The tube a spore fires, and the nucleus forced through it

A spring inside a four-micrometre spore

Microsporidia are single-celled, obligate intracellular parasites. They build no mycelium and no fruiting body; the whole stage that carries an infection from one host to the next is a spore a few micrometres long. Where they belong on the tree of life is still argued over. A phylogenomic study published in 2012, drawing on thousands of gene trees and on 53 concatenated genes, places them as a sister group to all other sequenced fungi — while stating plainly that its own sampling at the base of the tree was too thin to test the rival proposal linking microsporidia to the chytrid Rozella. Whether they sit inside the kingdom or immediately alongside it therefore remains an open question rather than a settled one.

On 18 September 2020, PLOS Pathogens published a three-dimensional reconstruction of their invasion apparatus by Pattana Jaroenlak and colleagues at the Skirball Institute of Biomolecular Medicine, New York University School of Medicine. Spores of Anncaliia algerae were sliced serially at 50 nanometre intervals and imaged by serial block-face scanning electron microscopy. The coiled section of the polar tube proved to have seven turns on average, where earlier work had suggested eight to eleven, and in every spore examined it formed a right-handed helix set at an angle to the spore's front-to-back axis. For all its visual dominance, the tube occupies only about 3.5 per cent of the spore's volume. An A. algerae spore measures 3.9 ± 0.4 μm along that axis and encloses 8.8 ± 1.4 μm³, and its outer wall layer is a uniform 0.16 ± 0.03 μm thick. An umbrella-shaped anchoring disc sits at the front, beneath the thinnest part of the wall, and that, on the authors' reading, is where firing begins.

Timing the discharge

Germination was triggered chemically. A. algerae spores were held at 30 °C in glycine-NaOH buffer at pH 9.0 with 100 mM potassium chloride; the two Encephalitozoon species were held at 37 °C in a buffer containing 5 per cent hydrogen peroxide at pH 9.5. The firing was recorded on an sCMOS camera running between 14 and 50 frames per second, twenty events per species.

The films separate into three phases: the tube extends, then holds static at full length, then the cargo emerges at the far end. Maximum velocity — a peak, not an average for the whole movement — was 235 ± 61 μm per second in A. algerae, 336 ± 142 in E. hellem and 290 ± 38 in E. intestinalis. Maximum accelerations were 1,503 ± 1,082, 5,219 ± 2,521 and 5,045 ± 2,242 μm per second squared respectively. Time to 90 per cent extension separates the species more sharply still: 160 ± 20 ms for E. intestinalis, 290 ± 200 ms for E. hellem, 830 ± 170 ms for A. algerae. In the static phase the tube stands at a mean maximum length of 101 μm in A. algerae, 53 μm in E. hellem and 36 μm in E. intestinalis. The whole sequence, from the start of firing to cargo ejection, closes in under 500 ms in both Encephalitozoon species but takes roughly 1.6 seconds in A. algerae, which is why a blanket phrase such as "a fraction of a second" does not fit all three.

A nucleus in a hundred-nanometre channel

A. algerae carries two nuclei, each about 0.7 μm across. The bore of the tube is approximately 100 nm — a figure the authors take from earlier negative-stain imaging rather than from their own measurements. They describe the mismatch carefully, as at least sevenfold. Live-cell imaging shows the nuclei genuinely making that passage, deforming severely on the way: their aspect ratio is 0.11 ± 0.04 inside the tube and 0.87 ± 0.07 once expelled, so they recover something close to a sphere afterwards. They move at up to 270 ± 115 μm per second. It is not a clean shot, though. In all seven films recorded, the nuclei paused inside the tube during the crossing.

What the recordings do not establish

They are not the first measurement of firing speed. The authors state that their value for A. algerae is approximately double the figure previously reported for the species by Frixione and co-workers in 1992, and attribute the difference to spore provenance and to better temporal resolution. What is genuinely new is the three-dimensional architecture of the tube inside an intact spore, together with the complete germination sequence captured for three species at once.

Nobody filmed a host cell being pierced here. Germination was triggered in buffer, and the paper's own abstract says only that the tube is thought to penetrate the host cell; its introduction keeps both options open, allowing that the extended tube may penetrate the target membrane or may simply latch onto it and anchor the spore there. The word "harpoon" comes from that same abstract and is the authors' metaphor, not a description of the structure. The tube is hollow, and because cargo only begins to move at roughly half extension, the authors read their own data as consistent with a model in which the tube everts, turning itself inside out as it emerges. A barbed head shot from a gun does not behave like that.

The figures also cover three human-infecting species — species that matter chiefly in immunocompromised patients — with twenty events each, and they represent peaks rather than sustained rates. Incomplete germination occurred in all three: sometimes the tube stopped short, sometimes it extended with no cargo at the tip. The familiar line that the tube is about twenty times the length of the spore is a rounding from the abstract; the paper's own measurements give roughly twenty-six times for A. algerae and nineteen for E. hellem. And this is a study of mechanics, not of clinical risk — it says nothing about mushrooms collected in a wood.

Worth looking at yourself

The article is open access under CC BY, and the germination films are attached as supporting information, so the numbers in figures four and five can be checked frame by frame against the footage they came from. The story has moved on since: in 2024, cryo-electron tomography of the Vairimorpha necatrix tube described cargo-filled tubes with ribosomes in spiral arrays along the inside, and emptied ones that were narrower and wrapped in a markedly thicker protein layer.

This catalogue already carries an entry on the Pilobolus spore gun and its acceleration of around 180,000 g. It is tempting to set the two numbers side by side as a league table, and it would be wrong: one is a spore thrown through air by turgor pressure, the other a tube extending through liquid, and the two were measured by entirely different methods. The comparison the authors themselves draw is the more useful one. Motor proteins haul organelles about at roughly one micrometre per second, chytrid zoospores swim at about 104 μm per second, and the polar tube exceeds 300 μm per second. All of it happens at a scale no eye will resolve without a microscope.