2.1. Screening of Chitin-Containing Materials as C/N for α-Glucosidase Inhibitors Production
The effect of chitin-containing materials on aGI production by Paenibacillus
sp. TKU042 was investigated with six kinds of chitinous materials: demineralized crab shell powder (deCSP), demineralized shrimp shell powder (deSSP), squid pen powder (SPP), shrimp head powder (SHP), fresh shrimp shell powder (frSSP), and cicada shell powder (CiSP) as the sole sources of C/N with the concentration of 1% (w
). The protein–chitin–mineral salts compositions of shrimp shells and squid pen were 48%, 38%, and 14% and 61%, 38%, and 1%, respectively [26
]. It is obvious that the mineral salts content in squid pen (1%, w
) is much lower than in shrimp shell (14%, w
). Therefore, only the crab shells and shrimp shells were decalcified and used as the C/N sources for aGI production.
The results of maximum aGI activity, aGI productivity, and cell growth were investigated, and are shown in Figure 1
. After four days of fermentation, the maximum aGI activity (Figure 1
A) of most C/N sources reached the highest level, 91–100% with the priority sequence of deCSP, deSSP, SHP, and SPP. The maximum aGI activities of deCSP, deSSP, and SHP remained thereafter; only SPP decreased dramatically to a level lower than 40% on the 6th day. The reason might be related to the higher content (around 60%) of protein in SPP. It is also possible that an additional inhibitory component was made and accumulated with increased time that affected the activity of the active compound. The other possibility could be the decreased stability of the active ingredient in SPP after the fourth day. A similar phenomenon was also found in Bacillus subtilis
TKU007, which showed that SHP was a better C/N source than SSP for the production of chitosanase and protease [27
] and nattokinase [28
]. Although the protein content of SPP was higher than that of SSP, the resultant enzyme productivity was not raised [27
]. This might be related to the ratio of chitin and protein in the C/N sources.
The maximum aGI activity of frSSP and CiSP was lower than 20%, even though the culture time was lengthened to six days. The maximum aGI activity was speculated to be affected by the presence of high levels of mineral salts (in the case of frSSP and CiSP) or some additional factors that were eliminated during demineralization (in the case of deCSP and deSSP). As shown in Figure 1
B, the aGI productivity reached its highest level after 4–5 days of fermentation. The highest was found in deCSP (5010 U/mL), deSSP (2476 U/mL), and then in the other four (SPP, SHP, frSSP, and CiSP) with aGI productivity lower than 660 U/mL. The same as the phenomenon of maximum activity found in Figure 1
A, the aGI productivity of deCSP and deSSP remained steadily even after six days of fermentation.
To examine the relationship of aGI productivity and bacterial growth, the culture broth after removing the residual chitinous materials (C/N source) by centrifugation (Kubota 5922, Japan) at 150× g
for 10 min was used for analyzing cell growth. As shown in Figure 1
C, deCSP and deSSP were not the best C/N sources for cell growth. These results showed that the production of aGI may not have a direct relationship with cell growth.
The abovementioned comparisons of aGI production by Paenibacillus
sp. TKU042 in different chitin-containing media are summarized in Table 1
. The aGI activity of fermented nutrient broth (FNB) [2
] and acarbose (an antidiabetic drug) were also compared. The aGI activities of the C/N sources all reached their highest level on the 4th day. The IC50
value was found in the priority sequence of deCSP (38 μg/mL), FNB (81 μg/mL), deSSP (108 μg/mL), SPP (422 μg/mL), SHP (455 μg/mL), and acarbose (1095 μg/mL). The maximum aGI activities were in the priority sequence of deCSP (98%), SPP (98%), FNB (93%), SHP (92%), SSP (89%), and then acarbose (74%).
As for the productivity of aGI after fermentation under optimal conditions, deCSP showed the best yield of 26,316 kU/g, followed consecutively by FNB (12,346 kU/g), deSSP (9259 kU/g), SPP (2370 kU/g), and SHP (2198 kU/g). These tested chitinous materials were all potential C/N sources for aGI production by Paenibacillus sp. TKU042, due to their higher productivity than that of acarbose (913 kU/g).
The weight yields of the obtained culture supernatants were also compared. The weight yields of deCSP (2.03 g/L) and deSSP (1.60 g/L) were lower than that of FNB (6.5 g/L). These results showed that there were fewer contaminants and greater ease for further purification of the aGI from the culture supernatants of deCSP and deSSP than from that of FNB. These results showed that the deCSP and deSSP were the most remarkable C/N sources for aGI production by Paenibacillus sp. TKU042.
2.2. The Effect of Protein Supplement on α-Glucosidase Inhibitors Production
To analyze whether extra protein may enhance aGI production, the supplementation of 0.2% and 0.4% of protein (polypeptone/yeast extract = 4/6) in the deCSP (1%)-containing medium was studied. As shown in Figure 2
, an inverse relationship between protein concentration and aGI productivity was found. The supplementation of free protein increased the bacterial growth (Figure 2
A) but did not enhance the aGI productivity (Figure 2
B). In the first two days, no significant difference in aGI productivity was found. The difference started to appear on the fourth day; the aGI productivity of these three C/N sources were in the order of 1% deCSP (4833 U/mL), 1% deCSP/0.2% protein (2512 U/mL), and then 1% deCSP/0.4% protein (1374 U/mL). According to the results, the ratio of protein/chitin appeared to play an important role in aGI production by Paenibacillus
sp. TKU 042. The C/N sources (such as crab shell chitin or shrimp shell chitin) with chitin but only less than 1% of protein, were also found to be unsuitable for use as an inducer for the protease production by B. subtilis
] and Bacillus
sp. TKU004 [29
To investigate whether the protein in deCSP was sufficient to play an important role in aGI production by Paenibacillus
sp. TKU 042, the protein in the deCSP was removed by heat-alkali-deproteinization treatment [29
]. The obtained deproteinized deCSP was then used as the chitin sample for investigating the effect of ratio of protein vs. chitin (0.1/1–0.8/1) on aGI production. As shown in Figure 2
C, the best results were found in the ratio of protein vs. chitin of 0.2/1 (5700 U/mL) and 0.1/1 (4500 U/mL). The optimal ratio of protein vs. chitin (0.2/1) investigated in this study was approximately the same as the deCSP, which showed comparable aGI productivity of 5500 U/mL. Another source of chitin (chitin obtained from SPP) was fermented by Paenibacillus
sp. TKU 042 at the same ratio of protein vs. chitin and cultivation conditions as above. Similarly, this bacterium strain showed the same manner of aGI production with aGI productivity of 5600 U/mL.
Nutrient broth (NB) containing proteins of polypeptone and yeast extract showed lower aGI productivity (3000 U/mL) than the ratio of protein vs. chitin (0.2/1), deCSP, and ratio of protein vs. chitin (0.1/1). These results showed that the supplementation of protein vs. chitin with a ratio higher than 0.2/1 decreased the aGI production. Therefore, deCSP was chosen for further optimization study.
2.3. Production of aGI from Demineralized Crab Shell Powder by Different Bacteria
To examine whether other bacteria strains, especially the strains of Paenibacillus
species, would also produce aGI in the deCSP-containing medium, 16 stocked chitinolytic bacteria which were all isolated from Taiwanese soils were tested. As shown in Table 2
, only the bacteria of Paenibacillus
species produced aGI against the tested three α-glucosidases. The maximum aGI activity against α-glucosidases of Saccharomyces cerevisiae
(S), Bacillus stearothermophilus
(B), and rat (R) were all above 90%.
Many strains of Paenibacillus
species have been reported to be potentially used in agricultural, industrial, health food, and medical utilizations, such as the biosynthesis of antifungal agents [30
], biofertizlizers [32
], enzymes [22
]. Recently, Paenibacillus
species have also been explored to use SPP as the sole C/N source for the production of exopolysaccharides [6
], antimicrobial biosurfactants [7
], antioxidants [6
], and homogentisic acid [16
sp. TKU042 was then chosen as the aGI-producing strain for further investigation.
2.4. Optimization of Culture Conditions
The deCSP was confirmed as a potential C/N source and chosen for an optimization study of some parameters, including the cultivation temperature (25, 30, 34, and 37 °C), culture volume (50, 70, 100, 130, and 160 mL), the seed culture volume (0.5, 1, 2, and 4 mL), and the concentration of deCSP (0.5, 0.75, 1, 1.25, 1.6, and 2%) (the data are shown in Supplementary Figure S1
). Overall, aGI were effectively produced by Paenibacillus
sp. TKU042 within the 1.6% deCSP-containing medium (130 mL medium in 250 mL-Erlenmeyer flask) at 30 °C in a reciprocal shaker (Yi Der LM-570R, Jun Yang, New Taipei City, Taiwan) at 150 rpm for four days (Figure S1
). The culture conditions before and after the optimization study are summarized in Table 3
. After the optimization was studied, the aGI productivity was increased from 5000 to 12,400 U/mL (2.48-fold), and the IC50
value was reduced from 81 to 6.7 µg/mL (12.1-fold). The culture supernatant (fermented deCSP) obtained by culturing Paenibacillus
sp. TKU 042 in the optimized condition was then used for further study.
2.5. Specific α-Glucosidase Inhibitors Activity of Fermented Demineralized Crab Shell Powder
To evaluate the potential of Paenibacillus sp. TKU 042 aGI to be developed as antidiabetic drugs, the inhibitory specificity of fermented deCSP was tested against six kinds of commercial enzymes, including α-glucosidases (S. cerevisiae, B. stearothermophilus, rice, rat intestine) and α-amylases (B. subtilis, porcine pancreas).
As shown in Table 4
, fermented deCSP showed lower IC50
(15.9 µg/mL) and higher maximum inhibitory activity (97%) than those of acarbose (78 µg/mL, 91%) against rat intestinal α-glucosidase. The IC50
and maximum inhibitory activity against S. cerevisiae
α-glucosidase of fermented deCSP (6.7 µg/mL, 99%) were also better than those of acarbose (1095 µg/mL, 74%). Acarbose showed better results of IC50
(0.042 µg/mL and 3.04 µg/mL) against α-glucosidases from B. stearothermophilus
, and rice, respectively, compared to fermented deCSP (6.6 µg/mL, 6.7 µg/mL). No inhibitory activity was found for fermented deCSP against porcine α-amylase and B. subtilis
α-amylase. For the evaluation of the potential inhibitors using as antidiabetic drugs, α-glucosidase from rat has been suggested as the most valuable resource among the tested enzymes, since it is a mammalian enzyme closer to that of human [2
]. In this study, fermented deCSP showed stronger inhibition against rat α-glucosidase than that of acarbose. This result suggested that fermented deCSP may be a good candidate of antidiabetic drugs.
2.6. Confirmation of α-Glucosidase Inhibitors Containing in Fermented Chitin-Containing Media
The same concentration (20 mg/mL) of the unfermented deCSP and 1–4 day-fermented deCSP solutions were analyzed by high-performance liquid chromatography (HPLC). The difference of the HPLC fingerprints of deCSP before and after fermentation can be clearly observed in Figure 3
A. After fermentation, some new main peaks appeared at the retention time of 8–8.5, 23, and 27.5 min, or enhanced its area (the peak at the retention time of 27.5 min). Unfermented deCSP was tested for α-glucosidase inhibition, but no activity was observed. The differences in HPLC fingerprints and the inhibitory activity between unfermented and fermented deCSP led to the conclusion that the aGI were produced by fermentation and had not previously existed in the deCSP.
To further compare which peaks especially belong to Paenibacillus
sp., the four day-fermented deCSP by Paenibacillus
sp. TKU042, Paenibacillus mucilaginosus
sp. TKU041, B. subtilis
TKU007, and Serratia marcescens
TKU011 were analyzed. As shown in Figure 3
B, only the peak at the retention time of 27.5 min was found in deCSP fermented by Paenibacillus
sp. TKU042 and P. mucilaginosus
TKU032. Combining these results with the data shown in Table 2
, this peak may signify the active compounds of aGI. The isolation and identification of these aGI will be performed soon.