GBF5 resulted in a

Dramatic and synergistic effect on DIN6-LUC

Activation. Both the GBF5 effect and its synergism with KIN10 were abolished

By the specific DIN6

Promoter mutation in the G-box g1

(fig. 2f). Similar results were obtained with closely related bzip S-group

Members expressed in mesophyll cells (bzip11/ATB2, at4g34590; bzip53,

At3g62420; and related bzip1, at5g49450; supplementary table 1 and fig. 2g,

H). Functional redundancy of the transcription factors was evident also from

The lack of overt phenotypes in the available loss-of-function single


Mutants (data not shown). Because the S- and C-group bzips have been shown

brain anoxia

To form functional heterodimers [16], we also examined the activity of

C-group bzips. Interestingly, one member of the C-group, bzip63, showed

Little activity alone but exhibited moderate synergism with KIN10 through the

G-box (fig. 2h). These studies identified novel functions of specific

GBFs/bzips and provide further evidence for convergent responses in stress

Signalling through KIN10 and KIN11.

To explore the extent of KIN10 transcriptional regulation and identify its

Early target genes, we scaled-up transient protoplast expression experiments,

Which could rule out secondary or long-term effects of metabolism and growth,

And circumvent limitations caused by redundancy and embryonic lethality

Observed in mammals and plants [15, 17, 18].Brain anoxia gene expression profiles with or

Without KIN10 expression were compared using arabidopsis

Whole-genome ATH1 genechips (GEO accession number GSE8257; supplementary

Methods). To filter candidate KIN10 target genes (at least twofold change

With P

Value 0.0004), we combined independent experimental validation using

Quantitative real-time reverse transcriptase PCR (qrt-PCR) with vigorous

Statistical analyses of genechip data from biological replicates, using

Parametric and non-parametric methods (see supplementary methods and

Supplementary fig. 6 for details). Guided by the experimentally validated


Transcription factor gene and 24 other marker genes, we were able to define

1,021 putative KIN10 target genes that exhibited reproducible activation or

brain anoxia

Repression (supplementary tables 1-3).

A final filtering step was applied to remove genes with inconsistent

Expression patterns in any of the seven data sets. The stringent and

Multi-step filtering processes selected a reliable list of 278 genes

Co-activated by KIN10 and sugar starvation conditions, and co-repressed in

Sugar-treated seedlings or CO 2

-fixing adult leaves (fig. 3a, and supplementary table 4). A second list of

322 genes was also identified on the basis of their co-repression by KIN10

And sugar starvation conditions, but co-activation in sugar-treated seedlings

Or differential CO 2

-fixing adult leaves (fig. 3b, and supplementary table 4). As predicted by


Gene response (fig. 1), hypoxia-treated protoplasts also displayed gene

brain anoxia

Expression profiles similar to KIN10 overexpression on the basis of ATH1

GeneChip (GEO accession number GSE8248) and triplicated analyses of selected

Marker genes by qrt-PCR (supplementary fig. 7a, b). The correlation across

The data sets (supplementary table 5) was remarkable considering the

Diversity of materials and experimental conditions used for the comparison

And filtering (see supplementary methods). This suggests that KIN10

Regulation is a cell-autonomous and ubiquitous event and highlights its

Physiological relevance.

Previous studies have identified specific snrk1 target genes, one each in

Potato and wheat [3, 5]. The present work provides for the first time a

Complete overview of the surprisingly broad genome-wide transcript changes

brain anoxia

Induced by these conserved protein kinases in a multicellular organism. The

Most prominent KIN10-activated (and sugar-repressed) genes represented a

Variety of major catabolic pathways, including cell wall, starch, sucrose,

Amino acid, lipid, and protein degradation that provide alternative sources

Of energy and metabolites (supplementary fig. 7c, and supplementary tables 3,

4). KIN10 also induced several APG/ATG


) orthologues and plant-specific trehalose metabolism genes (


), which potentially alter or sense the level of trehalose-6-P, a regulator

Of plant carbohydrate metabolism, growth and development [23, 24].

Conversely, a large set of genes involved in energy-consuming ribosome

brain anoxia

Biogenesis (87 genes) and anabolism (supplementary fig. 7d, e) were

Co-ordinately repressed by KIN10 (supplementary tables 3, 4). Although many

Plant metabolic pathways have been well characterized at the biochemical

Level, their molecular regulatory mechanisms are still poorly understood.

Thus, besides the well-known roles of snrk1 in phosphorylating and modulating

Enzymes important for carbon and nitrogen metabolism [3, 14, 25, 26], the

Present work identifies KIN10 and KIN11 as global regulators of gene

Expression involved in primary and secondary metabolism and protein synthesis

[11, 12, 13, 19, 20, 21, 22, 27]. The results also reveal central and

Previously unrecognized regulatory roles of the plant snrk1 (and possibly

brain anoxia

Also mammalian AMPK) in controlling a large number of genes encoding

Transcription factors, chromatin remodelling factors, and a plethora of

Signal transduction components (supplementary fig. 7c, e, and supplementary

Tables 3, 4).

To study the long-term effects of KIN10 activity at the whole-plant level, we

Generated arabidopsis

Transgenic plants overexpressing KIN10

( O1

And O2

) or with reduced KIN10

Expression using RNA interference (rnai; 10-1

And 10-2

) (supplementary fig. 8a). The KIN10

Overexpression seedlings displayed some advantage in primary root growth and

Development under low light with limited energy supply (supplementary fig.

8b, e). With exogenous sucrose, KIN10

Overexpression plants showed reduced growth in shoots and roots, possibly

brain anoxia

Owing to KIN10 repression of biosynthetic activities (supplementary fig.

8c-e). In contrast, the KIN10

Silenced lines were able to use exogenously supplied sucrose (supplementary

Fig. 8c-e) and glucose (data not shown) more efficiently and exhibited

Enhanced shoot and root growth. Growth on 3% sucrose also increased

Anthocyanin accumulation in wild-type, 10-1

And 10-2

Plants, but not in the KIN10

Overexpression lines (supplementary fig. 8f). Consistently,


Overexpression repressed the expression of MYB75/PAP1

(at1g56650) (supplementary table 3), a key transcription factor for

Anthocyanin biosynthesis (supplementary methods).

To circumvent the experimental limitation of obtaining double

Loss-of-function mutants [3, 17, 18], we generated kin11

brain anoxia

Single- and kin10 kin11

Double-mutant plants, using virus-induced gene silencing (VIGS) [28] of


In wild-type and 10-1

Or 10-2

Transgenic seedlings. Reduction of KIN

Transcripts was confirmed by RT-PCR and immunoblotting (using KIN10 and

P-AMPK antibodies; supplementary fig. 9a) before extensive phenotypic and

Molecular studies. Three weeks after infiltration, double mutants


And d-2

Showed a surprisingly dramatic growth defect with small leaves and short

Petioles under 13 h light/11 h dark (fig. 4d, and supplementary fig. 9b) or

Constant illumination conditions (data not shown). A strong accumulation of

Anthocyanins, curling of new leaves (leaves 11-12) and symptoms of early

Senescence were visible (fig. 4d).Brain anoxia the impact on growth and senescence in


And d-2

Lines was most striking five weeks after infiltration, when the transition

To the reproductive phase occurred in wild-type plants (fig. 4e). The most

Severely silenced plants senesced before flowering (fig. 4e), whereas those

With milder silencing bolted but failed to produce viable primary

Inflorescences and axillary floral meristems (fig. 4f). Unlike the

Snf1a snf1b

Double mutant in moss [29], the growth phenotypes could neither be rescued

By continuous light irradiance nor by supplementation with 1% sucrose (data

Not shown). This argues against the growth defects being merely due to

Impaired catabolism, and supports a more fundamental and unexpected role of

brain anoxia

KIN10/KIN11 in normal vegetative and reproductive growth and development in

Flowering plants.

To establish a definitive molecular and quantitative link between KIN10/11

Action and the ability to mount transcription activation in response to

Stresses and energy deprivation, we examined the response to hypoxia, dark or

DCMU in mesophyll cells and intact leaves of wild-type, single- and

Double-mutant plants. All KIN10 marker genes tested could no longer be

Activated by any of these stresses in the d-1

And d-2

Double mutants (fig. 4g, h; supplementary fig. 9c, d, and supplementary

Table 6). Consistent with a role of KIN10/11 in promoting catabolic

Processes, leaf starch–as a major carbon source in the absence of

brain anoxia

Photosynthesis–remained high at the end of night in the d-1

And d-2

Double mutants (fig. 4i). This suggests that KIN10/11 may have a previously

Unrecognized regulatory role in starch mobilization at night [30].

Alternatively, starch accumulation may reflect a reduced energy demand for

Growth. In the future, it will be interesting and informative to extend the

Analysis of gene expression reprogramming to changes in enzyme activities and

Metabolite homeostasis in response to environmental cues associated with

KIN10/KIN11 regulation.

Our studies provide molecular, biochemical, genetic and genomic evidence for

The essential functions of KIN10/KIN11 in plant protection and survival under

Stress, darkness and sugar deprivation conditions.Brain anoxia surprising roles of

KIN10/KIN11 in vegetative and reproductive growth and developmental

Transition under normal conditions are uncovered (supplementary fig. 1).

Their roles as energy sensors and integrators, orchestrating global

Transcription, may also be found in mammals and thus our findings in plants

May contribute to the understanding of AMPK functions in relation to

Diabetes, cancer, obesity and longevity. Further dissection of the components

And molecular mechanisms of the KIN10/KIN11 signalling cascades will provide

Valuable information on the metabolic control of plant growth and

Development, which may enable a more targeted genetic modification of plant

Development, architecture, carbon allocation, and stress resistance–all

brain anoxia

Major determinants of crop yield and renewable energy production.

Protoplast transient expression assays were carried out as described [32, 34,

38], using 35S-GUS


Reporter gene fusions as transfection controls [32, 34]. Data were generated

From at least three independent experiments with consistent results.

Protoplasts (1-4 x 10 4

) were incubated for 6 h in 1 ml of mannitol buffer in 6-well tissue culture

Plates (1 mm depth, excluding hypoxia effects) [32, 34, 38] or submerged in 1

Ml of mannitol buffer in a 1.5 ml microfuge tube (25-mm depth) for hypoxia

Treatment. Establishment of the hypoxic condition was confirmed by the

Induction of well-known hypoxia marker genes encoding ADH and PDC (refs. 38,

brain anoxia

39) in qrt-PCR and microarray analyses (data not shown). These genes are

Induced under hypoxic and anoxic conditions, but not under other types of

Cellular energy stress. Unlike DIN1

And DIN6,

Their induction cannot be repressed by sugar [40]. For dark treatment,

Plates were covered with aluminium foil for 6 h. If not otherwise indicated,

DCMU, glucose and sucrose, were added to final concentrations of 20 [mu]M, 25

MM and 50 mm, respectively.

The fifth and sixth leaves of 4-week-old plants, protoplasts and one-week-old

Seedlings were treated with darkness (leaves and protoplasts, 6 h; leaves of

VIGS plants, 10 h; seedlings, 20 h), DCMU (leaves and seedlings, 100 [mu]M,

20 h; leaves of VIGS plants, 100 [mu]M, 10h; protoplasts, 50 [mu]M, 6h) or

brain anoxia

Submerged in water (hypoxia treatment, leaves and seedlings, 20 h;

Protoplasts, 6 h) in the absence or presence of 3% sucrose, 25 mm glucose, or

A protein kinase inhibitor k252a (2 [mu]M). RNA samples were isolated and

Analysed by RT-PCR for DIN1

And DIN6

Expression using gene-specific primers (supplementary table 7).


Was used as a control gene. Total RNA (1 [mu]g) was converted to

Single-stranded complementary DNA by reverse transcriptase (invitrogen).

Quantitative real time RT-PCR (qrt-PCR) was performed using PCR master mix

SYBR green (biorad) in a biorad icycler (3 min 95 [degrees]C, 40 cycles at 10

S 95[degrees]C, 45 s at 59[degrees]C, and 1 min 95[degrees]C, 1 min

55[degrees]C). In the protoplast assays, marker gene expression was

brain anoxia

Normalized to CIPK23

(at1g30270) expression, chosen on the basis of its steady levels in the

KIN10 and hypoxia genechip experiments (supplementary table 1). In the leaf

Assays, marker gene expression was normalized to TUB4