The soil is one of the world’s most important resources, playing a crucial role in supporting the lives of humans, animals, plants and other living organisms (Bardgett and van der Putten, 2014; Bahram et al., 2018; Geisen et al., 2019; Guerra et al., 2020). However, due to rapid industry development and intense agriculture, soils in many ecosystems have been polluted, severely threating the health and sustainability of ecosystems (Nouri et al., 2009; He et al., 2014; Bandowe et al., 2019; Wang et al., 2019; Titaley et al., 2020; Alengebawy et al., 2021; Ali et al., 2022). As one of the most common pollutants in the soil, heavy metals have attracted global attentions (Chibuike and Obiora, 2014; Cristaldi et al., 2017; Shahid et al., 2017; Kumar et al., 2019; Eijsackers et al., 2020; Alengebawy et al., 2021). For example, cadmium can be taken up by plants and its accumulation in plant organs can inhabit plant growth due to toxicity (Raskin et al., 1994; Podar et al., 2004; Shahid et al., 2017; Wang et al., 2019). Soils are also frequently influenced by organic pollutants introduced by, e.g., oil exploitation and utilization of pesticides (Reid et al., 2000; Cristaldi et al., 2017; Sharma et al., 2019; Tudi et al., 2021). In particular, pesticides may persist in the soil for a long time and thus may be harmful to plants as they can alter the ecological process such as photosynthesis, enzyme activity, root growth and leaf formation (Boutin et al., 2014; Parween et al., 2016; Yadav and Devi, 2017; Aveek et al., 2019; Siddiqui et al., 2022). More often, soils in nature are co-contaminated by heavy metals and organic pollutants, which may enhance the negative effect of single pollutant on plant growth due to additive or synergic effects (Singh et al., 2017; Khudur et al., 2018; Zhao et al., 2018; Alengebawy et al., 2021; Ali et al., 2022). Therefore, it is crucial to test the response of plants to soil co-contaminated by heavy metals and organic pollutants to guide phytoremediation.

Biodiversity experiments have generally revealed a positive relationship between plant species richness and community productivity (e.g., Loreau and Hector, 2001; Tilman et al., 2001; van Ruijven and Berendse, 2003). These positive relationships are attributed to complementarity effect, i.e., more diverse communities can utilize resources more complementarily via niche differentiation and species facilitation, and/or selection effect, i.e., more diverse communities have a higher chance to include productive species (Loreau and Hector, 2001; Polley et al., 2003; van Ruijven and Berendse, 2005; Godoy et al., 2020). Therefore, increasing the number of species may potentially help the recovery of plant communities in contaminated soils. Similarly, increasing genotypic richness of a plant population may also promote the complementary utilization of resources and increase probability of the presence of particularly productive genotypes, as different genotypes of the same species also vary in their abilities to acquire resources (Fridley et al., 2007; Lankau and Strauss, 2007; Hughes et al., 2008; Baron et al., 2015). Thus, increasing genotypic richness of a plant population may increase its productivity (Crutsinger et al., 2006; Lankau and Strauss, 2007; Fridley and Grime, 2010; Kotowska et al., 2010; Schöb et al., 2015; Begum et al., 2022) and consequently the ability to remediate the contaminated soil. However, how genotypic richness influences the population productivity in soil co-contaminated by heavy metals and organic pollutants was largely unknown.

Hydrocotyle vulgaris is an introduced species in the Araliaceae family, and occurs in many natural and semi-natural wetlands and grasslands in China (Dong et al., 2015). Different genotypes of H. vulgaris can co-exist in the field, but often only one genotype is dominant (Wang et al., 2020). Previous studies have shown that H. vulgaris exhibits rapid asexual reproduction ability, high degree of phenotypic plasticity and strong tolerance ability (Dong, 1995; Dong et al., 2015; Liu et al., 2021; Xue et al., 2022). It has been used in the water purification for the removal of heavy metals and organic pollutants (Morand et al., 2011; Vafaei et al., 2013; Ni et al., 2018; Li et al., 2020). Therefore, to explore the population growth of H. vulgaris regulated by genotypic richness in contaminated soils may have significant implications for the application of this species in remediation of these contaminated soils. However, whether H. vulgaris can thrive in the soil co-contaminated by heavy metals and organic pollutants and thus may potentially be used for the remediation of contaminated soils is largely unknown.

To answer these questions, we conducted a greenhouse experiment by planting H. vulgaris populations consisted of one, four or eight genotypes in soil contaminated by cadmium, cypermethrin or both. Specifically, we tested the following hypothesis: (1) plant populations grown in the soil co-contaminated by cadmium and cypermethrin produces lower biomass than that grown in the soil contaminated by either cadmium or cypermethrin, due to additive or synergic effects of the two pollutants (Singh et al., 2017); (2) plant populations with more genotypes produces more biomass than those with less genotypes, due to higher utilization efficiency of soil nutrients; (3) the negative effect of co-contamination is weaker in populations of greater genotypic richness compared to those with lower genotypic richness.

In this study, we show that co-contamination by cadmium and cypermethrin reduced growth of H. vulgaris population, compared to contamination by cadmium only. However, increasing the number of genotypes of the population did not alter this effect. Despite that, co-contamination by cadmium and cypermethrin altered soil nutrients depending on the number of genotypes. These results indicate that genotypic richness may play little role in helping H. vulgaris in the soil co-contaminated by cadmium and cypermethrin.

We observed a lower plant growth in the soil co-contaminated by cadmium and cypermethrin than in the soil contaminated by cadmium, indicating that co-contamination may suppress the growth of H. vulgaris ( Figure 1 ). This result supported our first hypothesis partly, as we did not observe a growth difference in the co-contaminated soil and the soil contaminated by cypermethrin only. Therefore, the presence of cypermethrin may be the main driver of the negative effect of co-contamination in our study. As a pesticide mainly used for controlling insects, cypermethrin may also have directly negative influences on plant growth (Chauhan et al., 1999; Parween et al., 2016; Aveek et al., 2019). In this study, we observed that H. vulgaris population had a smaller leaf area (i.e., lower light capture ability) and a shorter, thinner internode (i.e., a poor escaping ability from harsh environment) in the two soils with cypermethrin. Besides, cypermethrin may also influence plant growth indirectly via changing soil abiotic and biotic characteristics (Gundi et al., 2007; Zhuang et al., 2011; Jia et al., 2019; Zhu et al., 2022). In our study, soil nitrogen was generally lower in the two soils with cypermethrin ( Figure 3A ), which may be associated with the poor plant growth in the two soils. Soil biota is also sensitive to cadmium and cypermethrin (Zhuang et al., 2011; Tejada et al., 2015; Liu et al., 2020; Zhu et al., 2022). The accumualtion of cadmium and cypermethrin in the soil can reduce the activity and diveristy of soil microbes, consequently reduce the performance of plants (Al-Ani et al., 2019; Jia et al., 2019; Cheng et al., 2022). However, we were unable to disentangle these effects in our study, as we did not measure soil biotic properties in this study.

We hypothesized a positive effect of genotypic richness on the growth of H. vulgaris population, as increasing the number of genotypes may promote the complementary utilization of resources and increase the occurrence probability of productive genotypes (Crutsinger et al., 2006; Hughes et al., 2008; Kotowska et al., 2010; Begum et al., 2022). However, we found that genotypic richness did not influence plant growth in this study and a similar result has also been reported in several other studies (Fridley and Grime, 2010; Prieto et al., 2015; Schöb et al., 2015). Our soil analysis revealed that increasing genotypic richness reduced total nitrogen in the soil contaminated by cadmium ( Figure 3A ), indicating that plant populations with greater genotypic richness have greater efficiency of nitrogen utilization than that with lower genotypic richness. However, the effect of genotypic richness on soil nitrogen did not alter the growth of H. vulgaris, despite that plant growth was significantly positively correlated to soil nitrogen ( Table 4 ). Therefore, genotypic richness may help the removal of nitrogen but cannot promote the growth of H. vulgaris in the contaminated soil.

As the effect of co-contamination may have overwhelmed the effect of genotypic richness, it is not surprising that increasing the number of genotypes failed to reduce the negative effect of co-contamination on H. vulgaris populations. Hence, our third hypothesis was not supported. Soil contamination may have acted as a “filter” that drives the convergence of H. vulgaris populations (Laliberté et al., 2014; Kraft et al., 2015). Our analysis confirmed it by showing that plant populations with different genotypic richness had consistent morphology in the contaminated soils ( Figure 2 ).

In conclusion, co-contamination by cadmium and cypermethrin can suppress the growth of H. vulgaris population compared to contamination only by cadmium, but increased genotypic richness cannot help to reduce this negative effect. Although we detected a neutral genotypic richness effect, it does not mean that increasing genotypic diversity is not beneficial for the remediation of contaminated soils. It should be noted that the relative abundance of genotypes may be more important than the number of genotypes in driving the positive diversity effects at local scales, which we failed to investigate in this study (Huang et al., 2022). Therefore, we recommend a more diverse plant population/community by considering both the number of genotypes/species, and their relative abundance in the remediation of contaminate soils, as highly diverse populations/communities are typically more stable than less diverse ones in facing the rapid global changes such as soil contaminations (Tilman et al., 2006; Isbell et al., 2015). We also acknowledged that the result of our short-term experiment is not comparable to long-term ones since both soil pollutants and plant population structures may change over time due to complex plant-soil interactions. Hence, further studies on long-term plant responses to soil contamination may be crucial for guiding the remediation of contaminated soils.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

WX and F-HY designed the study. LH, S-MY, and YJ conducted the experiment and collected the data. LH and WX analyzed the data. LH, WX, and F-HY wrote the first version of the manuscript. All authors commented on and reviewed the final draft. All authors contributed to the article and approved the submitted version.