# Cover Letter — Paper IV: Metabolic Divergence Across the Arecaceae

**Journal:** *Plant Physiology*
**Manuscript Type:** Research Article (Regular)
**Title:** Gene Copy Number Does Not Predict Oil Yield: Metabolic Divergence Across the Arecaceae Reveals a Hardware–Software Model of Carbon Partitioning

---

Dear Editor,

We submit "Gene Copy Number Does Not Predict Oil Yield: Metabolic Divergence Across the Arecaceae" for consideration in *Plant Physiology*. This work completes the fourth and final installment of a four-study series on palm metabolic evolution (companion Studies I and II, submitted to *New Phytologist* and *Mol. Biol. Evol.*, respectively; Study III, in revision for *Trends Ecol. Evol.*).

**What this paper shows — in one paragraph**

Twelve oil-bearing species spanning six angiosperm orders were surveyed for copy-number variation and expression in 16 core lipid and carbon-metabolism enzyme families. The central finding is that DGAT1 (PF03982) — the rate-limiting gatekeeper of triacylglycerol biosynthesis — is strictly capped at 1–3 copies across all six angiosperm orders surveyed, even after whole-genome duplications (WGDs), defying the long-standing dogma in oil crops that gene dosage scales with metabolic output. Instead, palms evolved a distinct genomic strategy: ancestral amplification of upstream flux enzymes (PlsC-type acyltransferase, biotin-lipid enzyme) shared by all four Arecaceae surveyed, while DGAT1 remains transcriptionally quiescent in oil-storing tissues. We call this the **hardware–software model**: the ancestral lipid toolkit (hardware) is conserved across the family, but lineage-specific transcriptional and regulatory programs (software) partition fixed carbon into divergent sinks — oil, sugar, or starch–defense. The DGAT1 bottleneck is the boundary condition on all palm oil yield, with implications for metabolic engineering.

**Why we believe this fits *Plant Physiology***

The manuscript addresses a question at the intersection of comparative genomics, metabolic evolution, and plant biochemistry: *Why is the final committed step of oil biosynthesis never amplified, even in the world's most productive oil crops?* The 12-species cross-order design, the audited Pfam-domain enzyme classification, and the independent proteomic corroboration (PRIDE PXD036949) together provide a level of data transparency that we believe meets *Plant Physiology*'s evidentiary standards. The 4-figure scheme — copy-number landscape → palm-specific expansion and expression → copy-expression decoupling → dual-sink carbon-partitioning model — sits within PP's 4–5 main-figure sweet spot for metabolic/evolutionary articles.

**Data transparency and reproducibility**

All copy-number values (192 cells in the Fig. 1 heatmap, plus DGAT1 across 12 species) were verified cell-by-cell against `copy_number_v2.json`, a version-controlled, ground-truth file generated from a uniform HMMER3.4 pipeline (E ≤ 1e-10, 16 Pfam models, deduplicated). The B-plan audit (2026-07-17) independently confirmed all Pfam-domain identities against real HMM NAMEs, eliminating annotation errors inherited from legacy enzyme dictionaries. Both stranded RNA-seq datasets are publicly archived: coconut DRR129244 (vdb-validated, 25.67 M paired reads) and oil palm SRR1019970 (2.63 M reads). The independent coconut endosperm proteome from a distinct tall-type cultivar (PRIDE PXD036949) corroborates the protein-level absence of DGAT1 across different coconut germplasms. HMMER parameters, tree-building details, and proteome PSM data are provided as Tables S1–S3.

**On formal dN/dS tests**

We deprioritized formal codeml-based selection tests in this manuscript. The non-monophyly of most families (only 1 of 13 shows strict 4-genus palm monophyly) suggests recurrent duplication and loss rather than stable directional selection — a pattern better suited to the regulatory/expression framework developed here. Targeted codeml on the DGAT1–CHS–Oleosin axis is currently planned as a follow-up within the companion *New Phytologist* study, where the regulatory basis of flavonoid canalization provides a natural home for selection analyses.

**What the four companion studies together establish**

- **Study I (*New Phytologist*):** PEPC1 exaptation blocks C4 photosynthesis in palms — the upstream carbon-entry lock.
- **Study II (*Molecular Biology and Evolution*):** PEPC gene-family evolution enforces a "Two-Lock" model retaining C3 across Arecaceae — the regulatory lock.
- **Study III (*Trends in Ecology & Evolution*):** Palm C3 retention spans 105–120 Myr — the evolutionary timescale.
- **Study IV (this manuscript, *Plant Physiology*):** The same palm family locked out of C4 photosynthesis retains an ancestrally expanded lipid toolkit whose terminal step — DGAT1 — is never amplified. Shared genetic hardware coupled with divergent regulatory software dictates carbon sink allocation, collectively closing the loop on carbon partitioning from photosynthetic entry to storage in Arecaceae.

**Suggested reviewers**

We suggest experts in plant lipid metabolism, comparative genomics of oil crops, and metabolic evolution. Specific names withheld to avoid conflicts of interest; available upon request.

**Cover letter declarations**

- This manuscript has not been published or submitted elsewhere.
- All authors have approved the manuscript and agree with its submission to *Plant Physiology*.
- The authors declare no competing interests.
- Data availability: RNA-seq (DRA/SRA: DRR129244, SRR1019970), proteome (PRIDE: PXD036949), copy-number matrix (Table 1, ground-truthed against copy_number_v2.json), HMMER models (16 Pfam domains, E ≤ 1e-10).

We thank you and the reviewers in advance for your time and expertise.

Respectfully,
Sun Chengxu (on behalf of all authors)
Coconut Research Institute, Chinese Academy of Tropical Agricultural Sciences (CRI-CATAS)
hnsuncx@qq.com