First record of the mosquito control potentiality of Stigmatogobius sadanundio (F. Hamilton, 1822) Gobiidae, Perciformes in laboratory condition.

BACKGROUND AND OBJECTIVES: In recent years, resurgence of mosquito-borne diseases has become a serious health problem in India. In the present study, Stigmatogobius sadanundio, a common indigenous fish, has been tested for its biocontrol potentiality for controlling Culex quinquefasciatus larvae. This small larvivorous fish can consume large number of Culex larvae even in the presence of alternate prey. This is the first report on the mosquito control ability of this fish.
MATERIALS AND METHODS: Experimental fishes were captured from tidal canals of Rupnarayan River in Purba Medinipur district, West Bengal. Mosquito larvae, pupae, and chironomid larvae were collected from Tamralipta municipality drainage system. Predation efficacy of the fish was evaluated on C. quinquefasciatus larvae and pupae as well as on Chironomus ramosus larvae which were collected from the drainage system of Tamralipta municipality and reared in the laboratory maintaining similar water parameters. Prey were offered to the fish separately and in paired combination to study its dietary preference.
RESULTS: S. sadanundio is a diurnal predator consuming significantly large number of prey during daytime. It prefers mosquito and chironomid larvae over mosquito pupae. The rate of predation was very high during 1st h of predation. It consumed more chironomid larvae in the presence of mosquito larvae during daytime but consumed large number of mosquito larvae as compared to other larvivorous fish.
CONCLUSION: S. sadanundio, an indigenous fish, is an effective biocontrol agent for the larvae of C. quinquefasciatus in laboratory condition. Even though the presence of alternate prey chironomid larvae influences the predation rate, it consumed large number of mosquito larvae. However, careful controlled field trials must be conducted before this fish is used as a biocontrol agent. Copyright:

INTRODUCTION

The use of fish in mosquito control has been well known for more than 100 years. Petr[1] reported that the use of larvivorous fishes for vector control is simple and inexpensive and should be considered as a component of integrated strategies. Use of native fish should be preferred because use of exotic fish has raised environmental concern in past as it results in significant elimination of native fish[2] and adversely affects biodiversity of the region.[3] The indigenous larvivorous fishes coexisting in mosquito larval habitat naturally offer an alternative in this regard. The use of indigenous fishes could reduce the reliance on insecticides and may provide a cost-effective, ecofriendly, safe, and target-specific vector control device. According to Lloyd,[4] the suitability of a native fish for mosquito control will depend to a great extent on its effectiveness as a predator and should be evaluated under laboratory condition before field trials are taken. The use of indigenous larvivorous fishes for biocontrol has been documented from different parts of the world.[567] In India, such attempts have been undertaken by Sharma and Ghosh[8] and Chandra et al.[9] The present study explores the suitability of an indigenous fish, Stigmatogobius sadanundio (F. Hamilton, 1822),[10] in controlling mosquito larvae under laboratory condition. The present study is aimed at evaluating the prey preference of S. sadanundio using Culex quinquefasciatus Say 1823 (Diptera: Culicidae) as the target prey and Chironomus ramosus Chaudhuri, Das and Sublette, 1992 (Diptera: Chironomidae: Chironomonae) as an alternative prey.

MATERIALS AND METHODS

Collection of fish

Stigmatogobius sadanundio (F. Hamilton, 1822) [Figure 1] was collected from intertidal canal system of Rupnarayan River by gill net and locally used hand net. Collected specimens were immediately transported to the laboratory in a 20 l plastic container filled with canal water. These fishes were released in a large (120 cm × 90 cm × 60 cm) aquarium in the laboratory filled with tap water after adding four teaspoons full common salt. Fishes were fed with zooplankton collected from a pond situated in the college campus.

Figure 1

Photograph of Stigmatogobius sadanundio (F. Hamilton, 1822)

Rearing of fish and acclimatization

Fishes were then transferred to a glass aquarium (120 cm × 60 cm × 60 cm), filled with mixture of canal water and tap water in 4:1 ratio; raising the volume of water to 300 l, Lemna minor L. and Pistia stratiotes L. were placed in the aquarium to simulate the natural habitat. The canal water was prefiltered by a plankton net (mesh size 72 μm) to remove plankton and unwanted materials. The aquarium was placed beside a glass window, to provide sufficient sunlight. The temperature of the water was maintained at 22°C–24°C. Salinity level of <0.5 parts per thousand and pH between 8.1 and 8.4 were maintained. The floor of the aquarium was covered with pebbles and sand. Fishes were fed with zooplankton everyday. In this condition, the experimental fishes were kept for 1 month to acclimatize. Every 5th day, one-third water was replaced maintaining the same temperature, pH, and salinity.

Collection and maintenance of mosquito larvae and pupae

The mosquito (C. quinquefasciatus) larvae were collected from different sewage drains using a hand net (mesh size 200 μm) within Tamralipta municipality. The collected mosquito larvae were stored in a glass tank (60 cm × 30 cm × 30 cm) with sewage water. The density of larvae in the tank was maintained at approximately 10,000/1800 cm2. No artificial food was given to the larvae. Every 2nd day, the tank sewage water was replaced with the sewage water from where these were collected.

Collection and maintenance of chironomid larvae

The chironomid (C. ramosus) larvae (1.5–15 mm in length) were collected from different sewage drains. The chironomid larvae were stored in a tank (60 cm × 30 cm × 30 cm) with sewage water and sediments.

Experimental aquaria

Five equal-sized (30 cm × 30 cm × 30 cm) glass aquaria were used for experimental purpose. Each tank was filled with 10 l of filtered canal and tap water mixed in 4:1 ratio. These aquaria were placed beside the window. Temperature, pH, and salinity of water were maintained as mentioned earlier.

After 1 month of acclimatization, randomly five fishes were chosen with equal length and weight for pilot and trial experiment.

Experimental protocol

Twenty-four hours before the start of experiments, five acclimatized fishes of approximately same size (5.2–5.6 cm length) and weight (1.858–2.097 g) were randomly picked from the stock aquarium and released one fish each in five experimental glass aquaria at 06:00 h and were starved for 24 h. The experiment commenced at 06:00 h next morning when 100 prey were released. A number of prey consumed were recorded at an interval of 1, 6, 12, and 24 h, i.e., at 07:00 h, 12:00 h, 18:00 h, and 06:00 h. Next morning, by counting unconsumed prey, 100 prey were supplemented at an interval of 1, 6, and 12 h. There were five sets of experiments as mentioned below, and each experiment was repeated for three times. As such, there were 15 observations for each experimental set.

Experimental sets

First set: five aquaria, each containing one fish and only mosquito larvae as prey. Repeated three times.

Second set: five aquaria, each containing one fish and only mosquito pupae as prey. Repeated three times.

Third set: five aquaria, each containing one fish and only chironomid larvae as prey. Repeated three times.

Fourth set: five aquaria, each containing one fish and mosquito larvae + mosquito pupae as prey. Repeated three times.

Fifth set: five aquaria, each containing one fish and mosquito larvae + chironomid larvae as prey. Repeated three times.

Thus, the prey were offered separately and together in paired combination.

Statistical analysis

For statistical analysis of the data, Prism 5 and Microsoft Excel were used.

Dietary preference index was analyzed using the formula of Chesson as in Krebs.[11]

where, = Manly's alpha (preference index) for prey type i; p, p = proportion of prey i or j remaining at the end of the experiment (i = 1, 2, 3…m) (j = 1, 2, 3...m) = e; e = number of prey type i remaining uneaten at end of experiment; n = initial number of prey type i in experiment; m = number of prey types.

Ethical statement

No ethical issues were involved in this study.

RESULTS

Findings reveal that when prey were offered separately, the rate of feeding was very high in the 1st h for all the prey. Thereafter, the rate of feeding gradually declined particularly so in case of chironomid larvae after 6 h [Figure 2]. Experiments also revealed that diurnal predation was significantly higher than the nocturnal predation [Figures2–6 and Table 1]. The fish consumed significantly more number of mosquito larvae as compared to pupae irrespective of the fact whether those were offered separately [Figure 2 and Table 2] or simultaneously [Figure 3 and Table 3]. S. sadanundio consumed mosquito larvae and chironomid larvae with high efficiency [Table 5 and Figure 4] but consumed significantly more mosquito larvae during night when those were offered separately [Table 4 and Figure 6]. However, when these were offered together, it consumed significantly more chironomid larvae except during night time when the fish consumed less number of chironomid larvae, but the difference was insignificant [Table 5 and Figure 7].

Figure 2

A comparison of the consumption of prey by Stigmatogobius sadanundio when offered separately

Figure 6

Number predated (χ̄ ± standard error) by Stigmatogobius sadanundio when mosquito larvae and chironomid larvae were offered together

Table 1

A comparison of the diurnal and nocturnal predation Stigmatogobius sadanundio

Test (x̅±SE)	Day	Night	t	P
Mosquito larvae	187.47±2.12	114.40±1.60	27.4753	<0.0001***
Mosquito pupae	100.10±1.341	78.00±1.45	11.0000	<0.0001***
Chironomid larvae	266.73±3.89	52.33±1.12	52.9530	<0.0001***

***Extremely significant. SE: Standard error

Table 2

Number of prey consumed (x̅±standard error) by Stigmatogobius sadanundio when Culex quinquefasciatus larvae and pupae were offered separately

Time (h)	Culex quinquefasciatus		t	P
	Larvae	Pupae
1	64.73±2.01 (53-79)	35.0 0±0.87 (31-42)	13.5249	<0.0001***
6	112.2±1.72 (99-124)	70.53±1.22 (63-77)	19.7136	<0.0001***
12 (day)	187.46±2.11 (175-201)	100.13±1.34 (93-109)	34.8146	<0.0001***
12 (night)	114.4±1.60 (102-124)	78.00±1.45 (66-87)	16.7850	<0.0001***
24	301.86±2.74 (290-317)	178.13±2.49 (159-194)	33.3700	<0.0001***

***Extremely significant

Figure 3

A comparison of the consumption of Mosquito larvae and pupae by Stigmatogobius sadanundio when offered together

Table 3

Number of prey consumed (x̅±standard error) by Stigmatogobius sadanundio when Culex quinquefasciatus larvae and pupae were offered together

Time (h)	Culex quinquefasciatus		t	P
	Mosquito larvae	Mosquito pupa
1	45.46±1.05 (37-52)	16.06±0.88 (11-21)	21.3916	<0.0001***
6	74.26±2.53 (57-87)	30±1.10 (23-36)	15.9826	<0.0001***
12 (day)	142.46±1.91 (128-156)	44.46±1.16 (41-51)	43.7964	<0.0001***
12 (night)	118.86±1.45 (110-128)	49.33±1.16 (41-56)	37.3181	<0.0001***
24	261.33±2.38 (253-282)	93.8±1.84 (80-103)	55.5271	<0.0001***

***Extremely significant

Table 5

Number of prey consumed (x̅±standard error) by Stigmatogobius sadanundio when Culex quinquefasciatus larvae and Chironomus ramosus larvae were offered together

Time (h)	Feeding rate		t	P
	Culex quinquefasciatus larvae	Chironomus ramosus larvae
1	45.53±2.31 (32-66)	71.13±1.56 (62-81)	−9.1566	<0.0001***
6	58.73±3.65 (41-93)	99±3.06 (78-117)	−8.4427	<0.0001***
12 (day)	85.6±3.16 (61-109)	139.06±2.14 (121-151)	−13.9900	<0.0001***
12 (night)	40.13±2.22 (21-51)	39.4±1.11 (32-46)	0.2946	<0.7705*
24	125.73±4.76 (93-156)	178.06±2.89 (150-195)	−9.3890	<0.0001***

*Insignificant, ***Extremely significant

Figure 4

A comparison of the consumption of Chironomid and Mosquito larvae by Stigmatogobius sadanundio when offered together

Table 4

Number of prey consumed (x̅±standard error) by Stigmatogobius sadanundio when Culex quinquefasciatus larvae and Chironomus ramosus larvae were offered separately

Time (h)	Feeding rate		t	P
	Culex quinquefasciatus larvae	Chironomus ramosus larvae
1	64.73±2.017 (53-79)	134.13±1.28 (126-141)	−29.0240	<0.0001***
6	112.2±1.72 (99-124)	236.13±2.19 (224-250)	−44.4822	<0.0001***
12 (day)	187.46±2.11 (175-201)	266.73±3.88 (246-288)	−17.6709	<0.0001***
12 (night)	114.4±1.60 (102-124)	52.33±1.12 (46-59)	31.6241	<0.0001***
24	301.86±2.74 (290-317)	319.06±3.85 (295-345)	−3.6390	<0.0011***

***Extremely significant

Figure 7

Number of predated (χ̄± standard error) by Stigmatogobius sadanundio when mosquito larvae and pupae were offered together

Number predated (χ̄ ± standard error) by Stigmatogobius sadanundio when prey were offered separately

One-way ANOVA revealed a significant difference in the consumption of the prey types (F (2,42) = 610.7; P < 0.0001). Tukey's multiple comparison test based on 24 h has revealed significant differences (P < 0.05) in relative consumption of mosquito larvae and chironomid larvae (|q| = 5.607), mosquito larvae and pupae (|q| = 39.72), and between mosquito pupae and chironomid larvae (|q| = 45.33]. Chesson index values also revealed that S. sadanundio has a high dietary preference for chironomid larvae (0.40), followed by mosquito larvae (0.33) and mosquito pupae (0.26) in this order.

DISCUSSION

In the present study, potentiality of S. sadanundio, an indigenous fresh water fish, has been investigated as a potent biocontrol for the first time. This species is native to south Asia, from India to Indonesia including Sri Lanka and Andaman islands. S. sadanundio satisfies all the criteria of a larvicidal fish such as small in size, hardy in nature, and capable of living in shallow water as stated by Kim et al.[12] Findings of the present study revealed that in starved S. sadanundio, the predation rate was very high during 1st h under laboratory condition. Similar observations have also been made by Griffin[13] for Australian mangrove fish (Pseudomugil signifer, Hypseleotris galii, and Pseudogobius sp.). S. sadanundio on an average consumed 300 Culex larvae in 24 h, which is considerably higher than the daily consumption of Colisa fasciatus (100 larvae/day) and by Oreochromis mossambica, Aplocheilus panchax, Rasbora elegans, Rasbora daniconius, Puntius ticto, Puntius sophore, Esomus danricus, Danio rerio, Oryzias melastigma (75–99 larvae/day). Next to this category comes Channa gachua, Anabas testudineus, Notopterus notopterus, Puntius sarana, Puntius phutunio, Chela aptar, Channa punctatus (50–74 larvae/day), Mystus bleekeri, Colisa lalia, Oxygaster bacaila, Chela laubuca (25–49 larvae/day) and Nemacheilus savona, and Nemacheilus aureus (1–24 larvae/day) as reported by Das et al.[14] This makes S. sadanundio a very efficient and potent biocontrol agent of C. quinquefasciatus. Like C. gachua[15] and R. daniconius,[16] S. sadanundio is also a diurnal predator as it predates significantly more on larvae and pupae during daytime, but has a greater preference for larvae over pupae. Larvivorous predators have a wide range of prey choice, and the presence of alternative prey may adversely affect the target prey consumption. In the present study, it was observed that S. sadanundio has a preference for Chironomus larvae over mosquito larvae, particularly during daytime. This finding is also confirmed by the higher Chesson index for chironomid larvae. However, during nocturnal feeding, such preference could not be noticed. During daytime, the fish relies more on the visual cue for predation as has been suggested by Mills et al.[17] for yellow perch, Perca flavescens. Perhaps, S. sadanundio was attracted more to the chironomid larvae during daytime because of its bright red color.

The presence of alternative prey considerably influences mosquito larvae consumption rate of larvivorous fishes.[1819] In the presence of alternative prey, biocontrol potentiality is considerably reduced in dytiscid beetles,[20] odonate larvae[2122] and in heteropteran bugs.[232425] Relative abundance of alternative prey can also alter the mosquito larval consumption rate of the fish.[26]

CONCLUSION

Findings of the present study indicate that S. sadanundio, a common freshwater fish, may be effectively used in the biocontrol of C. quinquefasciatus larvae due to its high larval consumption rate. Being indigenous may be recommended as an ideal alternate to other larvivorous fish such as Gambusia affinis and Poecilia reticulata on ecological consideration. Its effectiveness, however, must be tested beforehand in complex natural condition on the basis of detailed in depth field trials.

Financial support and sponsorship

Department of Science and Technology, Government of West Bengal Research Project (Memo No. 172 (Sanc.)/ST/P/S & T/1G-70/2017 Dated March 16, 2018).

Conflicts of interest

There are no conflicts of interest.