Rapid phosphatidic acid accumulation in response to low temperature stress in Arabidopsis is generated through diacylglycerol kinase

A plant sits on a lab bench — five days old, roots in a small tube, doing nothing remarkable. Then someone moves it to ice water. Within two minutes, before a single cold-response gene has switched on, a lipid molecule inside its cells has already spiked. That two-minute window is where this story begins. The molecule is phosphatidic acid, or PtdOH for short — a lipid second messenger that normally accounts for roughly two percent of total phospholipids in plant cells. Arisz and colleagues set out to track it using radioactive phosphorus, which is phosphorus thirty-two, allowing them to follow newly made molecules in real-time. Seedlings were incubated overnight with thirty-two phosphorus-orthophosphate, then plunged into zero degrees Celsius. What the team observed was a fast, large rise in labeled PtdOH — peaking within two minutes in seedlings and within five minutes in leaf discs cut from mature plants. In seedlings, the signal stayed elevated for at least two hours. In leaf discs, it climbed sharply and then drifted back toward baseline over the same period. The timing is what makes this finding striking. Cold-responsive gene transcription — the CBF and DREB1 transcription factors that plants use to gear up for freezing — begins rising around fifteen to thirty minutes after cold exposure. The downstream protective proteins, encoded by COR genes, accumulate only after about two hours. Full cold acclimation takes six to seven days. PtdOH appears in two minutes. That means no new proteins are involved. Whatever is making this molecule was already there, waiting. So the question becomes: which pre-existing enzyme is responsible? The paper lays out three candidate pathways that could, in theory, generate PtdOH so quickly. The first is de novo biosynthesis — PtdOH forms naturally as cells build new membrane phospholipids by sequentially adding fatty acid chains to a glycerol backbone. The second is phospholipase D, or PLD, which can directly cleave structural phospholipids like phosphatidylcholine to release PtdOH. The third involves a two-step sequence: phospholipase C, or PLC, hydrolyzes a phosphoinositide called PtdInsP, releasing a molecule called diacylglycerol, or DAG; then diacylglycerol kinase, or DGK, grabs that DAG and phosphorylates it — sticks a phosphate group on it — to produce PtdOH. Three routes, one product, and they leave different forensic traces. The team used two assays to differentiate the pathways. The first is a transphosphatidylation assay, which exploits a quirk of PLD: in the presence of a primary alcohol like n-butanol, PLD transfers its phosphatidyl group onto the alcohol instead of water, producing a diagnostic molecule called phosphatidylbutanol rather than PtdOH. So if PLD is active, you will see phosphatidylbutanol accumulate. Seedlings were pre-labeled overnight, butanol was added, and then the cold treatment was applied for five minutes. The cold produced a clear spike in PtdOH. Phosphatidylbutanol did not change. PLD was ruled out. The second assay exploited labeling kinetics. If PtdOH is made by DGK phosphorylating DAG using ATP, then the phosphate group on the new PtdOH comes directly and quickly from the cellular ATP pool — so labeled PtdOH appears fast, even when cells have only been exposed to radioactive phosphate for a short time. If PtdOH came instead from PLD breaking down structural phospholipids like phosphatidylethanolamine or phosphatidylglycerol, those structural lipids would first need to become labeled, which takes many hours. The team labeled seedlings for just twenty, sixty, or one hundred and eighty minutes — far shorter than the overnight labeling used to equilibrate the structural pools — and then applied the cold shock. Even after only twenty minutes of labeling, when structural phospholipids were barely radioactive, cold-induced labeled PtdOH was already robust. That pattern matches DGK, not PLD, and rules out de novo synthesis as well. Supporting this, a simultaneous and closely correlated decrease in labeled PtdInsP occurred at the same time as the PtdOH increase — matching in timing, temperature dependence, and magnitude. That fingerprint suggests PLC consuming PtdInsP to generate DAG, which DGK then converts to PtdOH. The picture that emerged was a two-enzyme cascade: PLC first, DGK second, and cold-induced PtdOH as the output. With the biochemical route identified, the natural next move was genetics. Arabidopsis has seven DGK genes, and the team obtained T-DNA insertion lines — plants with individual genes disrupted — for all seven and tested whether any single knockout killed the cold response. The answer was a flat no. In wild-type seedlings, labeled PtdOH rose from about one percent to three percent of total phospholipids after five minutes at zero degrees Celsius — roughly a threefold increase. The single-gene knockouts either matched that response or, intriguingly, exceeded it. The dgk five one knockout showed a four point four-fold increase; the dgk four two line also reached four point four-fold. In leaf disc assays, dgk seven two knockout plants produced a four point five-fold increase compared to two point nine-fold in wild type. Some knockouts appeared to overshoot, which the authors interpret as compensatory upregulation by the remaining DGK family members. That pattern of compensation is called functional redundancy. It is a common feature of plant signaling systems but still a genuine puzzle for anyone trying to identify which enzyme matters most. Arisz and colleagues are clear that single-gene knockouts are not sufficient here. Double or triple mutants — dgk one and dgk two combinations, for instance — will be needed to tease apart contributions. The seven DGK genes, combined with twelve PLDs and nine PLCs encoded in the Arabidopsis genome, give the plant a deeply layered and overlapping signaling network. To understand where in the cold-signaling hierarchy this lipid response sits, the team also tested a panel of established cold-stress mutants — plants with defects in known regulators of the cold response. The hos one, los one, and fry one mutants all showed normal PtdOH responses. The transcription factor mutants ice one and snow one, along with a Super-ICE1 overexpression line, showed slight differences, but these were not cold-specific and were likely pleiotropic effects — meaning they reflected general disruption rather than specific cold-response wiring. The lipid signal is not being switched on by the gene network; it is part of what feeds into it. Arisz and colleagues assemble these findings into a model of cold sensing at the membrane. Cold temperature changes the physical properties of membrane lipids — that much was already known. The new piece is that this physical change appears to activate a pre-existing PLC enzyme, which hydrolyzes PtdInsP to release DAG, and DGK is then right there to phosphorylate that DAG into PtdOH. No new proteins, no gene expression required — just existing enzymes responding to a membrane that has suddenly become colder and stiffer. The PtdOH produced acts as a second messenger, a molecular signal, though what it does next — which proteins it binds, which downstream responses it triggers — was not resolved in this work. The model is a proposal, not a proof. The genetic redundancy problem means the field still lacks a clean mutant that eliminates the response. But the biochemical case is coherent and specific: two independent assays ruling out PLD and de novo synthesis, converging on DGK, with the PtdInsP drawdown providing upstream confirmation. And the speed it explains is the point. Two minutes is not time for a gene. It is time for a protein. Understanding exactly how membrane physical state couples to PLC and then DGK activity, and identifying where PtdOH sends its signal from there, remains the work ahead — work that matters not just for basic cell biology, but for understanding, and eventually improving, how plants survive cold. This lecture was created by ennepō. Go to https://ennepo.ai to Discover, Create and Follow the latest research in your field. Read when you can. Listen when you want to.