Samples of nodules and underlying sediments were collected using a multicorer at depths of 5302-5562 m from zone M2 (MC02, BC38A, and BC78) in the western Pacific during August and September 2022 aboard the research vessel Dayangyihao on voyage 75 ( Figure 1 , Table 1 ). The M2 area in the northern Magellan Seamounts of the western Pacific is a geologically stable region with limited volcanic and tectonic activity (Hein et al., 1997). The surface sediments are mainly composed of brownish-yellow clay. The study area represents a typical tropical oligotrophic environment, with low surface productivity (Jiang et al., 2020a), moderately high bottom water oxygen concentrations (175–200 mmol/m³) (Dutkiewicz et al., 2020; Ren et al., 2022), and a bottom seawater temperature of approximately 1.5°C and salinity of around 34.6. The nodules in this area are primarily formed through hydrogenic processes, characterized by Co enrichment and a spheroidal morphology.

Samples were also collected using a multicorer at depths of 5177-5267 m from zone KW1 (MC04 and BC05) and zone A5 (BC12) in the eastern Pacific during voyage 73 of the research vessel Dayanghao within the same timeframe ( Figure 1 , Table 1 ). KW1 and A5 are adjacent blocks, both located in the western part of the CCZ in the eastern Pacific. The feature surface sediments with a brown hue, is primarily consisting of siliceous oozes. The dissolved oxygen concentration in the bottom seawater is approximately 150 mmol/m³ (Dutkiewicz et al., 2020), with a temperature of 1.5°C and salinity around 34.7 (Washburn et al., 2021). The nodules are predominantly poly-nodules and have a hydrogenetic or mixed-type origin (Reyss et al., 1985).

The collected nodules and sediments were stored at −80°C and transported on dry ice to the laboratory.

The polymetallic nodules were dried, crushed, weighed (5 g) and then ground into a homogeneous powder of 200 mesh size using an agate pestle and mortar. Elemental analyses were conducted at the Key Laboratory of Submarine Geosciences, Second Institute of Oceanography, Ministry of Natural Resources. The major elements in the nodules and sediments were determined using X-ray fluorescence spectroscopy (XRF, AxiosMAX, PANalytical, Netherlands). The trace elements in the nodules and sediments were determined using inductively coupled plasma-mass spectrometry(ICP-MS, Elan DRC-e, Perkin Elmer). All measured concentrations exceeded detection limits, with a relative standard deviation of laboratory precision less than 10%.

The nodules were mounted to epoxy and then cut in along the maximum axis to make thin sections. SEM images were performed using a TESCAN MIRA3 field-emission scanning electron microprobe (FE-SEM) at the Testing Center, Tuoyan Analytical Technology Co. Ltd. (Guangzhou, China). After the samples were carbon-coated, SEM images were acquired under an acceleration voltage of 20 kV, a beam current of 15 nA and a magnification of 300–500×. The surface elemental composition of carbon coated sample was characterized by energy dispersive X-ray analysis (SEM-EDAX) EDS detector (EDAX Element EDS detector) device attached to an SEM operating at 20 kV. Observing SEM images provides information about the internal structures within nodules, and further analysis can be conducted by combining the corresponding EDS results.

Under sterile conditions, the surface deposits adhering to the nodules were scraped away. The nodules were then fragmented using a chisel and then ground with a mortar. The nodules and surrounding sediments were individually weighed to 0.5 g. DNA was extracted using an Advanced Soil DNA Kit (MOBIO, Solana Beach, USA) according to the instructions.

The extracted DNA was used as a template for amplifying the V1-V9 regions with the primer set 27 F (5’-AGRGTTYGATYMTGGCTCAG-3’) and 1492 R (5’-RGYTACCTTGTTACGACTT-3’). Polymerase chain reaction (PCR) was conducted using the BioRad (S1000, Bio-Rad Laboratories, USA). Each sample was run in triplicate, and the PCR products from the same sample were pooled. Library preparation followed the 16S Amplification SMRTbell^® Library Preparation protocol, and sequencing of the amplicon library was performed by Guangzhou Meige Biotechnology Co., Ltd., utilizing the PacBio Sequel II platform (PacBio, USA).

Fastp (v0.14.1) removed sequences over 2000 bp, and Cutadapt (v1.14) eliminated primers, resulting in valid fragments. Uparse defined operational taxonomic units (OTUs) at a 97% similarity threshold, with a confidence level of 0.8. Usearch-sintax (v10.0.240) aligned OTU representative sequences against the Silva v132 database.

Alpha diversity indices, including abundance-based coverage estimator (ACE) and Shannon, were calculated using usearch-alpha div (v10) based on the relative abundances of OTUs in each sample. Rarefaction curves were drawn by usearch-alpha div rare (v10) based on a richness index to estimate the sampling efforts. Community compositions were performed using R software.

Disparities were observed in the external morphological characteristics and internal structures of polymetallic nodules from the western and eastern Pacific ( Figure 2 ). The nodules from the western Pacific were near-spherical ( Figure 2A ), with a homogeneous texture and no distinct core ( Figure 2B ). In contrast, nodules from the eastern Pacific were primarily poly-nodules containing more than three nuclei, exhibiting rough surfaces ( Figure 2C ) and distinct cores, which were encased by surrounding conduits ( Figure 2D ).

Nodules exhibited a significant enrichment of metal elements such as Mn, Fe, Ti, Co, Cu and Ni compared to the sediments ( Table 2 ). Mn, Cu and Ni were generally more enriched in nodules from the eastern Pacific, while Fe, Ti and Co were more enriched in nodules from the western Pacific. Based on the Mn/Fe ratio, nodules from the western Pacific were indicative of a pronounced hydrogenetic origin (Mn/Fe ≈ 1) (Verlaan et al., 2004; Hein and Koschinsky, 2014), whereas nodules from eastern Pacific generally show a mixture of diagenetic and hydrogenetic origin, with a predominantly diagenetic input (2.09-2.77) (Wegorzewski and Kuhn, 2014; Hein et al., 2020).

Both western and eastern Pacific nodules exhibit regularly arranged laminated structures ( Figure 3 ). The nodules from the western Pacific are characterized by laminated and concentric rims (He et al., in press), where compact layers (brighter) and loose layers (darker) alternate in a ring-like pattern, forming a rhythmic structure ( Figures 3A–C ). In contrast, nodules from the eastern Pacific predominantly display stromatolite-like rims with a loose texture and abundant pores (darker areas) distributed irregularly ( Figures 3D–F ). The compact structures in both regions show clear Mn enrichment on their surfaces, identified from the Mn elemental mapping, where lighter purple areas indicate higher Mn concentrations ( Figures 3C, F ). This is based on qualitative EDS analysis. The Mn distribution follows a rhythmic pattern, corresponding to concentric or stromatolite-like laminations.

Furthermore, many biological-like structures were observed from the fissures and cavities in the thin sections. For example, some bacteria-like microspheres ( Figures 4A ) exhibited a ring-like arrangement, with morphologies similar to those considered bacteria by Jiang et al. (2019) and Reykhard and Shulga (2019). Additionally, another type of bacteria-like microsphere, characterized by a flattened, centrally concave shape, was densely distributed within the channels ( Figure 4B ). Larger biological-like structures (>20 μm), including porous spheroidal ( Figure 4C ) and annular ( Figures 4D, E ) structures were also observed. Furthermore, a transparent and highly reflective extracellular polymeric substance (EPS)-like structure ( Figure 4F ) (Ren and Jones, 2021).

To further elucidate the elemental composition of these structures, qualitative EDS analysis was performed. Higher peak intensities indicate greater element enrichment ( Figure 4 ). Mn, Fe, and Ti are the primary metals that exhibit substantial enrichment within the nodules, and due to their considerable economic value, they have been the subject of extensive research (Hein et al., 2020). Furthermore, Mn and Fe play pivotal roles in cellular processes such as development, metabolism, and enzymatic activity (Bruins et al., 2000; Helmann, 2014), and are also known to facilitate the formation of bacterial biofilms (Avidan et al., 2010; Mhatre et al., 2016). Of particular interest, Mn, Fe, and Ti are also found to exhibit markedly high concentrations on the microstructural surfaces, which warrants further attention. The surfaces of microspheres exhibited the ability to enrich metals (Mn, Fe, Ti), particularly Mn ( Figures 4A, B ). Surfaces of other types of structures have similar elemental enrichments (Fe, Mn, Ti), particularly Fe ( Figures 4C–E ). EPS-like structure displayed comparable levels of enrichment for both Mn and Fe ( Figure 4F ).

The sequencing depth was sufficient, capturing most of the community diversity with a rich variety of bacterial taxa ( Supplementary Figure S1 ; Figure 5 ) (He et al., in press). Proteobacteria (38.2%-46.9%) was predominant across the samples, with the dominance of the class Gammaproteobacteria (18.1% to 32.3%). Woeseia exhibited dominance in WNA (12.26%) and WSA (9.56%), while Arcicella was the dominant genus in ENB (28.02%) and ESB (22.22%).

The experiment identified some unique genera that appeared to be exclusive to specific samples ( Table 3 ; Supplementary Table S1 ). For example, Leptospirillum and Lentisphaera were found only in western Pacific samples (WNA, WSA), while Psychrobium and Sulfitobacter were detected exclusively in eastern Pacific samples (ENB, ESB). Halomonas was enriched in ENB but was not detected in WNA. Similarly, wb1-A12, Blastocatella, and IS-44 were enriched in sediment samples (WSA, ESB) and were not identified in WNA. However, reliance on relative abundances alone was insufficient to definitively confirm the presence or absence of these genera, though it suggests the possibility of such patterns. It is important to note that these results are based on sequencing data from a single sample, highlighting the need for further replicates to robustly assess microbial enrichment preferences across different sample types.

The 16S full-length sequencing technique enabled species identification 29 species were detected from the nodule samples. Excluding the unsigned and uncultured species, 9 species could be identified ( Table 4 ). These species, belonging to the Gamma or Alpha classes of Proteobacteria, were from various genera ( Supplementary Table S2 ), indicating potential variations in their physiological characteristics and functions. It is also noteworthy that the occurrence of these species may be influenced by experimental factors, which may hinder the definitive determination of their exclusivity to a specific sample. Consequently, the zero abundances observed ( Table 4 ) cannot be conclusively interpreted as the absence of these species but instead represent the potential for these species to be unique to certain samples.

Previous studies utilizing SEM-EDS have identified manganese-oxidizing bacteria (MOB) and biological structures involved in nodule mineralization (Wang and Müller, 2009; Jiang et al., 2020a; Shulga et al., 2022). In this study, metal-enriched biological-like structures were observed, including bacteria-like structures ( Figures 4A, B ) and other biomophs ( Figures 4C–F ), which are also likely involved in the mineralization process. SEM-EDS may have resolution limitations in analyzing biological structures, which can hinder accurate characterization of fine biological features and trace element distributions, making these findings speculative in the absence of direct evidence for their biological origin. However, careful comparisons with similar structures reported in the literature were made during the SEM-EDS analysis.

In addition to the microspheres potentially representing mineralizing bacteria ( Figures 4A, B ), other biomorphs that may contribute to the mineralization process include porous structures, likely remnants of radiolarian skeletons, characterized by high Si content and features such as spines or spine-like openings ( Figure 4C ) (Jiang et al., 2019);annular, skeleton-like structures resembling the solid shells produced by coccolithophores ( Figures 4D, E ), as proposed by Wang et al. (2012); and transparent, highly reflective structures capable of enriching Mn and Fe, similar to extracellular polymeric substances (EPS) ( Figure 4F ) (Ren and Jones, 2021).The surfaces of these structures seem to serve as sites for the precipitation of metal oxides, resulting in the accumulation of Fe and Mn minerals (Jiang et al., 2019), which fall under the category of biologically induced mineralization (BIM). BIM occurs when organisms secrete metabolites that react with ions or compounds in the environment, leading to the growth and deposition of mineral particles (Frankel and Bazylinski, 2003). Jiang et al. (2019) suggested that these biomorphs can induce mineralization processes through surface enrichment and proposed a mineralization process of mixed colloids of Mn- and Fe-oxide-hydroxides through biological-induced oxidation of Mn(II) and Fe(II) on the surfaces of biological structures. The significant enrichment of metals observed on the surfaces of biological structures in the present study provides further evidence for this idea ( Figures 4 ).

However, it is crucial to acknowledge that the proposed biological structures are primarily qualitative interpretations based on the literature, rather than quantitative determinations. As such, their identification remains speculative, with EDS serving only as a supplementary tool. Consequently, the possibility of non-biological processes contributing to the observed morphological features cannot be excluded. For example, abiotic processes such as mineral precipitation or physical deposition could produce structures that resemble those associated with biological mineralization, complicating the interpretation of the results (Ren and Jones, 2021; Zhang et al., 2024). Moreover, the relationship between biological forms and mineral types in BIM remains unclear. Therefore, further research should aim to establish a more definitive link between specific biological structures and the minerals they may induce, with a focus on acquiring direct evidence of biological processes, while also critically considering the potential role of abiotic mechanisms in shaping these structures.

The growth environments of polymetallic nodules in the western and eastern Pacific may differ. In the western Pacific, polymetallic nodules are thought to be primarily hydrogenetic (Mn/Fe ~ 1), with metals precipitating from oxygen-rich near-bottom seawater, enriching Fe and Co, while nodules from the eastern Pacific appear to be more significantly influenced by diagenetic processes, with metal precipitation occurring in sediment pore spaces, enriching Mn, Cu, and Ni (Halbach et al., 1981, 1988). The elemental composition of nodules is likely influenced by factors such as surface primary productivity, organic carbon content in the seafloor, and redox conditions. In the western Pacific, low primary productivity and high dissolved oxygen in bottom waters may contribute to the rapid oxidation of Fe and Co to stable +3 oxidation states (Liu and Millero, 2002; Dutkiewicz et al., 2020; Jiang et al., 2020a). In contrast, in the eastern Pacific, higher primary productivity and organic carbon content in sediments may promote the formation of metal-organic complexes, potentially enriching Mn, Cu, and Ni in the sediments (Verlaan et al., 2004). As organic matter degrades, reducing conditions at the seafloor may release Mn and other elements in their 2+ states into pore waters, which could further contribute to the enrichment of these elements in the nodules (Jorgensen, 2001; Arndt et al., 2013).

The differing genesis of polymetallic nodules may also suggest variations in the hydrodynamic and geological conditions under which the nodules form. Nodules of hydrogenetic origin in the western Pacific typically develop in relatively stable aquatic and geological environments, facilitating the growth of near-spherical nodules characterized by rhythmic, dense, and uniform laminae ( Figure 2A , Figures 3A, B ) (Li et al., 2020). In contrast, nodules from the eastern Pacific are predominantly influenced by diagenetic processes, displaying stromatolitic-like growth structures with broader fissures and pores ( Figure 2D , Figures 3D, E ). These features indicate that nodule growth in this region was influenced by ongoing environmental modifications, including seafloor bioturbation and bottom currents (Veillette et al., 2007; Wegorzewski and Kuhn, 2014).

Furthermore, the differences in microbial community composition across nodule fields in different marine regions may also reflect variations in the environmental conditions. For example, notable distinctions were observed when comparing the bacterial communities identified in this study with those from other Mn nodule fields, Such as the Korea Deep Ocean Study (KODOS) (Cho et al., 2018) and the German license area (Blöthe et al., 2015) in the Clarion-Clipperton Fracture Zone (CCFZ), as well as the Kara Sea (Vereshchagin et al., 2019)and the South Pacific Gyre (Shiraishi et al., 2016). For instance, Marinobacter and Idiomarina, two genera that are mostly strictly aerobic, were abundant in the KODOS area (Ivanova et al., 2000; Green et al., 2006). However, they were not detected in samples from the western Pacific in this study, and their abundances in samples from the eastern Pacific were extremely low (<0.2%), potentially linked to variations in environmental oxygen levels. Moreover, Shewanella and Colwellia, as potentially important genera in the nodule formation in this study, were not detected in the KODOS area and were only found in nodules from the German license area. However, Colwellia exhibited widespread enrichment in sediments from the Kara Sea (2%) and the South Pacific Gyre (>50%). Both of these two regions are characterized by extremely low surface primary productivity (D’Hondt et al., 2009; Demidov et al., 2017), suggesting that Colwellia may accumulate in environments with exceptionally low carbon content due to its capability to degrade recalcitrant substances (Tully and Heidelberg, 2013). The acquisition of more detailed environmental data in future studies may provide further empirical support for our hypothesis.