Response surface methodology to optimize inulinase production by a newly isolated Penicillium amphipolaria strain by solid-state fermentation of Saccharum arundinaceum

In the present investigation, a new fungal inulinase producer Penicillium amphipolaria KAS 2555 has been isolated from the soil of dead mangroves litter area, followed by the inulinase production and optimization by solid-state fermentation using a low-cost substrate – hardy sugarcane (Saccharum arundinaceum). While screening, only Penicillium amphipolaria KAS 2555 showed the hydrolysis zone on the plate containing inulin media. The exoinulolytic nature of inulinase and its form of action was conrmed by thin-layer chromatography (TLC). After 96 h of the fermentation period, an activity of 2.45 U/gds was obtained. The I/S ratio of 0.59 proved that the enzyme is inulolytic in nature. Media optimization was performed to obtain a regression model using Central Composite Design (CCD). For optimization, ve signicant media components viz., inulin, (NH 4 ) 2 SO 4 , K 2 HPO 4 , KH 2 PO 4 and, NaCl were used. A 3.10-fold increase in activity of inulinase (7.59 U/gds) was obtained under the optimal settings of (g/gds) inulin- 0.1, (NH 4 ) 2 SO 4 - 0.002, K 2 HPO 4 - 0.1, KH 2 PO 4 - 0.02 and NaCl- 0.02.


Introduction
A variety of hydrolysing enzymes of industrial signi cance from various sources of microbes have been identi ed. Inulinases belong to one of these signi cant class of enzymes that acts on β, 2 − 1 linkages of inulin to yield different industrial products [1]. The β, 2 − 1 inulin linkages are affected by inulinases differently. They are therefore categorized as: exoinulinase and endoinulinase according to their pattern of action on inulin. Exoinulinase breaks inulin sequentially, while endoinulinase breaks the polymer internal links randomly to create varying chain length fructooligosaccharides [2].
Being a low calorie sweetener, fructose is bene cial over traditional sucrose and fructooligosaccharides are effective prebiotics [8]. Usually, production of fructose from starch involves the use of three enzymes namely, α-amylase, amyloglucosidase and glucose isomerase which leads to only 45% yield. In contrast to this, inulinases are used in a single enzymatic inulin hydrolysis step which yields 95% of fructose [9].
Inulinase producing microbes have been widely isolated from sources like marine soil [10], rotten wood [11], inulin rich plants [12], etc. Mangrove soil is also proved to be an e cient source of inulinase producing fungi [13]. A mangrove habitat consists of two parts, production-based where living mangrove trees are found and dead-based where dead mangroves are found. Dead mangroves litter is made up of leaves, barks, seeds, branch, wood, etc. A wide range of microbes inhabits the different parts of the deadbased mangrove environment like soil, leaf, wood, algae, and sand. In the present study, topsoil from the rotten wood of mangrove trees has been used for the isolation of inulinase producing microbes [14].
Mangrove soil is a rich source of microbes that produces enzymes like chitinase, cellulase, protease, and amylase [15]. To date, there is no report on the production of inulinase or any other enzymes from Penicillium amphipolaria species. This is the rst report on the solid state production of inulinase enzyme using this new fungal species isolated from the mangrove soil in the dead mangroves area.
Microbial sources of inulinases are bacteria, yeast and lamentous fungi. Nevertheless, there has been a considerable increase in fungal sources because of thermal and pH stability and growth on low cost substrates to produce important industrial products. Some active fungal inulinase producers are Aspergillus sp., Rhizopus sp., Penicillium sp., etc [16]. In the present study a new fungal species "Penicillium amphipolaria" has been isolated. Till date there is no report on the application of this fungus in any eld of research.
In general, inulinase is produced by submerged fermentation, nonetheless, nowadays, solid-state fermentation (SSF) is gaining a lot of attention because of the low cost process. Several cheap substrates such as sugarcane bagasse, peels of onion, pressmud, wheat bran and rice bran have been used for the production of inulinase enzyme [17]. Very recently, our research group have documented the use of a cost-effective solid substrate -"Saccharum arundinaceum", usually called as hardy sugarcane, for the inulinase production by employing Aspergillus avus var. avus strain ATCC 16883 [13].
In the process of fermentation, media optimization is necessary to achieve optimal growth of microbes and enhance the synthesis of metabolites. The relevant statistical method of optimization i.e., response surface methodology (RSM) uses analysis of variance (ANOVA) and multiple regression analysis to analyse the optimum model [18]. RSM determines the optimum process conditions using Central composite design (CCD) and Box-Behnken design (BBD) [19]. Therefore, this method was used to optimize the media components of inulinase production by SSF from the rotten wood mangrove soil isolated fungus, P. amphipolaria. This is the rst report on the production of inulinase from P. amphipolaria using dead-based mangrove soil as the chief source of fungal isolation.

Isolation and screening of inulin-degrading fungal strains
The method of isolation was performed using the surface soil of the rotten wood of the mangrove detritus situated in the estuary of Halady river, on the Herikudru island, Kundapura Taluk (13°38'28"N, 74°42'01"E), India. Here, the mangrove area is mostly comprised of mangrove trees that have died due to urbanization and deforestation. The soil collection was done in July 2017. Firstly, a serial dilution of the soil sample was carried out followed by plating 0.1ml of the sample on the Czapex Dox Agar medium (CDA). CDA medium consists of (g/l): sucrose -30, K 2 HPO 4 -1, NaNO 3 -2, KCl -0.50, MgSO 4 -0.50, FeSO 4 -0.01 and agar -15. It is used for the isolation of fungal species. The pH of the media was kept at 7.4 and autoclaved for 45min at 121℃ and 15-20 psi. The plates were incubated for 3-4 days at 30℃ to get diverse colonies of fungal species [13]. In the process of primary screening, individual colonies from different fungal colonies were selected and plated aseptically on inulin media. The inulin media consists of (g/l): inulin -6.00, Na 2 HPO 4 -33.9, NH 4 Cl -5, NaCl -2.50, KH 2 PO 4 -15, and agar -15. The plates were incubated for three days at 30℃. The pH of the media was kept at 6.7 [21]. Lugol's plate assay was performed to screen the microorganisms producing extracellular inulinases [22].

Identi cation of the fungal strain
The fungal strain which gave the highest activity was recognized using 18s rRNA gene sequencing ( 571bp) at National Centre for Microbial Resource (NCMR) Pune, India. Using the Tamura-Nei model, the evolutionary history was inferred by using the Maximum Likelihood method. The tree was constructed using 1000 bootstrap. The bootstrap support is 95 for the parent node. Tree construction method was selected based on the best model tting option in MEGA6 [41].

Inoculum preparation
The fungal spores were totally removed from the agar plates using 0.05% of sodium chloride solution. A haemocytometer (Improved Neubauer, Rohem, India) was used to count the number of cells. For SSF, 2×10 6 spores/ mL of spore suspension was used [23].

Pretreatment of solid substrate
Stems of S. arundinaceum collected from Chitpady colony, Udupi, India were cleaned properly couple of times in running water. Small cubes of the stems were prepared and sun-dried for a couple of days to remove moisture. Subsequently, ne powder was made by grinding the dried cubes in a mixer grinder and passed through a 1.18mm sieve. After sieving, the particles with larger sizes were removed and the ne powder was used in SSF as a solid substrate.

Solid-State Fermentation
SSF was carried out by mixing 2g of S. arundinaceum powder and 5ml of inulin media in 250ml conical asks. Moisture on a dry basis was maintained at 77.8%. All the asks were tightly covered with hydrophobic cotton and autoclaved. The asks were allowed to cool followed by the addition of 2ml of the spore suspension prepared earlier. Proper mixing of the solid substrate, fermentation media and the spore suspension was done to ensure the availability of nutrients uniformly. The asks were incubated at 30℃ for ve days [42]. During the process of preliminary screening, it was found that the maximum production of inulinase occurred at 96 th h. All the subsequent experiments were thus carried out for 96 h.

Extraction of Inulinase enzyme
For the extraction of enzyme after the completion of fermentation, the fermented contents were mixed with 30ml of sodium acetate buffer (0.1molar) having a pH of 4.75. The asks were then kept for 1 h on a rotary shaker (180rpm) to extract the enzyme. Subsequently, a clean muslin cloth was used to lter out the enzyme extract followed by centrifugation at 844 G-force (10,000 rpm in rotor 9, Tabletop Superspin Centrifuge -V/F M, Plasto Crafts., Mumbai) at 4℃ for 10 min. The determination of reducing sugars was done by the dinitrosalicylic acid (DNS) method using the cell-free supernatant [25].

Inulinase assay
For the assay of inulinase enzyme, 2g inulin was added to 10ml of 0.1M sodium acetate buffer to make a substrate solution. Then the reaction mixture which consisted of 0.9 ml of substrate solution and 0.1 ml of the cell-free supernatant was incubated for 10 min, followed by the addition of 1ml of DNS reagent. The mixture was heated for 10 min in a water bath. For the blank, the enzyme was deactivated by preheating the reaction mixture. The quantity of enzyme which produces 1μmol of fructose per min is de ned as one unit of inulinase activity.
Similar to inulinase assay, the invertase assay was determined using sucrose in the reaction mixture and the activity of invertase was determined with respect to the sucrose activity [21].

Inulinase to Sucrose activity (I/S ratio)
Usually, the inulinase activity is characterized from the activity of invertase using the I/S ratio. An I/S ratio higher than 10 -2 proves to have a high production of inulinase whereas an I/S ratio lower than 10 -4 proves to have a high production of invertase enzyme in the culture media [2].

Optimization of Inulinase Production
Proper understanding and former knowledge of bioprocess are necessary to obtain a model with accuracy. The data obtained from the experiments were studied using ''Minitab 16.2.2.0", a design software. Using Central composite design, a quadratic model was obtained. A fractional CCD 2(5-1) was used here and the signi cance level was kept at 95%. CCD consists of factorial trials and star points which assist the de ning of quadratic effects. The centre points assess the pure variance with inulinase production as the output variable [26]. The optimization of the ve signi cant media constituents which increased the inulinase production was done using RSM. A set of thirty-two experiments were performed by studying the ve components at 5 different levels, keeping zero as the central coded value. The experimental results of all the 32 experiments -the inulinase activity (Y) were recorded and submitted to ANOVA to carry out the analysis of data using multiple regression [27]. Lastly, an empirical model showing the relationship between the independent components and the measured response was achieved.

Analysis of end products
The pattern in which inulinase enzyme works on inulin was gured out by thin-layer chromatography (TLC) analysis. To perform the reaction of hydrolysis, 1.0 ml of the cell-free supernatant was mixed with 0.01g of inulin. The reaction mixture was incubated at 56℃ for six h. Then, the reaction was stopped by boiling the reaction mixture for 10 min. Along with the sample test (T), four standards namely, deactivated inulinase (I), fructose (F), glucose (G), sucrose (S), were spotted on the TLC plates. It was then air-dried and dipped in a developer solution composed of butanol:ethanol:water (50: 30:20). The plate was again air-dried and sprayed with diphenylamine reagent (1% diphenylamine, 10% phosphoric acid and 1% aniline). This reagent develops colors on reaction with sugars. Finally, the plate was kept in a hot air oven for 10 min at 120℃ for drying [21].

Seliwanoff's test
To distinguish between aldoses and ketoses, Seliwanoff's color test is carried out. Ketose sugar can be differentiated from aldose sugar by the formation of a burgundy or cherry red color with resorcinol on condensation, whereas aldose sugar doesn't give any color [28]. Heating for a long time is avoided as aldose sugars will also give a red color [29]. To perform this test, three controls namely fructose, glucose and deactivated inulinase with inulin was used and a test sample containing the supernatant along with inulinase and inulin was used. 0.5 ml of the test sample was mixed with 1ml of Seliwanoff's reagent and heated for 2 min. To 1ml of Seliwanoff's reagent, 1ml each of the following, namely, standard fructose, standard glucose, and a pre-heated mixture of deactivated inulinase and inulin were also mixed and heated for 2 min.

Osazone formation test
Osazone test is performed to determine and identify the sugars qualitatively. It is a simple test in which sugars form distinctive crystals of osazone on reaction with phenylhydrazine. Distinctive broom and needle-shaped osazones are formed by monosachharides like glucose and fructose. The following test was carried out using the optimized cell-free liquid of inulinase and inulin mixture along with the fructose control. 2 ml of the control fructose and 2 ml of the supernatant were mixed with phenylhydrazine hydrochloride solution followed by heating for 30 min in a water bath [30]. After cooling down of the solutions, the distinctive images of the osazone crystals were visualized by a light microscope (Euromex iscope, Holland) at 40x magni cation.

Microorganism
The fungal isolate that has grown on the inulin media were chosen for Lugol's plate assay. The fungal strain which had formed a clear halo in the plate assay has been indicated as DD MSF 1 in Fig. 1a and Fig. 1b. Figure 1a shows that the isolate has moderately deep colonies, white mycelia, and occose texture. Visagie et al., (2016) reported the same features for Penicillium amphipolaria KAS 2555. They isolated this species from the quarter main mountains of the dry valleys of Antarctica [31]. Till date, there is no work reported in any eld on this fungal species. Figure 2 shows DD MSF 1 represented in the phylogenetic tree as sequenced by NCMR Pune, India. DD MSF 1 was found to have a 99% similarity with Penicillium amphipolaria strain KAS 2555. This study reports the isolation of this fungal species from detritus mangrove soil for the rst time. In the GenBank under the (Accession number KT887872.1), the gene sequence of the ribosomal RNA of DD MSF 1 has been deposited. Among the fungal producers of inulinases, the Penicillium genus is one of the most common. According to the literature survey, species such as P. brevicompactum [11], P. rugulosum [32], P. citrinum [23], have been used for the production of inulinase enzyme.

Fermentation and optimization of the signi cant nutrients
After the completion of 96 h of fermentation, DD MSF 1 yielded a maximal activity of 2.45 U/gds. The activity gradually decreased after 96 h with the increase in the incubation period possibly due to the carbon source exhaustion in the fermentation media, catabolic repression, or due to the secretion of proteolytic proteins causing the denaturation of proteins. The experimental time of fermentation is consistent with the reported values. A maximal inulinase activity of 3.48 U/gds was obtained from A. avus ATCC 16883 [13], A. terreus gave a maximum activity of 5.680 ± 0.183 U/gds [11], a maximal activity of 319 U/ml was obtained from A. niger ATCC 20611 in shake ask [33]. Penicillium amphipolaria KAS 2555 gave an I/S ratio of 0.59, hence, proved the speci city of the enzyme inulinase. The result obtained in the present study had similarities with other microbial producers of inulinases as reported in the literature. An I/S ratio of 2.56 was obtained from A. avus ATCC 16883 [13], a ratio of 3.38 was obtained from P. oxalicum BGPUP-4 [21], a ratio in the range of 0.53-0.66 was obtained from Penicillium sp. NFCC 2768 [34], a ratio of 0.57-2.49 was obtained from Rhizopus microspores 13aIV [35].
During the preliminary process of traditional screening, ve nutrient components viz., inulin, (NH 4 ) 2 SO 4 , K 2 HPO 4 , KH 2 PO 4, and NaCl were screened out to be signi cant and their levels were determined for CCD as shown below in Table 1a. Thirty two experiments were performed according to the combinations given by the software. The number of runs were kept randomized. The process parameters were kept constant as follows: temperature (30℃), moisture (77.77%), inoculum size (2 ml of 2 × 10 6 fungal spores), amount of solid substrate (2 g), fermentation media (5 ml), pH (7) [36]. Five signi cant factors namely, inoculum size, moisture, the ratio of the quantity of wheat bran to rice husk, pH and temperature were screened by Sheng et al., (2009) for the optimization of inulinase enzyme using Cryptococcus aureus G7a in SSF [37].
Among the signi cant term, the optimum values were determined by CCD. In run 10, the highest activity of 7.38 U/gds and in run 22, the lowest activity of 5.74 U/gds were obtained as presented in Table 1b. The data obtained from the experiments are reproducible signi cantly as there is a small difference between the central points. ANOVA was used for the design of the experiment to determine the model e ciency and tness as shown in Table 2. F value of the model is shown as 10.24 which proves that it is signi cant.

Page 8/24
At a 95% level of con dence and a (p-value < 0.05), X 1 is signi cant among the linear model factors, X 2 2 and X 3 2 are signi cant among the square interactions and X 1 *X 2 , X 2 *X 4 are signi cant among the two way interactions. The coe cient of determination (R 2 ) for the activity of the inulinase enzyme was calculated as 0.9490 which is responsible for causing a 94.90% variation in the response.  Inulinase activity (U/gds) = 7.61-121.2 X 1 + 70.9 X 2 − 200.7 X 3 + 161.9 X 4 -44.2 X 5 − 111.1 X 1 2 -712 X 2 2 + 5000 X 3 2 + 3911 X 4 2 + 3086 X 5 2 + 1578 X 1 *X 2 -99X 1 *X 3 + 163 X 1 *X + 815 X 1 *X 5 + 1244 X 2 * X 3 − 4187 X 2 *X 4 − 1067 X 2 *X 5 + 2519 X 3 *X 4 -2037X 3 * X 5 + 2370 X 4 *X 5 The main source of energy and skeletal support to the cells of microbes are provided by the carbon sources. Thus, inulin came out to be the most signi cant among the other variables. Penicillium amphipolaria KAS 2555 produces inulinase using inulin as the only carbon substrate which induces the production process. Yet a large amount of inulin leads to a decrease in the inulinase activity as shown in Fig. 3 due to catabolite repression. K 2 HPO 4 and KH 2 PO 4 act as a buffer and helps in basic cell growth.
The inulinase activity decreases with an increase in KH 2 PO 4 as presented in Fig. 4. This inhibitory effect on the inulinase activity could be the result of phosphate ions released in excess. Ammonium sulphate is an important source which provides nitrogen in the growth of P. amphipolaria KAS 2555. However, the inulinase activity decreases in higher concentration possibly due to excess release of ammonium ions [43]. Another important factor, NaCl helps to maintain the osmotic balance in the growth media. The inulinase activity decreases as the concentration of NaCl increases possibly due to salinity stress which slows down the water transportation rate in and outside the fungal cells [44]. inulinase from Aspergillus terreus [11]. As the concentration of the ve media components is increased above the peak level, the activity of the inulinase enzyme is suppressed although there is continued slow growth of the cells. On the other hand, if the concentration of the media components decreases, the growth of the microbes and the activity of the inulinase enzyme will rapidly decrease [13].
The experimental inulinase activity using these values was found out to be 7.59 U/gds which was in agreement with the predicted inulinase activity of 7.53 U/gds. The inulinase activity obtained in the current study is 3.1 times higher than the un-optimized value. A comparison of the inulinase activity with the existing literature for few fungal species is given in Table 3. There is no literature to date, on the production of inulinase from Penicillium amphipolaria KAS 2555 both by submerged and solid-state fermentation. For the rst time, using hardy sugarcane, this new fungal strain has shown the viability of producing the enzyme inulinase on a laboratory scale by SSF.
Optimized inulinase yield (7.59U/gds) as obtained from the present study is comparable with the optimized inulinase yield (8.57U/gds) from our previous study. Only a yield difference of 0.11U/gds is observed between them. This study proved the e ciency of the mangrove soil containing dead mangroves litter also has good inulinases producing microbes like the soil from living mangroves litter.

Analysis of hydrolysis products
The TLC plate showed spots of fructose standard (F), glucose standard (G), sucrose standard (S) and the test sample (T) containing the products of hydrolysis. No spot was observed for the control (I) containing inactivated inulinase and inulin. A single-colored spot obtained for the test sample proved that only fructose and a minute quantity of glucose is present in the hydrosylate. No fructooligosachharides were present which concludes the exoinulotytic nature of Penicillium amphipolaria KAS 2555. Some of the reported fungal microbes like A. fumigatus [40], A. avus ATCC 16883 [13], P. oxalicum BGPUP-4 [21], etc are exoinulolytic in nature.

Seliwanoff's test
In the Seliwanoff's test, as depicted in Fig. 6, the test solution (4) and the fructose control (1)

Conclusion
In the present study, a potential new inulinase producing fungus Penicillium amphipolaria KAS 2555 was isolated from the detritus mangrove environment. According to our previous study using Aspergillus avus var. avus ATCC16883, living mangrove environment proved to be a rich source of organisms capable of exhibiting inulinase activity. In the present study, dead mangrove environment also proved to be a rich source of microbes producing inulinase. This fungal species gave a good yield of extracellular inulinase in SSF using hardy sugarcane. An increase of 3.10 fold was obtained in SSF using the statistical design of experiment. The study by CCD helped in determining the signi cance of the nutrients present in the media. Among the ve factors, inulin was found to be signi cant. As per the cited literature, no work has been done on Penicillium amphipolaria KAS 2555 for the production of inulinase or any other enzymes. Thus, the newly isolated fungal species, Penicillium amphipolaria KAS 2555 proved to be an e cient producer of inulinase enzyme, leading to the production of fructose in a single step. Further, scale-up studies can be done to enhance the enzyme production. The original idea was suggested by MRB and RS. The experimental work was carried out by DD with guidance from MRB and RS. DD drafted the article. All the authors involved in the critical revision of the article for important intellectual content. All the authors read and approved the nal manuscript.

Competing interests:
The authors have declared no competing interests 6.5 Availability of data and materials The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. 6.6 Ethics approval and consent to participate Not applicable.

Consent for publication
Not applicable.    The distinctive colours are given by the Seliwanoff's test, 1-standard fructose, 2-standard glucose, 3inactivated inulinase + inulin, 4-inulinase + inulin. The characteristic needle-shaped osazones is shown from fructose standard as 1 and fermented broth as 2.