Ammonia-oxidizing Archaea (AOA) are Winners to Survive in Oxygen-limited Habitat Compared to Ammonia-oxidizing Bacteria (AOB)

Purpose: Both ammonia oxidizing archaea (AOA) and bacteria (AOB) perform the ammonia oxidation together. These two kinds of microbes present a convenient model for studying niche specialization. To date, few surveys concentrated on the inuence of oxygen concentration on niche specialization of AOA and AOB in intertidal zones. Methods: Here, high-throughput sequencing by Illumina MiSeq and qPCR were applied to detect the change of abundance, diversity as well as community structure of both AOA and AOB with 0-60 cm sediments depth in the intertidal zone in Qingdao. Results and Conclusion: The AOA/AOB amoA gene copy numbers and AOA/AOB OTU numbers were all increased as sediment depth went deeper, which indicated that AOA were more adaptive to oxygen-limited niches compared to AOB and oxygen indeed led to the niche specialization of AOA and AOB in intertidal sediments. The dominant AOA and AOB were the Nitrosopumilus and Nitrosospira clusters, respectively, which indicated an ecological success in intertidal zone. Oxidation-reduction potential (ORP) was signicantly positively correlated with AOB abundance and AOB OTU numbers (P < 0.01). In addition, both TN (P < 0.01) and pH (P < 0.05) were signicantly and negatively correlated with AOB abundance. TN was also signicantly and negatively correlated with AOB OTU numbers (P < 0.05).


Introduction
Nitri cation, a key role in the ecosystem nitrogen cycle, is the conversion of ammonia to nitrate through nitrite (Beeckman et al., 2018). Three kinds of microorganisms of ammonia oxidant, nitrite oxidizers and complete ammonia oxidizers were involved in the nitri cation process (Stein and Klotz, 2016). Ammoniaoxidizing bacteria (AOB) was long thought to the only microorganism that performed ammonia oxidation, the rst and rate-limiting step in nitri cation. However, the ammonia oxidation theory was changed since the discovery of ammonia-oxidizing archaea (AOA) (Konneke et al., 2005). To date, several strains of AOA such as Candidatus Nitrososphaera gargensis (Hatzenpichler et al., 2008) (Wang, et al., 2020). Environmental factors play a key role in the niche differentiation between AOA and AOB. And the AOA and AOB relative contribution to ammonia oxidation in various ecosystems gradually become the hot spots in the eld of ammonia oxidation.
As reported previously, a serious of environmental factors may cause the niche specialization between AOA and AOB, such as pH, temperature, salinity, ammonia concentrations and oxygen concentrations.
AOA seems to be more adaptive in niches with lower pH and more likely to dominant the process of ammonia oxidation in acidic niches (Gubry-Rangin et al., 2011;Prosser and Nicol, 2012). AOB were more adaptable than AOA with increasing abundance but no alteration of composition at elevated temperature (Zhang et al., 2019b). Microcosm tests also veri ed that pH and temperature were key factors that leaded to the niche specialization of AOA and AOB (Aigle et al., 2020). Ammonia, as one of the substrate for ammonia oxidation, could cause niche specialization of AOA and AOB. The concentration of ammonia half saturation constant of some AOB strains (Jung et Park and Noguera 2007). The a nities of AOA for oxygen were much higher than that of AOB, which means that AOA had competitive advantages over AOB in the oxygen-limited environments. The AOA lived better in many oxygen-limited environments. AOA dominated the transcriptome and probably dominated the ammonia oxidation process in the OMZ (Stewart et al. 2012). The increased of AOA OTU number and abundance ratio (AOA: AOB) in ooded soil proved that AOA could better adapt to low oxygen condition (Liu et al., 2015).
The intertidal zone is an ecological crisscross zone with important environmental and ecological functions affected by marine and terrestrial ecosystems (Community structure and organization of tidepools.). Tidal daily rhythm so that sediment is submerged or exposed to air. The Shazikou wharf was built in 1976 and was one of the biggest wharf in Qingdao. Near the wharf, was the biggest sh market in Qingdao. The human activities and the decayed sh provided enough ammonia nitrogen for the ammonia-oxidizing microbes in the intertidal zones nearby. The ecological distribution, including the community structure and of abundance AOA and AOB in intertidal zones have been discussed previously (Hu et al., 2019). However, the relationship between oxygen concentration and niche specialization of AOA and AOB was not investigated. So, the main objectives of this research are: (1) to study the difference of abundances and community structure of AOA and AOB in different layers/depth of the sediments in the intertidal zones, (2) to explore which, AOA or AOB, was the main driver performing ammonia oxidation in different layers/depth of the sediments in intertidal zones, (3) to evaluate how the oxygen concentration in uence the niche specialization of AOA and AOB in intertidal zones.

Materials And Methods
Description of sampling sites, collection and physicochemical properties analysis of sediment The samples of sediment were collected using soil cylindrical sampler from the intertidal zone near the Shazikou wharf in Qingdao, Shandong Province. Three sampling sites (biological triplicates) along the sandy beach were selected. For each site, different depth (0-10, 10-20, 20-30, 30-40, 40-50 and 50-60 cm, namely S 0 − 10 , S 10 − 20 , S 20 − 30 , S 30 − 40 , S 40 − 50 , S 50 − 60 , respectively) were sampled and sediments in the same depth were mixed. Totally six sediment samples were obtained. Each sample was divided into two subsamples: one, stored at 4°C, was used for the physicochemical properties analysis; another, frozen at -80°C, was applied to molecular analysis. The pH and oxidation-reduction potential (ORP) were both measured in situ. The moisture content, inorganic nitrogen contents (including total nitrogen (TN), total inorganic nitrogen (TIN), ammonium (NH 4 + -N), nitrite (NO 2 − -N) and nitrate (NO 3 − -N)) were measured as described previously .

DNA extraction and quantitative PCR of amoA genes
We followed the manufacturer's instructions of a Power Soil DNA kit (Mo Bio Laboratories, Carlsbad, California, USA) to extract the DNA using 0.25g soil. The extracted DNA quality was assessed on 1% agarose gel using a nanometer droplet spectrophotometer (ND-1000; Isogen Life Science, Netherlands) to measure the concentration of DNA.
The copy numbers of AOA/AOB amoA genes were determined by the two primer sets (amoA-1F/amoA2-R for AOB amoA genes and CamoA-19f/CamoA-616r for AOA amoA genes). According to the previously described embodiment, the use of real-time iCycleriQ5 thermal cycler and detection system (CA Bio-Rad) for qPCR. (Hu et al., 2014).
High-throughput sequencing and bioinformatic analysisfor the amoA genes The functional genes (amoA genes) of AOB and AOA were ampli ed using the two primer sets as mentioned above. The protocols used to amplify the two functional genes have been described previously (Shen et [30][31][32][33][34][35][36][37][38][39][40] ) and deep layer (S 40-50 , S 50-60 ), which probably resulted from the higher nitri cation rates in the upper layer sediments. The ORP of the upper layer sediments were much higher than the rest of the samples. For the moisture, the middle layer was the highest. The ammonium content of the six sediment samples ranged from 1.32 to 6.21 mg/kg and the peak was found in the sample of S [30][31][32][33][34][35][36][37][38][39][40] . All the sediment samples had low nitrite content as expected. The nitrate content ranged from 4.99 to 16.89 mg/kg, and the nitrate content in the deep layer were higher than that of the upper layer. As for the TN, the content ranged from 857.77 to 1604.05 mg/kg, and the TN content increased as the depth went deeper.  (Table S1). The number of AOA and AOB amoA genes observed in this work was similar to previous study in the intertidal zones in Zhoushan, Zhejiang province ( where the environment became anoxic, the higher a nities of AOA for oxygen could help them to win the competition with AOB. AOA outnumbered AOB in oxygen-limited environment was also previously mentionedin Zhoushan intertidal zones. In the middle tidal zone where the oxygen was relatively su cient, the number of AOB amoA genes was higher than that of AOA. In the subtidal zones where oxygen was limited, AOA outnumbered AOB (Hu et al., 2019). From the point of abundance, oxygen concentration indeed led to the niche specialization of AOA and AOB.

Diversity of AOA and AOB
High-quality AOA and AOB sequences after quality control procedures were applied to diversity analysis. Aa for AOA, a total of 17144 high-quality sequences were obtained for the 6 sediment samples. Using the 15% cut-off, which was recommended previously (Pester et al., 2012), a total of 18 OTUs were obtained for the amoA gene of archaeal. The coverages of the 6 AOA amoA gene clone libraries were all higher than 99.0%, ranging from 99.80% to 99.96% (Table S2). The AOA OTU numbers of the 6 sediment samples ranged from 10 to 15, with S 30-40 (10 OTUs) and S 50-60 (15 OTUs) displayed the lowest and highest diversity, respectively (Table S2). The OTU numbers in the deep layers were higher than that in the upper and middle layer (Fig. 2). Overall, the change rule of the AOA OTU numbers was similar to the AOA abundance, both of them increased with sediment depth. AOA showed higher diversity as oxygen concentration became lower, which was in accordance with previous work in water-level-uctuating zones in Three Gorges Reservoir (Liu et al., 2015). At higher oxygen concentration of non-ooded areas, AOA OTU were lower value; In the relatively low oxygen concentration in ooded areas, AOA OTU values higher. This indicated that AOA was more adaptive to the oxygen-limited environments. A total of18865 highquality AOA sequences were obtained form 6 sediment samples. Using 15% as cutoff value (Purkhold et al., 2000), 18865 sequences were assigned to 11 OTUs. The coverages of the 6 AOB amoA gene clone libraries were ranging from 99.96% to 100.00% (Table S3). The sample S 0-10 showed the highest diversity, holding 11 OTUs in total, which was higher than the rest 5 sediment samples (Fig. 2). The OTU numbers in the rest of the 5 samples were all the same, holding 8 OTUs (Table S3). Similar to the change rules of abundance, the AOB OTU numbers decreased as sediment depth went deeper, which indicated that AOB was not tolerant or adaptable to the hypoxic environments. The research in the area of water-leveluctuating zones in Three Gorges Reservoir also showed the similar pattern. The AOB OTU numbers were always higher in the non-ooded zones where the oxygen concentration was su cient than that in the ooded zones where the environment was anoxic. The diversity of AOA was always higher than AOB in all the sediment samples (Fig. 2) and the ratio of AOA/AOB OTU numbers increased from 1.09 to 1.88 as the sediment depth went deeper, which was similar to the change pattern of the abundance.

Phylogenetic analysis and community structure of AOA and AOB
After all the quality control procedures, for AOA, totally 17144 high-quality sequences were obtained for the six sediment samples. According to the amoA genes classi cation of Archaea (Pester et al., 2012), the representative sequences of the 18 OTUs were grouped into four different clusters as shown in Fig. 3. Neither Nitrosotalea nor Thermal-related AOA was found in all the sediment samples. Nitrosophaera cluster contained 3 OTUs (OTU 12, OTU15, OTU 18), totally 9 sequences. OTU 12(containing 3 seqs) belonging to Nitrosophaera cluster was retrieved from sample S 20-30 and S 40-50 , OTU 15 (containing 2 seqs)was detected in sample S 0-10 and S 50-60 , and OTU 18 (containing 4 seqs) was only retrieved from deep layer samples (S 40-50 , S 50-60 ). The rest of the 15 OTUs were all a liated with Nitrosopumilus cluster.
Among the 17144 sequences, 17135 sequences were a liated with Nitrosopumilus cluster, accounted for 99.94% of all the sequences obtained. Allthe six sediment samples were composed dominantly by sequences belonging to Nitrosopumilus cluster. The dominance of the Nitrosopumilus cluster over the Nitrosophaera clusterin this study was similar to previous studies in Chongming eastern intertidal sediments and in the intertidal zones in Zhoushan Island (Zheng et al., 2013;Hu et al., 2019). Phylogeny and meta-data analyses of archaeal amoA sequences also showed that the most ofAOA belong to the Nitrosopumilus cluster (73%), which was nearly twice as much as the Nitrososphaera cluster (37%) (Alves et al., 2018). The failure of the Nitrososphaera cluster was more likely to form due to their inability to adapt to salty conditions or the uctuation of intertidal environmental factors. AOB obtained 18865 sequences from 6 sediment samples by the same quality control procedure as AOA Using the 15% cut-off that was recommended in a previous study (Purkhold et al., 2000), 18865 sequences were assigned to 11 OTUs. The representative sequences of the 11 OTUs were grouped into two clusters, with 2 OTUs (OTU 9, OTU 10) clustered into the Nitrosomonas cluster and the remaining 9 OTUs clustered into the Nitrosospira cluster (Fig. 4). Only 35 sequences were a liated with the Nitrosomonas cluster. All the six AOB communities were primarily composed of Nitrosospira-related sequences. The dominance of the Nitrosospira cluster over the Nitrosomonas cluster in this study was similar to previous studies in the coastal Pearl River estuary (Cao et al., 2011). The lower concentration of ammonia nitrogen (0.03 ~ 0.11 mm NH4 +) in the intertidal zone of Qingdao may help to explain the absolute dominance of nitrosoma and nitrosoma. The Nitrosomonas cluster has a lower a nity for the substrate and adapts to higher substrate concentrations. In contrast, the Nitrosospira cluster have relatively higher a nity with the substrate and are more likely to survive environments with low substrate concentrations (Zheng et  Environmental factors that in uence the diversity and communities of AOA and AOB Pearson correlation coe cient was used to analyze the linear relationship between different environmental factors and amoA gene abundance, OTU number, AOA, AOB diversity index, as shown in Table 2.. ORP was signi cantly positively correlated with AOB abundance and AOB OTU numbers (P < 0.01), which mean that AOB preferred the niches with higher oxygen concentrations. On the contrary, ORP was negatively correlated with AOA abundance, which mean that AOA preferred the niches with lower oxygen concentrations. All this could be explained by the higher a nity of AOA to oxygen compared to AOB. MC was signi cantly negatively correlated with AOB abundance (P < 0.01) and AOB OTU numbers (P < 0.05). TN was also found to be signi cantly negatively correlated with AOB abundance (P < 0.01) and AOB OTU numbers (P < 0.05). The pH value was negatively correlated with the abundance of AOB (P < 0.05).

Conclusion
In summary, our results showed the change pattern of the abundance and diversity of AOA and AOB with sediment depth. In the upper layers, the abundance of AOBwas higher than that of AOA, and in the deep layers, AOA outnumbered AOB. Both the ratio of AOA/AOB amoA gene copy numbers and AOA/AOB OTU numbers increased as sediment depth went deeper, which indicated that oxygen indeed led to the niche specialization of AOA and AOB and AOA won the competition with AOB in the oxygen-limited niches. AOA may dominant the ammonia oxidation process in the deep layers. The Nitrosopumilus and Nitrosospira clusters were the absolute dominant AOAand AOB, respectively, indicating an ecological success in the intertidal zone. Figure 1 Quantitative analysis of AOA and AOB in the six sediment samples The OTU numbers of AOA and AOB in the six sediment samples  Maximum-Likehood phylogenetic tree showing the phylogenetic a liations of the AOB sequences recovered from the sediment samples. The numbers at the nodes are percentages that indicate thelevels