Millipedes were collected in a forest belt near Rostov-on-Don, southern Russia (47.320584 N, 39.813120 E) on 12.IV.2026, using the sift method. Three species of julids common in the Rostov Region were collected: Nopoiulus kochii, Brachyiulus jawlowskii and Byzantorhopalum rossicum (Evsyukov and Golovatch 2013).

After collection, live diplopods were transferred to the laboratory. Three adult females of each species were used for the experiment.

We used two types of plastic to obtain particles low-density polyethylene (LDPE) and polylactide (PLA).

MPs were obtained by physically grinding larger fragments. Larger fragments were crushed using a metal grater, after which the resulting particles were washed through a 5 mm sieve. The particles collected from the sieve were then washed through a 2.9 mm sieve. Deionised water was used to rinse the sieves. The resulting particle mass was collected from the surface of a second screen and air-dried. This resulted in particles with a size of 3–5 mm, which, according to previous studies, are more readily consumed by diplopods (Evsyukov and Al-Kinani 2025).

At the end of the experiment, we needed to preserve the living cells on the cryosections of diplopods for further analysis. Therefore, we were unable to stain the microplastics on the slices with Nile Red (Solarbio, China), as this process requires dehydrating the samples. Instead, we opted to pre-stain the microplastic particles in advance. Additionally, many plastics, including those used in our study (LDPE and PLA), possess autofluorescence under UV light, making them easier to detect, even in the absence of Nile Red molecules absorbed on their surfaces.

After drying, the resulting particles were stained with Nile Red dissolved in n-hexane (10 mg/l). This dye stains various types of microplastics (Ivanova et al. 2020). The particles were then placed in glass Petri dishes and covered with the dye solution. They were kept in a fume hood until the solvent had completely evaporated.

We used a modified experimental design as described in Helmberger et al. (2022). In particular, we did not use a plaster bottom in the experimental arenas and extended the exposure time.

The experimental arenas were 100 ml glass jars filled with moistened sand. The sand had been wrapped with aluminium foil and heated in a muffle furnace at 500°C for 24 hours to burn away any contaminating MPs (Helmberger et al. 2022).

Prior to the experiment, the diplopods were placed in experimental arenas without microplastics and food for 2 days at a temperature of 23–25°C under natural light to cleanse their intestines of microplastic particles that they could have absorbed in habitat.

Microplastic particles of LDPE and PLA, stained with Nile Red, were sprinkled on the surface of the sand, along with oat flakes as a food source. Published data suggests that the food source does not interfere with the ingestion of microplastic particles by soil saprophages and actually stimulates it (Yang et al. 2021, Helmberger and Grieshop 2022). Diplopods, one individual per jar, were placed in the prepared jars and left there for 5 days at a temperature of 23–25°C under natural light. After this period, the diplopods were euthanised using ethyl acetate vapour, which does not dissolve microplastic particles, unlike ether or chloroform.

One specimen of each species was placed in 0.9% sodium chloride solution. The intestines of the other specimens were removed and also placed in the solution. After that, both the untreated individuals and the isolated intestines were processed on a cryostat.

Cryosections were prepared from whole millipede bodies as well as from the entire isolated intestine dissected on a cryostat (Leica CM1850, Germany). Prior to embedding, specimens were briefly blotted to remove excess moisture in order to minimise ice crystal formation during freezing. The material was then embedded in Tissue-Tek optimal cutting temperature compound (Sakura Finetek, USA) and the blocks were equilibrated to a stable temperature before sectioning the dense chitin-containing tissue. The cryochamber and specimen holder were maintained at −20°C. After trimming the blocks and setting the working thickness, sections of 5–7 µm were cut and collected on cold adhesive glass slides.

Sections were stained with the fluorescent nucleic acid dye Sytox Green (1:200, Invitrogen, USA), treated with an antifade mounting medium (Abberior, Germany) and coverslippe. We used an Olympus BX53 fluorescence microscope (Olympus Corporation, Tokyo, Japan), equipped with a high-resolution digital camera (EXCCD01400KPA, Hangzhou ToupTek Photonics Co., Ltd., Hangzhou, China) and fluorescence cube sets including: U-FUW (excitation 340–390 nm, dichroic 410 nm, barrier filter 420 nm), U-FBNA (excitation 470–495 nm, dichroic 505 nm, barrier filter 510–550 nm) and U-FRFPFL (excitation 535–555 nm, dichroic 565 nm, barrier filter 570–625 nm). All three fluorescence channels were used to compare the secondary emission of coloured microplastics and diplopod tissue sections. For subsequent analyses, only the UV excitation channel (U-FUW) was used.

The microscope we used had three different fluorescence modes. At the first stage of our study, we visually compared photographs taken using different modes to examine the stained sections of the diplopods. For comparison, we used photographs of MP particles stained with Nile Red before the experiment, as well as slices stained with Sytox Green.

In the obtained images, we compared the fluorescence colours of the MP and cell nuclei. For reliable visual discrimination, the fluorescence colours should be clearly distinguishable. As shown in Fig. 1, both blue and green excitation produced the same fluorescence colour for the MP and cell nuclei, which were green and orange, respectively. When excited with UV light, the MP fluorescence was blue or blue-white and the nuclei were bright green, making it possible to clearly differentiate both the intestinal wall and MP particles.

Therefore, we decided to use UV fluorescence for our next experiment.

We studied sections of whole body diplopods. That is, the diplopods were completely subjected to cryotomy, without dissection of the gut. From a microtomic perspective, this experiment was unsuccessful, as the cuticle fractured and crumbled during the slicing process. The cuticle exhibits autofluorescence under UV light, a property that is not due to chitin itself, but to endogenous fluorophores such as pterins and porphyrins (Marek 2017), making it visually similar to MPs.

At the same time, the glow of the cuticular fragments resembles that of MPs and is visually difficult to differentiate from them. Given that the cuticle breaks down during cutting, its fragments may fall on to the entire surface of the slice, including the intestinal cavity and may be mistaken for MPs (Fig. 2).

Studies of the prepared digestive tracts of diplopods have yielded promising results for imaging microplastics. The foregut and hindgut are lined with a cuticular intima, which is similar to the external cuticle and also exhibits autofluorescence under UV light. However, this cuticular lining is thinner and easier to visually distinguish from microplastic particles. Visual observations of microplastics in the digestive tract have shown that they are mainly located near the gut walls in the foregut and hindgut (Fig. 3) and in the lumen of the midgut (Fig. 4). This may be due to a peritrophic membrane that surrounds food lumps, separating them from the gut wall (Fontanetti et al. 2015).

The autofluorescence of the millipede cuticle under UV light is a well‑known challenge when working with arthropods (Marek 2017). The bright UV‑induced glow observed in millipedes is primarily due to endogenous fluorescent molecules, such as pterins and coproporphyrins, which are embedded within the cuticle. The millipede cuticle is a complex, multilayered structure (comprising epicuticle, exocuticle and endocuticle) built of a chitin‑protein matrix and, in most Diplopoda, is also mineralised (Makarov 2015). Thus, the autofluorescence signal originates from the entire cuticular complex.

In whole‑body sections, not only does the exoskeleton crumble during cutting, but its fragments may also fall into the intestinal lumen, leading to false‑positive identification of MPs. We therefore recommend that future studies on MP ingestion in arthropods with heavily sclerotised exoskeletons should either isolate the digestive tract before sectioning or use alternative approaches, such as chitinase treatment or differential staining of chitin (Helmberger et al. 2020, Tapia-Bahena et al. 2026). Our protocol – prestaining MPs with Nile Red, isolating the gut, cryosectioning and counterstaining nuclei with Sytox Green – proved effective and preserves the spatial context of MP within the foregut, midgut and hindgut, while avoiding contamination by cuticular fragments from the exoskeleton.

We observed a clear difference in MP localisation between the foregut, hindgut and midgut.

The foregut and hindgut, which are of ectodermal origin, are lined internally by a cuticular intima. This cuticular lining is thinner than the exoskeleton, but shares a similar complex composition (a chitin-protein matrix) and exhibits autofluorescence under UV light. In these parts of the intestine, MP particles are usually found close to the intestinal wall. This may be due to the rough or structured surface of the cuticle, which can physically trap particles.

In contrast, in the midgut – which lacks a cuticular lining, but secretes a peritrophic membrane that envelops the food bolus (Fontanetti et al. 2015) – MP particles remained mostly in the lumen, separated from the epithelium. To our knowledge, this is the first report of differential MP distribution amongst gut regions in diplopods.

The functional significance of this pattern is twofold. First, the peritrophic membrane in the midgut acts as a barrier that prevents direct contact of MPs with the midgut epithelium (Lehane 1997). Second, the gut microbiome of arthropods is compartmentalised: different microbial communities reside in the foregut, midgut and hindgut (Engel and Moran 2013). As MP accumulation sites differ, the potential for MP degradation by gut bacteria or fungi may also differ. Siddiqui et al. (Siddiqui et al. 2024) suggested that the hind-gut of saprophagous arthropods harbours a particularly diverse microbiota capable of breaking down recalcitrant polymers. In addition, densest populations of microbes inhabit cuticular surface of the hind-gut (Nardi et al. 2016). Our finding that MPs are located near the hind-gut wall supports the hypothesis that this region could be a hotspot for plastic biodegradation.

Several limitations should be acknowledged. Our observations are qualitative; we did not quantify the number of MP particles per unit area or the proportion of ingested MPs retained in each gut section. Only one exposure time (5 days) and one MP size class (3–5 mm) were tested. Smaller MPs (e.g. 1–100 µm) may behave differently, possibly penetrating the peritrophic matrix or even entering the epithelial cells. The sample size was relatively small (n = 3 individuals per species) and we used three julid species without comparing interspecific differences. Despite the visual distinction between autofluorescent cuticle and MPs under UV light, we cannot completely exclude the possibility that some very small cuticular fragments from the foregut or hindgut cuticular intima were misinterpreted as MPs – a more rigorous approach would involve spectral imaging or Raman spectroscopy (Al-Kinani and Evsyukov 2025).

Diplopods ability to ingest and fragment MPs (Evsyukov and Al-Kinani 2025) implies that they can serve as vectors for MP transport from the soil surface into deeper soil layers and into the food web (e.g. to predatory centipedes, birds or small mammals). Knowing that MPs accumulate in different gut regions may help predict their residence time and the likelihood of excretion versus retention.

Based on our results, we propose the following research lines: (1) quantitative assessment of MP distribution using automated image analysis; (2) long‑term exposure experiments to track MP fragmentation and translocation; (3) testing of smaller MP sizes (down to 1 µm) and other polymer types; (4) coupling MP visualisation with fluorescence in situ hybridisation (FISH) to identify gut bacteria attached to plastic surfaces; (5) ex vivo assays using isolated gut segments to measure MP degradation activity.