Positive effects and mechanisms of simulated lunar low-magnetic environment on earth-worm-improved lunar soil simulant as a cultivation substrate

2. Materials and methods 2.1 Mixed substrate, earthworms and magnetic field environments The organic solid waste used in this study was derived from our "Lunar Palace 365" experiment. To ensure representativeness, we mixed solid waste samples collected on July 4, 2017 (early phase of the experiment) and May 15, 2018 (late phase). The lunar soil simulant used was CUG-1B, provided by Prof. Long Xiao's research group at China University of Geosciences (Wuhan). The solid waste and lunar soil simulant were mixed at a 3:7 ratio, adjusted to 70% humidity, and placed in plastic bottles with ventilation holes. Eisenia fetida, a commonly used laboratory model species, was obtained from a worm breeding farm (Changzhou, Jiangsu, China). Prior to the experiment, the earthworms were acclimatized for two weeks to adapt to the experimental conditions. Mature adult earthworms (0.5 ± 0.39 g) were selected for the study and underwent thorough disinfection to ensure their effectiveness in improving soil properties and preventing the introduction of pathogenic pests into the cultivation substrate. All magnetic field treatments utilized a standardized cylindrical growth pot (65 cm diameter × 200 cm height, total volume ≈663,662 mL) constructed from polypropylene. The container was specifically designed to maintain consistent physical conditions across all magnetic field environments while accommodating the different field generation apparatus. Three magnetic field conditions were designed for comparative experiments, with each magnetic field treatment having 3 identical growth pots as replicates: (1) Low magnetic field group: growth pots were housed within a four-layer permalloy magnetic shielding device (Fig. 1a). Residual magnetic fields were measured at nine representative locations using a calibrated fluxgate magnetometer (Mag03 sensor, resolution 0.1 nT, bandwidth 3 kHz, range ±100 μT at 25°C). All positions maintained field levels below 5 nT (Fig. 1b). (2) Earth magnetic field control group: growth pots was placed on non-magnetic wooden tables (residual magnetism <5 nT) in an electromagnetic shielding chamber (background noise <10 nT), maintaining 25.3 ± 0.8 μT ambient field. Field uniformity was verified using the same magnetometer. (3) High magnetic field group: twenty N52-grade NdFeB permanent magnets (30×20×5 mm) were arranged in a Halbach array surrounding growth pots (Fig. 1c). Field uniformity was verified at multiple points (Fig. 1d), with precise spacing adjustments achieving 7,000 ± 250 μT central field strength (gradient <50 μT/cm). Ten earthworms were placed in each substrate for a 10-day observation period. Post-experiment, physiological indices of earthworms and physicochemical properties of the substrate were analyzed. 2.2 Earthworm physiological indices Histopathological evaluation (HE) staining was performed following standard histopathological methods. Intestinal tissue morphology and inflammation levels were observed and imaged at 200× magnification. Paraffin sections were dehydrated with xylene and absolute ethanol, stained with hematoxylin and eosin, and mounted with neutral resin for microscopic examination and image analysis. After gut content removal, intestinal tissue samples were homogenized in saline and analyzed for SOD and MDA content using commercial kits (Nanjing Jiancheng Bioengineering Institute, China). Whole brain tissue from each earthworm was homogenized in saline, and acetylcholine levels, Ca²⁺-ATPase, and Na⁺-ATPase activities were quantified using commercial kits (Nanjing Jiancheng Bioengineering Institute, China). 2.3 Substrate physicochemical properties Lunar soil simulant samples (25 ± 2.5 g) were extracted using undisturbed soil samplers. Extract pH and electrical conductivity (EC) were measured with pH (FE28, METTLER TOLEDO, Switzerland) and EC meters (FE38, METTLER TOLEDO). Soil organic matter (SOM) was determined via potassium dichromate titration. Humus content was measured after extraction with 0.1 mol/L sodium pyrophosphate and 0.1 mol/L sodium hydroxide, followed by dichromate oxidation. Total carbon (C) and nitrogen (N) were analyzed using an elemental analyzer (UNICUBE, Elementar, Germany). Hydraulic conductivity (HC) was determined via constant-head permeability tests and Darcy's law. 2.4 Microbial sequencing of earthworm gut and substrate Comprehensive information about the bacterial community composition in cultivation substrate was obtained through 16S rRNA high-throughput sequencing. DNA extraction, PCR amplification, and sequencing of cultivation substrate microbial communities were performed by Biomarker Tech. Corp., Beijing, China. Briefly, DNA was extracted with the TGuide S96 Magnetic Soil/Stool DNA Kit (Tiangen Biotech (Beijing) Co., Ltd., China) according to the manufacturer's instruction. DNA concentration was measured with the Qubit dsDNA HS Assay Kit and Qubit 4.0 Fluorometer (Invitrogen, Thermo Fisher Scientific, Eugene, Oregon, USA). The 338F: 5'-ACTCCTACGGGAGGCAGCA-3' and 806R: 5'-GGACTACHVGGGTWTCTAAT-3' universal primer set was used to amplify the V3-V4 region of 16S rRNA gene. Both the forward and reverse primers were tailed with sample-specific Illumina index sequences. The PCR was performed in a total reaction volume of 10 μl: DNA template 5-50 ng, *Vn F (10 μM) 0.3 μl, *Vn R (10 μM) 0.3 μl, KOD FX Neo Buffer 5 μl, dNTP (2 mM each) 2 μl, KOD FX Neo 0.2 μl, and ddH₂O up to 10 μl. Vn F and Vn R were selected according to the amplification area. The initial denaturation at 95°C for 5 min was followed by 25 cycles of denaturation at 95°C for 30 s, annealing at 50°C for 30 s and extension at 72°C for 40 s, and a final step at 72°C for 7 min. PCR amplicons were purified with Agencourt AMPure XP Beads (Beckman Coulter, Indianapolis, IN, USA) and quantified using the Qubit dsDNA HS Assay Kit and Qubit 4.0 Fluorometer (Invitrogen, Thermo Fisher Scientific). The quantified amplicons were then pooled together in equal amounts as the library. The constructed library was sequenced using Illumina NovaSeq 6000 (Illumina, Santiago, CA, USA). Downstream sequencing analysis was performed on BMK Cloud (Biomarker Technologies Co., Ltd., Beijing, China). Raw sequences were first processed using Trimmomatic and FLASH, with a moving window of 50-bp and a quality threshold score of 30. Singletons were then removed. Next, high-resolution amplicon sequence variants (ASVs) were identified from the reads using DADA2 (version 2020.06). Lastly, a representative sequence of each ASV was annotated through SILVA ribosomal RNA gene database (version 132) with a confidence score of 0.7. 2.5 Molecular ecological network analysis With 16S rRNA gene (bacteria) ASVs, ITS ASVs and environmental factors pooled together as the input, phylogenetic molecular ecological networks (pMENs) were constructed based on random matrix theory (RMT)-based network (Deng et al., 2012b). The threshold of similarity coefficients (r values of the Spearman’s rho correlation) for network construction was automatically determined when the nearest-neighbor spacing distribution of eigenvalues transitioned from Gaussian orthogonal ensemble to Poisson distributions (Deng et al., 2012b). Random networks corresponding to all pMENs were constructed using Maslov-Sneppen procedure with the same network size and average link number to verify the system-specificity, sensitivity and robustness of the empirical networks(Maslov and Sneppen, 2002). Network graphs were visualized with Cytoscape 3.8 software. 2.6 Process of microbial community assembly To explore the structure of bacterial and fungal community assembly processes by deterministic or stochastic processes (Stegen et al., 2012), the β-nearest taxon index (β-NTI; was calculated using the R package “picante” (version 1.8.2). The β-NTI value, calculated using null-model expectations with consideration for phylogenetic distance, provides insight into the turnover of microbial communities. A null distribution of Beta Mean Nearest Taxon Distance (β-MNTD) is performed by randomizing OTUs across the phylogeny and recalculating β-MNTD 999 times (Stegen et al., 2012). Β-NTI quantifies the number of standard deviations that the observed β-MNTD is from the mean of the null distribution. 2.7 Statistical analysis Shannon index (α-diversity) was calculated and displayed with the package "ieggr" (version 4.17) using R software (version 4.3.3). Principal coordinate analysis (PCoA) was calculated and displayed with the package "ape" (version 5.0) using R software. Generalized linear models were constructed by "glm" function in R to investigate relationships among relative abundance of key-microbes, and wheat agronomical parameters. Redundancy analysis (RDA) among microbial communities and environmental factors was performed using BMKCloud (www.biocloud.net, Biomarker Tech. Corp.). Analysis of variance (ANOVA) and post-hoc Fisher's Least Significant Difference (LSD) test with Bonferroni P-adjust determining the difference among cultivation groups were calculated with the package "agricolae" (version 1.3-7) using R software. All experiments were performed 3 or more times and experimental results were analyzed for statistical significance using 2-tail test.