The effects of Fe- and Mn-oxides and imogolite in the presence of kaolinite on organic nitrogen mineralization and soil enzyme activities

ABSTRACT An incubation experiment was conducted to study the effect of clay content and composition on organic nitrogen mineralization. The experiment measured the mineralization of organic nitrogen from alfalfa residues, enzyme activities, and microbial biomass nitrogen in mixtures of sand, kaolinite, and non-layered colloids (NLCs) with sand as a control. The study found that as the contents of kaolinite and NLCs increased, the mineralization of organic nitrogen and enzyme activity decreased, but microbial biomass nitrogen increased. The maximum decrease in organic nitrogen mineralization was 88.6%, and microbial biomass nitrogen increased from 4.7 to 15.5%. The acid and alkaline phosphatase activities also decreased by 86.0 and 93.6%, respectively, with an increase in clay content. The specific surface area (SSA) of the mixtures showed an inverse relationship with enzyme activity and mineralization of organic nitrogen. Inactivation of extracellular enzymes by adsorption on the surfaces of kaolinite and NLCs, and decreased accessibility of organic nitrogen substrate molecules due to adsorption, reduced the mineralization of organic nitrogen. Microbial biomass nitrogen increased as the water holding capacity of the mixtures increased, indicating the importance of water-filled pores in accommodating active microbial biomass and protecting it from desiccation and predators.


Introduction
Soil organic nitrogen is the biggest terrestrial pool of nitrogen, and mineralization is the nitrogen primary source for plants in most ecosystems (Batjes 2014). Even in agricultural systems with fertilizer applied, soil organic nitrogen is important. Various assays of mineralizable nitrogen have been shown to predict growth and response to fertilizers (Ros et al. 2011;Curtin et al. 2017), but they are time-consuming and expensive. Also, better management of the nitrogen cycle is crucial for global sustainability due to agricultural production, greenhouse gas emissions, soil acidification, and impacts on aquatic ecosystems (Suddick et al. 2013;San Martín 2020). So, we need a better understanding of the mechanisms of nitrogen cycling, particularly the factors influencing the stability and mineralization of organic nitrogen (i.e. production of NH 4 + -N plus NO 3 − -N). Soil microorganisms are responsible for 85-90% of organic matter decomposition (Lavelle et al. 1993), and microbial biomass supplies about 80% of soil organic matter (Liang and Balser 2011;Schmidt et al. 2011;Miltner et al. 2012). Organic compounds may be resistant to microbial degradation and have a long residence time in soil. This could be due to inherent chemical recalcitrance, inaccessibility of microbial population to organic materials due to physical protection, or stabilization by reaction with clays, metal oxides, and other organic molecules (Hassink and Whitmore 1997;Mikutta et al. 2006;Rakhsh et al. 2020). For example, decomposition of organic macromolecules to smaller molecules can lead to reactions with mineral particles and stabilization of organic matter in the soil (Cotrufo et al. 2013;Lehmann and Kleber 2015;Basile-Doelsch et al. 2020). However, the relative importance of these mechanisms in the stabilization of organic nitrogen compounds is not clear (Kögel-knabner et al. 2008).
Adsorption of organic matter to clay and oxide surfaces is an important chemical protection mechanism in carbon and nitrogen cycles. Several factors such as pH, redox conditions, characteristics of organic materials, and mineral surfaces affect the complexation and the degree of sorption of organic materials by mineral surfaces (Jastrow et al. 2007) and hence the mineralization/immobilization of organic nitrogen (Nikolaidis and Bidoglio 2013). Compared to clay minerals, clay size Aland Fe (hydr)oxides have a greater affinity for soil organic matter due to their higher SSA and reactivity (Sanjay and Sugunan 2008;Wallenstein et al. 2010). For example, goethite has a strong association with dissolved organic matter through ligand exchange reactions resulting in Fecarboxylate bonding (Kaiser and Zech 1999;Chorover and Amistadi 2001). In soils, organo-mineral complexes are also formed through the binding of organic ligands with Fe 3+ and Al 3+ on the exchange sites of clay minerals (Oades 1988;Boudot et al. 1989). The formation of these complexes stabilizes soil organic matter (Lopez-Sangil and Rovira 2013;Feng et al. 2014;Yang et al. 2021).
Non-crystalline materials such as allophane have exceptionally large SSA and porosity that results in preservation of organic matter via a combination of retention mechanisms (Mikutta et al. 2006). In allophanic soils, organic matter reacts slowly with allophane, and forming organic-allophane complexes causes organic matter to decompose more slowly than in other soils. In allophanic soils, enzymatic activities per unit of organic carbon are less than in other soils (Ross et al. 1982). Mulder et al. (2001) showed that increasing the concentration of Al in the forest floor caused a 30-40% decrease in the decomposition rate of soil organic matter.
Amino sugars and amino acids are important components of soil organic nitrogen (Hu et al. 2018;Farzadfar et al. 2021). Some studies have suggested that minerals influence the turnover rate of amino sugars and other nitrogen-containing compounds. In contrast, others have shown that minerals (Fe-and Mn-oxides and Al hydroxide) affect the proportion of nitrogen sequestered within microbial biomass but not the cycling of nitrogen through the amino sugar pool . So, while abiotic processes influence amino sugar cycling, adsorb mechanisms remain obscure. Some studies showed enzymes had a strong affinity for clays and oxides (Burns 1986;Naidja et al. 2000), which can affect their activity. Clays can impact enzymes in two ways. One is stabilizing and protecting enzymes that prevent them from being degraded and denatured (Nasseau et al. 2001). As a result, Allison (2006) found that clay minerals, especially allophane, intensified enzymatic activities in young volcanic soils. On the other hand, clays can inhibit the activity of enzymes due to blockage of active sites upon adsorption. Kobayashi and Aomine (1967) and Tietjen and Wetzel (2003) reported a decrease in enzyme activities after adding allophanic and montmorillonite clays. Rakhsh and Golchin (2018) showed that enzyme activities decreased significantly as the clay content increased. Marx et al. (2005) and Yan et al. (2010) observed that enzyme activities decreased in a claysized fraction of soils. Since adsorption of enzymes on clay surfaces occurs by different mechanisms, its effect on enzyme activities may differ accordingly.
Previous studies have only focused on the effect of clay minerals and metal oxides alone on organic matter mineralization or enzyme activity. Our knowledge of these processes in the soil matrix is lacking. Of course, there are some studies on enzyme activity and stabilization of organic matter in natural soils, which are primarily interpreted based on differences in soil texture. Studying the effect of metal oxides on enzyme activity and stabilization of organic nitrogen in natural soils is difficult due to confounding factors, including the type and amount of clay minerals and metal oxides. By using artificial soils, we are able to directly compare the stabilizing ability of various clay minerals and metal oxides and understand the mechanisms influencing organic matter accumulation and stabilization in soils. Many mechanisms have been reported for the effects of clay content and composition, but their relative importance is unclear because basic physical characteristics like specific surface area (SSA) and pore size distribution were not addressed. This research aimed to study the effects of clay content and composition on organic nitrogen mineralization, activities of acid and alkaline phosphatase and CM-cellulase, and microbial biomass nitrogen. Hence, it could conceivably be hypothesized that by reducing the confounding factors such as different soil conditions, variation in the type of metal oxide and the type of clays, and diversity of microbial biomass the exact effect of each of the investigated factors (type and amount of metal oxide in the presence of kaolinite clay) be realized. We did this by studying the decomposition of alfalfa residues in artificial soils with different contents of kaolinite and the non-layered colloids (NLCs) goethite, pyrolusite, and imogolite.

Kaolinite, non-layered colloids (NLCs), and chemical characterization
Kaolinite was acquired from Sigma Aldrich company (CAS No. 1318-74-7, EC No. 215-286-4). The < 2 μm fraction of kaolinite was prepared according to the method presented by Rakhsh et al. (2017). Goethite (α-FeOOH) was synthesized by the method reported by Schwertmann and Cornell (2008). A Fe(NO 3 ) 3 .9 H 2 O solution was slowly neutralized with the KOH solution, and the precipitate aged at 70 °C for 60 h. Manganese oxide (pyrolusite: β-MnO 2 ) was prepared by adding ethanol to KMnO 4 solution at room temperature (Subramanian et al. 2008). Imogolite (Al 2 SiO 3 (OH) 4 ) was synthesized as described by Kijima (2010). The Na 2 SiO 3 .5 H 2 O solution was slowly mixed with the AlCl 3 solution at room temperature for 20 min. The pH of the suspension was adjusted to 4 using NaOH and HCl solutions and kept at 100 °C for 4 d. The suspension was centrifuged, and imogolite particles were freeze-dried.
The mineralogy of kaolinite and NLCs was confirmed by X-ray diffraction (D8 ADVANCE/Bruker, Germany) ( Figure 1). The SSA of kaolinite, NLCs, and NLC-clay mixtures was determined by nitrogen adsorption at 77 K (P/P 0 range of 0.05 to 0.29) with a BELSORP-MINI II (BEL, Japan). Measurement of SSA and pore volume of kaolinite and NLC samples was performed in triplicate ( Table 1). Images of kaolinite and NLCs by transmission (TEM) and field emission scanning electron microscopy (FESEM) were collected using a CM 120 microscope (Philips, Eindhoven, Netherlands) and MIRA3 ×MU (TESCAN, Brno, Czech Republic) (Figures 2 and 3), respectively. Additional information is provided in Rakhsh et al. (2020). Characteristics of kaolinite, NLCs, and NLC-clay mixtures (Table 1) were determined after autoclaving moist of these materials. Because the artificial soils were then sterilized, these materials were also autoclaved to ensure that the temperature and pressure did not affect the materials' properties and that all materials had the same conditions.
The pH (1: 5 soil: water) and EC (1: 5 soil: water) of kaolinite and NLC-clay mixtures were measured with a glass electrode pH meter (Jenway 3510) and EC meter (Jenway 4510) (Rayment and Lyons 2011). Cation exchange capacity (CEC) of the materials was determined using the ammonium acetate (pH = 7) method (Mikutta et al. 2006). All measurements had three replications and their means are reported ( Table 1).

Preparation of artificial soils and inoculum
The sand (quartz) had a particle size of 0.05-2 mm and was prepared as described by Rakhsh et al. (2017). Before the incubation experiment, the alfalfa (Medicago sativa) residues were dried in a fanforced oven at 55 °C for 72 h and then passed through a 1-mm sieve (Rakhsh et al. 2017). The organic carbon and total nitrogen contents of the alfalfa residues were 48.2% and 3.7%, respectively, and the calculated C/N ratio was 13.
Samples of artificial soils (50 g each) were prepared in triplicate by mixing the sand with 0, 15, 30, or 45% (w: w) of the prepared clays, which had kaolinite: NLC ratios of 1: 0, 10: 1, or 5: 1. To perform the incubation experiment, the alfalfa residues were added to artificial soils at 50 mg kg −1 , equivalent to an organic nitrogen addition of 1.85 mg g −1 in artificial soil.
The water holding capacity of all artificial soils was measured in triplicate using a pressure plate apparatus (Pressure plate and ceramic plate cell, ELE International) (Klute 1986). The artificial soils were sterilized by autoclaving moist at 120 °C for three times 2 h. Before incubation, the sterilized soil samples were inoculated with microorganisms by adding a solution prepared from natural soil (Rakhsh et al. 2017).

The incubation experiment
Samples of artificial soils (equivalent to 50 g oven-dry artificial soil) were placed in incubation jars (750-mL) in triplicate and 5 mL inoculum together with enough distilled water was added to each jar to wet the sample evenly. The inoculated artificial soil samples were air-dried to prevent overwetting and then remoistened with distilled water to 60% of water holding capacity for incubation. To keep the moisture content of the artificial soil samples constant during incubation, distilled water was added to the bottom of the incubation jars (750-mL), which were sealed. The jars were opened to prevent anaerobic conditions once a week. The artificial soils were incubated for a total of 60 d in the dark at 23°C (in a constant temperature room) (Rakhsh et al. 2017(Rakhsh et al. , 2020Rakhsh and Golchin 2018).
Mineralization of organic nitrogen in the alfalfa residues was studied by measuring the concentrations of NH 4 + -N and NO 3 − -N in the soil samples on the 15th, 30th, 45th, and 60th d of incubation. Mean pore radius: 2Vp/As, Vp: total pore volume (cm 3 g −1 ) and As: SSA (m 2 g −1 ).
Activities of acid and alkaline phosphatases and CM-cellulase and microbial biomass nitrogen content were determined at 60 d of incubation.
To measure the concentration of NH 4 + -N and NO 3 − -N, a moist soil sample equivalent to 10 g of dry soil was placed in a 250-mL flask, 100 mL of 1% K 2 SO 4 was added and the suspension was shaken on a mechanical shaker for 60 min at 300 rpm (Ika Ks 260/Germany). The sample was then centrifuged at 3000 rpm for 6 min (Labnet Hermle Z400) and the supernatant was filtered (0.45 µm). To measure NH 4 + -N the filtered extract (50 mL) was treated with dilute NaOCl and phenol + NaOH solution (4 mL) to convert the ammonium ions to a blue indophenol, whose concentration was determined by measuring absorbance at 630 nm (UV/VIS Perkin Elmer-lambda 25) after 90 min. (Alef and Nannipieri 1995). To measure NO 3 − -N, 10 mL of clear filtrate and 2 mL of 10% H 2 SO 4 were transferred to each of two 100-mL flasks and the volume was adjusted to 100 mL with K 2 SO 4 solution and mixed. 2-3 granules of zinc were added to one of the flasks and allowed to stand for 4-5 h for the reduction of nitrate (control). The absorbance of the samples was measured at 210 nm against the reagent blank (Alef and Nannipieri 1995). Microbial biomass nitrogen was measured using the fumigation-extraction method. Two samples of moist soil, each equivalent to 5 g of dry soil, were placed in separate beakers, one of which was used for fumigation and the other kept non-fumigated as a control (Brookes et al. 1985). The fumigated and non-fumigated soil samples were extracted with 25 mL of 0.5 M K 2 SO 4 (three replicates). Ten mL of a reducing agent (KCr(SO 4 ) 2 ) along with 300 mg of pure zinc powder was added to a 250-mL digestion tube. Then 30 mL of filtered extract was added to the digestion tube and placed at room temperature for 2 h. Before starting the digestion, 0.6 mL of 0.19 M CuSO 4 solution and 8 mL of concentrated H 2 SO 4 were added to the sample. The samples were homogenized and gently heated to 80°C for 2 h to evaporate the water in the samples. The samples were then heated at 280 °C for 3 h. The digestion tubes were cooled, then samples were distilled in the presence of NaOH and H 3 BO 3 and the concentration of nitrogen was obtained in each sample. A K EN factor of 0.54 was used to convert the organic nitrogen flush to microbial biomass nitrogen (Joergensen and Mueller 1996).
The activities of acid and alkaline phosphatase in soils were determined as described by Tabatabai (1994), p. 1 g soil was incubated in a 50-mL test tube with 4 mL of Modified Universal Buffer (pH = 6.5 for assay of acid phosphatase and pH = 11 for assay of alkaline phosphatase) and 1 mL of 0.025 M p-nitrophenyl-phosphate (C 6 H 6 NO 6 P) for 1 h at 37°C (Incubator Nuve ES 110-250). The reaction was ended with the addition of 4 mL of 0.5 M NaOH and then 1 mL of 0.5 M CaCl 2 solution was added to flocculate the soil. The solution was filtered (0.42 µm) and the amount of p-nitrophenol produced during the incubation period was determined in the filtrate by spectrometry at 400 nm (UV/VIS Perkin Elmer-Lambda 25).
The activity of CM-cellulase was measured as described by Schinner and Von Mersi (1990), p. 5 g soil samples were incubated in 50-mL test tubes with 15 mL of 0.7% sodium carboxymethylcellulose solution and 15 mL of 2 M sodium acetate solution (pH = 5.5) for 24 h at 50 °C. After incubation, all tubes were shaken briefly and filtered immediately and 0.5 mL of filtrate was added to 20 mL of distilled water in test tubes. In the test tubes, 1 mL of reagent A and 1 mL of reagent B were added to 1 mL of diluted filtrate and the test tubes were sealed, mixed well, and retained in a boiling water bath for 15 min (Memmert WNE 45). The test tubes were then cooled in a cold-water bath for 5 min, 5 mL of reagent C was added to each tube, blended, and allowed to stand at room temperature for 60 min for color development. After 30 min, the amount of glucose produced during the incubation period was quantified in the filtrates by spectrometry at 690 nm (UV/VIS Perkin Elmer-lambda 25), against the reagent blank (Schinner and Von Mersi 1990).

Statistical analyses
A one-factor ANOVA experiment with three replications was used to assess the effects of artificial soils containing 15, 30 and 40% of kaolinite without and with NLCs as independent variables, on organic nitrogen mineralization, microbial biomass nitrogen, and soil enzyme activities, as dependent variables. The control was pure sand and the NLCs were goethite, pyrolusite and imogolite and the ratios of kaolinite to NLCs were, 5: 1 and 10: 1. The data were analyzed using a one-way ANOVA and means were compared by Duncan's Multiple Range Test (DMRT). Before applying ANOVA, the normality of the data and the homogeneity of variance were checked using Kolmogorov-Smirnov, and Levene tests, respectively. For data analysis, the SPSS (windows version 25.0, SPSS Inc. Chicago, U.S.A) and SAS (version 9.4, SAS Institute Inc. Cary, NC) software were employed.

Mineralization of nitrogen in the alfalfa residues
Mineralization of organic nitrogen occurred throughout the first 60 d of the incubation and was affected significantly (p < 0.0001) by the treatments (Table 2). Mineralization rate decreased with increasing clay and NLC content, with the lowest rate occurring in soil with the highest clay content, i.e. 0.8% of added nitrogen mineralized in soil with 45% clay comprised of 5: 1 kaolinite: imogolite (Figure 4), and the highest rate (7% of added nitrogen) occurring in control (Figure 4). The soils with imogolite had the lowest mineralization rate (across clay and NLC contents) (Figure 4). The mineralization of organic nitrogen was closely related to the SSA, with linear regression accounting for

Microbial biomass nitrogen and enzyme activity
Microbial biomass nitrogen content was significantly affected (p < 0.0001) by the treatments (Table 2), increasing as clay and NLC contents increased. The content of microbial biomass nitrogen, as a proportion of the nitrogen added in the alfalfa residues, ranged from 4.7% in the sand to 15.5% in the soils with 45% clay comprised of 5: 1 kaolinite: goethite ( Figure 6). It was highest in the soils containing goethite (across the clay contents) (Figure 6). Microbial biomass nitrogen was closely related to SSA, with linear regression (Figure 7).  Enzyme activities were significantly affected (p < 0.0001) by the treatments (Table 2), decreasing as clay and NLC contents increased. The acid and alkaline phosphatase activities ranged from 55.29 and 48.2 μg p-nitrophenyl g −1 dry soil h −1 in the soil with 45% clay comprised of 5: 1 kaolinite: imogolite to 396.24 and 752.79 μg p-nitrophenyl g −1 dry soil h −1 in the sand, respectively (Figures 8  and 9). The activity of CM-cellulase was similarly affected by the treatments (Figure 10). Among the NLCs, the lowest and highest enzyme activity was measured in soils with imogolite and goethite (across the clay and NLC contents), respectively (Figures 8-10). In general, the presence of NLCs reduced the activity of enzymes. Second, to the sand, the highest enzyme activity was observed in soils containing kaolinite but not NLCs (all three kaolinite contents) (Figures 8-10). The effects of the treatments on the activity of enzymes were related to the SSA with linear regression (Figure 11).

Discussion
Mineralization of organic nitrogen (Figure 4) and the activity of enzymes (Figures 8-10) decreased as the content of clay increased. In contrast, soils with higher clay contents had higher contents of microbial biomass nitrogen ( Figure 6). The decreased enzymatic activity with increasing soil clay content shown in this study may be one of the reasons for the reduction of organic nitrogen mineralization in soils with high clay (>15%) contents. Additionally, the adsorption of organic substrates containing nitrogen may decrease microbial access. Thereby mineralization of organic nitrogen was reduced. The adsorption capacity of soils for nitrogenous compounds depends on the type and content of clay and the chemistry of nitrogenous compounds (Nikolaidis and Bidoglio  2013). Also, this research showed that mineralization of organic nitrogen, enzyme activities, and microbial biomass nitrogen contents were influenced by NLC types and contents (Figures 4-11).
The close correlation of organic nitrogen mineralization with SSA of mixtures in this study ( Figure 5) and a similar correlation between organic carbon mineralization with mixtures' SSA in our previous research (Rakhsh et al. 2020), as well as an inverse relationship between SSA of mixtures and enzymatic activity, implies that the influence of clay type and content (texture) on the mineralization of organic carbon and nitrogen and enzyme activities are imposed through the surface  supplied by soil minerals for adsorption. The interaction of clay minerals and extracellular enzymes results in the adsorption, the intensity of which is determined by the chemical structure of the enzymes and the SSA, surface charge, and cation exchange capacity of the clays. The activity of enzymes decreases when they are bound to clay surfaces, as adsorption blocks their active sites or causes deforms (Zimmerman et al. 2011;Olagoke et al. 2019). Nevertheless, the activity of enzymes can be intensified or at least maintained if binding to clay surfaces stabilizes their structure and allows them to retain their catalytic activities (Sanjay and Sugunan 2008). Most of the activity of some enzymes in the soil may be related to stabilized forms rather than free forms (Marx et al. 2005;Schimel et al. 2017). Soil-stabilized enzymes may be an important reservoir of enzymatic activity that acts when the substrate becomes available and stimulates the soil microbial community. In addition, it enables decomposition of various compounds in the soil when enzyme production by the microbial community is low or not possible due to environmental stresses (Stursova and Sinsabaugh 2008). Olagoke et al. (2019) studied the effects of montmorillonite content on the activity of extracellular enzymes. They showed that the catalytic activity of α-glucosidase reduced by 76% as the content of montmorillonite increased from 0 to 10%. These results are consistent with the effects of kaolinite content in our study.
Earlier work has shown the effects of oxides on enzyme adsorption and activity, but our work suggests that the observed effects are more controlled by SSA than by the mineralogical composition of the clays. The presence of oxides and hydroxides of Fe, Al, and Mn may alter the adsorption of enzymes on clay surfaces . For example, the presence of Alhydroxide on montmorillonite surfaces resulted in the lower activity of urease, invertase, and tyrosinase but the higher activity of phosphatase (Rao et al. 1996;Gianfreda et al. 2002). In our study, the presence of goethite, pyrolusite, and imogolite decreased the activity of all enzymes (acid and alkaline phosphatase and CM-cellulase) compared to kaolinite (Figures 8-10). Imogolite had the most considerable effect and goethite the smallest (Figures 8-10). The result of pyrolusite was close to that of imogolite (Figures 8-10). Olagoke et al. (2020) reported that soil weathering and the formation of secondary and pedogenic minerals alter the mineralogical composition of soils and affect soil microbial processes by changing enzymatic activity. Our findings suggest that this effect occurs due probably to an increase in the number of adsorption sites and the amount of surface area available for the adsorption of enzymes.
Contradictory results have been reported for the effect of clay content on the content of microbial biomass carbon and nitrogen in soils, some showing a positive impact (Rakhsh et al. 2020) and others a negative impact (Liddle et al. 2020). The positive relationship between microbial biomass carbon and clay content has been attributed to the physical protection of microbial biomass by clays (Ladd et al. 1996). The microbial biomass nitrogen increased as the water holding capacities of mixtures increased (Figure 12), indicating that water-filled pores probably play an important role in protecting microbial biomass from desiccation and predators. A larger quantity of soil water means more spaces in which microbes can be active, mineralize organic matter, grow and incorporate nitrogen into their biomass. Our results also suggest that more microbial biomass does not necessarily mean more microbial degradation, as long as the extracellular enzymes secreted by soil microorganisms are inactivated through the adsorption to the surfaces of inorganic colloids ( Figures 5, 11 and 12). A greater abundance of micropores might imply more protection for soil microorganisms. Furthermore, kaolinite and kaolinite: imogolite mixture seem not to follow the general trend of increasing biomass with water content. It is possible that the presence of Al and toxicity to microorganisms also reduced the amount of microbial biomass in the same water content compared to other NLCs. The acidic pH of imogolite compared to other NLCs (Table 1) causes more release of Al in soil and its toxicity for microorganisms (Jansen et al. 2003).
The protection of soil organic matter by fine particles and reduction in the amount of carbon and nitrogen mineralized in fine-textured soils are well known (Sorensen 1981;Cai et al. 2016;Rakhsh et al. 2017Rakhsh et al. , 2020Rakhsh and Golchin 2018;Riaz and Marschner 2020). As the clays enhance the formation of aggregates and provide more surfaces for adsorption, physical protection of organic nitrogen through the occlusion within aggregates inaccessible to soil microorganisms and, or adsorption onto the clay surfaces may have been partially responsible for the reduced nitrogen mineralization in soils with high clay content. Organic matter can sorbed onto the surface of clays (Tisdall and Oades 1982). As a result, the organic carbon in the fine pores associated with clay minerals is not available for soil organisms (Elliott and Coleman 1988). A comparison of the diameters of pores in clay and sandy soils with the diameters of bacteria and fungi clearly shows this. The diameters of pores in sandy soils were 6-30 µm and in clay soils 0.2-1.2 µm (Bouwman et al. 1993). At the same time, the diameters of bacteria and fungi are about 0.5 and 5 µm, respectively (Bae et al. 1972).
The formation of the Al-organic matter complexes probably provides a mechanism for the stability of organic matter against microbial decomposition. Increasing the initial aluminum to carbon ratio to more than 0.1 resulted in the reduction of organic matter mineralization by more than 50% (Schwesig et al. 2003). In the presence of amorphous Fe-rich materials and Fe 3+ , organic matter is preserved against microbial decomposition, but the inhibitory effect of Al is more than that of Fe (Boudot et al. 1989). The effect of Al is partially due to the toxicity of Al 3+ , which leads to reduced growth and impaired respiration rate of microorganisms (Illmer and Erlebach 2003;Heckman et al. 2013). Our study results show that nitrogen mineralization decreased in the presence of NLCs. In the presence of imogolite, organic nitrogen decomposition was less than goethite and pyrolusite. Probably, Al in imogolite led to reduced activity of microorganisms. Aomine and Kobayashi (1964) observed that the microbial breakdown of cellulose and albumin in Andisols containing allophane was less than in Alfisols having illite and montmorillonite.
In the present study, organic nitrogen mineralization decreased with increasing soil clay content, but the content of microbial biomass nitrogen increased, indicating that sufficient fine pores were protecting organic nitrogen from microbial decomposition and also larger pores to protect microbial biomass from predators and desiccation. Incorporation of a higher proportion of soil total nitrogen into microbial biomass, better protection of microbial biomass from grazing and desiccation, and lower specific microbial respiration in the finer-textured soils also has been reported as the reasons for the higher content of microbial biomass nitrogen in fine-textured soils (Bauhus et al. 1998).

Conclusions
Clay minerals and NLCs are among the active components of soil that play a key role in the dynamics of organic nitrogen and its persistence in soil. However, their simultaneous presence in the soil makes it difficult to examine the contribution of each of them to organic nitrogen persistence. Increasing the amount of clay minerals and NLCs in the soil increases the reactive surface area and thus the interactions of organic nitrogen and extracellular enzymes with these clay minerals. It seems that increased SSA, due to increased clay content or the addition of NLCs to kaolinite, increased the physical protection of organic matter through the inactivation of enzymes, despite the increase in microbial biomass. The enzyme activities decreased sharply with increasing SSA of the mixtures. The reduced enzyme activity in mixtures with high SSA reduced the decomposition of organic matter, thus the mineralization of organic nitrogen. We showed that microbial biomass nitrogen increased as the water holding capacities and SSA of the mixtures increased, indicating that water-filled pores play an important role in protecting microbial biomass from desiccation and predators. A larger quantity of soil water means more space in which microbes can be active, mineralize organic matter, grow and incorporate nitrogen into their biomass. These results also suggest that more microbial biomass does not necessarily mean more microbial degradation, as long as the extracellular enzymes secreted by soil microorganisms are inactivated through the adsorption to the surfaces of inorganic and, or organic colloids. Kaolinite: imogolite mixture seem not to follow the general trend of increasing biomass with water content. It is possible that the presence of Al and toxicity to microorganisms also reduced the amount of microbial biomass in the same water content compared to other NLCs. It is feasible that soils dominated by these types of NLCs may develop different organic matter compositions in the long term, with microbial-derived carbon and nitrogen, which will affect the decomposition of organic matter. Furthermore, it will be fascinating to test this theory in soil with variability in microbial biomass and different NLCs to investigate soil organic matter dynamics.