Nuclear genome profiling of two species of Epidendrum (Orchidaceae): genome size, repeatome and ploidy

This study provides the first nuclear genome profiling of *Epidendrum anisatum* and *Epidendrum marmoratum* by integrating flow cytometry and k-mer analysis to reveal that both species are diploid with highly repetitive genomes dominated by Ty3-gypsy retrotransposons, where the significant 2.3-fold genome size difference is primarily driven by the expansion of these transposons and a species-specific 172 bp satellite in *E. anisatum*.

The Gist

Imagine the genome of a plant as a massive, chaotic library. Some libraries are small and tidy, while others are sprawling, filled with endless copies of the same book, scattered notes, and empty pages. This study is like a team of detectives trying to figure out the size and contents of two specific orchid libraries—Epidendrum anisatum and Epidendrum marmoratum—without actually reading every single page (which would take forever and cost a fortune).

Here is the story of what they found, told in simple terms:

1. The Mystery of the Two Orchids

The scientists wanted to know: How big are these orchid genomes, and what's inside them?
They picked two cousins:

• The Big One (E. anisatum): A giant library.
• The Small One (E. marmoratum): A much smaller, more compact library.

Surprisingly, the big one is 2.3 times larger than the small one. But why? Are there more "real" books (genes) in the big one? Or is it just full of junk?

2. The Two Detective Tools

To solve this, the team used two different methods to measure the libraries:

• Method A: The Flow Cytometry (The "DNA Scale"):
  Imagine putting the orchid's cell nuclei on a super-sensitive scale that weighs their DNA. This is the "gold standard" lab technique. They used special tissues (like ovaries and pollen) that are less likely to be "confused" by the cell's own internal copying errors.

  • Result: The scale confirmed the big orchid is indeed much heavier (larger genome) than the small one.

• Method B: The K-mer Analysis (The "Word Count"):
  Since they didn't have the whole library assembled yet, they used a computer trick. They chopped the DNA into tiny, fixed-length snippets (like cutting a sentence into 21-letter chunks). They then counted how often each chunk appeared.

  • The Logic: If a chunk appears 100 times, it's a common phrase (repetitive DNA). If it appears once, it's a unique story (genes). By counting these "words," they could mathematically estimate the total size of the library.
  • The Challenge: This method is tricky. If you cut the words too short, you get confused by typos. If you cut them too long, you miss the big picture. The scientists tested many different "cutting sizes" to find the perfect fit.

3. The Big Discovery: It's All About the "Junk"

Once they measured the size, they looked inside to see what made the big library so big.

• No Extra Copies of the Whole Library: They checked if the big orchid was a "polyploid" (meaning it had double or triple the number of chromosomes, like a library that accidentally got two copies of every single book). Result: No. Both orchids have the exact same number of chromosomes (20 pairs). They are both "diploid" (two sets).
• The Real Culprit: Repetitive DNA: The difference wasn't in the useful books (genes); it was in the repetitive DNA.
  • The "Ogre" Transposons: Both libraries were filled with a specific type of jumping gene called "Ty3-gypsy Ogre." Think of these as photocopiers that got stuck in "copy" mode. They keep making copies of themselves and pasting them randomly into the library. This makes up about a third of both genomes.
  • The "AniS1" Satellite: This is the real game-changer. The big orchid (E. anisatum) has a massive amount of a specific 172-letter DNA sequence called AniS1. It's like a giant billboard that takes up 11% of the entire library. The small orchid (E. marmoratum) barely has this billboard at all.

The Verdict: The big orchid isn't "smarter" or more complex; it just has a much bigger "junk drawer" filled with photocopied transposons and a massive satellite billboard.

4. What This Means for Science

This study is a blueprint for how to study plants that we don't know much about yet.

• Don't trust just one tool: The "DNA Scale" (lab) and the "Word Count" (computer) need to check each other's work.
• Tune your computer: If you are analyzing a messy, repetitive genome, you have to adjust your computer settings (like the "k-value") carefully, or you'll get the wrong answer.
• Evolution is weird: These two orchids are so similar in appearance and chromosome count, yet one has a genome more than twice the size of the other. It proves that evolution can change the size of a library without changing the number of books, just by filling the shelves with different amounts of repetitive noise.

In a Nutshell

The scientists found that the difference between a small and a large orchid genome isn't about having more "important" genes. It's like comparing a small, tidy room to a huge room filled with the same furniture but covered in layers of old newspapers and posters. The "Ogre" transposons and the "AniS1" satellite are the newspapers and posters that made the big orchid's genome explode in size.

Technical Summary

1. Problem Statement

The neotropical orchid genus Epidendrum is highly diverse (over 1,800 species) and ecologically significant, yet its nuclear genome characteristics remain poorly understood. Accurate genome profiling (size, ploidy, heterozygosity, and repetitive DNA content) is a critical prerequisite for whole-genome sequencing and assembly, particularly for non-model organisms.

The study addresses two main challenges:

1. Methodological Discrepancies: There is often a lack of agreement between laboratory-based genome size estimation (Flow Cytometry) and bioinformatics-based estimation (k-mer analysis). Factors such as high heterozygosity, repetitive DNA content, sequencing depth, and endoreplication (specifically partial endoreplication common in orchids) can skew k-mer results.
2. Genomic Variation: The evolutionary drivers behind the significant genome size differences observed within Epidendrum species are unknown. It is unclear whether these differences are driven by polyploidy, repetitive DNA expansion, or specific satellite DNA accumulation.

2. Methodology

The researchers employed a multi-faceted approach combining wet-lab cytogenetics and advanced bioinformatics on two closely related endemic Mexican species: Epidendrum anisatum and E. marmoratum.

A. Flow Cytometry (Gold Standard)

• Tissues: Used leaf, ovary, and pollinia tissues. Ovaries and pollinia were prioritized to identify the 1C (gametic) and 2C (somatic) nuclear populations, minimizing errors caused by endoreplication common in leaves.
• Protocol: Nuclei isolation using LB01 buffer, RNAse treatment, and Propidium Iodide staining.
• Standards: Used Human PBMCs and Pisum sativum (Pea) as internal references.
• Analysis: Performed on a CytoFLEX S flow cytometer with FlowJo software.

B. DNA Sequencing and K-mer Analysis

• Sequencing: Illumina HiSeq 2500 (150 bp paired-end) generated ~760M reads for E. anisatum and ~765M reads for E. marmoratum.
• Subsampling: Data was subsampled to simulate coverage depths of 5×, 10×, 20×, 30×, and 40× to determine minimum required coverage.
• K-mer Parameters: Tested a wide range of k-values (15 to 45) and two maximum k-mer coverage limits (10,000 vs. 2,000,000) to assess their impact on histogram resolution.
• Tools: Genome size was estimated using three distinct algorithms:
  • CovEst: Uses truncated Poisson distributions.
  • GenomeScope: Uses non-linear minimum square optimization with negative binomial distributions.
  • findGSE: Uses a mixed model with skewed normal distributions (tested in "hom" and "het" modes).
• Ploidy Estimation: Used nQuire (Gaussian Mixture Model) on mapped reads against low-copy nuclear genes to test diploid, triploid, and tetraploid models. Chromosome counting was also performed on root tips.

C. Repeatome Characterization

• Pipeline: RepeatExplorer2 (Galaxy platform) using both individual and comparative protocols.
• Annotation: Combined de novo graph clustering (TAREAN for satellites) with homology-based annotation (RexDB database).
• Metrics: Calculated the proportion of repetitive DNA, identified specific transposon families (LTRs), and quantified satellite DNA abundance.

3. Key Results

A. Genome Size and Ploidy

• Flow Cytometry: E. anisatum has a significantly larger genome (2.59 Gb) compared to E. marmoratum (1.13 Gb), a 2.3-fold difference.
• Ploidy: Both species are confirmed diploid (2n = 40). No evidence of partial endoreplication (PE) was found, distinguishing them from some other orchid subtribes.
• K-mer Analysis:
  • Coverage Depth: A depth of ≥20× was required to clearly distinguish error, heterozygous, and homozygous peaks.
  • K-value & Tools: Higher k-values (≥17) and a maximum k-mer coverage limit of 2 million yielded the most accurate results.
  • Tool Performance: GenomeScope and findGSE-het (heterozygosity-aware) provided the most consistent estimates with flow cytometry. findGSE-hom significantly overestimated E. anisatum due to its high heterozygosity (1.7% vs. 0.46% in E. marmoratum).

B. Repeatome Composition

• Repetitive Content: Both genomes are highly repetitive (63–73%).
• Dominant Element: Ty3-gypsy retrotransposons (specifically the Ogre family) are the most abundant repeats, occupying ~33% of E. anisatum and ~32% of E. marmoratum.
• Key Driver of Size Difference:
  • Satellite DNA (AniS1): A 172 bp satellite repeat (AniS1) accounts for 11.5% of the E. anisatum genome but is virtually absent (<0.01%) in E. marmoratum.
  • Conclusion: The ~1.46 Gb size difference is primarily driven by the massive expansion of the AniS1 satellite in E. anisatum, followed by variations in Ty3-gypsy abundance.

4. Key Contributions

1. First Genomic Profiling of Epidendrum: Provides the foundational genomic data (size, ploidy, repeat content) for this large and diverse orchid genus.
2. Methodological Optimization for Orchids:
   • Demonstrates that pollinia and ovaries are superior to leaves for flow cytometry in orchids to avoid endoreplication artifacts.
   • Establishes that for highly repetitive and heterozygous orchid genomes, k-mer analysis requires high coverage (≥20×), high k-values (≥17), and high maximum k-mer coverage limits (2M) to avoid underestimation.
   • Validates that findGSE-het and GenomeScope are the most reliable bioinformatic tools for these specific genomic characteristics.
3. Mechanism of Genome Size Evolution: Identifies that satellite DNA expansion (AniS1), rather than polyploidy, is the primary driver of the massive genome size difference between these two closely related species.

5. Significance

• Evolutionary Insight: The study highlights that rapid genome size evolution in orchids can occur through the massive amplification of specific satellite repeats and retrotransposons without changes in ploidy levels.
• Conservation of Resources: By validating bioinformatic profiling methods, the study enables researchers to estimate genome properties using museum specimens or biological collections where fresh tissue for cytometry may be unavailable.
• Future Sequencing: The identification of high heterozygosity and repetitive content (especially Ogre elements) provides critical parameters for designing future de novo genome assembly strategies for Epidendrum, preventing assembly failures due to unaccounted complexity.
• General Applicability: The findings offer a robust framework for genome profiling in other non-model plant lineages with complex, repetitive, and heterozygous genomes.