Research Article |
Corresponding author: Christian Laforsch ( christian.laforsch@uni-bayreuth.de ) Academic editor: Jaimie T.A. Dick
© 2024 Frederic Hüftlein, Jens G. P. Diller, Heike Feldhaar, Christian Laforsch.
This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Citation:
Hüftlein F, Diller JGP, Feldhaar H, Laforsch C (2024) Riparian invader: A secondary metabolite of Impatiens glandulifera impairs the development of the freshwater invertebrate key species Chironomus riparius. NeoBiota 92: 155-171. https://doi.org/10.3897/neobiota.92.119621
|
Invasive species represent a significant threat to native biodiversity. The Himalayan Balsam Impatiens glandulifera is an annual plant, which is invasive in Europe and often inhabits the riparian zone. It produces several secondary metabolites causing, for example, growth inhibition of terrestrial plants and invertebrates. One of these metabolites is the quinone 2-methoxy-1,4-naphthoquinone (2-MNQ). The compound gets washed out from the above-ground parts of the plant during precipitation and may then leach into nearby waterbodies. Despite some evidence for the allelopathic effect of plant secondary metabolites on terrestrial invertebrates, little is known about how 2-MNQ affects the survival or development of aquatic dipteran larvae, despite the importance of this functional group in European freshwaters. Here, we investigated the effects of 2-MNQ on larvae of the river keystone species Chironomus riparius in acute and chronic scenarios. The toxicity of 2-MNQ towards the first and the fourth larval stage was determined in a 48-hour acute exposure assay. We show that 2-MNQ has a negative impact on the development, growth and survival of C. riparius. The LC50 of 2-MNQ was 3.19 mg/l for the first instar and 2.09 mg/l for the fourth instar. A ten-day chronic exposure experiment, where the water was spiked with 2-MNQ, revealed that 2-MNQ had a significantly negative impact on larval body size, head capsule size, body weight, development and survival. These results demonstrate the negative impact of the secondary metabolite 2-MNQ from the terrestrial plant I. glandulifera on a crucial macroinvertebrate inhabiting the adjacent stream ecosystem in riverine ecosystems. This may lead to a decline in population size, resulting in cascading effects on the food web.
Allelopathy, benthic macroinvertebrates, ecotoxicity, invasive species, 2-methoxy-1, 4-naphthoquinone
The riparian zone, the transition zone between terrestrial and freshwater ecosystems, is amongst the most diverse habitats worldwide. The vegetational and structural diversity acts as a refuge for small mammals hiding in shrubs, trees serve as perching and nesting sites for birds and fallen wood debris provides resources for terrestrial as well as aquatic invertebrates (
Invasive plants can impair native species by producing allelopathic metabolites. The Japanese knotweed Fallopia japonica, for example, produces resveratrol, amongst other chemicals, which has been found to have inhibitory effects on seed germination and seedling growth of various plant species, potentially influencing the structure and composition of plant communities in invaded areas (
Another well-known example of an invasive alien species in riparian habitats is the Himalayan Balsam Impatiens glandulifera. It belongs to the family of the Balsaminaceae, reaches a height of up to 2.5 m, can disperse up to 2500 seeds per mature plant in a radius of 10 m and achieves up to 90% cover of invaded plots (
Amongst running waters, rivers belong to the most diverse ecosystems, providing the potential for various ecological niches due to the richness of different and heterogeneous habitat patches (
This paper, therefore, aimed to examine the effects of the allelopathic secondary metabolite 2-MNQ on the growth, development and survival of Chironomus riparius. We performed acute immobilisation tests, as well as low-dose chronic exposure experiments using concentrations that are comparable to those released during rain events in nature (
The starting culture, consisting of 10 egg ropes, was provided by Dr. Philipp Egeler from the ECT Oekotoxikologie GmbH (Flörsheim am Main, Germany). The organisms were then transferred into a self-built breeder (68 cm high × 42 cm wide × 55 cm deep), located in a Rubarth P 850 climate cabinet (Rubarth Apparate GmbH, Laatzen, Germany) with constant conditions of 20 ± 0.1 °C and 12 h light-dark cycle. The breeder consisted of gauze on three of the four sides and an acrylic glass plate on the front side, with two holes for gloves and a smaller hole to fit, for example, conic centrifugal tubes or exchange the medium, so that the cage never had to be opened. Inside the cage, two white bowls were placed, filled with quartz sand (average grain size: 0.16 mm, purchased from Quarzwerke GmbH, Frechen, Germany) and 1.5 litres M4-Medium (
Solid 2-MNQ was purchased from Sigma-Aldrich (Merck KGaA, Darmstadt, Germany), with 98% purity. In order to make it soluble in water, it was solved in 100 µl DMSO (Dimethylsulphoxide 99.7% purity; Bernd Kraft GmbH, Duisburg, Germany) per litre medium. The tests were conducted according to the OECD guidelines (OECD Test No. 235, 2011) for the first and adapted for the fourth instar larvae as those rely on sediment, which is not required in the guideline. The tests were performed in 6-well plates with a volume of 10 ml (Eppendorf AG, Hamburg, Germany). In each well, five first instar larvae were randomly placed. The first instar larvae were exposed to two control treatments (control: pure M4-medium; solvent control: M4-medium with 100 µg/l DMSO) and seven different concentrations of 2-MNQ (2, 3, 4, 5, 6, 7 and 8 mg/l). These values were chosen according to run-off values from
The procedure for the acute immobilisation test with the fourth instar larvae was very similar to that of the first instar, with the difference that 3 g of quartz sand (average grain size: 0.16 mm, purchased from Quarzwerke GmbH, Frechen, Germany) were added to every well. Quartz sand was added to avoid any additional stress for the individuals, as fourth-instar larvae require sediment for building their characteristic living- and feeding tubes (
For the chronic test with C. riparius, 50 second instar larvae, as they are the first sediment-dwelling instar, per replicate (five for every treatment) were randomly placed in a 1 litre Weck- beaker (J. WECK GmbH u. Co., KG, Wehr, Germany) that was filled with 800 ml M4-medium and 120 g quartz sand (average grain size: 0.16 mm, purchased from Quarzwerke GmbH, Frechen, Germany). The control, the solvent control for DMSO and three different concentrations of 2-MNQ (1, 2 and 3 mg/l) were each replicated five times. The concentrations were chosen according to the results of the acute immobilisation test (LC50 for the first instar: 3.19 mg/l). The 25 beakers were randomly placed in a climate chamber with constant conditions of 20 ± 0.1 °C and 16 h:8 h light:dark cycle. All beakers were gently aerated through a pump-hose system, with two pumps aerating the beakers through an air distributor (3 × 12-way distributor, 6 mm diameter each, OSAGA Deutschland, Glandorf, Germany). The larvae were fed with 0.5 mg Tetramin fish food per larva per day. The test lasted ten days until the control individuals had reached the fourth instar. Subsequently, the larvae were fixed in 80% ethanol and photographed under a dissecting microscope (Leica M50, Wetzlar, Germany; light: Leica KL 300 LED, Wetzlar, Germany) equipped with a digital image analysis system (camera: OLYMPUS DP26, Hamburg, Germany; cellSens Dimension v.1.11, OLYMPUS, Hamburg, Germany). The mortality in every replicate was recorded at the end of the experiment and the mean of the five replicates was calculated for the whole treatment. One beaker in the 1 mg/l treatment cracked in the middle of the test and became leaky as a result, which is why it was excluded from the analysis.
The body length of surviving preserved larvae was measured with a digital image analysis system using a polygonal line from the posterior end of the head capsule (HC) to the last visible appendage. After the whole larvae were photographed and measured, they were decapitated for further analysis. The width of the HC was measured from the left margin to the right margin at the widest points of the head. Abnormal head capsules were defined as such when the HC was constricted in combination with heavy pigmentation due to difficulties in the moulting process and recorded (yes/no) (Suppl. material
To measure the dry weight, decapitated larvae and the respective heads were placed into disposable weighing pans (41 × 41 × 8 mm, neoLab Migge GmbH, Heidelberg, Germany) and put into a desiccator for three days, to allow the ethanol to evaporate entirely. After three days, the larvae and the pans were weighed on a semi-micro scale in mg to the nearest second decimal (OHAUS Explorer EX225D/AD, OHAUS Europe GmbH, Nänikon, Switzerland, ± 0.06 mg linearity deviation). Subsequently, the larvae were removed from the pan and the latter was measured without the larvae to determine the dry weight of the total number of larvae per replicate. For comparing the mean dry weight per larva, the total dry weight was divided by the number of larvae that survived until the end of the experimental period.
The distribution of the larval stages in the treatments was determined following the method of
The data were analysed using the statistic programme R Version 4.0.4 (R Core Team 2020). The LC50-value, the plots and the dose-response curves for the acute immobilisation tests for L1 and L4 larvae were calculated with the built-in R package “drc” (
After 48 hours of exposing the first instar larvae, there was no observable mortality in both the control and solvent control medium and the treatment exposed to 2 mg/l 2-MNQ. The animals in the treatment exposed to 3 mg/l 2-MNQ showed 44% mortality and the animals in the 4 mg/l treatment showed already 80% mortality. Mortality reached 100% in the 5 mg/l treatment (Fig.
Dose-response curves with the fitted regression curve for the effect of 2-MNQ on the mortality of A first instar and B fourth instar larvae of C. riparius and the calculated LC50 with standard error for both instars.
The 48-hour acute immobilisation test for the fourth instar larvae revealed a calculated LC50 of 2.09 mg/l (Fig.
The body length of the individuals was significantly different between the treatments (one-way ANOVA: X2 = 862.23; df = 4, p < 0.001). The body length of the individuals treated with 2 mg/l 2-MNQ (mean ± SE 8.33 ± 0.05 mm; n = 5) and 3 mg/l 2-MNQ (mean ± SE 7.05 ± 0.38 mm; n = 5) was significantly smaller than the control (mean ± SE 14.04 ± 0.22 mm; n = 5), the solvent control (mean ± SE 14.17 ± 0.18 mm; n = 5) and the individuals exposed to 1 mg/l 2-MNQ (mean ± SE 13.47 ± 0.21 mm; n = 4) (p < 0.001 for all comparisons). The individuals of the 2 mg/l treatment had a significantly larger body length than those of the 3 mg/l treatment (p < 0.001). There was no significant difference between the control and the solvent control (p = 0.996), the control and the 1 mg/l treatment (p = 0.46) and the solvent control and 1 mg/l 2-MNQ (p = 0.26) (Fig.
Body length (A) and head capsule width (B) of larvae from C. riparius exposed to different concentrations of 2-MNQ (mean +/- SE; ANOVA; p < 0.05). Letters indicate significance between treatments. Framed values represent the mean of each group. Only significant differences between treatments and control are indicated.
The width of the head capsules (HCs) was significantly different between treatments (one-way ANOVA: X2 = 30.562; df = 4, p < 0.001). The HC-width of the individuals treated with 2mg/l 2-MNQ (mean ± SE 424.03 ± 28.60 µm) was significantly smaller than the control (mean ± SE 547.01 ± 3.46 µm) (p = 0.012), the solvent control (mean ± SE 542.55 ± 2.23 µm) (p = 0.02) and the 1 mg/l (mean ± SE 533.88 ± 3.35 µm) treatment (p = 0.03). The HC-width of the individuals treated with 3 mg/l 2-MNQ (mean ± SE 349.45 ± 33.20 µm) was significantly smaller than the HC of the individuals of all other treatments (p < 0.01), except from the individuals of the 2 mg/l treatment (p = 0.15). The HC of the control individuals was significantly larger than the HCs of the 1 mg/l treatment (p = 0.05). There was no significant difference between the control and the solvent control (p = 0.54) and the solvent control and 1 mg/l 2-MNQ (p = 0.71) (Fig.
Individuals exposed to 2 and 3 mg/l 2-MNQ showed significantly more abnormalities in form of conspicuous constrictions of the head capsule compared to the control (one-way ANOVA of Bayesian binomial regression: X2 = 37.711; df = 4, p < 0.001) (Fig.
There was a significant difference between the treatments for the mean dry weight per larva (one-way ANOVA: X 2 = 238.6; df = 4; p < 0.001). The animals exposed to 3 mg/l 2-MNQ (mean ± SE 0.17 ± 0.02 mg) showed a significantly lower mean dry weight per larva than the animals of the control treatment (mean ± SE 0.86 ± 0.07 mg) (p < 0.001), the individuals from solvent control (mean ± SE 0.84 ± 0.05 mg) (p < 0.001) and the individuals exposed to 1 mg/l 2-MNQ (mean ± SE 0.67 ± 0.03 mg) (p < 0.001). The animals treated with 2 mg/l 2-MNQ (mean ± SE 0.21 ± 0.01 mg) showed no difference in the dry weight per larva (p = 0.94), compared to the animals exposed to 3 mg/l 2-MNQ. The individuals exposed to 2 mg/l 2-MNQ had a significantly lower dry weight per larva than the controls, the solvent controls and the animals exposed to 1 mg/l 2-MNQ (C: p < 0.001; DMSO: p < 0.001; 1 mg/l: p < 0.001). The animals of the control treatment, the animals from the solvent control and those exposed to 1 mg/l 2-MNQ did not differ significantly in dry weight per larva (Fig.
The distribution of the larval instars differed significantly between the treatments (X2 (8, N = 960) = 421.91, p < 0.001). The larval instars’ distribution showed that 100% of the control individuals reached the fourth instar at the end of the test. In the solvent control, 97.6% of the individuals reached the fourth instar, while.1.6% only reached the third instar and 0.8% did not moult and stayed in the second instar. In the 1 mg/l treatment, 4% of the individuals reached the third instar at the end of the test and 96% reached the fourth instar. In the 2 mg/l treatment, 47.4% of the individuals reached the fourth instar, while 50.5% reached instar three and 2.1% stayed in the second instar. In the 3 mg/l treatment, 36% of the individuals reached the fourth instar, 56% reached the third instar and 8% did not moult at all (Fig.
Distribution of larval instars from C. riparius exposed to different concentrations of 2-MNQ. Letters indicate significance between treatments.
The distribution of larval instars differed significantly between the individuals exposed to the control treatment and all other groups (1 mg/l: p = 0.005; all other comparisons: p < 0.001), except with the solvent control (p = 0.08).
The mortality of C. riparius in the 10-day chronic exposure test showed a significant difference between the treatments (one-way ANOVA: X2 = 285.66; df = 4; p < 0.001). The animals exposed to 3 mg/l 2-MNQ (mean ± SE 32.6% ± 2.42) showed significantly higher mortality than the animals of the control (mean ± SE 1% ± 0.45) (p < 0.001), the solvent control (mean ± SE 1.2% ± 0.49) (p < 0.001) and the ones exposed to 1 mg/l 2-MNQ (mean ± SE 1% ± 0.41) and2 mg/l 2-MNQ (mean ± SE 11.6% ± 2.58) (p < 0.001). In addition, the animals exposed to 2 mg/l 2-MNQ expressed significantly elevated mortality compared to the control, the DMSO treatment and 1 mg/l 2-MNQ (p < 0.001 for all comparisons). The other treatments showed no significant difference in mortality (Fig.
Mortality in percent of the C. riparius larvae exposed to different concentrations of 2-MNQ (mean +/- SE; ANOVA; p < 0.05). Letters indicate significance between treatments. Framed values represent the mean of each group. Only significant differences between treatments and control were indicated.
Our results show that 2-MNQ has the potential to impair the survival and development of C. riparius after acute 48 hour and chronic 10-day exposure. We determined the LC50 after 48 h for the first instar larvae of C. riparius at a 2-MNQ concentration of 3.16 mg/l and 2.09 mg/l for the fourth instar larvae. Larvae of C. riparius exposed to a concentration of 2 and 3 mg/l 2-MNQ in the 10-day chronic exposure experiment had significantly increased mortality, reduced body- and head capsule size, as well as reduced body weight. They were further delayed in their development and showed a significantly higher proportion of individuals with deformed and abnormal head capsules.
The doses applied in the acute (max. 8 mg/l) and chronic (max. 3 mg/l) toxicity tests were below the concentration reported to be leached from one single plant after rain events (12.21 mg/l) (
It has already been shown that low concentrations of 1.5 mg/l 2-MNQ can significantly impair the growth and survival of individuals of the freshwater key species Daphnia magna (
The requirement of sediment of fourth instar larvae could be a reason for the higher toxicity of 2-MNQ, compared to the first instar. Naphthalene, for example, a structurally related compound to 2-MNQ, is known to be easily oxidised and interact with a SiO2/air interface (
Even though some chironomid species are known for their extreme tolerance towards environmental conditions like pH, temperature, oxygen content and even salinity, they are susceptible to anthropogenically induced pollution, drugs and other endocrine-disrupting substances (
For the assessment of the impact of 2-MNQ on riverine ecosystems, it might be essential to investigate the potentially different sensitivity of various macroinvertebrates, as C. riparius is known to display a comparatively greater tolerance towards deteriorating water quality (
This study reveals substantial acute and chronic toxicity of 2-MNQ towards the larvae of C. riparius. Individuals exposed to concentrations of 2 mg/l upwards showed a significantly reduced body size and head capsule size, a significantly reduced dry weight per larvae, developmental abnormalities and increased mortality compared to unexposed individuals. I. glandulifera is spreading extensively around the world, building monocultures across riverine ecotones and even invading forest ecosystems. The exposure risk to 2-MNQ could be highly increased when larger areas are covered by the plants at high densities along riverbanks. This can result in higher amounts of 2-MNQ leaching into aquatic ecosystems after precipitation, ultimately increasing its concentration within the waterbody. Our findings underscore the critical need for monitoring this neophyte, emphasising the imperative to focus on controlling its spread. This attention is vital to safeguard ecosystem functions of flowing waters.
Future research should include how riverine communities adapt to and are influenced by allelopathic substances, addressing also species interactions and resilience of these ecosystems.
The authors have declared that no competing interests exist.
No ethical statement was reported.
Jens Diller was funded by the German Environmental Foundation (DBU) (AZ 20017/509).
Conceptualization: FH, CL. Data curation: JGPD, FH. Formal analysis: FH. Funding acquisition: CL. Investigation: JGPD, FH. Methodology: FH. Project administration: CL. Resources: CL. Supervision: CL. Visualization: FH. Writing – original draft: FH. Writing – review and editing: JGPD, FH, HF, CL.
Frederic Hüftlein https://orcid.org/0000-0001-6267-165X
Jens G. P. Diller https://orcid.org/0000-0002-3018-2226
Heike Feldhaar https://orcid.org/0000-0001-6797-5126
Christian Laforsch https://orcid.org/0000-0002-5889-4647
All of the data that support the findings of this study are available in the main text or Supplementary Information.
Supporting information with figures and the R-Script
Data type: docx