Research Reveals How Cells Protect Against Stress

14 August 2002 -- Stress happens, and over the eons all species of living things have evolved all sorts of ways to cope. Now, new research has revealed that organisms as diverse as humans and plants share a common set of stress-protection maneuvers that are choreographed by the metabolic machinery in their cells.

The research led by Sarah M. Assmann, the Waller Professor of Plant Biology at Penn State, will be published in the 15 August 2002 issue of the journal Nature.

"We have shown, in more detail than was known before, the chain of cellular events that begins with an environmental stress and ends with an organism's protective response to that stress," Assmann says. "We also have discovered some previously unknown steps in that process."

Among the team's discoveries is that one cellular-processing step that originally was discovered in human cells also occurs in plant cells. "A human autoimmune disease and a disorder associated with breast cancer are known to result from a defect in this process, " Assmann says.

Specifically, the Assmann team studied a process triggered in plants by abscisic acid (ABA), a hormone that plants produce when they are stressed by drought. Assmann's lab discovered two years ago that the ABA hormone activates a type of protein called a kinase, which attaches phosphate groups to other proteins. The resulting cascade of events ultimately causes closure of microscopic pores on the plants' leaves in an effort to limit the loss of moisture.

In the present research, Assmann's group found that one of the targets of this ABA-activated kinase is a specific protein that binds RNA. Assmann's group further discovered that the ABA-induced phosphorylation of the RNA-binding protein caused its association with the RNA encoding dehydrin, a protein known to confer stress-resistance to plant cells.

Scientist have long known that, in both plant and animal cells, proteins designed to do particular jobs are produced from the genetic blueprint contained in the DNA inside the nucleus. In a process known as transcription, nuclear machines first copy the genetic code from the DNA molecules into a "transcribed" RNA molecule and then moves the RNA from the nucleus into the cell's cytoplasm, where it is "translated" into a protein. But Assmann and other researchers are discovering that RNA-binding proteins mediate a lot of cut-and-paste processing of the newly transcribed "raw" RNA before it is remodeled into "messenger" RNA and allowed to leave the nucleus carrying the blueprint for making a protein.

"A new paradigm that our research suggests is that the ABA hormone regulates the protein complement of a cell not only by controlling the initial transcription process but also by controlling the proteins involved in post-transcriptional remodeling of RNA molecules, including RNAs that encode stress-protective proteins," Assmann explains.

Another of Assmann's discoveries is that ABA regulates the formation of mysterious islands within the cell's nucleus called "nuclear speckles." Scientists do not yet know a lot about nuclear speckles in plants, but they do know that nuclear speckles in human cells contain proteins associated with the remodeling of RNA.

By expressing in plant cells the RNA-binding protein with a green fluorescent tag attached, Assmann's group was able to observe the localization of this protein within the living cell. As she watched through the microscope Assmann observed, for the first time, that ABA induced the relocation of the RNA-binding protein within the nucleus. Upon treatment of the plant tissue with ABA, the fluorescently-tagged RNA-binding proteins quickly gathered together into nuclear speckles that looked like green-glowing islands inside the cell's nucleus. "To our knowledge, such hormonally induced aggregation of RNA-remodeling proteins into nuclear speckles has not previously been observed either in plant or in animal cells," Assmann says.

In addition to giving researchers these and other important details about the processes that produce protective proteins, Assmann's research also eventually could give farmers more control over the moisture content of their crops." Our research points to a gene-regulation process that, if turned off after a crop matures, would assure that the pores on a plant's leaves would stay open, allowing it to dry more quickly in the field," Assmann explains. "In a crop like feed corn, for example, such control would be economically beneficial to farmers, who get a better price for their crop if it has reached its optimal moisture content."

In addition to Assmann, other members of the research team include Jiaxu Li, lead postdoctoral associate, postdoctoral associates Sona Pandey and Carl K.-Y. Ng, Ken-ichiro-Shimazaki and Toshinori Kinoshita at Kyushu University (Japan), and Steven P. Gygi of Harvard Medical School.

This research was supported by the National Science Foundation.

[B K K ]

Photos:
Figures from the paper published in Nature are available with their accompanying legends. Click on the image to download high-resolution images.

Contact:
Sarah M. Assmann: phone (+1) 814-863-9579, email sma3@psu.edu
Barbara K. Kennedy (PIO): phone (+1) 814-863-4682, email science@psu.edu

	Figure 1. AKIP1 encodes a single-stranded RNA-binding protein. The deduced amino acid sequence of AKIP1 cDNA aligned (Clustal method) with several putative single-stranded RNA-binding proteins. A.t CAC00749 and O.s BAA90354 are from Arabidopsis thaliana and Oryza sativa sequencing projects; H.s. hnRNP A/B and M.m AUF1 are from human and mouse with accession Nos. S17563 and NP_031542, respectively. Identities are shaded in black and gaps are indicated with dashes. The numbers on the right refer to the positions of the amino acids. RNP1 and RNP2 consensus sequences are indicated.
	Figure 2. AKIP1 specifically interacts with AAPK in the yeast two-hybrid system. Positive interactions are indicated by growth on media lacking histidine (left), or adenine (middle) and by the expression of b-galactosidase (right). AAPK-CD and AAPK-CT represent the N terminal region including the entire catalytic domain of AAPK (residues 1-267 of the 349 amino acids) and the C-terminal region including the acidic region of AAPK (residues 263-349), respectively. Yeast SNF1 and SNF4 are interacting positive controls.
	Figure 3. AKIP1 and AAPK exhibit overlapping nuclear localization in guard cells. a) AAPK-GFP localizes in the nucleus and cytoplasm. b) Brightfield image of a stomate showing the nucleus of left guard cell with gold particles evident. c) Fluorescence image corresponding to b) showing diffuse localization of AKIP1-GFP (no ABA treatment). d) Representative guard cell nucleus expressing AKIP1-GFP at 0 and 20 min following incubation in d) control buffer (n = 19) or e) 50 mM ABA (n = 14). f) Number of AKIP1-GFP labeled speckles following treatment with (n = 14) or without (n = 19) 50 mM ABA. At 20 min. +ABA differs from Control at P £ 0.05 (ANOVA). Scale bar = 10 mm.
	Figure 4. Phosphorylation of AKIP1 by ABA-activated AAPK regulates binding of AKIP1 to dehydrin mRNA. a) AAPK phosphorylates recombinant AKIP1 in an ABA-dependent manner. In vitro phosphorylation assay was performed as described in Methods. b) ABA-dependent phosphorylation of AKIP1 occurs in vivo. AKIP1 was immunoprecipitated from 32P-labeled guard cell protoplasts using AKIP1-specific antibodies. ABA was added at 25 mM with 0.25% DMSO as vehicle; the concentration of DMSO alone had no effect. c) ABA treatment does not alter AKIP1 protein abundance. AKIP1 was immunoprecipitated from guard cell protoplasts and detected by immunoblot. ABA treatment was as for b).
	Figure 5. Interaction of AAPK-phosphorylated AKIP1 with dehydrin transcript. a) AAPK-phosphorylated AKIP1 binds dehydrin mRNA. RNA binding assay was performed as described in Methods. A ~300 bp dehydrin PCR product was amplified from RNAs bound to AAPK-phosphorylated AKIP1 (lane 6, arrowhead). Molecular weight markers are shown on the right. b) AAPK-phosphorylated AKIP1 specifically interacts with sense dehydrin RNA. 32P-labeled sense dehydrin RNA was incubated with AKIP1-GST previously treated with inactive AAPK (lane 1), active AAPK (lanes 2), or active CDPK (lane 3). Alternatively, AKIP1 treated with active AAPK was incubated with 32P-labeled sense dehydrin RNA (lane 4), or first incubated with an excess of unlabelled (competitor) sense dehydrin then with 32P-labeled sense dehydrin RNA (lane 5), or with unrelated 32P-labeled RNA (non-competitor) from the RNA synthesis kit (lane 6), or with 32P-labeled antisense dehydrin RNA (lane 7). Binding of 32P-labeled RNA to AKIP1 was visualized by autoradiography.