Turn off MathJax
Article Contents

Lin Wang, Si Zhang, Jie Li. The role of biocrusts in nitrogen cycling on the tropical reef islands, South China Sea[J]. Acta Oceanologica Sinica. doi: 10.1007/s13131-021-1783-5
Citation: Lin Wang, Si Zhang, Jie Li. The role of biocrusts in nitrogen cycling on the tropical reef islands, South China Sea[J]. Acta Oceanologica Sinica. doi: 10.1007/s13131-021-1783-5

The role of biocrusts in nitrogen cycling on the tropical reef islands, South China Sea

doi: 10.1007/s13131-021-1783-5
Funds:  The Strategic Priority Research Program of the Chinese Academy of Sciences under contract Nos XDA13020301 and XDA13010500; the Fund of Innovation Academy of South China Sea Ecology and Environmental Engineering, Chinese Academy of Sciences under contract No. ISEE2018PY01.
More Information
  • Harboring polyextremotolerant microbial topsoil communities, biological soil crusts (biocrusts) occur across various climatic zones, and have been well studied in the terrestrial drylands. However, little is known about the functional metabolic potential of microbial communities involved in the biogeochemical processes during the early succession of biocrusts on the tropical reef islands. We collected 26 biocrusts and bare soil samples from the Xisha Islands and Nansha Islands, and applied a functional gene array (GeoChip 5.0) to reveal nitrogen (N) cycling processes involved in these samples. Both physicochemical measurement and enzyme activity assay were utilized to characterize the soil properties. Results revealed the composition of N-cycling functional genes in biocrusts was distinct from that in bare soil. Additionally, microorganisms in biocrusts showed lower functional potential related to ammonification, denitrification, N assimilation, nitrification, N fixation, and dissimilatory nitrate reduction to ammonium compared to bare soils. Although the abundance of nifH gene was lower in biocrusts, nitrogenase activity was significantly higher compared to that in bare soils. Precipitation, soil physicochemical properties (i.e., soil available copper, soil ammonia N and pH) and soil biological properties (i.e., β-glucosidase, fluorescein diacetate hydrolase, alkaline protease, urease, alkaline phosphatase, catalase and chlorophyll a) correlated to the N-cycling functional genes structure. Nitrate N and ammonia N were more abundant in biocrusts than bare soil, while pH value was higher in bare soil. Our results suggested biocrusts play an important role in N-cycling in coral sand soil, and will be helpful in understanding the development and ecological functions of biocrusts on tropical reef islands.
  • 加载中
  • [1] Acosta-Martínez V, Cruz L, Sotomayor-Ramírez D, et al. 2007. Enzyme activities as affected by soil properties and land use in a tropical watershed. Applied Soil Ecology, 35(1): 35–45. doi:  10.1016/j.apsoil.2006.05.012
    [2] Baran R, Brodie E L, Mayberry-Lewis J, et al. 2015. Exometabolite niche partitioning among sympatric soil bacteria. Nature Communications, 6: 8289. doi:  10.1038/ncomms9289
    [3] Barger N N, Belnap J, Ojima D S, et al. 2005. NO gas loss from biologically crusted soils in Canyonlands National Park, Utah. Biogeochemistry, 75(3): 373–391. doi:  10.1007/s10533-005-1378-9
    [4] Barger N N, Weber B, Garcia-Pichel F, et al. 2016. Patterns and controls on nitrogen cycling of biological soil crusts. In: Weber B, Büdel B, Belnap J, eds. Biological Soil Crusts: An Organizing Principle in Drylands. Cham, Switzerland: Springer, 257–285
    [5] Bates S T, Nash T H, Sweat K G, et al. 2010. Fungal communities of lichen-dominated biological soil crusts: Diversity, relative microbial biomass, and their relationship to disturbance and crust cover. Journal of Arid Environments, 74(10): 1192–1199
    [6] Belnap J. 1996. Soil surface disturbances in cold deserts: effects on nitrogenase activity in cyanobacterial-lichen soil crusts. Biology and Fertility of Soils, 23(4): 362–367. doi:  10.1007/BF00335908
    [7] Belnap J. 2002. Nitrogen fixation in biological soil crusts from southeast Utah, USA. Biology and Fertility of Soils, 35(2): 128–135. doi:  10.1007/s00374-002-0452-x
    [8] Belnap J. 2006. The potential roles of biological soil crusts in dryland hydrologic cycles. Hydrological Processes, 20(15): 3159–3178. doi:  10.1002/hyp.6325
    [9] Belnap J, Büdel B, Lange O L. 2001. Biological soil crusts: characteristics and distribution. In: Belnap J, Lange O L, eds. Biological Soil Crusts: Structure, Function, and Management. Berlin, Heidelberg: Springer, 3–30
    [10] Belnap J, Weber B, Büdel B. 2016. Biological soil crusts as an organizing principle in drylands. In: Weber B, Büdel B, Belnap J, eds. Biological Soil Crusts: An Organizing Principle in Drylands. Cham, Switzerland: Springer, 3–13
    [11] Beraldi-Campesi H, Hartnett H E, Anbar A, et al. 2009. Effect of biological soil crusts on soil elemental concentrations: implications for biogeochemistry and as traceable biosignatures of ancient life on land. Geobiology, 7(3): 348–359. doi:  10.1111/j.1472-4669.2009.00204.x
    [12] Bowker M A, Belnap J, Miller M E. 2006. Spatial modeling of biological soil crusts to support rangeland assessment and monitoring. Rangeland Ecology & Management, 59(5): 519–529
    [13] Brankatschk R, Fischer T, Veste M, et al. 2013. Succession of N cycling processes in biological soil crusts on a Central European inland dune. FEMS Microbiology Ecology, 83(1): 149–160. doi:  10.1111/j.1574-6941.2012.01459.x
    [14] Büdel B, Darienko T, Deutschewitz K, et al. 2009. Southern African biological soil crusts are ubiquitous and highly diverse in drylands, being restricted by rainfall frequency. Microbial Ecology, 57(2): 229–247. doi:  10.1007/s00248-008-9449-9
    [15] Castillo-Monroy A P, Bowker M A, Maestre F T, et al. 2011. Relationships between biological soil crusts, bacterial diversity and abundance, and ecosystem functioning: Insights from a semi-arid Mediterranean environment. Journal of Vegetation Science, 22(1): 165–174
    [16] Couradeau E, Giraldo-Silva A, De Martini F, et al. 2019. Spatial segregation of the biological soil crust microbiome around its foundational cyanobacterium, Microcoleus vaginatus, and the formation of a nitrogen-fixing cyanosphere. Microbiome, 7(1): 55. doi:  10.1186/s40168-019-0661-2
    [17] Darby B J, Neher D A. 2016. Microfauna within biological soil crusts. In: Weber B, Büdel B, Belnap J, eds. Biological Soil Crusts: An Organizing Principle in Drylands. Cham: Springer International Publishing, 139–157
    [18] de Caire G Z, de Cano M S, Palma R M, et al. 2000. Changes in soil enzyme activities following additions of cyanobacterial biomass and exopolysaccharide. Soil Biology and Biochemistry, 32(13): 1985–1987. doi:  10.1016/S0038-0717(00)00174-7
    [19] Dixon R, Kahn D. 2004. Genetic regulation of biological nitrogen fixation. Nature Reviews Microbiology, 2(8): 621–631. doi:  10.1038/nrmicro954
    [20] Duan Zhenghu, Xiao Honglang. 2000. Effects of soil properties on ammonia volatilization. Soil Science and Plant Nutrition, 46(4): 845–852. doi:  10.1080/00380768.2000.10409150
    [21] Elbert W, Weber B, Burrows S, et al. 2012. Contribution of cryptogamic covers to the global cycles of carbon and nitrogen. Nature Geoscience, 5(7): 459–462
    [22] Evans R D, Ehleringer J R. 1993. A break in the nitrogen cycle in aridlands? Evidence from δ15N of soils. Oecologia, 94(3): 314–317. doi:  10.1007/BF00317104
    [23] Evans R D, Johansen J R. 1999. Microbiotic crusts and ecosystem processes. Critical Reviews in Plant Sciences, 18(2): 183–225. doi:  10.1080/07352689991309199
    [24] Evans R D, Lange O L. 2001. Biological soil crusts and ecosystem nitrogen and carbon dynamics. In: Belnap J, Lange O L, eds. Biological Soil Crusts: Structure, Function, and Management. Berlin, Heidelberg: Springer, 263–279
    [25] Fisher K A, Yarwood S A, James B R. 2017. Soil urease activity and bacterial ureC gene copy numbers: effect of pH. Geoderma, 285: 1–8. doi:  10.1016/j.geoderma.2016.09.012
    [26] Ghiloufi W, Seo J, Kim J, et al. 2019. Effects of biological soil crusts on enzyme activities and microbial community in soils of an arid ecosystem. Microbial Ecology, 77(1): 201–216. doi:  10.1007/s00248-018-1219-8
    [27] Guan Pingting, Zhang Xiaoke, Yu Jun, et al. 2018. Soil microbial food web channels associated with biological soil crusts in desertification restoration: the carbon flow from microbes to nematodes. Soil Biology and Biochemistry, 116: 82–90. doi:  10.1016/j.soilbio.2017.10.003
    [28] Hardy R W F, Burns R C, Holsten R D. 1973. Applications of the acetylene-ethylene assay for measurement of nitrogen fixation. Soil Biology and Biochemistry, 5(1): 47–81. doi:  10.1016/0038-0717(73)90093-X
    [29] He Zhili, Deng Ye, Van Nostrand J D, et al. 2010. GeoChip 3.0 as a high-throughput tool for analyzing microbial community composition, structure and functional activity. The ISME Journal, 4(9): 1167–1179. doi:  10.1038/ismej.2010.46
    [30] He Mei, Wang Wenxiong. 2011. Factors affecting the bioaccessibility of methylmercury in several marine fish species. Journal of Agricultural and Food Chemistry, 59(13): 7155–7162. doi:  10.1021/jf201424g
    [31] Hooper D U, Johnson L. 1999. Nitrogen limitation in dryland ecosystems: responses to geographical and temporal variation in precipitation. Biogeochemistry, 46(1): 247–293
    [32] Hu Yigang, Zhang Zhishan, Huang Lei, et al. 2019. Shifts in soil microbial community functional gene structure across a 61-year desert revegetation chronosequence. Geoderma, 347(2019): 126–134
    [33] Johnson S L, Budinoff C R, Belnap J, et al. 2005. Relevance of ammonium oxidation within biological soil crust communities. Environmental Microbiology, 7(1): 1–12. doi:  10.1111/j.1462-2920.2004.00649.x
    [34] Kruskal J B. 1964. Nonmetric multidimensional scaling: a numerical method. Psychometrika, 29(2): 115–129. doi:  10.1007/BF02289694
    [35] Kuypers M M, Marchant H K, Kartal B. 2018. The microbial nitrogen-cycling network. Nature Reviews Microbiology, 16(5): 263
    [36] Lan Shubin, Wu Li, Zhang Delu, et al. 2015. Analysis of environmental factors determining development and succession in biological soil crusts. Science of the Total Environment, 538: 492–499. doi:  10.1016/j.scitotenv.2015.08.066
    [37] Li Jie, Chen Qi, Zhang Si, et al. 2013. Highly heterogeneous bacterial communities associated with the SCS reef corals Porites lutea, Galaxea fascicularis and Acropora millepora. PLoS One, 8(8): e71301. doi:  10.1371/journal.pone.0071301
    [38] Ligi T, Truu M, Truu J, et al. 2014. Effects of soil chemical characteristics and water regime on denitrification genes (nirS, nirK, and nosZ) abundances in a created riverine wetland complex. Ecological Engineering, 72: 47–55. doi:  10.1016/j.ecoleng.2013.07.015
    [39] Liu Weiqiu, Song Yongsheng, Wang Bin, et al. 2012. Nitrogen fixation in biotic crusts and vascular plant communities on a copper mine tailings. European Journal of Soil Biology, 50: 15–20. doi:  10.1016/j.ejsobi.2011.11.009
    [40] Liu Yubin, Zhao Lina, Wang Zengru, et al. 2018. Changes in functional gene structure and metabolic potential of the microbial community in biological soil crusts along a revegetation chronosequence in the Tengger Desert. Soil Biology and Biochemistry, 126: 40–48. doi:  10.1016/j.soilbio.2018.08.012
    [41] Lorenzen C J. 1967. Determination of chlorophyll and phaeopigments: spectrophotometric equations. Limnology and Oceanography, 12(2): 343–346. doi:  10.4319/lo.1967.12.2.0343
    [42] Marker A F H, Jinks S. 1982. Spectrophotometric analysis of chlorophyll a and phaeopigments in acetone, ethanol and methanol. Ergebnisse der Limnologie, 16: 3–7
    [43] Miralles I, Domingo F, Cantón Y, et al. 2012. Hydrolase enzyme activities in a successional gradient of biological soil crusts in arid and semi-arid zones. Soil Biology and Biochemistry, 53: 124–132. doi:  10.1016/j.soilbio.2012.05.016
    [44] Mush E A. 1980. Comparison of different methods for chlorophyll and phaeopigment determination. Archiv für Hydrobiologie, Beihefte, 14: 14–36
    [45] Ochoa-Hueso R, Manrique E. 2011. Effects of nitrogen deposition and soil fertility on cover and physiology of Cladonia foliacea (Huds.) Willd., a lichen of biological soil crusts from Mediterranean Spain. Environmental Pollution, 159(2): 449–457. doi:  10.1016/j.envpol.2010.10.021
    [46] Orellana L H, Rodriguez-R L M, Higgins S, et al. 2014. Detecting nitrous oxide reductase (NosZ) genes in soil metagenomes: method development and implications for the nitrogen cycle. mBio, 5(3): e01193–14
    [47] Pointing S B, Belnap J. 2012. Microbial colonization and controls in dryland systems. Nature Reviews Microbiology, 10(8): 551–562. doi:  10.1038/nrmicro2831
    [48] Sivaperuman C, Jaisankar I, Velmurugan A, et al. 2008. Tropical islands: ecosystem and endemism. In: Sivaperuman C, Velmurugan A, Singh A K, et al., eds. Biodiversity and Climate Change Adaptation in Tropical Islands. Amsterdam: Elsevier, 31–52
    [49] Van Den Ancker J A M, Jungerius P D, Mur L R. 1985. The role of algae in the stabilization of coastal dune blowouts. Earth Surface Processes and Landforms, 10(2): 189–192. doi:  10.1002/esp.3290100210
    [50] Vitousek P M, Porder S, Houlton B Z, et al. 2010. Terrestrial phosphorus limitation: mechanisms, implications, and nitrogen-phosphorus interactions. Ecological applications, 20(1): 5–15. doi:  10.1890/08-0127.1
    [51] Walvoord M A, Phillips F M, Stonestrom D A, et al. 2003. A reservoir of nitrate beneath desert soils. Science, 302(5647): 1021-1024
    [52] Wang Chao, Wang Xiaobo, Liu Dongwei, et al. 2014. Aridity threshold in controlling ecosystem nitrogen cycling in arid and semi-arid grasslands. Nature Communications, 5: 4799. doi:  10.1038/ncomms5799
    [53] Warshan D, Bay G, Nahar N, et al. 2016. Seasonal variation in nifH abundance and expression of cyanobacterial communities associated with boreal feather mosses. The ISME Journal, 10(9): 2198–2208. doi:  10.1038/ismej.2016.17
    [54] Weber B, Graf T, Bass M. 2012. Ecophysiological analysis of moss-dominated biological soil crusts and their separate components from the Succulent Karoo, South Africa. Planta, 236(1): 129–139. doi:  10.1007/s00425-012-1595-0
    [55] Wellburn A R. 1994. The spectral determination of chlorophylls a and b, as well as total carotenoids, using various solvents with spectrophotometers of different resolution. Journal of Plant Physiology, 144(3): 307–313. doi:  10.1016/S0176-1617(11)81192-2
    [56] Yang Yunfeng, Gao Ying, Wang Shiping, et al. 2014. The microbial gene diversity along an elevation gradient of the Tibetan grassland. The ISME Journal, 8(2): 430–440. doi:  10.1038/ismej.2013.146
    [57] Yang Hangyu, Liu Changzhong, Liu Yanmei, et al. 2018. Impact of human trampling on biological soil crusts determined by soil microbial biomass, enzyme activities and nematode communities in a desert ecosystem. European Journal of Soil Biology, 87: 61–71. doi:  10.1016/j.ejsobi.2018.05.005
    [58] Yeager C M, Kuske C R, Carney T D, et al. 2012. Response of biological soil crust diazotrophs to season, altered summer precipitation, and year-round increased temperature in an arid grassland of the colorado plateau, USA. Frontiers in Microbiology, 3: 358
    [59] Yue Haowei, Wang Mengmeng, Wang Shiping, et al. 2015. The microbe-mediated mechanisms affecting topsoil carbon stock in Tibetan grasslands. The ISME Journal, 9(9): 2012–2020. doi:  10.1038/ismej.2015.19
    [60] Zehr J P, Paerl H. 1998. Nitrogen fixation in the marine environment: genetic potential and nitrogenase expression. In: Cooksey K E, ed. Molecular Approaches to the Study of the Ocean. Dordrecht: Springer, 285–301
    [61] Zhao Lina, Liu Yubing, Wang Zengru, et al. 2020. Bacteria and fungi differentially contribute to carbon and nitrogen cycles during biological soil crust succession in arid ecosystems. Plant and Soil, 447(1): 379–392
    [62] Zhou Jizhong, He Zhili, Yang Yunfeng, et al. 2015. High-throughput metagenomic technologies for complex microbial community analysis: open and closed formats. mBio, 6(1): e02288–14
    [63] Zumft W G. 1997. Cell biology and molecular basis of denitrification. Microbiology and Molecular Biology Review, 61(4): 533–616. doi:  10.1128/.61.4.533-616.1997
  • 加载中
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Figures(5)  / Tables(7)

Article Metrics

Article views(47) PDF downloads(0) Cited by()

Related
Proportional views

The role of biocrusts in nitrogen cycling on the tropical reef islands, South China Sea

doi: 10.1007/s13131-021-1783-5
Funds:  The Strategic Priority Research Program of the Chinese Academy of Sciences under contract Nos XDA13020301 and XDA13010500; the Fund of Innovation Academy of South China Sea Ecology and Environmental Engineering, Chinese Academy of Sciences under contract No. ISEE2018PY01.

Abstract: Harboring polyextremotolerant microbial topsoil communities, biological soil crusts (biocrusts) occur across various climatic zones, and have been well studied in the terrestrial drylands. However, little is known about the functional metabolic potential of microbial communities involved in the biogeochemical processes during the early succession of biocrusts on the tropical reef islands. We collected 26 biocrusts and bare soil samples from the Xisha Islands and Nansha Islands, and applied a functional gene array (GeoChip 5.0) to reveal nitrogen (N) cycling processes involved in these samples. Both physicochemical measurement and enzyme activity assay were utilized to characterize the soil properties. Results revealed the composition of N-cycling functional genes in biocrusts was distinct from that in bare soil. Additionally, microorganisms in biocrusts showed lower functional potential related to ammonification, denitrification, N assimilation, nitrification, N fixation, and dissimilatory nitrate reduction to ammonium compared to bare soils. Although the abundance of nifH gene was lower in biocrusts, nitrogenase activity was significantly higher compared to that in bare soils. Precipitation, soil physicochemical properties (i.e., soil available copper, soil ammonia N and pH) and soil biological properties (i.e., β-glucosidase, fluorescein diacetate hydrolase, alkaline protease, urease, alkaline phosphatase, catalase and chlorophyll a) correlated to the N-cycling functional genes structure. Nitrate N and ammonia N were more abundant in biocrusts than bare soil, while pH value was higher in bare soil. Our results suggested biocrusts play an important role in N-cycling in coral sand soil, and will be helpful in understanding the development and ecological functions of biocrusts on tropical reef islands.

Lin Wang, Si Zhang, Jie Li. The role of biocrusts in nitrogen cycling on the tropical reef islands, South China Sea[J]. Acta Oceanologica Sinica. doi: 10.1007/s13131-021-1783-5
Citation: Lin Wang, Si Zhang, Jie Li. The role of biocrusts in nitrogen cycling on the tropical reef islands, South China Sea[J]. Acta Oceanologica Sinica. doi: 10.1007/s13131-021-1783-5
    • Biological soil crusts (biocrusts) are polyextremotolerant microbial topsoil communities, consisting of algae, bacteria, archaea, fungi, lichen and mosses in varying proportions that colonize the soil surface and get embedded together within a matrix of extracellular polymeric substance and soil particles to form a surface crust (Belnap et al., 2001, 2016). Biocrusts show different colors, ranging from white through green to black hues, depending on the different successional stages, reflected by the different dominating photoautotrophic organisms, including cyanobacterium-, lichen-, and moss-dominated types (Bowker et al., 2006; Büdel et al., 2009; Weber et al., 2012). As multifunctional communities, biocrusts play important ecological functions in various ecosystems that promote soil formation by increasing its nutrient and water contents (Evans and Johansen, 1999; Belnap, 2006; Pointing and Belnap, 2012), stabilize the soil and reduce soil erosion effectively by producing polysaccharides (Van Den Ancker et al., 1985; Belnap et al., 2001). These functions have a positive influence on seed germination, establishment and performance of plants, and population and behavior of animals (Evans and Johansen, 1999; Belnap et al., 2001; Lan et al., 2015; Guan et al., 2018).

      Nitrogen (N) is an essential macronutrient that is initially low in the soil and a vital limiting factor that influences productivity in arid terrestrial and rangeland ecosystems (Hooper and Johnson, 1999; Vitousek et al., 2010). Thus, N supply is one of the most important factors for further development of the initial ecosystem. Diverse N input processes (biological fixation and dust capture) and direct N loss processes (dissolution, vaporisation and erosion) have been found in the biocrusts developed in arid and semiarid ecosystems (Barger et al., 2016). Biological fixation of atmospheric N is an omnipresent biogeochemical transformation in the biological soil crusts. It possibly contributes to the dominant source of N in low-nutrient environments with few symbolic vascular plant N fixers (Evans and Ehleringer, 1993; Evans and Lange, 2001). Furthermore, N fixation by biocrusts leads to N release to the surrounding soil in a variety of N forms (inorganic and organic N) and improves the soil nutrient pools, which serves as a food source for plants, animals and other organisms in arid and semiarid ecosystems (Walvoord et al, 2003; Elbert et al., 2012; Darby and Neher, 2016). Studying the functional gene structure regarding N cycling in biocrusts is important to our broader understanding of the role of biocrusts in N input and loss in the soil ecosystem.

      Recently, numerous studies have focused on individual N processes (Barger et al., 2005; Johnson et al., 2005), N transfers (Baran et al., 2015), and N budget (Beraldi-Campesi et al., 2009; Brankatschk et al., 2013) in the biocrust ecosystem. Some of them used molecular tools (i.e., quantitative real-time PCR) to measure the abundance of N-cycling functional genes in biocrusts, and to evaluate the genetic potential of the microbial community catalyzing the related N-cycle processes, including nifH (N fixation), chiA (mineralization as proteolysis and chitinolysis), amoA (nitrification), and nosZ (denitrification) (Brankatschk et al., 2013; Couradeau et al., 2019). However, the metabolic potential related to global N-cycling has rarely been well evaluated in biocrusts (Liu et al., 2018).

      Located in the southernmost part of China, the South China Sea (SCS) contains three archipelagos (i.e., Dongsha Islands, Nansha Islands (NS) and Xisha Islands (XS)), and the predominant structures are coral reefs (Li et al., 2013). Besides, the soil matrix on the island is dominated by coral sand. Due to geographic isolation, climate cycling and ecological amplitude, the tropical reef islands in the SCS are endowed with unique biodiversity (Sivaperuman et al., 2008). However, biocrusts on tropical reef islands have been ignored to date. Here, we aim to present the characterization of N-cycling functional gene compositions of biocrusts and bare soil across NS and XS, and investigate the role of biocrusts in the N retention and balance on tropical reef islands.

    • Samples were collected from four different islands between May and June of 2017, two of them belonging to the NS, and the other two belonging to the XS. Affected by the tropical marine climate, the mean temperatures of the NS and XS were larger than 27°C and 26–27°C, respectively, and their mean annual precipitation were approximately 2 800 mm and 1 500 mm, respectively. The morphology of biocrusts in these islands revealed their characteristics as cyanobacteria-dominated crusts (the early succession of biocrusts). They were located in regions covering from large areas of coral sands without plants to those the interspaced between island plants, or even underneath shrubs. At each sampling site, biocrust samples (top 0–1 cm of biocrusts) and bare soil samples (top 0–1 cm of adjacent soil with no sign of biocrusts) were collected using sterile equipment, including spatulas, medicine spoons, and foil samplers. Among a total of 26 soil samples, 16 and 10 samples were collected from the NS and XS, respectively. Among NS samples, 8 and 8 were originated from biocrusts and bare soil, correspondingly, whereas 7 biocrust and 3 bare soil samples collected from XS.

      Bare soil and biocrust samples for studying soil properties (including soil enzyme activities) were collected in sealed plastic bags and stored in the refrigerator (4°C) until further measurements (Miralles et al., 2012; Ghiloufi et al., 2019; Yang et al., 2018). Samples for measuring chlorophyll a (Chl a) content were collected in sterile sampling opaque bags (EPN-4590, TWIRL’EM, Canada) and stored in the dark in a –20°C freezer. Samples for DNA extraction were collected using sterile equipment, and stored in 15 mL conical centrifuge tubes that were preserved using the LifeGuardTM soil preservation solution (MO BIO Laboratories, USA) according to the manufacturer’s instructions; they were transported to the laboratory as soon as possible and stored at −80°C until subsequent processing in the laboratory. Samples for nitrogenase activity tests were collected using a sterile foil sampler (size: 20 cm2×5 cm) and stored in sealed plastic bags at 4°C until subsequent measurements.

    • Physicochemical properties of the soil samples, including soil available boron, organic matter, organic carbon, soil available phosphorus, soil exchangeable calcium, soil available kalium, kalium, calcium, soil available zinc, soil available copper, soil available iron, soil available manganese, total water soluble salt, soil total phosphorus, soil available sulphur, soil total N, soil nitrite N, soil nitrate N, soil ammonium N, and soil ammonia N (a mixture of free ammonia and ammonium N), were measured according to the industry measurement standard of China, including agricultural trade standards, forestry industry standards, and national environmental protection standards (Table A1). Soil pH was determined by potentiometry in a 1:2.5 (soil:water, w:w) suspension (Acosta-Martínez et al., 2007) using a pH meter (pH 211, Hanna Instruments, Germany).

      Soil propertyTest methodMeasuring instrument
      Soil available boronNY/T 149-1990, Agricultural Trade Standards of PRCUltraviolet-visible spectrophotometer, Shimadzu, Japan
      Organic matterNY/T 1121.6-2006, Agricultural Trade Standards of PRCDigital Bottle-top Burette, Merck, Germany
      Organic carbonHJ 615-2011, National Environmental Protection Standards of PRCUltraviolet-visible spectrophotometer, Shimadzu, Japan
      Soil available phosphorusHJ 704-2014, National Environmental Protection Standards of PRCUltraviolet-visible spectrophotometer, Shimadzu, Japan
      Soil exchangeable calciumNY/T 1121.13-2006, Agricultural Trade Standards of PRCAtomic absorption spectrophotometer, Hitachi, Japan
      Soil available kaliumNY/T 889-2004, Agricultural Trade Standards of PRCFlame photometer, INESA, China
      KaliumNY/T 87-1988, Agricultural Trade Standards of PRCAtomic absorption spectrophotometer, Hitachi, Japan
      CalciumNY/T 296-1995, Agricultural Trade Standards of PRCAtomic absorption spectrophotometer, Hitachi, Japan
      Soil available zincNY/T 890-2004, Agricultural Trade Standards of PRCAtomic absorption spectrophotometer, Hitachi, Japan
      Soil available copperNY/T 890-2004, Agricultural Trade Standards of PRCAtomic absorption spectrophotometer, Hitachi, Japan
      Soil available ironNY/T 890-2004, Agricultural Trade Standards of PRCAtomic absorption spectrophotometer, Hitachi, Japan
      Soil available manganeseNY/T 890-2004, Agricultural Trade Standards of PRCAtomic absorption spectrophotometer, Hitachi, Japan
      Total water-soluble saltNY/T 1121.16-2006, Agricultural Trade Standards of PRCElectronic balance, Sartorius, Germany
      Soil total phosphorusHJ 632-2011, National Environmental Protection Standards of PRCUltraviolet-visible spectrophotometer, Shimadzu, Japan
      Soil available sulphurNY/T 1121.14-2006, Agricultural Trade Standards of PRCUltraviolet-visible spectrophotometer, Shimadzu, Japan
      Soil total nitrogenNY/T 53-1987, Agricultural Trade Standards of PRCUltraviolet-visible spectrophotometer, Shimadzu, Japan
      Soil nitrite nitrogenHJ 634-2012, National Environmental Protection Standards of PRCUltraviolet-visible spectrophotometer, Shimadzu, Japan
      Soil nitrate nitrogenHJ 634-2012, National Environmental Protection Standards of PRCUltraviolet-visible spectrophotometer, Shimadzu, Japan
      Soil ammonium nitrogenLY/T 1228-2015, Forestry Industry Standards of PRCUltraviolet-visible spectrophotometer, Shimadzu, Japan
      Soil ammonia nitrogenHJ 634-2012, National Environmental Protection Standards of PRCUltraviolet-visible spectrophotometer, Shimadzu, Japan

      Table A1.  Test method for soil properties and measuring instrument

      For Chl a content, 2.0 g samples (dried in a freeze dryer at –80°C) were extracted with 10 mL acetone (80%, v:v) at 4°C for 20 h (Marker and Jinks, 1982; Wellburn, 1994). Chl a content of the filtered solution was measured using spectrophotometric methods at 665 nm and 750 nm. Subsequently, 1 mol/L hydrochloric acid was added to the solution for acidification, followed by a final calculation to obtain results with high accuracy (Lorenzen, 1967; Mush, 1980).

      Enzyme assays were performed within a month of sample collection. The soil samples were used to analyze enzyme activities involved in N cycling, carbon cycling, phosphorus cycling and peroxide degradation system. Specifically, the activities of soil β-glucosidase, soil lipase, soil fluorescein diacetate hydrolase, soil alkaline protease, soil urease, soil alkaline phosphatase and soil catalase were tested using soil enzyme assay kits (Solarbio LIFE SCIENCE, Beijing Solarbio Science & Technology Co., Ltd., China) following a spectrophotometric method.

    • Nitrogenase activity, used to represent the rates of N2 fixation, was estimated as acetylene-ethylene reducing activity (Hardy et al., 1973) in all the collected soil samples. The samples, which had been reactivated by re-wetting, were placed in 100 mL conical flasks and sealed with a rubber stopper. Ten milliliters of air was withdrawn from the flask while adding an equal volume of acetylene and then incubated under laboratory conditions (temperature: 31°C, daylight: 16 h). The amount of ethylene in the headspace samples was determined by gas chromatography (Shanghaifenyi, China) equipped with a flame ionization detector. A standard mixture of 99.9×10–6 ethylene in N2 was used for calibration purposes. Nitrogenase activity was expressed as microgramme level of C2H4 per square meter of soil per hour (Belnap, 1996, 2002).

    • Total DNA was extracted using the HiPure soil DNA kit (Magen, China) according to the manufacturer’s instructions. Concentration and purity of genomic DNA were determined using a NanoVuePlus spectrophotometer (GE Healthcare, USA). The purified DNA was labeled with the Cy3 dye (GE Healthcare, USA) by random priming (Wang et al., 2014), and then purified by a QIA quick purification kit (Qiagen, USA), and dried in a labconco centrivap concentrator (Labconco, USA) at 50°C for 45 min. The dried and labeled DNA was diluted to the same concentration, incubated at 95°C for 5 min, and maintained at 42°C until hybridization. The labeled DNA was hybridized with GeoChip 5.0 (Zhou et al., 2015) in a hybridization station (MAUI, BioMicro Systems, USA), and scanned by a NimbleGen MS200 scanner (Roche Madison, USA) to obtain the optical signals. The scanned images of the hybridized GeoChip 5.0 were converted and extracted using the Agilent Feature Extraction 11.5 software (Agilent Technologies, USA). Raw data from the Agilent Feature Extraction were submitted to the laboratory’s Microarray Data Manager System (http://ieg.ou.edu/microarray/) and analyzed by the following major steps as described previously (He et al., 2010; Yue et al., 2015): (1) spots with a signal to noise ratio less than 2 were removed; (2) if the signal was detected in at least 2/3 of the repeat groups, the probes were considered positive; (3) the normalization of data was done through logarithmic transformation, the average signal intensity of each sample was determined, and intensity of each probe was proportionally enlarged by the mean intensity of the corresponding sample.

    • N-cycling functional gene diversity was estimated using gene number, Shannon index, and Simpson index. Statistical differences between the alpha diversity-related values and functional gene intensities from the different types of soil samples were analyzed by t-test (SPSS 18). A significance level of P-value less than 0.05 was adopted for all comparisons (He and Wang, 2011). Additionally, permutational multivariate analysis of variance was used to verify the pairwise statistical significance between biocrust and bare soil with vegan package in R. We graphically depicted the multivariate relationships of N-cycling functional gene compositions of different samples based on Bray-Curtis (Steinhaus) distance using non-metric multidimensional scaling (NMDS, vegan package in R) (Kruskal, 1964). The function envfit in vegan package was used to fit environmental vectors onto the ordination.

    • To extract the soil properties indicative of the biocrust formation process at the functional gene level, we analyzed 21 physicochemical and 8 biological (soil enzyme and Chl a content) soil properties (Tables A2 and A3). Besides, t-test was conducted to compare the value of soil pH, total N, ammonia N (a mixture of free ammonia and ammonium N), nitrate N, ammonium N, and nitrite N content (Fig. 1). Compared to that in biocrust (pH: 8.85 ± 0.34), soil pH was significantly higher in bare soil (9.64 ± 0.07), but soil ammonia N and soil nitrate N were significantly lower in bare soil (Fig. 1). Meanwhile, there was no significant change in soil ammonium N and nitrite N contents between bare soil and biocrust samples (Fig. 1). Among soil biological properties, except for soil lipase and soil alkaline protease, the rest showed higher levels of enzymatic activities and biomass in biocrust samples (Tables A2 and A3).

      Sample and type
      N5.01, BS(NS)N5.02, BS(NS)N5.03, BS(NS)N5.04, BS(NS)N7.01, BS(NS)N7.02, BS(NS)N7.03, BS(NS)
      pH9.619.579.599.679.629.659.71
      Soil available boron/(mg·kg–1)0.0130.0120.0130.0130.0280.0280.029
      Organic matter/(g·kg–1)8.018.058.027.982.892.892.95
      Organic carbon/%0.610.60.60.620.480.480.49
      Soil available phosphorus/(mg·kg–1)17.412.112.812.417.817.818.1
      Soil exchangeable calcium/(g·kg–1)46.346.145.447.4109108110
      Soil available kalium/(mg·kg–1)21.621.721.721.313.813.813.8
      Kalium/(mg·kg–1)70670970470476.576.476.3
      Calcium/(g·kg–1)285281286287361359363
      Soil available zinc/(mg·kg–1)5.515.575.635.335.185.175.2
      Soil available copper/(mg·kg–1)0.060.060.070.060.00600.0166
      Soil available iron/(mg·kg–1)7.97.97.87.92.92.92.9
      Soil available manganese/(mg·kg–1)0.20.20.20.20.20.20.2
      Total water soluble salt/(g·kg–1)3.63.63.93.22.62.92.5
      Soil total phosphorus/(mg·kg–1)191195193184156155157
      Soil available sulphur/(mg·kg–1)9.619.689.479.6874.574.374.1
      Soil total nitrogen/%0.0180.0190.0160.0180.0170.0170.017
      Soil nitrite nitrogen/(mg·kg–1)0.480.530.420.490.780.710.8
      to be continued

      Table A2.  Soil characteristics of the samples collected from the tropical reef islands, South China Sea

      Continued from Table A2
      Sample and type
      N5.01, BS(NS)N5.02, BS(NS)N5.03, BS(NS)N5.04, BS(NS)N7.01, BS(NS)N7.02, BS(NS)N7.03, BS(NS)
      Soil nitrate nitrogen/(mg·kg–1)8.778.98.868.5410.810.710.9
      Soil ammonium nitrogen/(mg·kg–1)3.063.063.113.010.270.310.26
      Soil ammonia nitrogen/(mg·kg–1)0.20.20.20.210.170.160.18
      Sample and type
      N7.04, BS(NS)YX.01, BS(XS)YX.02, BS(XS)ZX.01, BS(XS)N5.1, BTS(NS)N5.2, BTS(NS)N5.3, BTS(NS)
      pH9.59.719.79.688.519.128.88
      Soil available boron/(mg·kg–1)0.0270.0370.0380.0280.0420.0190.015
      Organic matter/(g·kg–1)2.834.114.1574.2159.613.2
      Organic carbon/%0.480.380.374.530.820.480.68
      Soil available phosphorus/(mg·kg–1)17.415.415.129576.99.29.3
      Soil exchangeable calcium/(g·kg–1)11030.831.137.759.267.238.1
      Soil available kalium/(mg·kg–1)13.717.817.721.118.214.68.7
      Kalium/(mg·kg–1)76.824625123933007621700
      Calcium/(g·kg–1)36124524822416895.785.7
      Soil available zinc/(mg·kg–1)5.165.275.16.0310.77.262.11
      Soil available copper/(mg·kg–1)00.040.040.130.70.170.41
      Soil available iron/(mg·kg–1)32.12.21.28.86.84.3
      Soil available manganese/(mg·kg–1)0.20.20.20.410.60.8
      Total water soluble salt/(g·kg–1)2.32.42.31.42.42.12.3
      Soil total phosphorus/(mg·kg–1)15549649654500351191347
      Soil available sulphur/(mg·kg–1)75.115.315.313.824.65.298.19
      Soil total nitrogen/%0.0180.0240.0250.5360.0690.0480.063
      Soil nitrite nitrogen/(mg·kg–1)0.830.981.030.511.480.70.48
      Soil nitrate nitrogen/(mg·kg–1)10.99.969.7711.260.610.329.9
      Soil ammonium nitrogen/(mg·kg–1)0.260.210.253.732.240.90.8
      Soil ammonia nitrogen/(mg·kg–1)0.180.30.310.260.454.584.92
      Sample and type
      N7.1, BTS(NS)N7.2, BTS(NS)N7.3, BTS(NS)N7.4, BTS(NS)N7.5, BTS(NS)YX.1, BTS(XS)YX.2, BTS(XS)
      pH9.59.719.79.688.519.128.88
      Soil available boron/(mg·kg–1)0.0270.0370.0380.0280.0420.0190.015
      Organic matter/(g·kg–1)2.834.114.1574.2159.613.2
      Organic carbon/%0.480.380.374.530.820.480.68
      Soil available phosphorus/(mg·kg–1)17.415.415.129576.99.29.3
      Soil exchangeable calcium/(g·kg–1)11030.831.137.759.267.238.1
      Soil available kalium/(mg·kg–1)13.717.817.721.118.214.68.7
      Kalium/(mg·kg–1)76.824625123933007621700
      Calcium/(g·kg–1)36124524822416895.785.7
      Soil available zinc/(mg·kg–1)5.165.275.16.0310.77.262.11
      Soil available copper/(mg·kg–1)00.040.040.130.70.170.41
      Soil available iron/(mg·kg–1)32.12.21.28.86.84.3
      Soil available manganese/(mg·kg–1)0.20.20.20.410.60.8
      Total water soluble salt/(g·kg–1)2.32.42.31.42.42.12.3
      Soil total phosphorus/(mg·kg–1)15549649654500351191347
      Soil available sulphur/(mg·kg–1)75.115.315.313.824.65.298.19
      Soil total nitrogen/%0.0180.0240.0250.5360.0690.0480.063
      Soil nitrite nitrogen/(mg·kg–1)0.830.981.030.511.480.70.48
      Soil nitrate nitrogen/(mg·kg–1)10.99.969.7711.260.610.329.9
      Soil ammonium nitrogen/(mg·kg–1)0.260.210.253.732.240.90.8
      Soil ammonia nitrogen/(mg·kg–1)0.180.30.310.260.454.584.92
      Sample and type
      YX.3, BTS(XS)YX.4,
      BTS(XS)
      YX.5, BTS(XS)YX.6,
      BTS(XS)
      ZX.1, BTS(XS)
      pH8.398.859.38.389.32
      Soil available boron/(mg·kg–1)0.0240.030.1341.220.063
      to be continued
      Continued from Table A2
      Sample and type
      YX.3, BTS(XS)YX.4,
      BTS(XS)
      YX.5, BTS(XS)YX.6,
      BTS(XS)
      ZX.1, BTS(XS)
      Organic matter/(g·kg–1)28.8225.4515.312.7
      Organic carbon/%1.441.390.510.790.93
      Soil available phosphorus/(mg·kg–1)44.626.110.61560.5
      Soil exchangeable calcium/(g·kg–1)25.98667.49867.4
      Soil available kalium/(mg·kg–1)10632.921.679.316
      Kalium/(mg·kg–1)33003211901000179
      Calcium/(g·kg–1)0.96254303292295
      Soil available zinc/(mg·kg–1)4.298.057.67.9910.9
      Soil available copper/(mg·kg–1)0.310.310.090.170.11
      Soil available iron/(mg·kg–1)0.99.11.92.53.3
      Soil available manganese/(mg·kg–1)7.50.90.40.70.6
      Total water soluble salt/(g·kg–1)2.426.310.52.5
      Soil total phosphorus/(mg·kg–1)12371110165529759
      Soil available sulphur/(mg·kg–1)23.623.3268334020.1
      Soil total nitrogen/%0.1560.0790.0290.0660.083
      Soil nitrite nitrogen/(mg·kg–1)0.890.45151.741.26
      Soil nitrate nitrogen/(mg·kg–1)9.648.0516.350.311
      Soil ammonium nitrogen/(mg·kg–1)4.440.528.663.810.92
      Soil ammonia nitrogen/(mg·kg–1)6.546.8219.29.8113.1
      Note: N5.1–N5.3 and N7.1–N7.5: biocrust topsoil samples (BTS) from the Nansha Islands (NS); YX.1–YX.6 and ZX.1: BTS from the Xisha Islands (XS); N5.01–N5.04 and N7.01–N7.04: bare soil samples (BS) from the NS; YX.01, YX.02 and ZX.01: BS from the XS.
      SCA/
      (μmol·d–1·g–1)
      SFHA/
      (μmol·d–1·g–1)
      APA/
      (μmol·d–1·g–1)
      SAPA/
      (μmol·d–1·g–1)
      SUA/
      (μg·d–1·g–1)
      SGA/
      (μmol·d–1·g–1)
      SLA/
      (μmol·d–1·g–1)
      Chl a/
      (μg·g–1)
      N5.15254.33126.9916.01813.961.2840.7727.34
      N5.227.1937.06295.0316.96715.63138.1917.9
      N5.332.3241.26292.2316.86749.920.9739.7816.19
      N7.139.4849.2771.7616.89851.592.3237.539.21
      N7.245.946.01173.4916.85817.31.4735.3520.12
      N7.355.4741.07100.7717.25861.311.6935.1534.35
      N7.443.3252.87226.9716.52811.232.2832.7718.11
      N7.549.346.74186.4216.94710.471.5538.5910.17
      YX.149.3154.3739.3614.49805.771.9630.6623.42
      YX.238.829.3527.489.64816.390.9435.1513.93
      YX.344.6946.232.4219.44893.171.193710.65
      YX.437.6155.075.4517.66844.311.5439.8410.73
      YX.518.2923.85141.687.67204.250.7537.7320.08
      YX.620.1650.884.3419.05620.640.4437.5327.99
      ZX.131.7632.3324.5616.2596.662.234.827.01
      N5.017.224.3175.61.4475.270.133.170
      N5.027.14.1893.671.6977.090.1335.150
      N5.037.034.6266.051.25780.133.170
      N5.047.524.1267.11.3770.710.0731.190.01
      N7.015.722.81157.060.0838.850.1233.30.04
      N7.025.522.82184.560.0643.70.2234.360.04
      N7.035.952.87162.880.0935.510.08300
      N7.045.692.73123.730.0937.330.0535.550.07
      YX.016.133.1116.861.29163.580.0238.260.05
      YX.026.42.98109.982.13179.620.0738.030.05
      ZX.0150.4230.9383.5318.47548.413.8335.090.05
      Note: N5.1–N5.3 and N7.1–7.5: biocrust topsoil samples from the Nansha Islands; YX.1–YX.6 and ZX.1: biocrust topsoil samples from the Xisha Islands; N5.01–N5.04 and N7.01–N7.04: bare soil samples from the Nansha Islands; YX.01, YX.02 and ZX.01: bare soil samples from the Xisha Islands. Abbreviations: soil catalase activity (SCA), soil fluorescein diacetate hydrolase (SFHA), soil alkaline protease activity (APA), soil alkaline phosphatase activity (SAPA), soil urease activity (SUA), soil β-glucosidase activity (SGA), soil lipase activity (SLA), and chlorophyll a (Chl a).

      Table A3.  Soil enzyme activities and Chl a contents of the samples collected from South China Sea

      Figure 1.  Alterations of physicochemical soil properties in the biocrusts and bare soil samples are presented in the boxplot. Boxes limit the 25th and 75th percentile with the median presented as lines inside. Error bars present the 1st and 99th percentile. Significant differences: ***P<0.001, *P<0.05.

      In the present study, compared to that in bare soil ((0.023 ± 0.019) μmol/(m2·h), we observed significantly higher nitrogenase activity (measured by acetylene reduction assay) in biocrust ((2.390 ± 1.280) μmol/(m2·h)) (Fig. 2). However, there was no significant difference in nitrogenase activity between NS-Biocrust and XS-Biocrust (Fig. 2).

      Figure 2.  Alterations of nitrogenase activity in the bare soil and biocrust of samples are presented in the boxplot. Boxes limit the 25th and 75th percentile with the median presented as lines inside. Error bars present the 1st and 99th percentile. Significant differences: ***P<0.001. Abbreviations: Nansha Islands (NS) and Xisha Islands (XS).

    • The number of detected genes, Shannon index and Simpson index were measured to understand the composition of N-cycling functional genes on the tropical reef islands (Fig. 3). A total of 37 416 N-cycling functional genes were detected by GeoChip 5.0, and the number of detected N-cycling functional genes ranged from 826 to 2 758 across the samples. A t-test analysis showed that the number of detected N-cycling functional genes, and Shannon and Simpson indexes in biocrust were significantly lower than in bare soil (Fig. 3). Higher proportions of common genes were observed in NS-Biocrust vs. XS-Biocrust (81.84%), NS-Bare soil vs. XS-Bare soil (78.52%), and 65.38%–72.88% of the genes were shared between biocrust and bare soil samples.

      Figure 3.  Alterations of richness and diversity of microbial N-cycling functional genes in the biocrust and bare soil samples are presented in the boxplot. Boxes limit the 25th and 75th percentile with the median presented as lines inside. Error bars present the 1st and 99th percentile. Significant differences: **P<0.01, *P<0.05.

    • We detected 25 types of N cycle-related genes by GeoChip 5.0, and abundance of the key functional genes involved in all 9 processes, including ammonification, anammox, denitrification, assimilatory N reduction, dissimilatory N reduction, nitrification, N fixation, assimilation and N assimilation (assimilation and N assimilation belong to N-transfer process), was significantly lower in biocrust compared to that in bare soil (Fig. 4a), except for two genes, namely nirA and p450nor (PR), which were involved in assimilatory N reduction and denitrification, respectively. These results showed that, with biocrust formation, the metabolic potential of N-cycling in soil decreased on the tropical islands. Moreover, the N-cycling functional genes structure in biocrust and bare soil were distinct (F= 6.339, Padjusted= 0.045, Table 1, Fig. 5).

      Figure 4.  Average gene abundance of N-cycling functional genes in the biocrust samples. All data are presented as Mean±SD calculated from biological repetition. a. Biocrust compared with bare soil; b. NS-Biocrust compared with XS-Biocrust. Abbreviations: glnA_fungi (GF), ammonium transporter (AT), nitrite transporter (NT), nitrate reductase (NR), p450nor (PR), Nansha Islands (NS) and Xisha Islands (XS). Significant differences: **P<0.01, *P<0.05.

      TypeFR2PPadjusted
      Biocrust vs. Bare soil6.3390.2090.0030.045
      NS-Biocrust vs. XS-Biocrust4.0760.2390.0280.375
      NS-Bare soil vs. XS-Bare soil2.1070.1900.0941.000
      Note: Nansha Islands (NS) and Xisha Islands (XS).

      Table 1.  Permutational multivariate analysis of microbial N-cycling functional genes composition among different soil types

      Figure 5.  Dissimilarity of microbial N-cycling functional genes compositions in biocrusts comparing to bare soil samples. Ordination using NMDS is derived from Bray-Curtis dissimilarity and applied to analyze the composition of microbial N-cycling functional genes. Circle shape represents soil samples from NS and XS. Different soil types are color coded. The function envfit from the R vegan package was used to fit environmental vectors onto the ordination (environmental factor significant correlation with NMDS, P<0.01). Abbreviations: non-metric multidimensional scaling (NMDS), Nansha Islands (NS), Xisha Islands (XS), soil β-glucosidase (SGA), soil fluorescein diacetate hydrolase (SFHA), soil urease (SUA), soil alkaline phosphatase (SAPA), soil catalase (SCA), and chlorophyll a (Chl a).

      The abundance of most N-cycle related functional genes was significantly lower in NS-Biocrust than in XS-Biocrust (Fig. 4b), except for nirA (assimilatory N reduction), hao (nitrification), PR, norB (denitrification), hzo, and hzsA (anammox). Further, we observed no difference in the N-cycling gene composition between NS and XS in both biocrusts and bare soil (Table 1, Fig. 5).

    • NMDS analysis revealed a total of one environmental factor and ten soil properties, including precipitation, soil pH, soil available copper, soil ammonia N, soil β-glucosidase, soil fluorescein diacetate hydrolase, soil alkaline protease, soil urease, soil alkaline phosphatase, soil catalase and Chl a, to be significantly correlated with the N-cycling gene compositions (Table 2); 7 properties (P≤0.01) were shown in Fig. 5.

      NMDS1NMDS2r2P (>r)
      Pp–0.532 91–0.846 170.247 00.036
      pH0.290 96–0.956 740.401 90.002
      SAC–0.434650.900 600.298 50.014
      SAN2–0.011 680.999 930.272 70.032
      SCA–0.366 420.930 450.636 80.001
      SFHA–0.282 770.959 190.605 70.001
      APA–0.890 96–0.454 080.267 10.028
      SAPA–0.301 460.953 480.631 40.001
      SUA–0.365 610.930 770.711 00.001
      SGA–0.292 660.956 220.355 60.006
      Chl a–0.456 150.889 900.539 30.001
      Note: Factors with P≤0.01 are shown in bold type. Abbreviations: non-metric multidimensional scaling (NMDS), precipitation (Pp), soil available copper (SAC), soil ammonia N (SAN2), soil β-glucosidase (SGA), soil fluorescein diacetate hydrolase (SFHA), soil urease (SUA), soil alkaline phosphatase (SAPA), soil catalase (SCA), soil alkaline protease (APA), and chlorophyll a (Chl a).

      Table 2.  Correlations between environmental factors and NMDS axes

    • We have presented the first study on N-cycling functional gene structure of microbial communities on the tropical reef islands, SCS. Notably, in both biocrust and bare soil samples, the abundances of nirK, nirS, narG, and nosZ (denitrification), ureC (ammonification), nifH (N fixation) genes were the highest (Fig. 4b), which suggested the corresponding pathways were dominant in the N-cycling processes. Moreover, we observed N-cycling functional gene composition to be significantly altered as biocrust formation on the islands.

      Specifically, the abundance of the most key functional genes involved in denitrification was significantly higher in bare soil than biocrust except for PR (Fig. 4a). Higher abundance of genes involved in denitrification in bare soil might lead to greater nitrate-reducing reactions. Moreover, higher abundance of nirK, nirS, and nosZ genes (Zumft, 1997; Ligi et al., 2014; Orellana et al., 2014) in bare soil might contribute to the higher N2O and N2 emission. A previous study demonstrated the alterations of N2O emission to negatively impact soil nitrate N concentrations (Yang et al., 2014). In our present study, we found the nitrate N content was significantly lower in bare soil than in biocrust, which might be attributed to the higher abundance of denitrification-related genes in the former.

      In the present study, we observed that ammonia N content, consisting of free ammonia and ammonium N, was significantly higher in biocrust than in bare soil. Ammonia N was the major product of biological N fixation and assimilation/dissimilation in N-cycling processes. As one of the most important functions in biocrusts, biological N fixation is the dominant source of N in the dryland ecosystem (Barger et al., 2016). Consistently, our result demonstrated that nitrogenase activity was significantly higher in biocrust compared with bare soil. Additionally, higher nitrate concentration (Fig. 1) might contribute to higher ammonia N content in biocrusts (Barger et al., 2016; Kuypers et al., 2018). No significant difference in the ammonium N content was observed between biocrusts and bare soil, which means the free ammonia content was likely significantly higher in biocrusts. Ammonium N has been shown to decrease pH by neutralizing alkaline soil, and elevate free ammonia content in biocrusts (Duan and Xiao, 2000; Barger et al., 2016), which was consistent with our result that pH value was lower in biocrust compared with bare soil (Fig. 1). In addition, a previous study reported that lower soil pH enhanced the hydrolysis activity of urease, an enzyme converting urea into the ammonia N (Fisher et al., 2017). It agreed with our observation that soil urease activity was higher in biocrust compared to bare soil. Thus, the higher activity of urease in biocrusts probably leads to the accumulation of ammonia N content.

      The nifH gene encodes a vital structural protein of the nitrogenase enzyme that catalyzes biological N fixation (Zehr and Paerl, 1998). In the present study, we observed nitrogenase activity and ammonia content were significantly higher compared with bare soil (Figs 1 and 2). In contrast, the abundance of nifH gene was significantly higher in bare soil (Fig. 4a). These observations are inconsistent with the previous study that N fixation rates were positively correlated with nifH gene abundances (Brankatschk et al., 2013). Although nitrogenase catalyzes biological N fixation, its magnitude of activity depends on the energy and availability of electron donors (Dixon and Kahn, 2004). Served as elemental sinks in different ecosystems, biocrusts uptake CO2 through photosynthesis and provide more energy than bare soil (Belnap et al., 2001). Thus, they might contribute to a higher level of nitrogenase activity in biocrusts. Moreover, according to Warshan et al. (2016), higher N fixation activity should be explained by higher nifH gene expression rather than higher nifH gene abundance, since not all genes have metabolic activity or present “real” functionality. Further, the biological N fixation process might be regulated by multiple enzymes, but the Geochip analysis based on known genes might lead to the limitation of the data.

      In the study, we observed that there was no difference in NS-Biocrust vs. XS-Biocrust and NS-Bare soil vs. XS-Bare soil in the N-cycling functional gene structure. However, compared to XS, the samples collected from NS were dispersally distributed (Fig. 5). This could probably be attributed to the larger spatial distances between the sampling sites in NS. We further investigated the correlation among environmental elements, soil biological properties and the N-cycling functional gene structure in biocrust and bare soil on the tropical reef islands in the SCS (Fig. 5). Yeager et al. (2012) demonstrated that seasonal shifts and variation of precipitation could alter the diazotroph community structure and therefore affect the abundance of nifH genes. In this study, we found precipitation was the crucial environmental variable influencing the composition of N-cycling functional genes (Table 2). Soil available copper has been proven to correlate with the N cycle in biocrusts (Ochoa-Hueso and Manrique, 2011; Liu et al., 2012). It was one of the key environmental drivers observed in our study. Furthermore, our results showed that activities of soil β-glucosidase, fluorescein diacetate hydrolase, urease, alkaline phosphatase and catalase were significantly correlated with the microbial N-cycling gene composition and enriched in biocrusts (Fig. 5). In agreement with our observation, Liu et al. (2018) had reported that soil enzyme activities were the significant factors correlated with the microbial functional gene structures. Previous studies have demonstrated that the microbial functional structure was correlated with the microbial community (Liu et al., 2018; Hu et al., 2019; Zhao et al., 2020), and the microbial abundance and composition could impact biocrust enzymatic activities (Bates et al., 2010; Castillo-Monroy et al., 2011). Our results suggest microbial consortia, which take part in N cycling in biocrust, contribute to hydrolytic activities of several kinds of enzymes, such as soil β-glucosidase, fluorescein diacetate hydrolase, urease, alkaline phosphatase and catalase.

      In sum, this study first comprehensively investigated the microbial N-cycling functional gene composition in biocrusts on tropical reef islands and observed that it varied significantly with biocrust formation. Abundances of N-cycling related functional genes were lower in biocrusts compared to that in bare soils. However, nitrogenase activity was significantly higher in biocrusts. Further, nitrate N and ammonia N were more abundant in biocrusts. Additionally, climatic factor (i.e., precipitation), soil chemical properties (i.e., soil available copper, soil ammonia N, and pH), and soil biological properties (i.e., β-glucosidase, fluorescein diacetate hydrolase, alkaline protease, urease, alkaline phosphatase, catalase and Chl a) significantly correlated to the microbial N-cycling functional structures. Our results suggest biocrusts played a vital role in N-cycling in coral sand soil on tropical reef islands. The higher N fixation efficiency, as well as ammonia N and nitrate N contents, in biocrust compared to in bare soil suggested that the N fixed by biocrusts was one of the N input approaches in tropical reef island soil. The role of biocrusts in the N retention and balance in tropical reef islands was worth studying over longer time scales. Given that precipitation was significantly correlated with the N-cycling functional structure observed in the present study, a comprehensive investigation of the impact of rainfall patterns on biocrust communities would be helpful in understanding the development and ecological functions of biocrusts on tropical reef islands in the context of climate changes.

    • We thank Jie Wang from the Guangzhou Institutes of Biomedicine and Health, Chinese Academy of Sciences for helpful data analysis and suggestion to make this manuscript greatly improved, Yiyang Zou from the South China Sea Institute of Oceanology, Chinese Academy of Sciences for his assistance in data analysis, Hongqiang Yang from the South China Sea Institute of Oceanology, Chinese Academy of Sciences for his suggestion in sample collection, the Xisha Marine Environment Observation and Research Station and Nansha Marine Ecological and Environmental Research Station, Chinese Academy of Science for sampling assistance.

Reference (63)

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return