Abstract

An attempt has been made to explore the stability of protease enzyme (isolated from Bacillus sp.) by statistical method. More than 100 isolates were screened for extracellular protease activity derived from various potable water samples of Mahabubnagar district, Telangana State, India. The activity of protease is found to be varying from sample to sample, the highest being reported by the isolate from water sample of Kalwakurthy mandal, Mahabubnagar district and therefore was selected for further studies. The 16S rRNA (ribosomal ribonucleic acid) gene sequence of the isolate showed closest similarity with Bacillus sp. and the sequence was submitted to National Center for Biotechnology Information (NCBI) gene bank (accession number GU566359) and the culture was deposited in three international culture deposition centers (KCTC-13725: MTCC-10465: JCM-16713). The present study revealed that, this Bacillus sp. showed a greater amount of protease production with the inherent characters of thermo, alkali and oxidant stability which makes it a potential alternative protease producing strain in various industrial applications.

INTRODUCTION

Proteases (EC: 3.4. 21-24, 99) are enzymes and about 2% of the entire genes code for proteases in higher organisms (Barrett et al. 2001). Since the plant and animal proteases could not meet the ever growing demand, more emphasis is given on microbial proteases globally (Sakinala et al. 2016). Bacteria are the promising source of protease enzyme for billion dollars of international trade. Microbial proteases are known for stability at broad range of pH, temperatures, and nature of substrates.

However, wild strains are unable to secrete huge amounts of extracellular enzymes. Several attempts like induced mutations by physical and chemical mutagens render to enhance the capacity. Recently, the statistical approaches gained prominence over a collection of independent variables (Kennedy & Krouse 1999). Industrial production requires less investment and more profit. Therefore, at least a 10–20% reduction in the input is considered to be cost-effective (Alnahdi 2012).

Enzyme purification is a critical task; the steps involved in all types of systems are strictly applied to minimize protein contamination. In light of the above aspects, an effort was made in the present investigation to assess the protease produced by the isolate. In order to specify the kinetic behavior of the protease, purification was done by employing a two-tier approach, ammonium sulfate or acetone solvent precipitation and homogenization by salting out with dialysis followed by polyacrylamide gel electrophoresis (Gupta et al. 2005).

MATERIALS AND METHODS

In the present study, drinking water samples were collected from different primary health centers (rural health care facilities which are run by government) of Mahabubnagar district for the screening of protease producing bacteria. About 1 L of potable water sample was collected in a manner to avoid the contamination of atmospheric microorganisms (Pindi et al. 2013a, 2013b, 2014).

Plating and isolation

0.1 mL of sample was layered on skimmed milk agar (SMA: poly peptone 5.0 g/L; yeast extract 5.0 g/L; glucose 10.0 g/L; KH2PO4 1.0 g/L; MgSO4 0.02 g/L; skimmed milk 2.0 g/L; agar 18.0 g/L) using sterile spreader under aseptic condition. The plates were labeled and incubated for 24–72 h at 25–30 °C (Chandra et al. 2015a).

Primary screening

The qualitative screening of the alkaline protease producing bacteria was performed by the method suggested by Chandra et al. (2015a). The proteolytic potential was calculated by the formula: 
formula
where Hz=Hydrolytic zone, Tzd=Total zone diameter, Cd = Colony diameter.

The enzyme producing efficiency of the selected isolates was screened and isolated. The colonies were maintained in SMA at 4 °C, and the efficiency for each isolate was tested at each subculture.

Cultural, biochemical, and molecular characterization

The characterization of the isolate was carried out according to the guidelines of the Bergey's Manual of Determinative Bacteriology (Holt et al. 1994).

Enzyme assessment

Enzyme assay was carried with casein as the substrate by the method described in Chandra et al. (2015b).

Purification of alkaline protease

The proteases were purified by ammonium sulfate or acetone solvent precipitation and homogenization by salting out with dialysis followed by polyacrylamide gel electrophoresis for the determination of their molecular weights (Anbu et al. 2013) (Figure 2).

Evaluation of enzyme kinetics

The purified proteases were evaluated for pH, temperature stability, solvent, oxidizing agents, and substrate specificity by varying the one component at a time (Palmer & Bonner 2007).

RESULTS AND DISCUSSION

Screening and characterization of isolates

More than 100 extracellular protease producing isolates were isolated from different potable water samples of Mahabubnagar district, Telangana State. Among them, the isolate RS from Kalwakurthy Civil Hospital drinking water sample has exhibited superior extracellular proteolytic activity based on primary screening and was selected for further studies. The 16S rRNA gene sequence results revealed that the isolate RS showed closest similarity with Bacillus sp. (Figure 1). The gene sequence was submitted to NCBI gene bank and the accession number was GU566359, the culture was deposited in three international culture deposition centers (KCTC-13725: MTCC-10465: JCM-16713). Biochemical characterization of isolate RS confirmed that the organism is gram positive, filamentous rod shaped, and motile bacteria. Out of all the Bacillus species, the superior protease activity with a greater degree of stability and strength was shown by isolate RS. Further investigations were carried out with RS isolate to secure stronger scientific evidence to support the study.

Figure 1

Phylogenetic analysis shows the isolate RS is Bacillus sp. (GU566359).

Figure 1

Phylogenetic analysis shows the isolate RS is Bacillus sp. (GU566359).

Figure 2

SDS-polyacrylamide gel electrophoresis of protease Bacillus sp. RS; where, lane M, protein markers; lane A, crude enzyme; lane B. Purified enzyme.

Figure 2

SDS-polyacrylamide gel electrophoresis of protease Bacillus sp. RS; where, lane M, protein markers; lane A, crude enzyme; lane B. Purified enzyme.

Design of experiments for enhanced production and bioprocess optimization

Bioprocess optimization by a conventional statistical approach is indispensable and widely applied for optimization of various industrially important microbial products like enzymes and chemicals. In this research few methods like OVAT (one variable at a time) and RSM (response surface methdology) were adopted based on the principle of design of experiment to obtain the optimal product. After OVAT experiment the yield was 470 ± 02 EU/mL and PBD (Plackett-Burman design) followed by RSM protease production yield was 15,000 EU/mL. The resulted OVAT increased to 31.9 fold which is highly significant in regulating the production of protease enzyme.

Based on the principle of design of experiment (DoE), the methodology encompasses use of various types of experimental designs, generation of polynomial equations, and mapping of the response over the experimental domain to determine the optimum product.

In contrast, Haddar et al. (2010) reported that the 14 fold increase in protease production by RSM with PBD followed by face centered composite design (FCCD). The other experiments performed by different authors including Sen & Satyanarayana 1993; Tari et al. 2006; Anbu et al. 2013 and Sakinala et al. 2016 also showed similar results of protease enzyme activity.

Enzyme purification of alkaline protease

The crude way purification of RS protein was 69.4 mg/mL, the acetone solvent precipitation yield was 115.4 mg/mL, the purity was increased to 12.195 U/mg fold, and resulted specific activity per mg of protein concentrate was 63.258 enzyme units. The purity of the enzyme from RS isolate yielded was 20.27% by acetone purification method. Another type of purity method performed was ammonium sulfate precipitation, which showed 124.7 mg/mL of crude total protein with the specific activity per mg of protein concentrate 71.371 enzyme units. The purity of the enzyme with this process of extraction was 13.75 fold and yield percentage was 24.728 which have a more purity when compared to acetone extraction method. Also, the size of the purified enzyme was found to be 28 kDa when compared to the standard marker.

After successful precipitation, salting out was carried out and the amounts were reduced to 10 mL. The dialyzed enzyme was purified with DEAE-C (diethylaminoethyl cellulose column) chromatography.

A critical perusal of Table 1 reveals that, among the different purification strategies employed herein, the crude enzyme of the wild strains exhibited a low amount of protein content in contrast to the acetone and ammonium sulfate precipitation methods. Though the purity was low in DEAE method, interesting fact is that the specific activity of the enzyme was increased to 2,025.6 U/mg of purified protein with 390.51 fold purity and the recovery percentage of yield was 227.88.

Table 1

Purification regimes and specifications of alkaline protease

Isolate Type of purification Total Protein mg/mL Total activity U/mL Specific activity U/mg Purification Fold U/mg Yield % 
RS Crude 69.4 360 5.187 100 
Acetone 115.4 7,300 63.258 12.195 20.27 
NH4SO4 124.7 8,900 71.371 13.75 24.72 
DEAE 40.5 82,037 2,025.6 390.51 227.88 
Isolate Type of purification Total Protein mg/mL Total activity U/mL Specific activity U/mg Purification Fold U/mg Yield % 
RS Crude 69.4 360 5.187 100 
Acetone 115.4 7,300 63.258 12.195 20.27 
NH4SO4 124.7 8,900 71.371 13.75 24.72 
DEAE 40.5 82,037 2,025.6 390.51 227.88 

Characterization of enzyme

Influence of pH on alkaline protease stability and activity

As in the case of a wild strain of RS, it has shown optimum (100%) of relative activity within the pH 9.0–10.0. In contrast, mutated RS strain has shown more than 80% of specific activity. Interestingly, the stability of the partially purified enzyme was further stably maintained even at an elevated pH 12.0 with 100% of specific activity. These results (Table 2) revealed that the enzymes can withstand the harsh industrial process and perform well.

Table 2

Performance of the purified protease in different parameters

pH Relative activity Temperature Relative activity Metal ions Relative activity Solvents Relative activity Oxidizing agents Relative activity Substrates Relative activity 
Control 100 Control 100 Control 100 Control 100 Control 100 Control 100 
49.03 ± 03 10 10.38 ± 01 KCl 88 ± 02 Methanol 80 ± 03 H2O2 105 ± 03 Gelatin 80 ± 03 
63.45 ± 03 15 47.43 ± 02 CaCl2 80 ± 02 Ethanol 80 ± 03 SDS 80 ± 03 Casein 100 ± 03 
100 ± 02 20 82.82 ± 02 NaCl 85 ± 02 Propanol 80 ± 03 Tween 20 120 ± 03 Hemoglobin 80 ± 03 
10 100 ± 02 25 86.36 ± 03 FeCl3 78 ± 02 Butanol 90 ± 03 Tween 80 117 ± 03 BSA 90 ± 03 
11 100 ± 02 30 100 ± 03 CoCl2 75 ± 02 Acetone 95 ± 03 Triton X − 100 80 ± 03 Soya protein 95 ± 03 
12 63.45 ± 03 35 100 ± 03 Mgcl2 130 ± 02 Chloroform 130 ± 03 Reducing agents  Gluten 130 ± 03 
– – 40 100 ± 03 Mncl2 78 ± 02 Hexane 82.5 ± 03 Control 100 ± 03 Egg albumin 82.5 ± 03 
– – 45 100 ± 03 ZnCl2 80 ± 02 Diethyl ether 68.8 ± 03 Mercaptoethanol 52 ± 03 – – 
– – 50 100 ± 03 – – – – Sodium thioglycolate 130 ± 03 – – 
– – 55 100 ± 03 – – – – Inhibitors  – – 
– – 60 100 ± 03 – – – – EDTA 68 ± 03 – – 
– – 65 47.43 ± 03 – – – – PMSF 00 – – 
– – 70 17.95 ± 03 – – – –   – – 
– – 75 17.95 ± 03 – – – –   – – 
– – 85 10.38 ± 03 – – – –   – – 
pH Relative activity Temperature Relative activity Metal ions Relative activity Solvents Relative activity Oxidizing agents Relative activity Substrates Relative activity 
Control 100 Control 100 Control 100 Control 100 Control 100 Control 100 
49.03 ± 03 10 10.38 ± 01 KCl 88 ± 02 Methanol 80 ± 03 H2O2 105 ± 03 Gelatin 80 ± 03 
63.45 ± 03 15 47.43 ± 02 CaCl2 80 ± 02 Ethanol 80 ± 03 SDS 80 ± 03 Casein 100 ± 03 
100 ± 02 20 82.82 ± 02 NaCl 85 ± 02 Propanol 80 ± 03 Tween 20 120 ± 03 Hemoglobin 80 ± 03 
10 100 ± 02 25 86.36 ± 03 FeCl3 78 ± 02 Butanol 90 ± 03 Tween 80 117 ± 03 BSA 90 ± 03 
11 100 ± 02 30 100 ± 03 CoCl2 75 ± 02 Acetone 95 ± 03 Triton X − 100 80 ± 03 Soya protein 95 ± 03 
12 63.45 ± 03 35 100 ± 03 Mgcl2 130 ± 02 Chloroform 130 ± 03 Reducing agents  Gluten 130 ± 03 
– – 40 100 ± 03 Mncl2 78 ± 02 Hexane 82.5 ± 03 Control 100 ± 03 Egg albumin 82.5 ± 03 
– – 45 100 ± 03 ZnCl2 80 ± 02 Diethyl ether 68.8 ± 03 Mercaptoethanol 52 ± 03 – – 
– – 50 100 ± 03 – – – – Sodium thioglycolate 130 ± 03 – – 
– – 55 100 ± 03 – – – – Inhibitors  – – 
– – 60 100 ± 03 – – – – EDTA 68 ± 03 – – 
– – 65 47.43 ± 03 – – – – PMSF 00 – – 
– – 70 17.95 ± 03 – – – –   – – 
– – 75 17.95 ± 03 – – – –   – – 
– – 85 10.38 ± 03 – – – –   – – 

An alkaline protease from Bacillus halodurans with pH optima of 9.0 was also reported (Dabonné et al. 2011). Also, an alkaline stable protease with pH optima of 9.0 by B. horikoshi (Joo & Chang 2005) has been reported. Pandey et al. (2012) documented a thermostable alkaline protease from Nocardiopsis alba OK-5 with extreme pH tolerance of 10.0–11.0. Salwan & Kasana (2012) have also reported an alkaline protease from Acenitobacter sp. MN MTCC (10786) with broad pH range of 7.0 to 11.0.

Influence of temperature on alkaline protease stability and activity

In the present investigation, an attempt has been made to assess the influence of different temperatures. The optimal range was found to be between 40–70 °C. Interestingly, mutant strain of RS has shown more than 70% activity at 30 °C and optimal over a broad temperature range 40–48 °C. Finally, it is concluded that the present enzymes are thermophilic in nature. Hence, they can be used in an industrial process wherein the rise in high temperature is a problem.

The above results are in agreement with Adidi et al. (2008) who reported an alkaline protease from Botrytis cinerea with an optimum temperature of 50 °C. Dabonné et al. (2011) also reported an alkaline protease from Bacillus halodurans with pH optima of 9.0 and 65 °C of temperature optima. Genckal & Tari (2005) reported an extreme alkaline serine protease with optimal pH 11.0 and 60 °C of temperature optima by Bacillus strain from the highly alkaline environment of Turkey. Chellappan et al. (2006) have reported an extreme alkaline protease with broad pH range from 5.0–12.0 and 60 °C temperature optima from Engyodontium album BTMFS10. Pandey et al. (2012) documented a thermostable alkaline protease from Nocardiopsis alba OK-5 with extreme pH tolerance from 10.0–11.0 and 60–80 °C of temperature tolerance. Salwan & Kasana (2012) have also reported an alkaline protease from Acinetobacter sp. MN MTCC (10786) with broad pH range of 7.0–11.0 with 40 °C of temperature optima.

Effect of various solvents and metal ions on alkaline protease stability and activity

The solvent-stability of proteases were assessed and presented in Table 2. It is evident from Table 2 that the proteases have shown more than 70% of the activity with the solvents tested. However, the purified proteases of all strains were found to be stable in the chloroform solvents with more than optimal (100%) activity. Thus, it is apparent from the Table 2 that the chloroform followed by acetone and butanol are the good solvent systems. Further, these solvents are excellent at 20% V/V ratio for the enzyme activity.

The protease of RS wild strain has shown more than 90% of the activity with the metal ions. However, CaCl2, MgCl2, and MnCl2 have enhanced the enzyme activity. Both CaCl2, and MgCl2 at 0.5 and 0.4 mM, respectively, were enough to enhance 130% of specific relative activity. Interestingly, the protease of RS mutant strain has shown more than 90% of the activity with the metal ions. CaCl2, MgCl2, and MnCl2 have enhanced the enzyme activity. The CaCl2 at 0.5 mM was enough to 142.82% of specific relative activity.

The findings of present investigations with regard to the influence of metal ions and solvents are in accordance with many reports (Usami et al. 2005; Akolkar et al. 2008; Habib et al. 2011).

Effect of various oxidizing, reducing, surfactants and inhibitors on alkaline protease activity and stability

Of the two oxidizing agents tested, H2O2 at 0.5 mM concentration has enhanced the proteases with more than 100% of relative activity. Moreover, SDS showed an inhibitory effect with less than 85% of specific relative activity. It is evident from Table 2 that the surfactants also greatly enhanced the enzymatic activity to over the 100% specific relative activity in all the cases except Triton X-100, which was inhibitory and resulted in less than 85% specific residual enzymatic action. It is noticeable that the reducing agents also greatly influenced the enzymatic catalysis over 100% specific residual activity in all the cases except mercaptoethanol, which is strongly inhibitory in the action and resulted in 52%, even zero of the specific residual enzymatic action. In all the cases, ethylene diamine tetra acetic acid (EDTA) significantly increased the enzymatic catalysis except phenyl methyl sulfonyl fluoride (PMSF), this is strongly inhibitory in action and resulted in zero of the specific relative activity. Hence, it can be concluded that all the above enzymes belong to serine proteases.

Thus, the observations on the characterization of alkaline protease are in accordance with the reports of the work done by Akolkar et al. 2008; Habib et al. 2011.

Substrate specificity of the partially purified enzyme

The enzyme activity of the purified protease was tested with different substrates, including cytochrome C, hemoglobin, casein, egg albumin, bovine serum albumin, gelatin, and soybean protein, and the results are presented in Table 2. Among all, cytochrome C was found to be the preferred substrate with high activity (110%), followed by casein (100%), hemoglobin (75%) gelatin (70%), wheat gluten (50%), egg albumin and bovine serum albumin (50%) and finally soybean meal (48%). Similar results were also made (Adinarayana et al. 2003; Habib et al. 2011).

Enzyme kinetics

The rate of the reaction was tested by incubating the reaction mixture (1:1 ratio of [S]:[E] W/W) at varying different parameters.

Influence of substrate concentration rate of reaction

Critical analysis of the enzyme showed a steady steep; after attaining the optimum range it showed a steady state, further increase in the substrate and steadiness in the activity was observed.

Michaelis–Menten behavior of the enzymes has been studied. Further analysis of the nature of the kinetic data revealed that as the Km increased, an increase in the Vmax has been observed at RS required 11–19 mg [s] of initial [s] concentration, on solving the Km it would be 5.1 mg/S for RS. The quadratic fit of the kinetic data also supports the hypothesis and the estimated correlation coefficient R2 was found to be more than 95.56% for the alkaline protease. Hence, the data is validated. The data presented in Figure 3 confirms the LB plot of enzyme kinetics. Critical analysis of Figure 3 reveals that the alkaline proteases of RS strains have Km 5.1 mg/S for RS.

Figure 3

L.B plot kinetics of the purified protease of Rs Bacillus sp. (GU566359).

Figure 3

L.B plot kinetics of the purified protease of Rs Bacillus sp. (GU566359).

CONCLUSION

The challenges which demand attention include loss of enzyme activity over a period of time due to the harsh conditions during industrial processes. The results confirmed the importance of optimization of the production parameters to achieve maximum yield during production of industrially important enzymes. The current four-way optimization resulted in the design of an economical medium with less investment and more yield. The purification by ammonium sulfate yields to a good amount of active protein. The kinetic properties of the enzyme of the isolate RS have gained prominence over other reported proteases by exhibiting thermo, alkali, and oxidant stability. Further screening for agro-based cheap materials and validation of the industrial process is sought. In view of the above-made unusual findings and properties exhibited by the current enzyme, more focus on experiments for industrial applications is in progress.

ACKNOWLEDGEMENTS

The authors are deemed to highly oblige the Honorable Vice-Chancellor, Prof. B. Rajarathnam, Palamuru University, Mahabubnagar for support and encouragement.

COMPLIANCE WITH ETHICAL STANDARDS

Conflict of interest: The authors declare that they have no conflicts of interest.

Ethical approval: This article does not contain any studies concerning human participants or animals by any of the authors.

REFERENCES

REFERENCES
Akolkar
A. V.
,
Deshpande
G. M.
,
Raval
K. N.
,
Durai
D.
,
Nerurkar
A. S.
&
Desai
A. J.
2008
Organic solvent tolerance of Halobacteriumsp. SP1(1) and its extracellular protease
.
J. Basic. Microbiol.
48
(
5
),
421
425
.
Alnahdi
H. S.
2012
Isolation and screening of extracellular proteases produced by new Isolated Bacillus sp
.
J. App. Pharm. Sci.
2
(
9
),
071
074
.
Barrett
A. J.
,
Rawlings
N. D.
&
O'Brien
E. A.
2001
The MEROPS database as a protease information system
.
J. Struc. Biol.
134
(
2–3
),
95
102
.
Chandra
S. J.
,
Pallavi
P.
,
Anuradha
B. S.
&
Reddy
S. R.
2015a
Optimization of bioprocess for enhanced production of alkaline protease by a Bacillus subtilis SHmIIIa through Plackett-Burman design
.
African J. Microbiol. Res.
9
(
28
),
1738
1744
.
Chandra
S. J.
,
Pallavi
P.
,
Reddy
K.
&
Reddy
R.
2015b
Alkaline protease bioprocess optimization through response surface methodology for alkaliphilic Bacillus subtilis SHmIIIa Mutant Strain from Warangal-Telangana
.
American J. Curr. Microbiol.
3
(
1
),
35
45
.
Chellappan
S.
,
Jasmin
C.
,
Basheer
S. M.
,
Elyas
K. K.
,
Bhat
S. G.
&
Chandrasekaran
M.
2006
Production, purification and partial characterization of a novel protease from marine Engyodontium album BTMFS10 under solid state fermentation
.
Proc. Biochem.
41
(
4
),
956
961
.
Dabonné
S.
,
Ahi
A. P.
,
Gonnety
J. T.
&
Kouamé
L. P.
2011
A soluble aggregated thermophile metallo amino peptidase produced by an alcalophile strain of Bacillus halodurans
.
Curr Res. J. Biological Sci.
3
(
1
),
5
30
.
Gupta
A.
,
Roy
I.
,
Patel
R. K.
,
Singh
S. P.
,
Khare
S. K.
&
Gupta
M. N.
2005
One-step purification and characterization of an alkaline protease from haloalkaliphilic Bacillus sp
.
J. Chromatogr. A
1075
(
1–2
),
103
108
.
Habib
S. M. A.
,
Fakhruddin
A. N. M.
,
Begum
S.
&
Ahmed
M. M.
2011
Production and characterization of thermo-alkaline extracellular protease from Halobacterium sp. AF1
.
Asian J. Biotechnol.
3
(
4
),
345
356
.
Haddar
A.
,
Fakhfakh-Zouari
N.
,
Hmidet
N.
,
Frikha
F.
,
Nasri
M.
&
Kamoun
A. S.
2010
Low-cost fermentation medium for alkaline protease production by Bacillus mojavensis A21 using hulled grain of wheat and sardinella peptone
.
J. Biosci. Bioengg.
110
(
3
),
288
294
.
Holt
J. G.
,
Krieg
N. R.
,
Sneath
P. H. A.
,
Staley
J. T.
&
Williams
S. T.
1994
Bergey's Manual of Determinative Microbiology
.
Williams and Wilkins
,
MD, USA
.
Kennedy
M.
&
Krouse
D.
1999
Strategies for improving fermentation medium performance: a review
.
J. Indus. Microbiol. Biotechnol.
23
(
6
),
456
475
.
Palmer
T.
&
Bonner
P. L.
2007
Enzymes: Biochemistry, Biotechnology, Clinical Chemistry, 2nd Edition
.
Pindi
P. K.
,
Raghuveer
Y. P.
&
Kodaparthy
A.
2013a
Bacteriological and physico-chemical quality of main drinking water sources
.
Pol. J. Environ. Stud.
22
(
3
),
825
830
.
Sakinala
J. C.
,
Anuradha
B. S.
,
Reddy
K.
&
Reddy
S. R.
2016
Screening and optimization of physical parameters for enhanced alkaline protease production by alkaliphilic Bacillus subtilis SH2 isolate
.
American J. Curr. Microbiol.
4
(
1
),
55
65
.
Sen
S.
&
Satyanarayana
T.
1993
Optimization of alkaline protease production by thermophilic Bacillus licheniformis S-40
.
Indian J Microbiol
33
,
43
47
.
Usami
R.
,
Fukushima
T.
,
Mizuki
T.
,
Yoshida
Y.
,
Inoue
A.
&
Horikoshi
K.
2005
Organic solvent tolerance of halophilic archaea, Haloarcula strains: effects of NaCl concentration on the tolerance and polar lipid composition
.
J. Biosci. Bioengg.
99
(
2
),
169
174
.