Field tests
Water samples were collected on 3 shrimp farms located in Jiangsu Province, China, in April. The first farm (near Huan Gangerzu Town; 32o 30’26” N; 121o10’11” E) used greenhouse-enclosed raceways for intensive shrimp production. Fresh groundwater was directly pumped into separated ponds under a greenhouse, each individual pond was 6x50 m in size. The shrimp age was 2 months. Water was sampled from well, 6 ponds, and a small creek filled with effluents from ponds.
The second shrimp farm (near Changsha City; 32o23’31” N; 121o18’09” E) had a system of open ponds (100 x 50 m each), filled with unsalted water from a local canal. Thee shrimps were 2 months old. Water was sampled from the canal and 6 ponds.
The third shrimp farm (near Yangkou Town; 32o35’44” N, 120o59’09” E) took water directly from the Yellow Sea. Open ponds were 250 x 60 m each. Water was collected from 6 ponds, and a canal supplied water to the ponds. The shrimps were 1 month old.
Samples were collected in triplicate in 100 mL plastic bottles early in the morning and were immediately transported to a laboratory to determine monosilicic acid, polysilicic acid, pH and microbial cell abundance.
The pH was analyzed on an Ion Meter PHS-3e (MRC, China). The concentration of monosilicic acid was determined using the modified molybdenum blue method (Mullin and Riley, 1955) with a spectrophotometer V5800 XZBELEC (China). This method tests Si only in the form of monosilicic acid, without interference from phosphorus.
To analyze polysilicic acid, 2 g of NaOH was added to 20 mL of centrifuged water and kept in a refrigerator at + 4oC for 2 weeks (Matichenkov et al. 1997). During this time all polymers of silicic acid are transformed into monomers. After that monosilicic acid was determined as described above. The concentration of polysilicic acid was calculated by the following formula:
Sipoly = Sitotal – Simono,
where Sitotal is the concentration of monosilicic acid determined after depolymerization;
Simono is the concentration of monosilicic acid determined before depolymerization.
Microbial abundance (cells mL− 1) was estimated on the day of sampling by binocular biological microscope (ML10, Guangzhou Mingmei, China) according to the method proposed by Newell and Newell (2006). The same method was used to calculate the cell density in the laboratory experiment.
Laboratory experiment
A laboratory experiment was conducted with collected farm water summarized samples and 3 commercial probiotics used in shrimp farming: dry probiotics “Ecopro” and “Ecopro Cold” (Ecomicrobial Co, USA) and liquid probiotic “HeJunMei” (Jiangsu Aijiafuru Soil Remediation Co, China). In dry and liquid probiotics, the amounts of bacteria and yeast spores were not less than 1x1012 and 1x1010cells kg− 1, respectively.
To activate dry probiotics, 1 g of probiotic was mixed with 1 L of sterilized distilled water (DW) and kept at + 24oC for 24 h. The liquid probiotic was diluted 1:10.
One liter of nutrient solution was prepared with K2HPO4 3.125 g; KH2PO4 3.125 g; (NH4)2HPO4 3.125 g; MgSO4.7H2O 0.25 g; FeSO4.7H2O 0.0125 g; MnSO4.7H2O 0.00875 g and sucrose 12.5 g (Vitullo et al. 2012). Eighty (80) mL of this nutrient solution was added to each flask. Ten (10) mL of pond water collected on the day of sampling, probiotic solution or DW was added to the flasks. Considering that Farm 3 used sea water, NaCl (35 g L− 1) was added to flasks with pond water from Farm 3. Then 10 mL of DW or monosilicic acid solution at 10 and 20 mM Si was added to reach the Si concentrations of 0, 1 and 2 mM. Monosilicic acid solutions were prepared from concentrated monosilicic acid (Fisher Scientific, CAS-No 7699-41-4). The pH in each flask was adjusted to 7 by adding 0.1 M HCl or 0.1 M NaCl.
Flasks were kept in a climatic chamber at 24 ± 1oC. The light/night regime was 12/12 hours with light intensity 600 µmol photons м-2 s-1. Flasks were aerated twice a day for 1 h (morning and evening). After 3 days, the concentration of monosilicic acid and the density of microorganisms were determined using the method described above.
Each treatment and each analysis were conducted in triplicate. All data obtained was subjected to a statistical analysis based on comparative methods using Duncan’s multiple range tests for mean separation at the 5% level of significance (Duncan 1957).