Herbarium genomics reveal signatures of colonization history, lineage turnover, and adaptation during invasion

By analyzing low-coverage whole genome sequencing from 513 herbarium specimens spanning 1838–2018, this study reconstructs the complex invasion history of North American white clover, revealing that its success was driven by multiple introductions, lineage turnover, and admixture that increased genetic diversity and facilitated adaptation.

The Gist

Imagine you are trying to solve a mystery about how a specific plant, White Clover, took over North America. Usually, scientists only look at the plants alive today to figure out their history. But that's like trying to understand a whole movie by only watching the final scene. You miss the plot twists, the character introductions, and the dramatic changes that happened along the way.

This study is different. The researchers acted like time-traveling detectives. Instead of just looking at modern clover, they dug into herbariums (plant museums) and pulled out 513 pressed, dried clover specimens collected between 1838 and 2018. By reading the DNA in these "fossilized" leaves, they built a slow-motion movie of the invasion, revealing three major plot points: Who brought the clover? Did the different groups mix? And how did the clover adapt to its new home?

Here is the story of the White Clover invasion, broken down simply:

1. The "Immigration Wave" Followed Human History

Think of North America as a giant party where different groups of people arrived at different times.

• The Early Arrivers (British & Belgian): The first clover to show up in the 1800s mostly came from Britain and Belgium. These plants set up shop in the Northern and Mid-Atlantic parts of the US (like New England and the Northeast). They were the "originals" in those regions.
• The Late Arrivers (French & Spanish): Later, as trade routes shifted, clover from France and Spain started arriving. These plants found a home in the Southern US.
• The Connection: The study found that the clover didn't just appear randomly; it followed the exact same paths as human colonizers. If the British settled an area, British clover was there. If the Spanish or French settled an area, their clover followed. It's like the plants were hitching rides on the same ships and wagons as the humans.

2. The "Mixing Bowl" vs. The "Takeover"

Once these different groups of clover arrived, two very different things happened depending on where they were:

• In the Mid-Atlantic (The Mixing Bowl): Here, the British clover and the newer French/Spanish clover met. Instead of fighting, they mixed. Imagine two different flavors of ice cream being stirred together. This "mixing" (called admixture) created a super-diverse population. The genetic variety went up, making the plants stronger and more adaptable.
• In the South (The Takeover): In the deep South, something dramatic happened. The original British clover didn't just mix; it was replaced. The newer Spanish/French clover was better suited for the hot, humid Southern climate. Over time, the British clover faded away, and the Spanish/French type took over completely. This is called lineage turnover. It's like a new band moving into a neighborhood and the old band moving out because the new one fits the vibe better.

3. The "Survival Suit" (Adaptation)

White clover has a cool superpower: it can produce cyanide (a poison) if an insect tries to eat it. However, making cyanide is expensive for the plant, so it only does it if it needs to.

• The Climate Rule: In the hot South, insects are active year-round, so the clover needs its poison suit. In the cold North, insects are dormant in winter, so the clover can save energy and drop the poison.
• The Evolution: The study showed that over the last 150 years, the clover in the South gradually learned to turn on its poison genes more often, while the North kept them off. This wasn't a sudden switch; it was a slow, steady adjustment. The plants were slowly "tailoring their suits" to fit the local weather, becoming better at surviving in their specific neighborhoods.

Why Does This Matter?

This paper is a game-changer because it proves that history matters.

• It's not just one story: The invasion wasn't a single event; it was a century-long saga of different waves arriving, mixing, and sometimes replacing each other.
• Herbariums are time machines: By using old dried plants, scientists can see the "failed" lineages (the British clover in the South) that are now gone. If they only looked at today's plants, they would think the South was always Spanish, missing the whole story of the British takeover and replacement.
• Adaptation takes time: Evolution isn't always a lightning strike. Sometimes, it's a slow, gradual shift over generations, which we can now see clearly thanks to these old specimens.

In a nutshell: White clover didn't just invade North America; it followed the footsteps of European colonizers, mixed its genes in some places, replaced its neighbors in others, and slowly learned to dress for the local weather. It's a perfect example of how nature and human history are deeply intertwined.

Technical Summary

1. Problem Statement

Invasion biology faces a critical challenge: disentangling the relative roles of colonization history, admixture (interbreeding between distinct lineages), and natural selection in driving invasion success.

• Limitations of Contemporary Data: Genomic studies relying solely on contemporary samples only capture the "endpoint" of an invasion. They often fail to reconstruct detailed timelines of colonization, obscure the history of failed lineages, and make it difficult to detect recent admixture events or the precise timing of adaptive shifts.
• Knowledge Gap: While Trifolium repens (white clover) is a model species for studying rapid adaptation (specifically regarding cyanogenesis clines), the spatiotemporal dynamics of its introduction to North America, the sources of these introductions, and the tempo of subsequent evolutionary processes remain unclear.

2. Methodology

The authors employed a herbarium genomics approach, leveraging low-coverage whole-genome sequencing (lcWGS) to create a genomic time series spanning nearly two centuries.

• Sample Collection:

  • Historical: 513 preserved T. repens specimens collected between 1838 and 2018 from 24 U.S. herbaria (covering 39 states).
  • Contemporary: 693 samples from 18 native-range populations across Europe (Britain, France, Spain, Belgium, etc.) and additional contemporary U.S. samples.
  • Filtering: Specimens collected after 1998 were excluded from downstream analyses to prioritize older, rarer material, resulting in a final dataset of 470 herbarium samples.

• Laboratory & Bioinformatics:

  • DNA Extraction: Modified CTAB protocol with increased CTAB and $\beta$-mercaptoethanol concentrations to handle degraded herbarium DNA.
  • Sequencing: Illumina HiSeq X and NovaSeq X Plus platforms (paired 150-bp reads).
  • Processing: Reads were trimmed to remove damage at termini (first/last 5 bp), mapped to a haplotype-resolved reference genome, and filtered for contaminants (fungal/bacterial) using a custom NCBI database.
  • Variant Calling: Genotype likelihoods were calculated using ANGSD (avoiding hard genotype calls to account for low coverage). The dataset was filtered for four-fold degenerate sites, resulting in 172,871 SNPs.

• Analytical Framework:

  • Colonization History: Principal Component Analysis (PCA), Random Forest classification, Discriminant Analysis of Principal Components (DAPC), and NGSadmix (model-based ancestry estimation) were used to infer source populations and track ancestry shifts over time.
  • Admixture & Turnover: Ancestry proportions were analyzed across 40-year cohorts and latitudinal zones (North, Mid-Atlantic, South). Metrics included Max Q (maximum membership coefficient to an ancestral cluster) to detect admixture (low Max Q) and lineage turnover (flipping dominance between clusters).
  • Diversity & Selection: Individual heterozygosity was calculated to track genetic bottlenecks. Read abundance at specific loci (Ac/ac and Li/li for cyanogenesis) and five putatively adaptive haploblocks were quantified to assess temporal shifts in allele frequencies.

3. Key Contributions

• Temporal Resolution: Provided the first high-resolution genomic timeline of a plant invasion, moving beyond static snapshots to dynamic historical reconstruction.
• Decoupling Processes: Successfully disentangled the effects of multiple introductions, admixture, and lineage turnover, which are often confounded in contemporary-only studies.
• Anthropogenic Link: Directly correlated genetic ancestry patterns with historical European colonial settlement routes (British, French, Spanish, Belgian).

4. Key Results

A. Colonization History Parallels European Settlement

• Multiple Introductions: The invasion was not a single event but a prolonged process involving distinct introductions from Britain, Belgium, France, and Spain.
• Temporal Sequence:
  • Early (1838–1877): Dominated by British and Belgian ancestry, primarily in the North and Mid-Atlantic.
  • Later (Post-1900): Increasing contributions from French and Spanish lineages, particularly in the Mid-Atlantic and South.
• Latitudinal Cline: A clear genetic gradient emerged over time. The South (<35°N) shifted from British dominance to **Spanish/French dominance** after 1950, while the North (>40°N) remained genetically stable and dominated by British lineages.

B. Admixture and Lineage Turnover

• Mid-Atlantic Region: Exhibited significant admixture. The "Max Q" metric declined over the 20th century, indicating ongoing mixing between British and Spanish/French lineages.
• Southern Region: Exhibited lineage turnover rather than admixture. British ancestry coefficients dropped from ~1.0 to ~0.09, while Spanish/French ancestry rose to ~0.91. This suggests the replacement of early British colonists by later Spanish-introduced lineages, likely driven by environmental filtering (adaptation to hot, high-evapotranspiration climates).
• Northern Region: Remained relatively stable with little evidence of turnover or significant admixture.

C. Genetic Diversity Dynamics

• Diversity Increase: Contrary to the expectation of a severe, permanent bottleneck, heterozygosity increased significantly over time in the South and Mid-Atlantic.
• Driver: This increase was strongly correlated with admixture (negative correlation between heterozygosity and Max Q), demonstrating that secondary contact between distinct introduced lineages replenished genetic variation.

D. Adaptive Evolution (Tempo and Mode)

• Cyanogenesis Loci (Ac/ac and Li/li):
  • Showed protracted, continuous shifts in allele frequencies over time rather than rapid fixation immediately after introduction.
  • Selection was strongest in the South, where read abundance for the cyanogenic alleles increased steeply over time, aligning with contemporary latitudinal clines.
• Haploblocks: Five large adaptive haploblocks identified in previous studies showed weak or non-significant temporal shifts in this dataset. This suggests selection on these regions is either weaker, more spatially variable, or that the study lacked sufficient power to detect modest frequency changes in low-frequency alleles.

5. Significance

• Methodological Validation: Demonstrates the power of herbarium genomics to reconstruct complex invasion histories that are invisible to contemporary sampling alone. It reveals "ghost" lineages (those that went extinct or were replaced) and the precise timing of admixture events.
• Evolutionary Insight:
  • Admixture as a Catalyst: Confirms that secondary contact between independently introduced lineages can restore genetic diversity and potentially facilitate adaptation.
  • Lineage Turnover: Provides empirical evidence that invasive populations can undergo rapid lineage replacement (turnover) driven by environmental mismatch, a process rarely documented in real-time.
  • Tempo of Adaptation: Suggests that adaptation to new climates (e.g., cyanogenesis clines) can be a protracted process occurring over decades rather than an instantaneous event.
• Historical Context: The study underscores that biological invasions are deeply intertwined with human geopolitical history, with species distributions mirroring centuries-old colonial trade and settlement patterns.

In conclusion, the invasion of Trifolium repens in North America was a dynamic, multi-century process characterized by sequential introductions, regional lineage turnover, and admixture-fueled diversity increases, all of which facilitated the species' rapid adaptation to diverse North American climates.