Kaasaegsed molekulaarsed meetodid ökoloogias

Loeng 5 — sampling and nucleic acid extraction.

Everything in Lectures 2–4 assumed a tube with usable nucleic acid in it. This lecture is about how that tube is filled — and about all the ways the answer is decided before any PCR is run.

Where the technology is now

Using the ClpX unfoldase to ratchet proteins through a CsgG nanopore — multi-pass, single-molecule nanopore reading of long protein strands (Motone et al. 2024)

eDNA in regular monitoring

Putting eDNA into regular monitoring

  • Numerous pioneer projects at national level, especially at regulatory policy level, with pilot proposals in several countries.
  • DNAqua-Net — establishment of the EU CEN Working Group on standardisation of eDNA methodologies, and input to ECO–STAT, the European Commission working group on the Water Framework Directive.
  • The UN Decade of Ocean Science for Sustainable Development (2021–2030) has eDNA initiatives.
  • The HELCOM Baltic Sea Action Plan — eDNA for non-indigenous species.

Databases needed — the example of GEOME

The Genomic Observatories MetaDatabase captures who, what, where, and when for biological samples and their associated sequences.

A sequence without its metadata is not a monitoring datapoint — it is an anecdote.

Different challenges by application and environment

Snow and ice, soil and sand, seawater, freshwater — each substrate changes capture, preservation, and inhibition.

Improvements to eDNA collection and extraction methodologies are crucial for eDNA-based monitoring.

Sampling

Material for extractions: sampling

  • Extraction of biomolecules — DNA, RNA, protein — is the starting point for all downstream analyses.
  • But first the sample must be obtained: from living or conserved tissue, cells, virus particles, faeces, and so on.
  • In certain situations sampling requires killing individuals — destructive sampling.
  • Marker systems that rely on PCR need only small quantities, which can be collected non-destructively:
    • invasive — blood, tissue clips;
    • semi-invasive — tissue swab;
    • non-invasive — shed or dropped material.

“SnotBot” — drone-collected blowhole exhalation (Sahu et al. 2023)

Challenges of non-invasive sampling

One of the great benefits of molecular methods is getting genetic information from very old or very tiny samples — invaluable for species that are hard to locate but leave traces. But:

  • DNA is generally low in quantity, low in quality (highly degraded), and often associated with inhibitory molecules — the same problem as museum specimens.
  • The slightest contamination of a PCR with DNA from another source can generate product from the contaminant instead of the sample.
  • All materials — tubes, tips — must be sterilised, and filter tips with piston plungers used to keep pipette barrels clean.
  • Negative controls are essential, and in the absence of contamination should of course generate no product.

Non-invasive sampling in practice

Some major means of non-intrusive eDNA sampling:

  1. chewed sticks · (B) faeces · (C) eggshells · (D) adhesive tape trap · (E) hair snare · (F) filtering of water · (G) snow imprints · (H) blowhole exhalation · (I) body parts · (J) non-lethal remote dart biopsy · (K) hair stripping discs · (L) oral swabbing.

Passive DNA samplers

Substrates left in the water to adsorb eDNA over time, instead of filtering a fixed volume at one instant — cheap, robust, and integrating over the deployment period rather than sampling a moment.

Digging the past out of sediments — aDNA

Marine sediments and glacio-lacustrine varved clays give an ordered archive: increasing depth is increasing age, and each layer holds its own community.

Automated sampling

RoCSI eDNA sampler

Environmental Sample Processor (ESP)

AUV developed by INESC TEC

eDNA sampling drone

RNA from the sea

The Automatic Fluid Injection Sampler (AFISsys) automates in situ water sampling and fixation, so mRNA profiles can be analysed from material that was stabilised at depth rather than on deck.

Question every step

Study design → in the field → in the laboratory → at the keyboard.

Each column is a set of questions that has to be answered before the previous one is executed — which is why sampling design is a data-analysis problem, not a logistics problem.

Extraction

Extraction essentials

Successful nucleic acid purification generally requires four things:

  • effective disruption of cells or tissue — cell lysis;
  • denaturation of nucleoprotein complexes;
  • inactivation of nucleases — RNase for RNA work, DNase for DNA work;
  • avoiding contamination.

The target should be free of protein, carbohydrate, lipid, and the other nucleic acid — DNA free of RNA, RNA free of DNA. Quality and integrity directly affect every downstream result.

Exonucleases — DNases or RNases that catalyse hydrolytic cleavage of phosphodiester linkages in the backbone

The age of genomic material

  • The eRNA:eDNA ratio decreases significantly through degradation — potentially giving an estimate of the age of the genomic material.
  • eRNA quantification can increase detection, because rRNA is present at high concentration.
  • It may improve interpretation of positive detections through the ratio, and by detecting low-abundance mitosis-associated mRNAs that degrade within ~24 h.

Marshall et al. 2021

Two decay rates, one ratio — and the ratio, unlike either quantity alone, does not depend on how much material was there to begin with.

Why is handling RNA trickier than DNA?

  • RNA is unstable and has a very short half-life once extracted from cells or tissue.
  • Naturally occurring RNA is mostly rRNA (80–90%), with mRNA (2.5–5%) and tRNA. For gene expression work the rRNA must be removed — rRNA depletion.
  • RNA is especially susceptible to degradation because of the ubiquitous RNases present in blood, all tissues, and most environmental bacteria and fungi.
  • Endogenous RNases must be inhibited with strong denaturants for intact RNA isolation.
  • RNases are difficult to inactivate: heat-stable — they can refold after heat denaturation — and they need no cofactors.

Nucleic acid extraction methods

Generally divided into solution-based or column-based protocols. Conventional methods:

  1. Guanidinium thiocyanate–phenol–chloroform extraction — the most commonly used, in many variations; guanidinium thiocyanate denatures proteins, including RNases.
  2. Alkaline extraction — mainly bacterial cultures for plasmid DNA; uses SDS.
  3. CTAB extraction — cetyltrimethylammonium bromide precipitates nucleic acids and acidic polysaccharides from low ionic strength solutions.
  4. EtBr–CsCl gradient centrifugation — complicated, expensive, time-consuming.
  5. Poly(A)⁺ RNA purification by oligo(dT)-cellulose chromatography — for eukaryotic mRNA.

Solid-phase extraction

  • Found in most commercial kits; quick and efficient compared with conventional methods.
  • The solid phase adsorbs nucleic acid depending on the pH and salt content of the buffer.
  • Normally performed with spin columns under centrifugal force.
  • Silica matrices, glass particles, and anion-exchange carriers are common solid supports.
  • Four key steps: adsorption, washing, elution — and the sample-specific lysis that precedes them.

New capture chemistries

CIFF — nucleic acids adsorbed onto glass beads, then pelleted through a fluorinated oil

Magnetic silica capture: incubation → washing → elution → final purification

Chitosan-modified filter paper — DNA binds at pH 5, releases at pH 9

Microvolume extraction

Microvolume methods for microscale amplicon and metagenomic studies: physical vs chemical lysis, from 2 L down to 1 µL of sample — and the community profile changes with the choice.

Extraction with robots

  • Purification of high-quality genomic DNA, pathogen DNA/RNA, cell-free circulating DNA, and cellular RNA.
  • 96 samples in 4 batches, from the smallest forensic samples to 10 ml sample volumes.

ElysION — Oxford Nanopore’s sample-to-answer solution

Hybridization-based methods

DNA hybridization probes

Oligonucleotide probes:

  • short synthetic DNA sequences, typically 15–30 bp, complementary to a specific target;
  • used to detect particular species, strains, or functional genes in environmental DNA;
  • applications include detecting microbial species in soil or water, monitoring biodiversity, and identifying invasive species;
  • fluorescently labelled probes are used in fluorescence in situ hybridization (FISH) to visualise microbes in their natural environment.

DNA probes

Label with a fluorescent dye, denature, hybridise, and view under the microscope — the signal is where the sequence is.

Basics of fluorescence

Fluorescence is the property of some atoms and molecules to absorb light at a particular wavelength and subsequently emit light of longer wavelength after a brief interval. The gap between the two — the Stokes shift — is what makes filters possible.

qPCR: multiplexing

Combining different probes and fluorophores, with different emission filters and detectors — the same principle as the FISH work that follows, read by a machine instead of an eye.

Fluorescence in situ hybridization (FISH)

  • Detects specific DNA/RNA sequences in intact cells.
  • The signal can be amplified for greater sensitivity: catalysed reporter deposition — CARD-FISH.

Fix → hybridize → wash → image.

FISH — spatial organisation of biofilms

Species in polymicrobial biofilms can be organised in three ways:

  • (A) separate microcolonies — neutralism;
  • (B) co-aggregation — cooperation;
  • (C) arranged in layers — competition.

FISH answers a question sequencing cannot: who is next to whom.

Phylogenetic probes

In the Richmond Mine acid mine drainage system, biofilm formation is initiated by Leptospirillum rubarum; L. ferrodiazotrophum appears at later developmental stages.

  • (A) DNA (blue; most small dots are Archaea) and general bacteria (green);
  • (B) general bacteria (green), Leptospirillum group IV (yellow);
  • (C) general bacteria (green), L. ferrodiazotrophum (red);
  • (D) plus Leptospirillum group II (blue);
  • (E), (F) Island 2 samples.

CLASI-FISH

CLASI-FISH — microorganisms are hybridized with two or more fluorochrome-conjugated probes. With 8 fluorochromes, combinatorial labelling gives up to 28 specific combinations.

choose(8, 2)          # pairs of fluorochromes
  [1] 28
choose(8, 1) + choose(8, 2)   # singles plus pairs
  [1] 36

Single-molecule mRNA FISH

Trypanosoma brucei, a single-celled flagellate, imaged with the QuantiGene ViewRNA ISH Cell Assay.

(A) polysome gradients of untreated, starved, and recovering cells; (B) cells co-expressing mChFP-DHH1 and PABP2-eYFP over 3 h starvation and 2 h recovery; (D) northern blots probed for total mRNA and rRNA; (E) the stress granule marker PABP2-mChFP.

Fritz et al. 2015

FRET

Tandem dyes are two covalently bonded fluorochromes — a donor and an acceptor. As the donor becomes excited it transfers energy to the acceptor by a radiationless process: Förster Resonance Energy Transfer.

Because transfer efficiency falls off as the sixth power of distance, FRET reports on proximity, not just presence.

DNA microarray technology

PhyloChip — 500,000 probes, bacterial and archaeal 16S rRNA genes, 8,741 known OTUs per chip, to rapidly assess community type. GeoChip — over 300,000 probes covering ~47,000 gene sequences from 292 gene families, to rapidly assess community function.

Flow cytometry

Flow cytometry

Forward scatter ∝ cell surface area; right-angle (side) scatter ∝ cell complexity; plus one or more fluorescence channels.

Core size and signal shape

Sheath flow squeezes the sample into a core of 10 to 40 µm — the trade-off between throughput and resolution.

Each particle crossing the laser gives a pulse with an area, height, and width — three numbers per channel, from which everything else is derived.

Bacterioplankton vertical profile

Baltic Sea, 14.05.2013 — ×1000 cells per ml

High- and low-nucleic-acid populations separate cleanly in scatter/fluorescence space. HNA cells account for 66–100% of dark CO₂ fixation (Jost et al. 2008).

Gotland Deep

Pump-CTD, 27.10.2015 — the spread of hypoxic and anoxic bottom water in the Baltic, 1906 → 2012

Fluorescence-activated cell sorting (FACS)

Gate a population on the cytogram, and the instrument physically deflects those cells into a tube — turning a statistical population into material you can sequence.

From a gate to an identity

Sorted population P2, then 16S rRNA gene amplicon sequencing and CARD-FISH with a specific probe — two independent lines of evidence that the gate corresponds to Sulfurimonas subgroup GD17.

Single-cell genomics

  • Cell sorting followed by genome amplification;
  • single amplified genomes (SAGs);
  • multiple displacement amplification (MDA) with a DNA stain such as SYBR Green I.

The amplification curve separates 1000 cells, 100 cells, 1 cell, and 0 cells — and the 0-cell control is the one that matters.

Automation: CytoBuoy and µFACS

  • The downsides of regular FC and FACS are high cost and relatively large size.
  • Much effort is going into portable, low-cost, automated techniques such as µFACS — microfluidic fluorescence-activated cell sorting.
  • Useful for environmental testing, and for medicine in low-resource settings.

Check yourself

Why are negative controls non-negotiable in non-invasive and eDNA work?

Because the target is present in trace amounts and any contaminant is not. Without a negative control you cannot distinguish a real low-copy detection from a laboratory artefact — and the artefact will amplify more efficiently.

Why is RNA work harder than DNA work?

RNases are everywhere, heat-stable, and need no cofactors, so they cannot simply be boiled away. Intact RNA requires strong denaturants at the moment of lysis and RNase-free technique throughout.

Why does the eRNA:eDNA ratio estimate the age of material?

RNA degrades much faster than DNA, so the ratio falls predictably with time since release — and because it is a ratio, it does not depend on how much material was shed in the first place.

What does FISH tell you that amplicon sequencing cannot?

Spatial arrangement and per-cell presence: whether taxa form separate microcolonies, co-aggregate, or layer. Sequencing homogenises the sample and loses that structure.

Why sort cells before sequencing?

To convert a mixed community into a defined subpopulation — a single amplified genome, or an amplicon library from one gate — so the sequence can be attributed to a physical, countable set of cells rather than to the bulk.

Thank you for your attention! Questions?

Tähtis

peeter.laas@taltech.ee · peeter.laas@ut.ee