Planta (2014) 240:137–146 DOI 10.1007/s00425-014-2069-3

Original Article

Arabidopsis thaliana PDX1.2 is critical for embryo development and heat shock tolerance Jan Erik Leuendorf · Sutton L. Mooney · Liyuan Chen · Hanjo A. Hellmann 

Received: 30 November 2013 / Accepted: 20 March 2014 / Published online: 19 April 2014 © Springer-Verlag Berlin Heidelberg 2014

Abstract  Main conclusion  PDX1.2 is expressed in the basal part of the globular-stage embryo, and plays critical roles in development, hypocotyl elongation, and stress response. Abstract The Arabidopsis thaliana PDX1.2 protein belongs to a small family of three members. While PDX1.1 and PDX1.3 have been extensively described and are well established to function in vitamin B6 biosynthesis, the biological role of PDX1.2 still remains elusive. Here, we show that PDX1.2 is expressed early in embryo development, and that heat shock treatment causes a strong up-regulation of the gene. Using a combined genetic approach of T-DNA insertion lines and expression of artificial micro RNAs, we can show that PDX1.2 is critically required for embryo development, and for normal hypocotyl elongation. Plants with reduced PDX1.2 expression also display reduced primary root growth after heat shock treatments. The work overall provides a set of important new findings that give greater insights into the developmental role of PDX1.2 in plants. J. E. Leuendorf and S. L. Mooney equally contributed. Electronic supplementary material The online version of this article (doi:10.1007/s00425-014-2069-3) contains supplementary material, which is available to authorized users. J. E. Leuendorf  Freie University Berlin, Albrecht‑Thaer‑Weg 15, 14195 Berlin, Germany e-mail: [email protected]‑berlin.de S. L. Mooney · L. Chen · H. A. Hellmann (*)  School of Biological Sciences, Washington State University, P.O. Box 644236, Pullman, WA 99164‑4236, USA e-mail: [email protected] L. Chen e-mail: [email protected]

Keywords Embryo development · Heat shock response · PDX family · Pyridoxal-5-phosphate synthase · Vitamin B6 Abbreviations amiRNA Artificial microRNA Estr Estradiol PDX Pyridoxine biosynthesis PLP Pyridoxal-5-phosphate vitB6 Vitamin B6 Gene accession numbers PDX1.1 At2g38230 PDX1.2 At3g16050 PDX1.3 At5g01410 PDX2 At5g60540 ACTIN At3g18780

Introduction The pyridoxine biosynthesis (PDX) family in plants is primarily known for its role in vitamin B6 (vitB6) biosynthesis (Leuendorf et al. 2008, 2010; Mooney et al. 2013). In Arabidopsis thaliana, it consists of three PDX1 members, PDX1.1, PDX1.2, and PDX1.3, and one PDX2 protein. PDX1.1 and PDX1.3 synthesize, in concert with PDX2, the active B6 co-factor pyridoxal-5-phosphate (PLP), which is primarily required for amino acid metabolism, but can also function in other metabolic pathways and as an important antioxidant (Bilski et al. 2000; Chen and Xiong 2005; Titiz et al. 2006; Denslow et al. 2007; Gonzalez et al. 2007; Mooney and Hellmann 2010). Consequently, loss of PDX1.1, PDX1.3, and PDX2 affects vitB6 biosynthesis, and plants affected in either both PDX1.1 and PDX1.3 or PDX2 develop an embryo lethal phenotype. In addition, plants

13

138

affected in either PDX1.1 or PDX1.3 have been described to be widely affected in development and abiotic stress sensitivities (Tambasco-Studart et al. 2005, 2007; Titiz et al. 2006; Wagner et al. 2006; Leuendorf et al. 2010). PDX1 and PDX2 proteins can be found in bacteria, archaea, and eukarya (Mooney et al. 2009), and crystallization studies in several non-plant organisms showed that they assemble to a cog-wheel-like structure that functions as a PLP synthase, and which consists of two hexameric PDX1 rings surrounded by 12 PDX2 proteins (Raschle et al. 2005; Strohmeier et al. 2006; Neuwirth et al. 2007, 2009). Although no detailed structural studies have been performed in plants, it is conceivable that Arabidopsis PDX1.1 or PDX1.3 in concert with PDX2 forms similarly composed PLP synthase complexes (Wagner et al. 2006; Leuendorf et al. 2010). While the roles of PDX1.1 and PDX1.3 are well established, it is still widely unknown what the function of PDX1.2 is. In vitro and in planta studies have emphasized that the protein is not required for vitB6 biosynthesis, but rather may function as a negative regulator in the biosynthetic pathway (Wagner et al. 2006; Leuendorf et al. 2008, 2010). This is most likely due to its ability to assemble with all other PDX1 proteins but not with PDX2 (Wagner et al. 2006), and thus may interfere with the assembly of a functional PLP synthase in plants. In plants, PDX1.2 can be found in large complexes that are of a similar size to the PLP synthase (Leuendorf et al. 2010). However, overexpression of PDX1.2 in Arabidopsis only marginally down-regulates vitB6 biosynthesis, indicating that the protein may not be found primarily in complex with PDX1.1 and PDX1.3 in plants. In addition, PDX1.2 gene expression is up-regulated by UV and oxidative stress (Denslow et al. 2007), indicating a function of the protein in stress response. However, to this point, it remains elusive whether PDX1.2 impacts plant development and to what extent it is required for a normal stress response. Here we show that PDX1.2 is required for normal embryo development and that the protein is expressed in a very distinct pattern in embryos. We also observe that in dark grown seedlings PDX1.2 impacts the hypocotyl cell elongation. In addition, the gene is rapidly up-regulated upon heat stress, and mutants affected in PDX1.2 show reduced root elongation after exposure to elevated temperatures. The work shown here assigns new roles to PDX1.2 that are related to developmental processes in plants.

Materials and methods

Planta (2014) 240:137–146

University, Columbus, OH, USA), and were kept under standard growth conditions (long day with 16:8 light:dark; 20 °C). When grown in sterile culture, minimal Arabidopsis growth medium (ATS) was used as described in Estelle and Somerville (1987) without additional supplement of sugar. Wild type (WT) and transgenic plants were separated into non-treated or treated with 10 μM estradiol (Cayman Chemical Company, Ann Arbor, MI, USA). Estradiol was added into ATS medium for plates or sprayed throughout the lifespan of the plant in soil daily unless otherwise indicated. The pdx1.2 SAIL line was backcrossed twice into WT Col-0 before further genetic analyses were performed. DNA constructs generated and used in this work For complementation assays, PDX1.2 was cloned into pCR8 through TA-cloning before recombination into pMDC7 through Gateway LR reaction. Artificial microRNA (amiRNA) constructs were designed through programs available on the (http://wmd3.weigelworld.org/), and generated using specific primer sets as well as the vector pRS300. amiRNA constructs were cloned into the binary vectors pGWB14 for constitutive repression, and into pMDC7 (Curtis and Grossniklaus 2003) for inducible repression of PDX1.2. For PDX1.2 promoter driven GUS expression, a previously described construct was used (Wagner et al. 2006). For primers used see Supplementary Table S2. Expression analysis For Northern blot analysis and quantitative GUS assays, standard procedures were used as we described before (Hellmann et al. 2000). Total Arabidopsis RNA extraction and reverse transcription were done through an Isolate RNA Kit (Bioline USA Inc., Taunton, MA, USA) and a High Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Grand Island, NY, USA) according to the manufacturers’ instructions, respectively. Quantitative RT-PCRs were performed following a standard protocol (95 °C for 7 min, then 40 cycles of 95 °C for 15 s and 60 °C for 1 min) on a 7500 Fast Real-Time PCR System (Applied Biosystems). ACTIN2 gene from Arabidopsis was used as a reference gene in qRTPCR for data modification. 50 ng of cDNA from different tissues was used from each template and for each reaction. Primers used for qRT-PCR are shown in Supplemental Table S2. GFP and autofluorescence were analyzed with a confocal laser scanning microscope from Leica Model TCS SP5 (Leica Microsystems, Buffalo Grove, IL, USA).

Plant material and growth conditions Silique analysis All plants used in this study were A. thaliana L. plants of the ecotype Col-0 (material originally obtained through the Arabidopsis Biological Resource Center, Ohio State

13

Mature siliques were harvested and cleared overnight in a 75 % EtOH: 25 % acetic acid solution. This solution was

Planta (2014) 240:137–146

exchanged with a 25 % EtOH storage solution. Cleared siliques were photographed on a light box with a Nikon D200 camera. Dark grown hypocotyl analysis Sterilized seeds were plated on ATS with or without estradiol (further referred to as Estr), wrapped in aluminum foil and stored at 4 °C overnight, prior to growth under standard conditions. Plates were unwrapped after 9 days and seedlings were photographed with a Nikon D200 camera. For individual cell measurements, a high-resolution Leitz Aristoplan microscope (Leica Microsystems) with a Leica DFC425-C camera was used. Quantitation was done through the Image J analysis program (http://rsb.info.nih. gov/ij/). Heat shock treatments Sterilized seeds were plated on ATS with or without Estr and stored at 4 °C overnight, prior to incubation at 21 °C for 5 days. For root elongation assays, root tips were marked prior to heat shock. Seedlings were incubated at three different conditions: 38 °C for 1 h, 45 °C for 1 h, or 38 °C for 1 h followed by 45 °C for 1 h and returned to 21 °C. Root tips were marked after 24 h, pictures were taken with a Nikon D200 camera, and growth was measured through Image J analysis.

Results Loss of PDX1.2 results in aberrant seed development and aborted embryogenesis In an effort to elucidate the role of PDX1.2, we searched for effective T-DNA insertions, and identified a single SAIL-T-DNA insertion line, SAIL_640_D1, further denoted as pdx1.2-1 that showed an insertion after the ATG. The line was originally wrongly annotated as having an insertion in the promoter region; however, PCR analysis and sequencing confirmed that the insertion is present 248 bp behind the ATG. Because PDX1.2 does not contain any introns, we expected that this insertion results in a knock-out mutant. The corresponding SAIL plants were grown, and 77 plants were analyzed through PCR on genomic DNA for the presence of the T-DNA insertion. We identified 34 WT and 43 heterozygous pdx1.2-1 plants, but strikingly no homozygous pdx1.2-1 plants were found (Table 1). To further investigate this, we took advantage of the BASTA resistance that is mediated by the T-DNA construct. Analysis of the offspring from a self-pollinated heterozygous SAIL parent showed that out of 211 plants, 84

139 Table 1  T-DNA segregation analysis of the filial generation that derived from a self-fertilized, heterozygous pdx1.2-1 parental plant Number of WT (PDX1.2/ Heterozygous for Homozygous for investigated T-DNA insertion T-DNA insertion PDX1.2) (PDX1.2/pdx1.2-1) (pdx1.2-1/pdx1.2-1) plants 34

43

0

77

Genotyping was performed using PDX1.2 gene and Sail T-DNA specific PCR primers

were BASTA sensitive, while 127 were resistant (Table 2). The observed F1 ratio of sensitive versus resistant plants was around 1:1.5, respectively, indicating that in addition to a problem in embryogenesis, there might also be a gametophytic problem present. Additional reciprocal crosses were performed between WT and heterozygous pdx1.2-1 mutants. In such a case, the expected segregation ratios for BASTA are either that all plants are BASTA sensitive if the T-DNA insertion causes a gametophyte abortion, or a 1:1 resistance: sensitivity ratio if the T-DNA insertion permits gametophyte development. As shown in Table 2, F1 seedlings derived from a reciprocal cross and tested for BASTA resistance showed a segregation of around 1:0.7 sensitive versus resistant plants, independently of whether the pollen donor was WT or a pdx1.2-1 mutant (Table 2). Although the ratios deviated from expected ‘gametophytic’ segregation patterns, they were still too low for a sole embryo-based problem, which should have been 1:1. Consequently, a Chi-square test was performed against the hypothesis of a sole embryogenic defect. The segregation analysis through PDX1.2 gene and T-DNA insertion-specific PCR and the BASTA resistance analysis confirmed that the hypothesis must be rejected at a significance level of P = 0.05 (Supplemental Table S1). It was, therefore, concluded that the T-DNA insertion in the PDX1.2 gene does affect embryo development; however, additional effects, such as a gametophytic problem with low penetrance, or a silencing effect of the T-DNA construct resulting in BASTA sensitivity, also need to be considered. We decided to focus on the embryo development, and closer investigation of siliques from heterozygous pdx1.2-1 plants showed a high number of aborted seeds. Microscopic analysis showed that aborted seeds contained embryos that had not gone beyond the globular stage, strongly supporting the notion that PDX1.2 is required for early embryo development (Fig. 1). In an attempt to complement the pdx1.2-1 mutant, an estradiol (Estr) inducible construct that allowed controlled expression of a PDX1.2 cDNA was introduced into heterozygous pdx1.2-1 plants. Several independent transgenic T1 plants were generated that were continuously treated with Estr to allow a high PDX1.2 transgene expression

13

140 Table 2  Segregation frequency of BASTA resistance in crossed and selfed heterozygous pdx1.2-1 plants

Planta (2014) 240:137–146 Crossing ♀

BASTA selection



Sensitive

Selfed hetero pdx1.2-1 I

22

31

53

Selfed hetero pdx1.2-1 II

26

26

52

Selfed hetero pdx1.2-1 III

15

38

54

Selfed hetero pdx1.2-1 IV

21

32

52

Selfed hetero pdx1.2-1 plants (total amount) WT I

84

127

211

32

20

52

30

22

52

Hetero pdx1.2-1 I

Hetero pdx1.2-1 II

WT I

Hetero pdx1.2-1 III

WT II

29

22

51

Hetero pdx1.2-1 IV

WT II

28

23

51

Hetero pdx1.2-1 WT I

119

87

206

hetero pdx1.2-1 I

32

20

52

WT I

hetero pdx1.2-1 II

34

18

52

WT II

hetero pdx1.2-1 III

26

26

52

WT (total amount)

WT II

hetero pdx1.2-1 IV

WT

hetero pdx1.2-1 (total amount)

WT

WT

throughout development, and thus to have a greater chance of successful complementation. In total, 10 T1 plants were identified that carried the Estr-inducible construct, and which were heterozygotic for the PDX1.2 T-DNA insertion. T2 seeds of the next generation were first selected on estradiol and antibiotic-containing ATS medium, brought into soil, and then consistently sprayed with estradiol to avoid any developmental problems from the absence of PDX1.2. Through PCR on genomic DNA, we were able to identify three independent homozygous T2 plants (Supplemental Fig. S1), demonstrating that a complementation was possible. However, we were not able to generate stable lines from any of these three plants in the T3 generation. Reduced PDX1.2 expression through artificial microRNA also causes aberrant seed development and reduced hypocotyl elongation Because we only had a single loss-of function allele for PDX1.2 and could not generate stable complementation lines, further confirmation was needed that the protein is indeed required for normal embryo development. For this purpose, we designed two artificial microRNA (amiRNA) constructs targeting areas in the 5′ and central region of the PDX1.2 gene, respectively (Supplemental Fig. S2). The corresponding constructs were cloned into the binary vector pGWB14, introduced into WT and expressed under the control of a 35S promoter. Several independent lines were generated, and analysis of siliques of the T1 generation showed a significantly high number of aborted seeds (Supplemental Fig. S2). The frequency of aborted seeds

13

Resistant

Number of investigated plants

29

23

52

121

87

208

25

0

25

ranged from 25 % to up to 50 %. Unfortunately, no transgenic plants were identified in the T2 generation on selective culture medium, indicating either loss or silencing of the T-DNA construct. However, the overexpression of the PDX1.2 amiRNA clearly caused aberrant seed development similar to that in the pdx1.2-1 mutant. To gain stable mutants affected in PDX1.2 expression, it was decided to again use the Estr-inducible construct. For this purpose, the central amiRNA construct was cloned into the binary vector pMDC7, and introduced into WT. Several independent transgenic plants were recovered, and further analyzed in the T2 generation. As shown in Fig. 2, treatment of T2 plants with Estr reduced expression of PDX1.2, and resulted in aberrant seed development comparable to pdx1.2-1 mutants. Since this was not observed in control plants, and it was reproducible in independent transgenic lines, it was concluded that PDX1.2 is required for normal seed development. Because PDX1.2 is expressed in various tissues with highest levels in root and flower (Supplemental Fig. S3; Wagner et al. 2006), we also tested amiPDX1.2 plants for any other developmental defects, but could not find changes in root, shoot, leaf or flower development. Since PDX1.2 is expressed specifically in the root elongation zone (Wagner et al. 2006), plants were further tested for aberrant auxin sensitivity using the synthetic auxin derivative 2,4-D, but no changed sensitivities were seen in the mutant when compared to wild type (Supplemental Fig. S4). The only significant changes that were observed occurred in seedlings germinated in the dark. Here, amiPDX1.2 plants developed a shorter hypocotyl in comparison to control plants when

Planta (2014) 240:137–146 Fig. 1  Analysis of seed development in heterozygotic SAIL_640_D1/pdx1.2-1 plants. a Schematic drawing of PDX1.2 with “1” indicating location of the start codon, and “1,379” bp the end of the coding region. The black triangle marks the location of the T-DNA insertion. b 4-day-old siliques from heterozygotic pdx1.2-1 plants contained a high frequency of gaps from collapsed seeds, indicated by arrows. c–h Isolated and cleared seeds analyzed from three- (c, d), four- (e, f), and five- (g, h)-day-old siliques show that embryos in seeds that appear to be normal develop from heart to torpedo stage (c, e, g), respectively, while this was not observed in collapsed seeds (d–f). i, j Close-up of embryos of 3-day old siliques in a normal (i) and mutant (j) seed showed that the embryos in the mutant seeds (pointed out by arrow) did not develop beyond the globular stage. Bars indicate 50 μm. Seeds analyzed were from two independent pdx1.2-1 plants with at least n = 30 seeds per plant analyzed

141

a 1

SAIL_640_D1/pdx1.2-1 (248 bp)

1379 bp

PDX1.2/At3g16050

b

grown in the presence of Estr (Fig. 2e, f), which correlated with reduced cell length (Supplemental Fig. S5), indicating that PDX1.2 is likely involved in cell elongation processes. The current results point out that PDX1.2 has very specific roles that appear to be most prominent in seed and hypocotyl developments, but that the protein may also affect gametophyte development. PDX1.2 displays distinct tissue‑specific expression at early stages in embryo development To gain a better understanding between mutant phenotypes and PDX1.2 expression, a GFP reporter construct was introduced into WT plants in which a GFP gene was C-terminally fused to the PDX1.2 cDNA under the control of the native PDX1.2 promoter (proPDX1.2:PDX1.2:GFP). Confocal microscopy of seeds at various developmental stages showed an interesting expression pattern of PDX1.2:GFP. Fluorescence was mainly detectable in embryos of the globular stage, and here the expression was restricted to the basal part of the embryo (Fig. 3a–d). However, later in development, when the plants entered the

c

d

e

f

g

h

i

j

heart to torpedo stages, this expression was gone, and GFP fluorescence was mainly detectable in the tip of the radicle (Fig. 3e). We did not observe expression of PDX1.2:GFP in any other seed tissue. The presence of PDX1.2:GFP in early globular stages further corroborates a critical role of PDX1.2 in embryo developmental, especially for the transition from globular to heart stage. PDX1.2 is regulated by abiotic stress and participates in heat stress tolerance Earlier work has shown that PDX1.2 expression is strongly, and rapidly, up-regulated by UV-B treatment, and also shows a significant increase under oxidative stress, while no significant change is observable under drought stress (Denslow et al. 2007). To investigate whether further stresses regulate PDX1.2, we tested heat shock treatment. A short pulse of 39 °C triggers a rapid up-regulation of PDX1.2 expression comparable to a heat shock response (Fig.  4). This was also observable in plants expressing a GUS reporter under the control of a 1.5 kbp PDX1.2

13

142 1.2

amiPDX1.2#5

c

16

8 6 4

+ Esstr

0.4

10 + Esstr

*

*

12

- Estr

0.6

14

- Estr

hypocotyl lenggth [mm]

0.8

2 0

0.2

+ amiPDX1.2#7

d

1

f 60

*

50

0.6

0.4

* 0.2

avverage seedd number

0.8 08

mock Estr+

*

40 30 20 10 0

0

viable collapsed

20 18 16 14

*

12 10 8 6 4

+ Estr

amiPDX1.2#7

hypocoty yl length [mm]

1.2

Estr

+ Estr

+

- Estr

-

- Estr

mock Estr+

WT

Expression level of PDX X1.2 (fold ch hange relativve to untreatted control)

18

1

0

b

20

e

1 cm

Expresssion level of PDX1.2 P (folld change rellative to untrreated control)

a

Planta (2014) 240:137–146

2

-

+ WT

+ amiPDX1.2 #5

+ Estr amiPDX1.2 #7

0

Fig. 2  Transient expression of amiPDX1.2 causes seed loss in an inducible manner. amiPDX1.2 #5 (a) and #7 (b) seedlings were treated for 24 h with Estr (10 mM). qRT-PCR analysis shows that both lines have reduced PDX1.2 expression when treated with Estr. Error bars represent standard error. This was not observed in WT plants in response to Estr treatment (data not shown). c amiPDX1.2#7 but not WT siliques develop less and with a higher frequency collapsed seeds after Estr treatment. d Quantification of viable versus

collapsed seeds in Estr-treated and untreated WT, amiPDX1.2 #5, and #7 plants. Plants were sprayed at the onset of flowering. Numbers represent seed average of ten independent siliques. Error bars represent standard deviation. e, f Treatment with Estr causes reduced hypocotyl growth of seedlings grown in the dark. Error bars represent standard deviation (n  = 20). Asterisks indicate in this and all subsequent figures a statistical significant difference (One-way ANOVA, P 

Arabidopsis thaliana PDX1.2 is critical for embryo development and heat shock tolerance.

PDX1.2 is expressed in the basal part of the globular-stage embryo, and plays critical roles in development, hypocotyl elongation, and stress response...
2MB Sizes 2 Downloads 3 Views