Sample collection

Silkworm eggs were incubated using conventional methods. The silkworm larvae were reared with mulberry leaves from first to third instar, and then were divided into two identical groups; one reared with mulberry leaves, the other with tricuspid cudrania leaves, both at the start of the fourth instar. Twenty silkworm larvae of each treatment group were obtained every other day, from the first day of fourth instar to the seventh day of fifth instar, and subjected to starvation for 24 hours. Subsequently, the larvae were surface sterilized by submersion into 75% ethanol for one minute and rinsed three times using sterile distilled water prior to dissection. The larvae were dissected using dissection scissors and the guts were withdrawn with sterilized fine-tipped forceps. The intestinal juice of each five larvae was collected using sterile syringes and placed in a 5 mL sterilized centrifuge tube as one unit of collection for processing.

Isolation of dominant bacteria

At the maximum dilution of intestinal juice, any bacterial colony that accounted for more than 10% the total colony was defined as a dominant microflora. Each sample collected could possess two or more types of dominant bacteria (Li 1999). According to the screening standard of dominant bacteria, 1 mL intestinal juice was added to 9 mL sterilized physiological saline, and tenfold serial dilution aliquots were used for the inoculation. On each agar plate, 0.1 mL of intestinal juice with maximum dilution was spread and incubated at 30 °C for 48–72 hours; all samples were repeated three times. The media used for the isolation of dominant bacteria included nutrient agar, potato dextrose agar, Gause's No. 1 agar, and Czaper's agar medium, which were autoclaved at 121 °C for 20 minutes, and pH value was adjusted to 9.2–9.8 (Huang 1999). Dominant colonies were picked out, purified three times by inoculating on the corresponding agar plates, and further transferred to agar slants.

Identification of dominant bacteria

The dominant intestinal bacteria were identified by bacteriological properties and 16S rRNA gene sequencing. Morphological tests were done by standard procedures. The physiological-biochemical characteristics were determined on the basis of Gram stain, spore stain, oxidase test, catalase test, motility, Voges-Proskauer test, methyl red test, indole test, glucose fermentation test, hydrogen sulfide production, and hydrolysis of starch test (Dong and Cai 2001).

DNA was extracted by boiling bacterial cell suspension in sterile distilled water (Brousseau et al. 1993). 16S rRNA genes were amplified with universal primers of 27F (5'-GAGTTTGATCCTGGCTCAG-3') and 1492R (5'-CGGTTACCTTGTTACGACTT-3'). PCR amplification was performed in a total volume of 50 µL containing 5 µL DNA extract, 2 µL of 10 µmol/L of each primer, 4 µL of 5 mmol/L dNTPs, 4 µL of 25 mmol/L MgCl₂, 5 µL of 10 × PCR buffer, and 0.3 µL of 5 U/µL Taq DNA polymerase using a thermal cycler. Cycling conditions were as follows: initial denaturation at 95 °C for five min, 29 cycles of 94 °C for one min, 55 °C for one min, 72 °C for one min, and a final extension at 72 °C for five min. PCR products were examined by electrophoresis in a 0.8% agarose gel, and bands were visualized by staining with ethidium bromide. PCR products were further purified with the QIAquick PCR purification kit (Quiagen,  www.quiagen.com) and cloned into pMD18-T vector followed by sequencing. Sequence analysis was performed using the BLAST algorithm in GenBank ( http://www.ncbi.nlm.nih.gov). Bacterial identifications were based on 16S rRNA gene sequence similarity. A phylogenetic tree was constructed with the neighbor-joining method (Saitou and Nei 1987) of MEGA 4.0 software using bootstrap values with 1000 replicates.

Screening of lipase-producing bacteria

Each dominant isolate was inoculated on Rhodamine B agar plate medium (0.5% beef extract, 0.5% peptone, 0.5% (NH₄)₂SO₄, 0.05% MgSO₄, 0.4% K₂HPO₄, 0.02% CaCl₂, 0.2% NaNO₃, 12.0% olive oil emulsion, 2.0% Rhodamine B (0.1mg/mL), 1.8% agar, pH = 9.2–9.8) using a sterilized toothpick, and incubated at 30 °C for 48 hours. Lipase production was observed as a pink zone of hydrolysis around bacterial colony.

Growth curve assay of lipase-producing bacteria

Each lipase-producing strain was inoculated in a 250 mL Erlenmeyer flask containing 25 mL of medium, consisting of 0.5% beef extract, 0.5% peptone, 0.5% sucrose, 0.3% NaCl, 0.2% K₂HPO₄, pH = 9.2–9.8, and incubated at 30 °C for 24 hours on a rotary shaker at 200 rpm. Subsequently, the inocula were transferred into production medium (2.0% peptone, 0.5% sucrose, 0.1% (NH₄)₂SO₄, 0.1% MgSO₄, 0.2% K₂HPO₄, 1.0% olive oil, pH = 9.2–9.8) at a rate of 2.0%, followed by incubation at 30 °C on a rotary shaker at 200 rpm. The samples were taken after every 12 hours until 72 hours. Cell density was measured by taking the optical density at 600 nm against the cell-free control.

Assay of lipase activity

The production media were inoculated with 2.0% of seed cultures and incubated at 30 °C for 48 hours on a rotary shaker at 200 rpm. The cultures were centrifuged at 10,000 rpm for 10 min, and the supernatant was examined for lipase activity using an ultrasensitive lipase detection kit (Nanjing Jiancheng Bioengineering Institute,  www.njjcbio.com/html_en/contact_us.php) by taking the optical density at 420 nm. One unit of lipase activity was defined as the amount of enzyme required for the consumption of 1 µmol substrate per minute under the standard assay condition.

Table 1.

Morphological characteristics of dominant bacteria in the intestine of the silkworm larvae reared with mulberry leaves and tricuspid cudrania leaves.

Isolation of dominant bacteria

Using the isolation procedure described above, a total of 56 dominant isolates were successfully collected from the intestine of the fourth and fifth instar silkworm larvae and classified into 12 phenotypes based on the colony color, size, and cellular morphology. Typical strains SW83, SW48, SW42, SW115, SW54, SW81, SW56, SW101, SW82, SW75, SW38, and SW41 represented the 12 phenotypes, respectively. Ten and seven phenotypes existed in the intestine of the larvae fed with mulberry leaves and tricuspid cudrania leaves, respectively; four of them were common in the intestine of both treatment groups (Table 1).

Table 2.

Physiological and biochemical characteristics of dominant bacteria in the silkworm intestine.

Identification of dominant bacteria

Based on morphological and physiological-biochemical characteristics (Tables 1, 2), strains SW82 and SW83 were similar to Pseudomonadaceae, strains SW75, SW81, and SW115 were similar to Enterobacteriaceae, and strains SW48, SW42, SW54, SW56, SW101, SW38, and SW41 were similar to Rhizobiaceae, Vibrionaceae, Brevibacteriaceae, Corynebacteriaceae, Caulobacteraceae, Micrococcaceae, and Bacillaceae, respectively (Buchanan and Gibbons 1984).

To further characterize the dominant intestinal bacteria, strain identification was performed using a molecular approach. The results showed that the 16S rRNA genes of all these isolates had ∼ 1500 bp DNA fragments by PCR amplification. All the sequences were deposited in the GenBank database. Accession numbers HM584787 to HM584790 corresponded to strains SW83, SW48, SW42, and SW115, accession numbers HM584792 to HM584798 corresponded to strains SW101, SW75, SW38, SW54, SW81, SW82, and SW41, and accession number HQ202802 corresponded to strain SW56. To clarify the phylogenetic position of these dominant bacteria, a phylogenetic tree was constructed based on 16S rRNA gene sequence homology (Figure 1). The results showed strains SW83, SW48, SW42, SW54, SW56, SW101, SW82, SW38, and SW41 had the highest homology with Pseudomonas, Agrobacterium, Aeromonas, Brevibacterium, Corynebacterium, Brevundimonas, Stenotrophomonas, Staphylococcus, and Bacillus, respectively. Strains SW115, SW81, and SW75 showed the highest homology with Klebsiella. However, together with the morphological and physiological-biochemical characteristics, strains SW115, SW81, and SW75 belonged to Citrobacter, Klebsiella, and Enterobacter, respectively.

Screening of lipase-producing bacteria

According to screening the isolates on Rhodamine B agar plate, nine lipase-producing bacteria were obtained and named as SW41, SW54, SW56, SW72, SW79, SW80, SW81, SW84, and SW82, respectively (Figure 2). Except for strain SW82, which was common in both, the other bacteria only existed in the intestine of the silkworm larvae reared with mulberry leaves.

As identified above, strains SW41, SW54, SW56, SW81, and SW82 were Bacillus, Brevibacterium, Corynebacterium, Klebsiella, and Stenotrophomonas, respectively. 16S rRNA gene sequencing further identified the other lipase-producing bacteria. Strains SW72, SW79, and SW80 (accession numbers HQ202803 to HQ202805) showed the highest similarity to Staphylococcus, and strain SW84 with accession number HQ202806 showed the highest similarity to Brevibacterium (Figure 1).

Growth curve assay of lipase-producing bacteria

The bacterial density was examined after every 12 hours until 72 hours; the maximum cell density of all the lipase-producing bacteria was observed at ∼ 48 hours of fermentation. After that, the bacterial density reduced, probably due to the consumption of nutrients. The maximum bacterial density showed a significant difference; strain SW41 had the maximum yield, whereas strain SW72 had the minimum yield (Figure 3).

Assay of lipase activity

Lipase activity was determined using a SmartSpecTM3000 UV spectrophotometer (Bio-Rad,  www.bio-rad.com). Results showed the maximum total lipase activity after 48 hours of fermentation in vitro. At their maximum bacterial density, strain SW41 showed the highest lipase activity (249.59 U/L), while strain SW54 showed the lowest activity (77.46 U/L) (Figure 4).