Flooding patterns shape microbial community in mangrove sediments

F looding patterns shape microbial community in mangrove sediments 1 2 Mir na Vázquez-Rosas-Landa1+, R osela Pérez-Ceballos2,3, Arturo Zaldívar-Jiménez4, 3 Stepha nie E Hereira-Pacheco5, L eonardo D. Pérez-González1, Alej andra Prieto‑Davó6, O mar 4 Celis-He rnández2, 3, Julio C. Canales-Delgadillo2, 3*+. 5 6 1 Uni dad Académica de Ecología y Biodiversidad Acuática, Instituto de Ciencias del Mar y 7 L imnología, Universidad Nacional Autónoma de México, Mexico city, Mexico 8 2 I nstituto de Ciencias del Mar y Limnología Estación El Carmen, Universidad Nacional 9 Aut ónoma de México, Ciudad del Carmen, México. 10 3 Conse jo Nacional de Humanidades de Ciencias y Tecnologías, México (CONAHCYT). 11 4 Ase soría Técnica y Estudios Costeros, Mérida, Yucatán, México. 12 5 Labor atorio de Interacciones Bióticas, Centro de Investigación en Ciencias Biológicas, 13 Uni versidad Autónoma de Tlaxcala, Tlaxcala, México. 14 6 Uni dad de Química en Sisal, Facultad de Química, Universidad Nacional Autónoma de 15 Mé xico, Puerto de abrigo s/n, Sisal, Yucatán, México 16 17 18 *Co rresponding Author: 19 Juli o C. Canales-Delgadillo2,3 20 2 I nstituto de Ciencias del Mar y Limnología Estación El Carmen, Universidad Nacional 21 Aut ónoma de México, Ciudad del Carmen, México. 22 3 Conse jo Nacional de Humanidades de Ciencias y Tecnologías, México (CONAHCYT). 23 24 Emai l address: 25 jcc [email protected] 26 27 + Both authors contributed equally to the manuscript 28 29 Abst ract 30 Back ground. Mangrove ecosystems located in the tropics and subtropics, are crucial for 31 reg ulating global weather patterns and sequestering carbon. However, they face threats from 32 huma n activities like altered water flow and deforestation. While the symbiotic relationship 33 betw een mangrove trees and surrounding microbes are essential for their survival, the impact 34 of human activity on these microbial communities remains incompletely understood. We 35 inve stigated how microbial communities change in degraded mangrove ecosystems due to 36 loss of hydrologic connectivity, aiming to elucidate consequences and inform restoration 37 strat egies. 38 39 Me thods. Employing 16S rRNA sequencing, we analyzed samples of sediment cores from 40 c onserved, moderately degraded, and degraded mangrove sites across dry and flood seasons 41 a t three sediment depths. 42 43 Re sults. Our analysis identified 11,469 Amplicon Single Variant (ASVs), revealing diversity 44 loss correlated with degradation levels. Notably, we observed shifts in microbial diversity 45 within s ediment layers, with conserved sites dominated by Vibrionaceae in upper layers, 46 pote ntially indicating urban contamination. In moderate-degradation sites, seasonal patterns 47 e merged, with Halomonas and Marinomonas dominating the dry season and 48 E xiguobacterium thriving during flooding. Interestingly, a community mainly composed of 49 F irmicutes persisted across all degradation scenarios in deeper sediment layers, suggesting 50 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.24.604998doi: bioRxiv preprint pote ntial for ecosystem restoration. Our findings provide insights into microbial responses to 51 huma n-induced stressors and highlight the role of core microbial communities in guiding 52 res toration efforts for degraded mangrove ecosystems. 53 54 Key words: soil, degradation, microbiome, holobiont, hidroperiod 55 56 Intr oduction 57 Ma ngrove ecosystems, found in tropical and subtropical regions worldwide (Lee et al., 2014), 58 a re vital components of coastal landscapes and play essential roles in global weather patterns, 59 incl uding carbon sequestration (Chatting et al., 2022; Song et al., 2023). Despite their 60 e cological significance, these ecosystems face severe threats from human activities such as 61 a ltered river flows and extensive deforestation (Akram et al., 2023). Within these ecosystems, 62 mic roorganisms play key roles in sustaining ecosystem health and functioning, however, their 63 ro le, particularly as a potential link for restoration, remains to be fully understood. 64 65 I n mangrove ecosystems, microorganisms participate in nutrient acquisition, disease 66 res istance, nutrient recycling, and stress tolerance (Palit et al., 2022; Holguin, Vazquez & 67 Bas han, 2001; Akram et al., 2023; B ashan & Holguin, 2002; Vovides et al., 2011; McKee & 68 F aulkner, 2000; Woodroffe et al., 2016). Mangrove trees and the microbial community 69 inte ract mutually to support each other's survival (Akram et al., 2023). For instance, 70 man groves shelter and nourish various microorganisms in their rhizosphere and root tissues 71 (Pura hong et al., 2019), and in return, these microorganisms facilitate nutrient cycling, such 72 a s nitrogen-fixation (Sjöling et al., 2005; Vovides et al., 2011), an essential nutrient often 73 limi ted in coastal environments, thereby promoting mangrove growth and productivity. These 74 dive rse microbial communities also serve as buffers enhancing resilience to environmental 75 stres sors such as salinity, flooding, and disease (Lai et al., 2022). For instance, adding 76 A zospirillum in restoration efforts promotes nutrient absorption and stress tolerance in 77 man grove roots by facilitating the colonization of diverse microbes (Domínguez-Núñez & 78 Berr ocal-Lobo, 2021). Additionally, a diverse array of bacterial genera, including 79 A grobacterium, Alcaligenes, Arthrobacter, Bacillus, Enterobacter, Erwinia, Pseudomonas, 80 R hizobium, Serratia, Stenotrophomonas, Streptomyces, and Xanthomonas, have shown 81 pr otective effects against fungal and bacterial pathogens in various plant systems by 82 c olonizing infection sites, competitively excluding pathogens, and secreting antimicrobials 83 (Bonat erra et al., 2022). Although reports of this phenomenon in mangroves are limited, 84 sim ilar mechanisms could operate within mangrove ecosystems. Leaning to the idea where 85 man grove trees and the surrounding microbial community form a resilient entity known as 86 holobio nt (Allard et al., 2020), challenging traditional views of organisms and emphasizing 87 the interconnectedness between hosts and their associated microbial communities (Morris, 88 201 8). 89 90 Rec ognizing the holobiont concept in mangrove ecosystems is key for conservation and 91 man agement efforts. Conservation strategies should prioritize preserving mangrove species 92 a nd safeguarding the diversity and integrity of associated microbial communities (Lai et al., 93 202 2). For instance, it has been shown that different tree species, especially Aviceina 94 ger minians had different microbial communities associated (Gomes et al., 2014), and also 95 c ore microbes and specialized microbiota to each tree has been observed (Wainwright et al., 96 202 3), therefore, the success of the restoration is related to the microbial communities present 97 in th e roots of plants (Gomes et al., 2010). Integrating hydrological dynamics into restoration 98 plan s is vital, as water availability, sediment deposition, and nutrient cycling profoundly 99 influe nce mangrove survival and regeneration (Pérez-Ceballos et al., 2020; Ellison, Felson & 10 0 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.24.604998doi: bioRxiv preprint F riess, 2020; Akram et al., 2023). Thus, holistic restoration initiatives that consider the entire 10 1 holobio nt and incorporate hydrological factors are essential for ensuring the long-term 10 2 res ilience and sustainability of mangrove ecosystems. 10 3 T idal fluctuations in mangrove ecosystems dictate crucial factors such as nutrients, oxygen 10 4 leve ls, dispersal mechanisms, and salinity. These elements are essential for the well-being 10 5 a nd distribution of mangrove species and their associated microbial community. Human-10 6 induc ed alterations to hydrological patterns can disrupt these ecosystems, leading to reduced 10 7 tida l flushing, elevated salinity, sedimentation, and nutrient imbalances (Kamali & Hashim, 10 8 201 1; Pérez-Ceballos et al., 2020). Conversely, prolonged flooding can raise water sulfide 10 9 leve ls, harming mangrove trees (Pérez-Ceballos et al., 2022). 11 0 11 1 Hyd rological dynamics also shape the distribution and composition of microbial communities 11 2 in m angrove habitats (Mai et al., 2021; Thomson et al., 2022). Tidal inundation introduces 11 3 dive rse microbial populations from neighboring areas, enriching soil microbiota. Changes in 11 4 hyd rological patterns can disrupt these microbial communities, affecting nutrient availability, 11 5 dec omposition rates, and ultimately mangrove health (Pérez-Ceballos et al., 2020; Akram et 11 6 a l., 2023). Therefore, mangrove rehabilitation efforts should prioritize actions like restoring 11 7 tida l connectivity and managing sedimentation to foster optimal conditions for mangrove 11 8 gr owth and the establishment of healthy microbial communities. 11 9 12 0 Bui lding upon the importance of microorganisms and hydrological dynamics in mangrove 12 1 e cosystems, our study aims to investigate the impact of ecosystem degradation on microbial 12 2 c ommunities within mangrove sediments. Samples were collected in the Laguna de 12 3 T érminos, south Gulf of Mexico, from non-degraded mangrove, moderately degraded, and 12 4 fu rther degraded mangrove across two seasons and three sediment depths, with microbial 12 5 dive rsity assessed using 16S rRNA sequencing. We hypothesize that increasing degradation 12 6 will reduce microbial diversity, anticipating seasonal and depth fluctuations in composition. 12 7 T hrough identification of key microbial species, our findings offer valuable insights to guide 12 8 res toration initiatives amid anthropogenic pressures. Estero Pargo within Laguna de 12 9 T érminos, our study site, serves as an urban mangrove model to study these impacts, 13 0 e xhibiting notable changes such as fish farming expansion and tree mortality. This mangrove 13 1 syste m's zonation comprises red (Rhizophora mangle), white (Laguncularia racemosa), and 13 2 blac k (Avicennia germinans) mangrove species. Estero Pargo provides a pertinent context for 13 3 und erstanding the effects of human activities on mangrove ecosystems. 13 4 13 5 Mat erials & Methods 13 6 Stu dy area 13 7 L aguna de Términos, designated as a Ramsar site of international and high biological 13 8 impor tance, is currently experiencing several pressures due to growing industry and 13 9 ur banization. The study area encompasses a mangrove forest situated within an estuary 14 0 kno wn as Estero Pargo on the southeast side of Isla del Carmen in Campeche, Mexico (18° 14 1 39' 05" N; 91° 45' 31" W and 18° 39' 03" N; 91° 45' 26" W, Supplementary figure 1). Estero 14 2 Par go, a tidal channel spanning approximately 6 km in length and averaging 14 m in width, 14 3 c overs a surface area of approximately 52 ha. The study area is located approximately 1-2 km 14 4 a way from the nearest urban settlements, with at least two untreated wastewater discharges 14 5 e ntering Estero Pargo. Additionally, over the past three years, fish farming activities have 14 6 e xpanded within the estuary, along with an increase in the size of a patch of dead trees. 14 7 Conse quently, Estero Pargo represents an urban mangrove that can serve as a model location 14 8 fo r studying anthropogenic impacts on biogeochemical cycles in non-pristine mangroves 14 9 (V ande et al, 2019). 15 0 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.24.604998doi: bioRxiv preprint 15 1 Ac cording to the zonation (fringe, basin, impaired area), three mangrove species comprise the 15 2 vege tation cover in Estero Pargo: at the fringe (approximately 20-30 m wide) along the edge 15 3 of the tidal channel, the dominant species is the red mangrove (Rhizophora mangle), with a 15 4 few individuals of white mangrove (Laguncularia racemosa) also present. Moving to the 15 5 bas in (about 300 m wide), situated behind the fringe, the dominant species shifts to the black 15 6 man grove (Avicennia germinans). Lastly, there exists a patch (approximately 2.8 ha) of an 15 7 impa ired mangrove area where most of the trees have died, yet a few individuals of A. 15 8 ger minans persist (Supplementary figure 1). 15 9 16 0 T he annual average rainfall in the study area is approximately 1,680 mm, with a mean annual 16 1 tem perature of 27 °C (Coronado-Molina et al., 2012). The tidal regime is about 0.33 m 16 2 (Co ntreras Ruiz Esparza, Douillet & Zavala-Hidalgo, 2014). There are two periods of 16 3 minimum (F ebruary to August) and maximum (September to January) flooding. The soil bulk 16 4 dens ity is 0.22±0.008 g cm-3 to 0.44± 0.02 g c m-3 and organic matter is 8.13 ±1.86% to 12.0 16 5 ±3% (Pérez-Ceballos et al., 2022). 16 6 16 7 Sam pling 16 8 T o characterize the bacterial community in the sediments of Estero Pargo, we conducted 16 9 sa mpling along a conservation gradient, ranging from the most conserved zone, hereafter 17 0 ref erred to as non-degraded (ND) or the fringe, followed by the moderately degraded (MD) 17 1 site or the basin, to the most damaged area, referred to as further degraded (D) or the 17 2 impa ired mangrove. The ND site exhibited no evidence of natural or anthropic factors 17 3 impa cting the forest structure, with most trees alive and the forest covering this zone 17 4 c ompletely, serving as our reference site. In the MD site, although most trees were alive, 17 5 signs of degradation were observed as several trees had died. Lastly, the D site experienced 17 6 fu rther tree mortality, with only a few living trees remaining (Supplementary figure 1). 17 7 178 We established three sampling plots, each measuring 100 m2, par allel to the tidal channel 17 9 within e ach sampling zone. At the midpoint of each plot, we collected sediment cores with a 18 0 dept h of 50 cm from the soil surface. Subsequently, we subsampled each core at intervals of 18 1 0-10 c m, 10-30 cm, and 30-50 cm (hereafter referred to as horizon levels) to obtain samples 18 2 fr om the central part of each subsection, corresponding to 5 cm, 20 cm, and 40 cm horizon 18 3 leve ls. To address seasonal variations, we collected three replicates from each site during two 18 4 dist inct seasons: the flooding season (January) and the dry season (May) in 2018. All samples 18 5 we re transported to the laboratory and stored at -20°C until processing. 18 6 18 7 DNA extraction 18 8 Af ter thawing and fully homogenizing the samples, we extracted DNA for molecular 18 9 a nalyses using 250 mg of sediment. To ensure a comprehensive representation of the 19 0 bac terial community at each horizon level, we pooled five replicates of DNA extraction per 19 1 c ore section for analysis. We purified DNA from 54 sediment samples according to the 19 2 man ufacturer's instructions using the E.Z.N.A Soil DNA kit (Omega Biotech, Inc. Georgia, 19 3 USA ). 19 4 19 5 Amp lification and Sequencing 19 6 Ge netic libraries were prepared using the primers 341F (CCTACGGGNGGCWGCAG) and 19 7 785 R (GACTACHVGGGTATCTAATCC) to amplify fragments of the V3-V4 regions of the 19 8 16S RNA gene (Ma et al., 2021). DNA samples were prepared for targeted sequencing with 19 9 the Quick-16S™ NGS Library Prep Kit (Zymo Research, Irvine, CA). The sequencing library 20 0 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.24.604998doi: bioRxiv preprint wa s prepared using real-time PCR machines to control cycles and limit PCR chimera 20 1 fo rmation. All PCR reactions were set to a final volume of 20 uL and treated with the 20 2 fo llowing thermal protocol: 10 min at 95°C for denaturation, then 20 cycles of 95°C for 30 20 3 se c, 55°C for 30 sec, and 72°C for 3 min. The final PCR products were cooled at 4°C prior to 20 4 quan tification with qPCR fluorescence readings and pooled together based on equal molarity. 20 5 T he final pooled library was cleaned up with the Select-a-Size DNA Clean & Concentrator™ 20 6 (Zym o Research, Irvine, CA), then quantified with TapeStation® (Agilent Technologies, 20 7 San ta Clara, CA) and Qubit® (Thermo Fisher Scientific, Waltham, WA). We used the 20 8 Z ymoBIOMICS® Microbial Community Standard (Zymo Research, Irvine, CA) as a positive 20 9 c ontrol for each DNA extraction. The final library was sequenced on Illumina® MiSeq™ 21 0 with a v3 reagent kit (600 cycles), and performed with >10% PhiX spike-in. 21 1 212 21 3 Dat a processing and taxonomy assignment 21 4 T he quality of the raw reads was analyzed using FastQC (v.0.12.0) (Andrews S, 2010). 21 5 Subse quently, TrimGalore (v.0.6.10) (Krueger et al., 2023) was used to remove Illumina 21 6 Uni versal Adapter (AGATCGGAAGAGC) from the reads. We utilized R (v4.2.2) (R Core 21 7 T eam., 2023) and DADA2 (v.1.16) (Callahan et al., 2016) to filter the reads with the 21 8 fo llowing parameters: the reads were truncated to 290 bases for forward and 200 bases for the 21 9 rev erse. This means that reads longer than these were not allowed. The maximum expected 22 0 e rrors (maxEE) which refers to the limit of errors in a sequence at a base level allowed per 22 1 rea d, was set at 5 errors per read. Reads with an expected error exceeding this value were 22 2 e xcluded. The ambiguous bases (N) were not allowed. For taxonomy assignment, we 22 3 e mployed the function assignTaxonomy from the package DADA2 (v.1.16) (Callahan et al., 22 4 201 6) with the database GreenGenes2 (v.2022.10) (McDonald et al., 2023). 22 5 22 6 P hylogenetic reconstruction 22 7 We utilized the AlignSeqs function from the DECIPHER package (Wright, 2017) to align the 22 8 16S RNA sequences. Subsequently, we employed the dist.ml function from Phangorn 22 9 (Sc hliep, 2011) to generate a distance matrix using the F81 substitution model. We then 23 0 e mployed the nj function to construct an unrooted tree using Neighbor Joining. 23 1 23 2 Comm unity analysis 23 3 We explored the microbial community accounting for horizon depth, zone and season 23 4 var iables. Visualization of each parameter option was achieved through non-metric 23 5 multi dimensional scaling (NMDS) analysis, employing Bray-Curtis distance. Based on the 23 6 e xploratory analysis, the sample zr2502_16_R1 which could potentially have skewed the 23 7 a nalysis, was excluded (Supplementary Figure 2). Subsequently, our dataset was filtered to 23 8 e ncompass only those amplicon sequence variants (ASVs) present in at least 10% of the 23 9 sa mples. 24 0 24 1 We employed the plot_richness function from phyloseq (McMurdie & Holmes, 2015) to 24 2 inve stigate alpha diversity patterns using the Chao1, Simpson, and Shannon indices. 24 3 Signi ficance testing was conducted through a Wilcoxon test, with adjustments for multiple 24 4 c omparisons using the Bonferroni method. Additionally, we employed ANOVA followed by 24 5 T ukey's test for further analysis. We employed microViz (v.0.12.1) (Barnett, Arts & Penders, 24 6 202 1), a comprehensive package to conduct both Principal Component Analysis (PCA) and 24 7 Princi pal Coordinate Analysis (PCoA) to elucidate the underlying patterns of variation in our 24 8 data set. Prior to analysis, the data underwent a centered log-ratio (clr) transformation for 24 9 PC A, enabling effective exploration of compositional differences while mitigating the impact 25 0 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.24.604998doi: bioRxiv preprint of spurious correlations. For PCoA, we did not apply any transformation, allowing for a 25 1 dire ct examination of dissimilarity among samples using Bray Curtis distance and UniFrac 25 2 dist ance. We corroborate the observed patterns performing a PERMANOVA test. These 25 3 a nalyses collectively provided insight into the structure and distribution of the data, 25 4 fac ilitating the interpretation of ecological dynamics. 25 5 25 6 Dif ferentially abundant bacteria were detected using the microbiomeMarker package (v.1.8) 25 7 (Cao , 2021), specifically by employing the run_aldex function with the glm_anova method, 25 8 a ssessing all taxonomic ranks. This analysis was conducted across all data pairs, resulting in a 25 9 list of differentially abundant taxa, which is available in Supplementary Table 1. 26 0 26 1 Me tabolic inference 26 2 PI CRUSt2 was used for the prediction of the functional abundances (Douglas et al., 2020), 26 3 whe re ASVs are placed into a reference tree with reference genomes; hidden-state prediction 26 4 a pproaches are used to infer the genomic content of sampled sequences using several 26 5 data bases such as enzymes (ECs), KEGG orthologues (KOs), cluster of orthologous genes 26 6 (CO Gs) and metabolic pathways (MetaCyc). To obtain and visualize differentially abundant 26 7 data (KOs), ggpicrust2 package (v.1.7.3) in R was used (Yang et al., 2023). The method 26 8 a pplied to asses differentially abundant data was ALDEx2 (Fernandes et al., 2013) 26 9 27 0 Re sults 27 1 Mic robial diversity dynamics across mangrove ecosystem degradation levels 27 2 Our study aimed to elucidate how microbial communities respond to various levels of 27 3 man grove ecosystem degradation, classified as non-degraded (ND), moderate degradation 27 4 (MD ), and degraded (D). Additionally, we accounted for seasonal variations and sampled 27 5 thre e different depths (5, 20, and 40 cm) as part of our experimental design (see Figure 1A). 27 6 We conducted a comprehensive analysis of microbial communities within the mangrove 27 7 e cosystem, utilizing 619,326 raw sequence reads, which led to the identification of 11,469 27 8 ASV s. Subsequently, we filtered out low-abundance taxa (present in less than 10% of the 27 9 sa mples), retaining 648 ASVs for further analysis. We hypothesized that increasing levels of 28 0 man grove ecosystem degradation would lead to a decline in microbial diversity, accompanied 28 1 by s hifts in microbial community structure influenced by both season and depth (Figure 1B). 28 2 T o assess this hypothesis, we analyzed alpha diversity patterns. The Chao1 index, revealed 28 3 significa nt differences between MD and D sites, where the MD site is more diverse. We 28 4 c onducted pairwise comparisons using the Wilcoxon rank sum test with continuity correction 28 5 to a ssess differences in Chao1 richness index among sites. The analysis showed a statistically 28 6 significa nt difference between the MD and D sites (Wilcoxon rank test, p = 0.017), while no 28 7 significa nt difference was observed between the ND and D sites. P-values were adjusted for 28 8 multi ple comparisons using the Bonferroni method. Consistent with the Chao1 index 28 9 fi ndings, the Shannon diversity index showed a noticeable decrease in diversity as 29 0 degr adation levels increased, which was significant among MD and D. However, the 29 1 Sim pson diversity index did not reveal significant differences between degradation sites; 29 2 how ever, it indicated elevated values, suggesting the dominance of certain microbial groups 29 3 (F igure 1C). 29 4 29 5 T o comprehend alpha diversity patterns, we divided the dataset by depth and assessed 29 6 dive rsity across seasons. Employing the Simpson index to quantify species dominance and 29 7 the Shannon index to consider species richness and evenness, we obtained a comprehensive 29 8 view of community diversity. Notably, the dry season consistently displayed higher diversity 29 9 than the flood season, supported by ANOVA and Tukey tests revealing significant seasonal 30 0 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.24.604998doi: bioRxiv preprint dif ferences (p-adjust = 0.007). Diversity at the 5 cm depth was linked to water loss, with 30 1 div ersity declining during the dry season according to degradation levels, contrasting with the 30 2 fl ood season. Statistical analysis confirmed significant mean differences between non-30 3 de graded and degraded sites for the Shannon diversity index (p-adjust = 0.011) and Simpson 30 4 ind ex (p-adjust = 0.023; Figure 2). At the 20 cm depth, diversity patterns were similar across 30 5 s easons, with degradation leading to decreased diversity, though not statistically significant. 30 6 N onetheless, the Simpson index highlighted higher dominance during the dry season, 30 7 ind icating prevalent groups during this period. At the deepest horizon of 40 cm, the dry 30 8 s eason appeared substantially more diverse than the flood season, possibly due to water 30 9 pr esence altering oxygen gradients, leading to collapse of anoxic microorganisms during 31 0 nor mal conditions. Statistical tests revealed significant diversity differences between 31 1 de graded and non-degraded sites (p-adjust = 0.060) using the Shannon index. However, in the 31 2 fl ood season, no significant differences were found using the Shannon index. Nevertheless, 31 3 s ignificant differences were noted between non-degraded and moderately degraded sites (p-31 4 a djust = 0.01), as well as between non-degraded and degraded sites (p-adjust = 0.009) using 31 5 the Simpson index. 31 6 31 7 31 8 F igure 1. Overview of Experimental Design and Alpha Diversity. A) Schematic 31 9 re presentation of the experimental design depicting three distinct mangrove sites, each 32 0 c haracterized by specific mangrove species. B) Hypothesized alterations in microbial 32 1 div ersity: anticipation of a reduction in microbial diversity (I), fluctuations based on seasonal 32 2 va riations (II), and shifts relative to depth (III). C) Comparative analysis of alpha diversity, 32 3 e mploying three distinct indices, across degradation sites, significance of the comparison * = 32 4 0. 05, ** = 0.01 and ns = not significant. 32 5 32 6 32 7 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.24.604998doi: bioRxiv preprint 32 8 F igure 2. Alpha diversity across degradation sites considering season and depth. The 32 9 Sh annon and Simpson indices depict diversity patterns, highlighting variations across 33 0 dif ferent degradation sites. 33 1 33 2 D epth and zone drive changes in microbial community structure 33 3 T o explore the microbial community dynamics, we used the clr-transformation and Principal 33 4 Com ponent Analysis (PCA), we delineated distinct microbial communities between ND and 33 5 M D sites, particularly noticeable during the dry season in the MD site (Figure 3A). Principal 33 6 Coor dinate Analysis (PCoA) on the untransformed data further revealed a distinct clustering 33 7 a ssociated with the ND site (Figure 3B), suggesting that there are characteristic microbial 33 8 gr oups in this condition. Moreover, depth analysis via PCoA's third dimension demonstrated 33 9 a consistent microbial community at the 40 cm horizon across degradation levels and seasons 34 0 (F igure 3C), contrasting with upper horizons where degradation level (Figure 3D), especially 34 1 in th e dry season (Rectangles in Figure 3D), significantly influences microbial composition. 34 2 PERMA NOVA analysis conducted with the adonis method reveal significant effects of the 34 3 thr ee environmental variables zone, season and depth (p = 0.0001), however based on the R-34 4 s quared value we observed that depth and zone explain large proportion of variation (0.07199 34 5 a nd 0.07264, respectively), while season only explained the 0.03531 (Supplementary Table 34 6 2) . 34 7 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.24.604998doi: bioRxiv preprint 34 8 F igure 3. Sample Distribution Based on Environmental Variables. A) Principal Component 34 9 A nalysis illustrating sample distribution by zone and season, differentiated by color and 35 0 s hape, respectively. B) Principal Coordinate Analysis showcasing the first and second 35 1 dim ensions. C and D) First and third dimensions of Principal Coordinate Analysis. Panel C 35 2 e mphasizes depth through color, while D highlights the zone. 35 3 35 4 W e examined the microbial community composition using UniFrac distance, which measures 35 5 ge netic distance and highlights differences in microbial communities based on lineage 35 6 c omposition. Our analysis, employing a principal coordinate analysis (PCoA) and 35 7 PERMA NOVA, demonstrated that our three variables play significant roles in shaping the 35 8 c ommunity structure (P < 0.001; see Supplementary Table 3). Regarding depth, microbial 35 9 lin eages from the 40 cm horizon stand out from the rest of the community. Additionally, they 36 0 a ppear to be closely associated with the flood season (Figure 4A,C). Meanwhile, our findings 36 1 s uggest that the degraded zone harbors a distinct composition of lineages compared to other 36 2 de gradation stages (Figure 4B). 36 3 36 4 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.24.604998doi: bioRxiv preprint F igure 4. Microbial community structure depicted using UniFrac distances. Each panel 36 5 re presents a different variable: A) Depth, B) Zone, and C) Season, color-coded accordingly. 36 6 36 7 Shif ts in Gammaproteobacteria and Firmicutes 36 8 T o elucidate the microbial diversity contributing to differences in community structure, we 36 9 re constructed a phylogenetic tree. This analysis unveiled distinct patterns among 37 0 G ammaproteobacteria and Firmicutes across degradation sites, with Gammaproteobacteria 37 1 pr evalent in both ND and MD sites, while Firmicutes exhibited greater diversity and spread 37 2 a cross all degradation stages (Figure 5). Notably, within Proteobacteria, particularly the 37 3 V ibrionaceae family, we observed heightened prevalence in both ND and MD sites. 37 4 H owever, a subset of Vibrionaceae was exclusively confined to the ND site, contrasting 37 5 s harply with the dominance of Marinomonadaceae and Halomonadaceae in the MD. 37 6 A dditionally, specific lineages of Nitrospirota were solely identified in the D site, suggesting 37 7 a potential association with our observed alpha diversity patterns that show high dominance. 37 8 Conv ersely, within the Firmicutes phylum, clusters associated with Lactococcus from the 37 9 fa mily Streptococcaceae were uniformly distributed across all sites, indicating a significant 38 0 ro le in the ecosystem. Regarding seasonal dynamics, Exiguobacteraceae and Planococcaceae 38 1 w ere predominantly concentrated in MD sites. Interestingly, the prevalence of 38 2 M arinomonadaceae and Halomonadaceae in MD sites correlated with the dry season, while 38 3 E xiguobacteraceae and Planococcaceae thrived during the flood season. Furthermore, our 38 4 s patial analysis revealed depth-related distribution patterns, particularly pronounced in ND 38 5 a nd MD sites. Gammaproteobacteria, linked with Marinomonadaceae, Halomonadaceae, and 38 6 V ibrionaceae, were predominantly found at the surface (5 cm depth), whereas Firmicutes 38 7 a ssociated with Lactococcus were notably prevalent at 40 cm depth, spanning all zones and 38 8 s easons. This observation suggests a central role for Lactococcus-related Firmicutes in 38 9 e cosystem dynamics. 39 0 39 1 F igure 5. The phylogenetic reconstruction of 16S rRNA Amplicon Single Variant (ASVs) 39 2 pr esent in Estero Pargo. The inner circle delineates the sampling zone, season, and depth. 39 3 E ach bar indicates the abundance of the corresponding ASV. Clade color indicates taxonomy 39 4 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.24.604998doi: bioRxiv preprint a t the family level. Above the names we highlighted the zone and season from which those 39 5 fam ilies were found. 39 6 39 7 We utilized the analysis of differential abundance considering sample and scale variations 39 8 (A LDEx2; see Supplementary Table 1) to examine disparities in abundance across diverse 39 9 e nvironmental parameters. Among sites categorized as ND and MD, certain taxonomic 40 0 gr oups emerged as distinctive to the ND site such as the phyla Desulfobacterota and 40 1 Chl oroflexota, along with the ASVs classified under the species Vibrio rumoiensis, as 40 2 pr eviously observed in the phylogenetic tree (Figure 4). When contrasting MD with D site, 40 3 line ages from Firmicutes such as Rossellomorea and Lactococcus predominated, alongside 40 4 Ha lomonadaceae, indicative of the MD site. Additionally, lineages from the phylum 40 5 Nitros pirota, specifically from the class Thermodesulfovibrionia, distinguished the D site. 40 6 Seas onal dynamics revealed a pronounced pattern of higher abundance of lineages from the 40 7 fam ilies Marinomonadaceae and Halomonadaceae during the dry season (Figure 5), 40 8 c orrelating with the MD site. On the contrary, during the flood season, Firmicutes and 40 9 De sulfobacteriota were predominant. Depth-related variations were notable, particularly 41 0 betw een the surface (5 cm) and deeper horizons (20 cm and 40 cm). Firmicutes, 41 1 Ac tinobacteriota, and Desulfobacteriota were predominant at the surface, while Nitrospirota, 41 2 e specially Thermodesulfovibriota, and Pseudomonadaceae were prevalent at 40 cm. 41 3 41 4 Sea son and zone influence on metabolism patterns 41 5 Phylog enetic placement of ASVs in a genome tree revealed that the dry season exhibited 41 6 highe r diverse presence of enzymes (ECs), metabolic pathways, KEGG orthologues (KOs) 41 7 a nd COGs compared to the flood season (Figure 6) which is in agreement with what we 41 8 obse rved regarding taxonomic diversity (Figure 2). The differential analysis of KO's 41 9 a bundance indicated a significantly greater statistical difference in several types of 42 0 met abolism such as amino acid metabolism and the degradation of xenobiotic compounds 42 1 (F igure 6B and C) in the non-degraded sites compared to the degraded and moderately 42 2 degr aded ones. 42 3 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.24.604998doi: bioRxiv preprint 42 4 F igure 6. Metabolism overview. A) Abundance of predicted enzymes categorized as COG, 42 5 E C, KO and Pathways. Differential abundance analysis (ALDEx2) in KO’s accounting for 42 6 the different zones: B) MD vs ND and C) D vs ND. 42 7 42 8 42 9 43 0 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.24.604998doi: bioRxiv preprint Disc ussion 43 1 Our study aimed to explore how microbial communities respond to varying degrees of 43 2 degr adation within mangrove ecosystems, characterized by altered hydrological connectivity 43 3 a nd increasing urbanization. Our findings show the dynamic nature of surface sediment 43 4 laye rs, which exhibit heightened susceptibility to change. This was particularly evident at the 43 5 non-degr aded and moderately degraded sites. At these sites, distinct microbial compositions 43 6 e merged, with one site dominated by Vibrionaceae and the other by bacteria like 43 7 Ma rinomonadaceae and Halomonadaceae, displaying clear seasonal patterns. Despite this, 43 8 c ertain bacteria persist in deeper sediment layers, potentially playing key roles as early 43 9 c olonizers in ecosystem restoration efforts. Furthermore, our observations highlight the 44 0 impa ct of water flow on microbial diversity, being the dry season more diverse than the 44 1 fl ood. 44 2 44 3 As anticipated, we noted a decrease in microbial diversity correlated with the degradation 44 4 leve l. However, we found no reduction in functional diversity according to the degradation 44 5 leve l. This led us to propose that the key molecular functions may be conserved through the 44 6 loss of hydrological connectivity. This could be understood by the term of functional 44 7 red undancy, which refers to the coexistence of multiple distinct taxa or genomes capable of 44 8 per forming the same focal biochemical function (Louca et al., 2018), facilitating the 44 9 mai ntenance of ecosystem functions despite a loss in taxonomic diversity (Li et al. 2021; 45 0 Ha roon et al. 2013; Waite et al. 2020; Wang et al. 2023). However, regarding seasonality, the 45 1 dr y season consistently exhibited higher diversity compared to the flooding season, with a 45 2 par allel pattern observed in metabolic diversity; this pattern may be related to the dominance 45 3 of microbial groups specifically related to Proteobacteria. For example, bacteria residing in 45 4 man grove sediments play a pivotal role in denitrification, the process through which they 45 5 c onvert nitrate (NO3-) a nd nitrite (NO2-) into nitrogen gas (N2). Various bacterial taxa, such as 45 6 Pseud omonas, Paracoccus, and Bacillus, contribute to this essential process at different 45 7 se diment depths within mangrove ecosystems (Haroon et al., 2013). Similarly, sulfate-45 8 red ucing bacteria (SRB) like Desulfobacter, Desulfovibrio, and Desulfuromonas are crucial 45 9 in m angrove sediments, where they convert sulfate (SO42-) in to sulfide (S2-) . These SRBs can 46 0 be found at various sediment depths, actively participating in sulfate reduction processes 46 1 (W aite et al., 2020). Moreover, bacteria involved in carbon metabolism, such as Clostridium, 46 2 Cellu lomonas, and Acetobacterium, contribute to cellulose degradation and fermentation 46 3 pr ocesses (Wang et al., 2023) within mangrove sediments. These bacteria also exhibit 46 4 fu nctional redundancy across sediment depths, indicating their importance in maintaining 46 5 e cosystem function. Furthermore, their distribution is influenced by spatial contiguity and 46 6 rec iprocating tidal flows (Li et al., 2021). We are aware of the bias that metabolic prediction 46 7 throug h 16S RNA comes with, however, we believe it gives us some general patterns 46 8 reg arding what we can expect and help us to raise new hypotheses regarding the metabolic 46 9 pote ntial in the community structure. 47 0 47 1 I n mangrove ecosystems, the interplay of water dynamics plays a crucial role in shaping 47 2 mic robial communities, with significant implications for their diversity and composition 47 3 (Lui s et al., 2019). During the flooding season, water flow fluctuations influence the 47 4 tr ansition between aerobic and anaerobic conditions, thereby impacting microbial diversity. 47 5 Obse rvations at different sediment horizons revealed interesting patterns: at 5 cm depth, 47 6 dive rsity tends to increase with degradation levels, possibly due to input from external water 47 7 sour ces like the ocean or lagoon. A possible source of diversity variation is the tidal 47 8 impor tation of organic matter. An increase in bacterial diversity has been observed in the top 47 9 laye rs of mangrove sediments due to the input of organic matter that tidal regimes allow, 48 0 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.24.604998doi: bioRxiv preprint c reating microhabitats that increase the diversity of heterotrophic bacteria (Zhu et al. 2018). 48 1 Sim ilarly, at 40 cm, diversity follows a similar trend, yet notably lower diversity at the ND 48 2 site suggests limitations imposed by anaerobic conditions induced by water saturation, which 48 3 like ly restrict the presence of aerobic-obligate bacteria. Hence, variation in hydrology 48 4 e merges as a key driver shaping microbial communities in these ecosystems. In the dry 48 5 se ason, when water flow ceases, bacterial diversity undergoes significant transformations 48 6 depe nding on the duration of isolation from external water sources. Prolonged isolation leads 48 7 to pr onounced shifts in microbial diversity. However, when floods reintroduce water flow, 48 8 mic robial diversity decreases as new species from external sources dominate the 48 9 e nvironment. Similar to our findings, other studies demonstrated that in freshwater systems, 49 0 the bacterial communities respond to dry and wet seasons by changing their functional and 49 1 taxo nomic diversity, showing an increased alpha diversity during the dry season (Ren et al., 49 2 201 9). Such results can be associated with higher nutrient concentrations, lower water 49 3 diluti on capacity, and reduced degradation of organic matter (Ren et al. 2019). The 49 4 a vailability and susceptibility to change of dissolved organic matter are critical factors 49 5 influe ncing bacterial community composition, particularly in upper soil horizons. 49 6 Und erstanding these dynamics provides insights into the intricate relationships between water 49 7 dyn amics and microbial communities in mangrove ecosystems. 49 8 49 9 Due to the intermittent flooding caused by tides in mangroves, the availability of nutrients 50 0 a nd the concentration of oxygen and salinity, among other factors, vary significantly (Basak 50 1 e t al., 2016). It is known that the availability of nutrients determines the structure of the 50 2 mic robial communities. For example, S concentration and anoxic conditions can increase the 50 3 a bundance of Deltaproteobacteria and Epsilonproteobacteria as organisms representative of 50 4 these phyla reduce or oxidize S compounds (Lin et al., 2019). It is known that the percentage 50 5 of organic matter, pH, and seasonal water saturation influence the composition and structure 50 6 of the microbial community and the abundance of specific functional groups (Ansola, Arroyo 50 7 & Sáen z de Miera, 2014; Arroyo, Sáenz de Miera & Ansola, 2015; Ding et al., 2015). 50 8 T herefore, seasonal variation in hydrological connectivity, water flow, or stagnation, mainly 50 9 a t site D, influenced the structure and composition of the bacterial community, where we 51 0 obse rved an increase in the abundance of Nitrospira, a bacteria that participates in the 51 1 nitrog en cycle, which was abundant on this site, suggesting that an increase of ammonium 51 2 e xist in this environment (Daims & Wagner, 2018; Meng et al., 2022). 51 3 51 4 I n non-degraded mangrove ecosystems, a higher abundance of KO's associated with amino 51 5 a cid metabolism reflects optimal environmental conditions. Amino acids serve pivotal roles 51 6 a s precursors for essential biochemical processes including protein synthesis, nucleotide 51 7 fo rmation, chlorophyll production, and hormone regulation. The significant difference in the 51 8 rel ative abundance of these KO's in the ND compared to MD and D zones, indicates efficient 51 9 nitrog en utilization for growth and development. Therefore, the greater presence of amino 52 0 a cid metabolism-related KO's in non-degraded mangroves signifies a balanced and healthy 52 1 e cosystem state supporting vigorous plant metabolism and ecological stability. (Shah et al., 52 2 2021; Ningsih et al., 2020). 52 3 52 4 On the other hand, our results were similar to those reported for riverine flood plains, where 52 5 the abundance of anaerobic bacteria increased when the system remained flooded for a more 52 6 e xtended period (Argiroff et al., 2017). We observed higher diversity and dominance of 52 7 a erobic bacteria such as Firmicutes and Proteobacteria in the dry season, which might be 52 8 signal s of a loss in hydrological connectivity and water flow. 52 9 53 0 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.24.604998doi: bioRxiv preprint T idal influence may shape bacterial diversity by introducing taxa from external habitats (Zhu 53 1 e t al. 2018). As the ND site was consistently influenced by tidal variation in Estero Pargo, it 53 2 showe d distinctive groups of Vibrionaceae. V ibrio species are natural inhabitants of aquatic 53 3 e nvironments, including sediments, estuaries, and marine coastal waters. They can live in 53 4 both sa line waters with 30-35 ppt and low saline environments (Wong & Griffin, 2018). As 53 5 we ll, Vibrio species have been found as part of the microbiota of R. mangle (Gomes et al., 53 6 201 4) which may explain its presence in the ND site, which is dominated by this family. 53 7 Vib rio species isolated from salt marshes have shown the capacity of fixing nitrogen, which 53 8 sugge sts that the presence of Vibrio in the ND site could be also playing this role (Criminger 53 9 e t al., 2007). And other Vibrio directly isolated from R. mangle have been shown to be able to 54 0 inhibit bac teria and fungal phytopathogens (Rameshkumar & Nair, 2009). Recent increases in 54 1 V ibrio abundance have been related to human activities that can negatively affect aquatic or 54 2 mar ine habitats (Narayanan et al., 2020). For example, aquaculture is thought to be an 54 3 a ctivity that can increase the abundance of Vibrio species because organisms such as rotifers 54 4 a nd Artemia, used as live feed, can be hosts of several Vibrio species (Sanches-Fernandes et 54 5 a l., 2022). In addition, high densities of cultured fish can, in turn, increase the abundance of 54 6 V ibrio species that can be released to natural environments when ponds are cleaned and water 54 7 is disc harged to water courses (Sampaio et al., 2022). In our study area, the growing fish 54 8 far ming activities could contribute to the recorded Vibrio abundance in Estero Pargo. 54 9 55 0 At the MD site, we observed a notable shift in microbial diversity. During the dry season, 55 1 Ha lomonas and Marinomonas were notably abundant, whereas during the flood season, 55 2 E xigoubacterium emerged as dominant. Although all three groups of bacteria are recognized 55 3 a s halophilic, changes in salt concentration may favor some lineages over others (Edbeib, 55 4 Wa hab & Huyop, 2016; Chen et al., 2017). Mangrove sediments show oxygen gradients; 55 5 typic ally, the upper layers have high oxygen concentrations, while the lower layers have less 55 6 oxy gen and high rates of other elements (sulfur, nitrogen, and methane), suggesting multiple 55 7 ver tical niches. Therefore, the diversity changes at the MD site could reflect these multiple 55 8 nich es. Initially, we speculated that the MD site might resemble a pristine mangrove 55 9 e nvironment due to the high presence of Vibrionaceae, as in the ND site. However, given the 56 0 tight c orrelation between degradation levels and decreased water flow, it is plausible that the 56 1 MD site reflects how arid environmental conditions act as a selection factor for halotolerant 56 2 line ages. 56 3 56 4 Hal omonas has been found in other mangrove ecosystems as part of the core microbiome, it 56 5 has been suggested that the production of ectoine an osmolyte molecule might help the host 56 6 to c ope with osmotic stress. Therefore, considering the holobiont hypothesis, this bacteria 56 7 c ould be interacting with other organisms within the ecosystem, probably the mangrove tree 56 8 to pr ovide this type of protection (Wainwright et al., 2023). On the other hand, Marinomonas 56 9 has been shown to participate in the conversion of dimethylsulfoniopropionate (DMSP) to 57 0 dime thyl sulfide (DMS) which has been shown in corals that they help them in the sulfur 57 1 c ycle, therefore we believe that could be also happening here (Wainwright et al., 2023). 57 2 57 3 Our study revealed enrichment of halophilic bacteria in certain conditions, this suggests the 57 4 pote ntial use of some halophilic bacteria to protect the plants from saline stress as it has been 57 5 shown before (Bharti et al., 2015). 57 6 57 7 T he MD site dominated by Avicieana germinias showed a different microbial composition 57 8 c ompared to what has been described previously, Desulfatiglans has been described as an 57 9 a bundant lineage associated to this type of tree, however in our study we did not find this to 58 0 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.24.604998doi: bioRxiv preprint be any abundant, similar thing happened with Ignavibacteriales, Phaeodactylibactes and 58 1 SAR324, which were associates to A. germinis (Gómez-Acata et al., 2023). This could be 58 2 rel ated to 1) we sampled a human-impacted ecosystem. Therefore, that could change the 58 3 mic robial structure compared to other studies; 2) differences in substrate composition, which 58 4 in pr evious studies is karstic; and 3) our low sequence coverage, which may mislead our 58 5 c onclusions. However, this also shows the uniqueness of microbial communities associated 58 6 with spe cific trees. 58 7 58 8 De spite the higher variability in species diversity observed in the upper sediment horizons, 58 9 we observed that the diversity at 40 cm was preserved in all the degradation levels. This 59 0 res ult suggests that environmental conditions stay stable at greater depths for longer, allowing 59 1 the bacterial community to remain unchanged over more extended periods. We believe this 59 2 c ommunity could be used in restoration efforts as a seed community to enhance the process. 59 3 I nterestingly, Lactococcus was observed as a dominant lineage at 40 cm, this lineage is 59 4 usua lly constrained to habitats like food related, however some strains have been isolated 59 5 fr om mangrove ecosystems and their production of bacteriocin which are molecules secreted 59 6 to pr event the growth of other bacteria, has been used in other plant systems to prevent 59 7 mic robial growth (Hwanhlem et al., 2013; Kleerebezem et al., 2020). 59 8 59 9 F urther research is needed to fully understand the metabolic potential underlying our 60 0 hyp othesis. We are eager to explore a metagenomic approach to testing our theories, delving 60 1 dee per into the microbial communities involved. Additionally, we aim to conduct culturing 60 2 e xperiments focusing on specific bacterial strains we believe could play a crucial role in 60 3 e cological restoration efforts. This multidimensional strategy will not only enrich our 60 4 und erstanding of the ecosystem dynamics but also pave the way for practical applications in 60 5 e nvironmental conservation. 60 6 60 7 Conc lusion 60 8 He re we investigated the impact of lost of hydrological connectivity in the microbial 60 9 c ommunities of a mangrove ecosystem situated at Ciudad del Carmen, Campeche in the Gulf 61 0 of Mexico, we found that this lost of connectivity at first in the moderate degraded site is 61 1 a ccompanied by the selection of key microbes that can cope with dry and highly halophilic 61 2 c onditions such as Halomonas, Marinomonas and Exiguobacterium, however, there were 61 3 patt erns associated with seasons. The upper levels of the sediments are highly influenced by 61 4 huma n activities and shape microbial communities, however deeper horizons, especially 61 5 dur ing the dry season are highly diverse and can be preserved longer. We believe this 61 6 c ommunity could be used in restoration efforts as a seed community to enhance the process 61 7 of a new mangrove forest restoration. 61 8 61 9 Compet ing Interests 62 0 Non e 62 1 622 Auth or Contributions 62 3 M. V.R.L, R.P.C, A.Z.J, and J.C.C.D conceived the study and analysis. R.P.C, A.Z.J, A.P.D, 62 4 O. C.H, and J.C.C.D conducted field work and DNA extraction. M.V.R.L and J.C.C.D wrote 62 5 the first draft of the manuscript. M.V.R.L and L.D.P.G performed bioinformatic analysis. 62 6 S.E.H .P performed the metabolic prediction analysis. M.V.R.L prepared figures and tables. 62 7 R.P.C, A.Z.J, and J.C.C.D got funding. All authors contributed and approved the final version 62 8 of the manuscript. 62 9 63 0 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.24.604998doi: bioRxiv preprint DNA Deposition 63 1 Amplicon sequences are publicly available through the MG-RAST project entitled Estero Pargo under 632 the link: https://www.mg-rast.org/linkin.cgi?project=mgp104729. 63 3 634 Dat a Availability 63 5 Amplicon sequences are publicly available through the MG-RAST project entitled Estero Pargo under 636 the link: https://www.mg-rast.org/linkin.cgi?project=mgp104729. 63 7 638 F unding. This research was self funded. 63 9 64 0 Ac knowledgements. 64 1 We thank Josefina Santos Ramírez, Tomás Zaldívar Jiménez, Ricardo Ortegón Herrera, 64 2 Ma rio Alejandro Gómez Ponce, Hernán Álvarez Guillén, Andrés Reda Deara their assistance 64 3 with logistic s and field data collection; added a Citlalli Garrido Abreu for their lab analysis. 64 4 64 5 Re ferences 64 6 Akr am H, Hussain S, Mazumdar P, Chua KO, Butt TE, Harikrishna JA. 2023. Mangrove 64 7 hea lth: A review of functions, threats, and challenges associated with mangrove 64 8 m anagement practices. Forests, Trees and Livelihoods 14:1698. 64 9 Alla rd SM, Costa MT, Bulseco AN, Helfer V, Wilkins LGE, Hassenrück C, Zengler K, 65 0 Zim mer M, Erazo N, Mazza Rodrigues JL, Duke N, Melo VMM, Vanwonterghem I, 65 1 Junca H, Makonde HM, Jiménez DJ, Tavares TCL, Fusi M, Daffonchio D, Duarte CM, 65 2 Pe ixoto RS, Rosado AS, Gilbert JA, Bowman J. 2020. Introducing the Mangrove 65 3 Mi crobiome Initiative: Identifying Microbial Research Priorities and Approaches To 65 4 Bett er Understand, Protect, and Rehabilitate Mangrove Ecosystems. mSystems 5. DOI: 65 5 10 .1128/mSystems.00658-20. 65 6 Ansol a G, Arroyo P, Sáenz de Miera LE. 2014. Characterisation of the soil bacterial 65 7 com munity structure and composition of natural and constructed wetlands. The Science 65 8 of the total environment 473-474:63–71. 65 9 Ar giroff WA, Zak DR, Lanser CM, Wiley MJ. 2017. Microbial Community Functional 66 0 Pot ential and Composition Are Shaped by Hydrologic Connectivity in Riverine 66 1 F loodplain Soils. Microbial ecology 73:630–644. 66 2 Ar royo P, Sáenz de Miera LE, Ansola G. 2015. Influence of environmental variables on the 66 3 str ucture and composition of soil bacterial communities in natural and constructed 66 4 wetl ands. The Science of the total environment 506-507:380–390. 66 5 Babr aham Bioinformatics - FastQC A Quality Control tool for High Throughput Sequence 66 6 Data . Available at http://www.bioinformatics.babraham.ac.uk/projects/fastqc/ (accessed 66 7 Janua ry 8, 2024). 66 8 Barn ett D, Arts I, Penders J. 2021. microViz: an R package for microbiome data visualization 66 9 and statistics. Journal of open source software 6:3201. 67 0 Bas ak P, Pramanik A, Sengupta S, Nag S, Bhattacharyya A, Roy D, Pattanayak R, Ghosh A, 67 1 Ch attopadhyay D, Bhattacharyya M. 2016. Bacterial diversity assessment of pristine 67 2 m angrove microbial community from Dhulibhashani, Sundarbans using 16S rRNA gene 67 3 t ag sequencing. Genomics data 7:76–78. 67 4 Bas han Y, Holguin G. 2002. Plant growth-promoting bacteria: a potential tool for arid 67 5 m angrove reforestation. Trees 16:159–166. 67 6 Bhar ti N, Barnawal D, Maji D, Kalra A. 2015. Halotolerant PGPRs Prevent Major Shifts in 67 7 I ndigenous Microbial Community Structure Under Salinity Stress. Microbial ecology 67 8 70 :196–208. 67 9 Bonat erra A, Badosa E, Daranas N, Francés J, Roselló G, Montesinos E. 2022. Bacteria as 68 0 Bi ological Control Agents of Plant Diseases. Microorganisms 10. DOI: 68 1 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.24.604998doi: bioRxiv preprint 10 .3390/microorganisms10091759. 68 2 Callah an BJ, McMurdie PJ, Rosen MJ, Han AW, Johnson AJA, Holmes SP. 2016. DADA2: 68 3 Hi gh-resolution sample inference from Illumina amplicon data. Nature methods 13:581–68 4 58 3. 68 5 Cao Y. 2021. microbiomeMarker. Bioconductor. DOI: 68 6 10 .18129/B9.BIOC.MICROBIOMEMARKER. 68 7 Chatt ing M, Al-Maslamani I, Walton M, Skov MW, Kennedy H, Husrevoglu YS, Le Vay L. 68 8 20 22. Future mangrove carbon storage under climate change and deforestation. 68 9 Fr ontiers in marine science 9. DOI: 10.3389/fmars.2022.781876. 69 0 Chen Y-H, Lu C-W, Shyu Y-T, Lin S-S. 2017. Revealing the Saline Adaptation Strategies of 69 1 t he Halophilic Bacterium Halomonas beimenensis through High-throughput Omics and 69 2 Tra nsposon Mutagenesis Approaches. Scientific reports 7:13037. 69 3 Contre ras Ruiz Esparza A, Douillet P, Zavala-Hidalgo J. 2014. Tidal dynamics of the 69 4 Te rminos Lagoon, Mexico: observations and 3D numerical modelling. Ocean Dynamics 69 5 64 :1349–1371. 69 6 Cor onado-Molina C, Alvarez-Guillen H, Day JW Jr, Reyes E, Perez BC, Vera-Herrera F, 69 7 Twille y R. 2012. Litterfall dynamics in carbonate and deltaic mangrove ecosystems in 69 8 t he Gulf of Mexico. Wetlands Ecology and Management 20:123–136. 69 9 Crimin ger JD, Hazen TH, Sobecky PA, Lovell CR. 2007. Nitrogen fixation by Vibrio 70 0 par ahaemolyticus and its implications for a new ecological niche. Applied and 70 1 env ironmental microbiology 73:5959–5961. 70 2 Da ims H, Wagner M. 2018. Nitrospira. Trends in microbiology 26:462–463. 70 3 Din g X, Peng X-J, Jin B-S, Xiao M, Chen J-K, Li B, Fang C-M, Nie M. 2015. Spatial 70 4 di stribution of bacterial communities driven by multiple environmental factors in a 70 5 bea ch wetland of the largest freshwater lake in China. Frontiers in microbiology 6:129. 70 6 Dom ínguez-Núñez JA, Berrocal-Lobo M. 2021. Application of microorganisms in forest 70 7 pl ant. In: Biofertilizers. Elsevier, 265–287. 70 8 Dou glas GM, Maffei VJ, Zaneveld JR, Yurgel SN, Brown JR, Taylor CM, Huttenhower C, 70 9 La ngille MGI. 2020. PICRUSt2 for prediction of metagenome functions. Nature 71 0 bi otechnology 38:685–688. 71 1 E dbeib MF, Wahab RA, Huyop F. 2016. Halophiles: biology, adaptation, and their role in 71 2 dec ontamination of hypersaline environments. World journal of microbiology & 71 3 bi otechnology 32:135. 71 4 E dlund A, Hårdeman F, Jansson JK, Sjöling S. 2008. Active bacterial community structure 71 5 alon g vertical redox gradients in Baltic Sea sediment. Environmental microbiology 71 6 10 :2051–2063. 71 7 E llison AM, Felson AJ, Friess DA. 2020. Mangrove rehabilitation and restoration as 71 8 expe rimental adaptive management. Frontiers in marine science 7. DOI: 71 9 10 .3389/fmars.2020.00327. 72 0 F ernandes AD, Macklaim JM, Linn TG, Reid G, Gloor GB. 2013. ANOVA-like differential 72 1 expr ession (ALDEx) analysis for mixed population RNA-Seq. PloS one 8:e67019. 72 2 Gom es NCM, Cleary DFR, Pinto FN, Egas C, Almeida A, Cunha A, Mendonça-Hagler LCS, 72 3 Sm alla K. 2010. Taking root: enduring effect of rhizosphere bacterial colonization in 72 4 m angroves. PloS one 5:e14065. 72 5 Gom es NCM, Cleary DFR, Pires ACC, Almeida A, Cunha A, Mendonça-Hagler LCS, 72 6 Sm alla K. 2014. Assessing variation in bacterial composition between the rhizospheres 72 7 of t wo mangrove tree species. Estuarine, coastal and shelf science 139:40–45. 72 8 Góm ez-Acata ES, Teutli C, Falcón LI, García-Maldonado JQ, Prieto-Davó A, Yanez-72 9 Monta lvo A, Cadena S, Chiappa-Carrara X, Herrera-Silveira JA. 2023. Sediment 73 0 m icrobial community structure associated to different ecological types of mangroves in 73 1 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.24.604998doi: bioRxiv preprint Celes tún, a coastal lagoon in the Yucatan Peninsula, Mexico. PeerJ 11:e14587. 73 2 Hol guin G, Vazquez P, Bashan Y. 2001. The role of sediment microorganisms in the 73 3 produc tivity, conservation, and rehabilitation of mangrove ecosystems: an overview. 73 4 Biol ogy and fertility of soils 33:265–278. 73 5 Hwa nhlem N, Biscola V, El-Ghaish S, Jaffrès E, Dousset X, Haertlé T, H-Kittikun A, 73 6 Ch obert J-M. 2013. Bacteriocin-Producing Lactic Acid Bacteria Isolated from 73 7 Mangr ove Forests in Southern Thailand as Potential Bio-Control Agents: Purification 73 8 and Characterization of Bacteriocin Produced by Lactococcus lactis subsp. lactis 73 9 KT2W 2L. Probiotics and antimicrobial proteins 5:264–278. 74 0 Ka mali B, Hashim R. 2011. Mangrove restoration without planting. Ecological engineering 74 1 37 :387–391. 74 2 Ka therine Vande Velde, Jean Hugé , Daniel A. Friess,Nico Koedam, Farid Dahdouh-Guebas. 74 3 20 19. Stakeholder discourses on urban mangrove conservation and management. Ocean 74 4 & coastal management 178:104810. 74 5 Klee rebezem M, Bachmann H, van Pelt-KleinJan E, Douwenga S, Smid EJ, Teusink B, van 74 6 Mas trigt O. 2020. Lifestyle, metabolism and environmental adaptation in Lactococcus 74 7 l actis. FEMS microbiology reviews 44:804–820. 74 8 L ai J, Cheah W, Palaniveloo K, Suwa R, Sharma S. 2022. A systematic review of the 74 9 ph ysicochemical and microbial diversity of well-preserved, restored, and disturbed 75 0 m angrove forests: What is known and what is the way forward? Forests, Trees and 75 1 Live lihoods 13:2160. 75 2 L ee SY, Primavera JH, Dahdouh-Guebas F, McKee K, Bosire JO, Cannicci S, Diele K, 75 3 F romard F, Koedam N, Marchand C, Mendelssohn I, Mukherjee N, Record S. 2014. 75 4 Ec ological role and services of tropical mangrove ecosystems: a reassessment. Global 75 5 ec ology and biogeography: a journal of macroecology 23:726–743. 75 6 L in X, Hetharua B, Lin L, Xu H, Zheng T, He Z, Tian Y. 2019. Mangrove Sediment 75 7 Mi crobiome: Adaptive Microbial Assemblages and Their Routed Biogeochemical 75 8 Proce sses in Yunxiao Mangrove National Nature Reserve, China. Microbial ecology 75 9 78 :57–69. 76 0 L ouca S, Polz MF, Mazel F, Albright MBN, Huber JA, O’Connor MI, Ackermann M, Hahn 76 1 AS, S rivastava DS, Crowe SA, Doebeli M, Parfrey LW. 2018. Function and functional 76 2 red undancy in microbial systems. Nature Ecology & Evolution 2:936–943. 76 3 L uis P, Saint-Genis G, Vallon L, Bourgeois C, Bruto M, Marchand C, Record E, Hugoni M. 76 4 20 19. Contrasted ecological niches shape fungal and prokaryotic community structure in 76 5 m angroves sediments. Environmental microbiology 21:1407–1424. 76 6 Ma i Z, Ye M, Wang Y, Foong SY, Wang L, Sun F, Cheng H. 2021. Characteristics of 76 7 Mi crobial Community and Function With the Succession of Mangroves. Frontiers in 76 8 mi crobiology 12:764974. 76 9 Ma X-X, Jiang Z-Y, Wu P, Wang Y-F, Cheng H, Wang Y-S, Gu J-D. 2021. Effect of 77 0 m angrove restoration on sediment properties and bacterial community. Ecotoxicology 77 1 30 :1672–1679. 77 2 Mc Donald D, Jiang Y, Balaban M, Cantrell K, Zhu Q, Gonzalez A, Morton JT, Nicolaou G, 77 3 Pa rks DH, Karst SM, Albertsen M, Hugenholtz P, DeSantis T, Song SJ, Bartko A, 77 4 Havu linna AS, Jousilahti P, Cheng S, Inouye M, Niiranen T, Jain M, Salomaa V, Lahti 77 5 L, Mir arab S, Knight R. 2023. Greengenes2 unifies microbial data in a single reference 77 6 t ree. Nature biotechnology:1–4. 77 7 Mc Kee KL, Faulkner PL. 2000. Restoration of biogeochemical function in mangrove forests. 77 8 Re storation Ecology 8:247–259. 77 9 Mc Murdie PJ, Holmes S. 2015. Shiny-phyloseq: Web application for interactive microbiome 78 0 ana lysis with provenance tracking. Bioinformatics 31:282–283. 78 1 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.24.604998doi: bioRxiv preprint Me ng S, Peng T, Wang H, Huang T, Gu J-D, Hu Z. 2022. Evaluation of PCR primers for 78 2 detec ting the distribution of nitrifiers in mangrove sediments. Applied microbiology and 78 3 bi otechnology 106:5811–5822. 78 4 Mor ris JJ. 2018. What is the hologenome concept of evolution? F1000Research 7. DOI: 78 5 10 .12688/f1000research.14385.1. 78 6 Palit K, Rath S, Chatterjee S, Das S. 2022. Microbial diversity and ecological interactions of 78 7 m icroorganisms in the mangrove ecosystem: Threats, vulnerability, and adaptations. 78 8 Env ironmental science and pollution research international 29:32467–32512. 78 9 Pér ez-Ceballos R, Zaldívar-Jiménez A, Canales-Delgadillo J, López-Adame H, López-79 0 Port illo J, Merino-Ibarra M. 2020. Determining hydrological flow paths to enhance 79 1 resto ration in impaired mangrove wetlands. PloS one 15:e0227665. 79 2 Pura hong W, Sadubsarn D, Tanunchai B, Wahdan SFM, Sansupa C, Noll M, Wu Y-T, 79 3 Bu scot F. 2019. First Insights into the Microbiome of a Mangrove Tree Reveal 79 4 Sig nificant Differences in Taxonomic and Functional Composition among Plant and Soil 79 5 Co mpartments. Microorganisms 7. DOI: 10.3390/microorganisms7120585. 79 6 Rame shkumar N, Nair S. 2009. Isolation and molecular characterization of genetically 79 7 di verse antagonistic, diazotrophic red-pigmented vibrios from different mangrove 79 8 rhiz ospheres. FEMS microbiology ecology 67:455–467. 79 9 Sch liep KP. 2011. phangorn: phylogenetic analysis in R. Bioinformatics 27:592–593. 80 0 Sjöl ing S, Mohammed SM, Lyimo TJ, Kyaruzi JJ. 2005. Benthic bacterial diversity and 80 1 nu trient processes in mangroves: impact of deforestation. Estuarine, coastal and shelf 80 2 scie nce 63:397–406. 80 3 Song S, D ing Y, Li W, Meng Y, Zhou J, Gou R, Zhang C, Ye S, Saintilan N, Krauss KW, 80 4 Cr ooks S, Lv S, Lin G. 2023. Mangrove reforestation provides greater blue carbon 80 5 bene fit than afforestation for mitigating global climate change. Nature communications 80 6 14 :756. 80 7 T he R Project for Statistical Computing. Available at https://www.R-project.org/ (accessed 80 8 Janua ry 8, 2024). 80 9 T homson T, Fusi M, Bennett-Smith MF, Prinz N, Aylagas E, Carvalho S, Lovelock CE, 81 0 Jones BH, Ellis JI. 2022. Contrasting Effects of Local Environmental and Biogeographic 81 1 F actors on the Composition and Structure of Bacterial Communities in Arid 81 2 Mono specific Mangrove Soils. Microbiology spectrum 10:e0090321. 81 3 T willey RR, Rivera-Monroy VH. 2009. Ecogeomorphic models of nutrient biogeochemistry 81 4 for mangrove wetlands. In: Perillo GME, Wolanski E, Cahoon DR, Brinson MM eds. 81 5 Co astal Wetlands: An Integrated Ecosystem Approach. Elsevier, 641–675. 81 6 Vov ides AG, Bashan Y, López-Portillo JA, Guevara R. 2011. Nitrogen Fixation in Preserved, 81 7 Refo rested, Naturally Regenerated and Impaired Mangroves as an Indicator of 81 8 F unctional Restoration in Mangroves in an Arid Region of Mexico. Restoration Ecology 81 9 19 :236–244. 82 0 Wa inwright BJ, Millar T, Bowen L, Semon L, Hickman KJE, Lee JN, Yeo ZY, Zahn G. 82 1 20 23. The core mangrove microbiome reveals shared taxa potentially involved in 82 2 nu trient cycling and promoting host survival. Environmental microbiome 18:47. 82 3 Woo droffe CD, Rogers K, McKee KL, Lovelock CE, Mendelssohn IA, Saintilan N. 2016. 82 4 Mangr ove Sedimentation and Response to Relative Sea-Level Rise. Annual review of 82 5 marin e science 8:243–266. 82 6 Wr ight E. 2017. DECIPHER. Bioconductor. DOI: 10.18129/B9.BIOC.DECIPHER. 82 7 Ya ng C, Mai J, Cao X, Burberry A, Cominelli F, Zhang L. 2023. ggpicrust2: an R package 82 8 for PICRUSt2 predicted functional profile analysis and visualization. Bioinformatics 39. 82 9 DO I: 10.1093/bioinformatics/btad470. 83 0 Z hou Z, Meng H, Liu Y, Gu J-D, Li M. 2017. Stratified Bacterial and Archaeal Community 83 1 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.24.604998doi: bioRxiv preprint i n Mangrove and Intertidal Wetland Mudflats Revealed by High Throughput 16S rRNA 83 2 Gene Sequencing. Frontiers in microbiology 8:300984. 83 3 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.24.604998doi: bioRxiv preprint