One hundred filamentous fungi isolated from the Brazilian Caatinga caves and 12 endophytic fungi isolated from coffee plant (Coffea arabica L.) shoots, belonging to the Culture Collection of BIOGEN, were evaluated for their ability to produce phytases. The cultures were maintained in MilliQ water at 4˚C. The isolates were reactivated on potato dextrose agar (PDA) and incubated at 25˚C for 7 days.

The selection of fungal phytase producers was carried out in medium containing (g/L) phytic acid C₆H₁₈ O₂₄P₆: 5; NaNO₃: 3; MgSO₄∙7H₂O: 0.5; KCl: 0.5; FeSO₄: 0.12; CaCl₂: 0.66; ZnSO₄: 0.17; agar: 30 [15] . The fungi that showed a clear zone (halo) around the colony and possessed an Enzyme Index (EI-halo diameter/colony diameter) greater than 2 were selected to evaluate the specific enzymatic activity.

The molecular identification of the selected fungi was carried out using sequences of the ITS region. The fungi were grown on PDA, and mycelium was scraped with a sterile toothpick. The extraction of total DNA was performed according to the “Mobio Ultra Clean Microbial^® kit”. Amplification reactions were performed in a volume of 30 µL containing 15 µL of Quiagen kit, 12 µL∙H₂O, 10 pmol primer F, 10 pmol primer R and 10ng DNA.

Primers were ITS1 (5’ TCCGTAGGTGAACCTGCGG 3’) and ITS4 (5’ TCCTCCGCTTATTGATATGC 5’) and amplification conditions were as follows: 95˚C 2 min, 95˚C 1 min, 50˚C 1 min, 72˚C 1 min and 72˚C 7 min, programmed for 35 cycles. The amplifications were performed in a thermocycler “MULTIGENE”, Labnet International Inc. The sequences were analyzed with the aid of the SeqAssem program, and alignment with other sequences available in the “GenBank” database was performed by the MEGA program.

The fermentation medium was prepared in 250 mL Erlenmeyer flasks by adding (g/L) of different additional carbon sources (wheat bran, soybean, sugarcane bagasse, oat, corn starch, and rye flour): 5, yeast extract: 3; NaNO₃: 3; MgSO₄∙7H₂O: 0.5; KCl: 0.5; FeSO₄: 0.12; CaCl₂: 0.66, ZnSO₄: 0.17, pH 5.5 [15] , previously autoclaved at 120˚C, 1.5 atm for 20 min. Three mycelium disks of approximately 8 mm of the selected fungus (Muscodor sp.) were inoculated in the sterilized fermentation medium.

The influence of physical parameters on enzyme production and fungal growth as agitation (100, 125 and 150 rpm), initial pH (4, 5, 6 and 7), temperature (25˚C, 30˚C and 35˚C) and incubation time (24, 48, 72, 96, 120, 144 and 168 h) were also analyzed.

After incubation, the cultures were filtered in vacuum Buchner funnel with Whatman filter paper N˚1 to give a cell free filtrate. The filtrate was dialyzed against distilled water for 24h at 4˚C and used for determining the extracellular enzymatic activity.

Phytase activity was determined according to Gulati, Chadha and Saini [16] . The reaction medium was composed of 50 μL of enzyme sample incubated with 50 μL of phytic acid solution (dodecasodium hydrated C₆H₆ Na₁₂O₄P₆.H₂O Sigma^®) 1% (w/v), dissolved in sodium acetate buffer 0.2 M, pH 5.0. The reaction was performed at 40˚C and then added with 100 μL of 15% TCA (trichloroacetic acid) and 300 μL distilled water in each test tube followed by addition of 0.9 μL of the chromogenic reagent (of sulfuric acid 0.76 M, 10% ascorbic acid and 2.5% ammonium molybdate; 3:1:0.5 v/v/v). The tubes were then incubated at 50˚C for 20 min and cooled down; after the reading was taken at 820 nm using a spectrophotometer. In each experiment we included inactivated enzyme controls to estimate the non-enzymatic hydrolysis of substrate. One unit of phytase activity was defined as the amount of phytase required to release one µmol of inorganic phosphorus (Pi) per minute under test conditions [13] .

The total protein concentration was determined according to Bradford method [17] using bovine serum albumin (BSA) as standard. The amount of protein was expressed as mg/mL of sample. Total protein (mg total) was calculated by multiplying the protein in mg/mL for the sample volume.

The thermal stability was determined incubating the phytase samples in aqueous solution at 40˚C, 50˚C and 60˚C. After time intervals of 2, 5, 10, 15, 20, 30, 45 and 60 min, aliquots were removed and maintained in an ice bath for phytase activity determination.

The pH stability of phytase was observed incubating the phytase samplesin citrate buffer 100 mM (pH 3, 4, 5 and 6). The enzyme and buffer were added at 1:1 (v/v), and after incubation for 1 h at 25˚C, aliquots were removed and the relative enzyme activity was determined.

The effect of different salts (1 mM) (NaCl, AlCl₃, CuCl₂, CuSO₄∙5H₂O, FeCl₃, HgCl₂, KCl, MgSO₄∙7H₂O, Zn (NO₃)∙7H₂O and CaCl₂), Urea and EDTA on phytase activity was investigated. The enzymatic reaction was performed as described above with addition of each compound. The results were expressed as relative activity (%).

All experiments were performed in triplicate, using inactivated enzyme controls to estimate the non-enzy- matic hydrolysis of substrate and taking into account the standard deviation for the construction of graphs.

The enzyme production by microorganism was affected by different culture conditions (Figure 2). Considering the influence of the initial pH of the culture medium, the peak of production of phytase by Muscodor sp. was obtained at pH 5.0 (Figure 2(a)) as well as observed for production of phytase by Aspergillus niger ATCC 9142 [24] and A. niger FS3 [25] . These results are in agreement with Vohra and Satyanarayana [26] that described the initial pH from 5.0 to 6.0 as characteristic for production of most microbial phytases.

According to the Figure 2(b), the agitation level also influenced the production of phytase by Muscodor sp. with maximal production obtained at 125 rpm. At 150 rpm, the fungal biomass was higher than that observed at 100 and 125 rpm, but the enzyme production was reduced, indicating that the increased biomass is not directly related to the enzyme production. It is known that the enzyme production is affected by the fungal morphology as well as the influence of agitation on this morphology, what can explain the result obtained for Muscodor sp. The production of phytase by Aspergillus niger under submerged fermentation was increased with stirring speed from 150 to 300 rpm. Moreover, the fungal morphology was influenced by the stirring with small pellets and tangles prevailing in the cultures at 150 rpm, while free filamentous form was obtained at 300 rpm [27] .

The optimum temperature for the production of phytase by Muscodor sp. was 30˚C coinciding with the best temperature for the fungal growth as previously determined in our laboratory. Under this temperature the enzymatic production was 2-times higher than that observed at 25˚C. At 35˚C the enzyme production decreased 20%

if compared to 30˚C (Figure 2(c)). Several studies have investigated the optimum temperature for reaction of the phytase and phytase reported with optimal activity between 40˚C and 60˚C, such as those obtained for the fungi A. fumigatus and A. niger NRRL 3135, with optimal activity at 37˚C [28] and 55˚C [29] , respectively. However, phytases from Aerobacter aerogenes and Candida krusei WZ-001 showed optimum temperatures of 25˚C and 40˚C respectively [30] [31] . Thus, this demonstrates that these enzymes can act on various temperatures can be applied in industrial processes which require low or high temperatures.

The production of phytase by Muscodor sp. as function of incubation period was evaluated according to the parameters previously established: temperature of 30˚C, pH 5.0, agitation at 125 rpm. The maximum phytase activity (26.51 U/mg) was achieved at 144 h of fermentation (Figure 2(d)). This period for enzyme production was minor than that mentioned for production of phytase by Rhodotorula mucilaginosa JMUY [32] . On the other hand, it was higher than that observed for enzyme production by Rhizopus microsporus var. microsporus [13] , by Aspergillus niger CFR 335 and by Aspergillus ficuum SGA [33] .

Under optimized parameters of cultivation (30˚C, initial pH 5.0, 125 rpm for 144 h), the specific activity of the phytase produced by the endophytic Muscodor sp. increased 11.32 times (from 2.34 U/mg to 26.51 U/mg).

The stability of Muscodor sp. phytase was evaluated after storage in 100 mM citrate buffer (pH 3.0, 4.0, 5.0 and 6.0) for 1 h at 25˚C (Figure 3(b)). At pH 5.0, it was observed maximal enzyme stability, but at pH 4.0 and 6.0, the enzyme activity was 20% - 30% from that observed at pH 5.0. The enzyme activity was fully inhibited at pH 3.0. Increase in the pH value reduced the phytase stability (data not shown). The pH stability is an important characteristic that should be considered aiming the phytase application. Efforts have been done to obtain enzymes with a wide pH range of stability as, for example, the recombinant phytases from Aspergillus japonicus

BCC18313 and Aspergillus niger BCC 18081 [34] . Phytase II of Aspergillus niger was stable from pH 3.5 to 9.0 at room temperature over 12 h [35] . Other studies show that phytase I of Monascus was stable at pH 5.5 - 6.5 and phytase II had a better stability from pH 6.0 to 7.0 [36] .

Another important aspect that has technological importance is the thermal stability. The Muscodor sp. phytase was fully stable at 40˚C through the period analyzed. However, at 50˚C and 60˚C the enzyme stability was reduced with half-life (T_1/2) of 10 min and 1.5 min, respectively (Figure 3(a)). Most microbial phytases reported in the literature has shown similar stability at temperatures from 45˚C to 55˚C [37] . However, according to Casey and Walsh [38] , most fungal characterized phytases is stable at temperatures in the range from 30˚C - 60˚C.

The enzyme produced by Aspergillus fumigatus was inactivated only at 70˚C [39] . The phytase produced by biofilm R. microsporus var. microsporus purified was completely stable at temperatures of 30˚C and 40˚C for 120 min [13] .

The influence of different compounds on the phytase activity is presented in the Table 3. The phytase produced by Muscodor sp. was totally inhibited by EDTA, K⁺, Mg²⁺, Zn²⁺, Ca²⁺ and urea (data not shown), and relatively inhibited by Na⁺, Hg²⁺, Cu²⁺, Al³⁺, Fe²⁺. On the other hand, in presence of FeCl₃ 86% of phytase activity was maintained. Phytase produced by Aspergillus niger was also inhibited by Cu²⁺, Zn²⁺, Hg²⁺ and Fe²⁺ [8] . The activity of phytase from Candida krusei WZ-001 was completely inhibited by Zn²⁺ and strongly inhibited by Mg²⁺, but the increase of the concentration of Fe²⁺ (5 mM) resulted in recovery of phytase activity [30] [31] . Increasing the concentration of sodium, the enzyme activity from Muscodor sp. was drastically inhibited. Ullah, Sethumadhavan and Mullaney [40] reported that 500 mM NaCl had an inhibitory effect on the activity of phytase from Aspergillus niger.

Some phytases can be inhibited by metal ions, but it is difficult to determine whether inhibition is a result of metal binding to the enzyme or training complex phytic acid-poorly soluble ion. The formation of precipitated by the addition of Fe²⁺ and Fe³⁺ in the enzymatic assay suggests that there is a reduction in the concentration of active substrate complex of phytic acid formation poorly-soluble Fe [41] . Maenz, Engele-Schaan, Newkirk and Classen [42] evaluated the inhibitory potential at neutral pH of various minerals on the activity of microbial phytases and described the following order of inhibition: Zn²⁺ > Fe²⁺ > Mn²⁺ > Fe³⁺ > Ca²⁺ > Mg²⁺. EDTA (ethylenediaminetetraacetic acid) is a chelating compound capable of forming stable complexes with various metal ions and most enzymes are adversely affected by this compound reducing the catalytic activity [43] .