OUP user menu

Lipid Utilization for Ovarian Development in an Autogenous Mosquito, Culex pipiens molestus (Diptera: Culicidae)

K. Sawabe, A. Moribayashi
DOI: http://dx.doi.org/10.1603/0022-2585-37.5.726 726-731 First published online: 1 September 2000


During the first ovarian cycle, autogenous female mosquitoes develop their ovaries in the absence of blood feeding. In autogenous Culex pipiens molestus (Forskal), complete yolk deposition was observed 2 d after emergence, even when no feeding was allowed (starved). Neutral lipids in Cx. p. molestus increased during the pupal stage, abruptly declined after emergence, and again increased on day 3. In contrast, neutral lipids decreased in anautogenous Anopheles stephensi (Liston) and Cx. p. pallens (Coquillett) and starved females died within 2–3 d after emergence. High ratios of two major neutral lipids, free fatty acid and triglyceride, were isolated by thin-layer chromatography (TLC) from the lipid contents of both Cx. p. molestus and An. stephensi fourth- instars and newly emerged females. Fatty acid analyses using gas liquid chromatography (GLC) and GLC-mass spectrometry (GLC-MS) showed higher proportions of unsaturated than saturated fatty acids in Cx. p. molestus at both stages and two major neutral lipids: free fatty acids and triglycerides. The percentage composition of linoleic acid (C18:2), which is a precursor of arachidonic acid, was higher in Cx. p. molestus than in An. stephensi. Our results indicated that elevated lipid content before emergence may play a role of inducing ovarian development in autogenous mosquitoes.

  • Culex pipiens molestus
  • Culex pipiens pallens
  • Anopheles stephensi
  • yolk deposition
  • lipids
  • fatty acids

Blood ingestion and ovarian development by vector mosquitoes play important roles in the transmission of several human pathogens. Most female mosquitoes take a blood meal to develop their eggs, whereas autogenous species initiate vitellogenesis in the absence of blood feeding. Biochemical differences between autogenous and anautogenous species may reveal mechanisms important in ovarian development that could be exploited for the control of mosquitoes as vectors.

Autogeny has been reported in at least 15 genera of mosquitoes (Clements 1963, Ellis and Brust 1973). Autogeny was first described in Culex pipiens, showing differing degrees of autogeny (Clements 1992). The common house mosquito in Japan, Culex pipiens pallens (Coquillett), is anautogenous, but Culex pipiens molestus (Forskal) is autogenous. Although often classified as intraspecific forms, Cx. p. molestus is now regarded as a physiological variant of Cx. pipiens (L.) (Barr 1982, Harbach et al. 1984).

Physiological differences between autogenous and anautogenous mosquitoes have been proposed based on differences in the content of proteins or amino acids (Corbet 1964, Lang 1978) and hormone release (Lea 1963, Fuchs et al. 1980, Borovsky 1982, Guilvard et al. 1984). Protein reserves are a very important component of yolk proteins (Uchida and Suzuki 1981, Uchida et al. 1990). Lipids play an important role in oogenesis as a principal component of the ovarian yolk (Clements 1992) and as an energy source (Twohy and Rozeboom 1957). However, lipids have not been investigated fully in ovarian development.

Mechanisms of oogenesis and vitellogenesis in mosquitoes have been studied mainly in Aedes (Hagedorn et al. 1973; Borovsky 1981, 1982; Dhadialla and Raikhel 1990) and information on Anopheles is limited (Hagedorn et al. 1977, Briegel 1990). Tsukamoto (1982) reported that some Cx. pipiens have large amounts of a natural inhibitor to some enzymes such as lactate dehydrogenase (LDH) and malate dehydrogenase (MDH) in the larval midgut, whereas in Anopheles stephensi (Liston), the amount is low. An. stephensi is not only anautogenous like Cx. p. pallens, but it possibly exhibits enzymatic activity different from Cx. p. molestus.

In our article the changes in lipid content and fatty acid composition were compared between two Culex pipiens species, Cx. p. molestus and Cx. p. pallens, and An. stephensi. We demonstrate the role of lipid reserves in ovarian development between autogenous and anautogenous females.

Materials and Methods


Cx. p. molestus was obtained from the Institute of Tropical Medicine, Nagasaki University, Nagasaki, Japan; Cx. p. pallens from The National Institute of Infectious Diseases, Tokyo, Japan; and An. stephensi (Beach strain, Pakistan) from the Sumitomo Chemical, (Takarazuka, Japan). Larvae were fed a powdered mixture of Ebios (dried brewer's yeast, Tanabe Pharmaceutical, Osaka, Japan), liver powder (baby food, Yukijirushi, Hokkaido, Japan), and Tetramin (food for tropical fish, Tetra-Werke, Melle, Germany). Adults were maintained at 25°C, 60–80% RH, and a photoperiod of 16:8 (L:D) h, and fed a 2% sugar solution. About 5-wk-old male ddY mice obtained from Kyudo (Fukuoka, Japan) were used as a blood source for Cx. p. pallens and An. stephensi, whereas Cx. p. molestus was maintained without a blood meal.

Changes in Neutral Lipids and Ovarian Development.

Ten fourth-instars and newly emerged adults were frozen at −80°C. Additional adults were allowed to feed on sugar (sugar-fed) or starved until death. Individuals were homogenized in 20 μl of distilled water using a micro-tissue grinder (Wheaton, Millville, NJ). Neutral lipids were measured by the GPO-p-chlorophenol method (Triglyceride G-test Wako, Wako Pure Chemical Industries, Osaka, Japan) with absorbance read at 505 nm using a spectrophotometer. Values were reported as triglyceride equivalents calculated from a standard curve for 300 mg/ml of triolein.

For female mosquitoes, ovarian development was observed under a light microscope. Ten females were dissected on each occasion, and yolk deposition was determined in more than two-thirds of the oocytes.

Dry Weight and Lipid Contents.

To determine dry body weight, fourth-instar larvae on the last day before pupation and 1-d-old females were thawed at room temperature, placed on filter paper (Toyo No. 2, Toyoroshi, Tokyo, Japan), and desiccated in a glass box with silica gel for 3 h. Fifty individuals were weighed and averaged.

Fifty individuals of each stage were homogenized in a mixture containing chloroform and methanol (2:1, vol:vol) using a glass homogenizer (Iwaki Glass, Chiba, Japan). The homogenates were centrifuged at 1,000 × g for 10 min at 4°C, the supernatant decanted into a v-vial (Wheaton), and the precipitate vortexed with a fresh chloroform-methanol mixture. This procedure was repeated three times to recover all the lipids from the precipitate. The lipid extract was evaporated under N2, allowed to dry under low air pressure for ≈1 h, and weighed. Total lipids for the 50 individuals was measured and then averaged.

Separation of Neutral Lipids by Thin-Layer Chromatography (TLC).

Neutral lipids from fourth-instars and 1-d-old females were separated from the total lipid extract on silica gel 60 F254 TLC plates (20 by 20 cm, Wako Pure Chemical Industries) as described by Sawabe and Mogi (1999).

Analyses of Fatty Acids by GLC and GLC-MS.

To identify the fatty acid composition of the lipids isolated from the TLC plates, fatty acid samples were scraped from the plates and methylated with 5% hydrochloric acid-methanol in a sealed tube at 110°C for 2 h. Methylated fatty acids were dissolved in a small amount of hexane and analyzed by GLC (GLC, Shimadzu Scientific Instruments, Kyoto, Japan) as was described by Moribayashi et al. (1996). Further analyses of methylated fatty acids were performed to confirm the location of unsaturated carbon chains in polyunsaturated fatty acid molecules by GLC-Mass spectrometry (GLC-MS, Hewlett-Packard, Palo, Alto, CA) as was described by Moribayashi et al. (1996).


Changes in Neutral Lipids.

Neutral lipids in Cx. p. molestus increased to >50 μl at the larval–pupal transition, declined immediately after emergence, and increased again on day 3 (Fig. 1). Cx. p. pallens had large amounts of neutral lipids (40.8–42.0 μl) before emergence, similar to Cx. p. molestus (32.0–51.6 μl), but the quantity of neutral lipids decreased to 27.6 μl after emergence. The level of lipids in Cx. p. pallens on day 1 was significantly greater than that of Cx. p. molestus (9.7 μl: t = 3.66, df = 18, P < 0.005). Neutral lipids at fourth-instar in An. stephensi was nearly equal to Cx. p. molestus (t = 0.94, df = 18, P > 0.1), but then declined by 1 d after emergence. Although lipid levels in An. stephensi were significantly less than in Cx. p. pallens at all stages (t = 3.59–9.67, df = 18, P < 0.005 in all cases), the same pattern of change in neutral lipids were found during development.

Fig. 1

Changes in the neutral lipid content at different stages of development among three species of mosquitoes. Cpm, Cx. pipiens molestus; Cpp, Cx. p. pallens; Ans, An. stephensi. Determination of yolk deposition (YD) is described in Materials and Methods. Ten individuals were measured and each value indicates the average with 95% CI.

Starved Cx. p. molestus females survived at least for 3 d after emergence and yolk deposition was observed on day 2, whereas An. stephensi and Cx. p. pallens females died within 2–3 d after emergence. In Cx. p. molestus, both sugar-fed and starved females showed the same time trend in changes in neutral lipid quantity, but the levels for sugar-fed females were significantly higher than starved females (t = 12.97–15.71, df = 18, P < 0.0005 in all cases; Fig. 2). For males, neutral lipids declined slightly by days 1 and 2 after emergence, and increased again on day 3 in starved and sugar-fed individuals, respectively. Lipid syntheses in both sexes of sugar-fed An. stephensi were similar to starved males of Cx. p. molestus. Starved individuals of both sexes died within 2–3 d after emergence.

Fig. 2

Changes in the neutral lipid content at different stages of development of An. stephensi (left) and Cx. p. molestus (right). Larvae used were at the last day of the fourth-instar, and adults were starved (open circles with a dashed line) and sugar-fed (closed circles with a solid line). Ten individuals of each sex were measured and each value indicates the average with 95% CI.

Lipid Content and Neutral Lipid Components.

Dry weights of Cx. p. molestus fourth-instars and 1-d–old females were heavier than An. stephensi; however, there were no significant differences in the lipid content per mg dry weight between species or developmental stages (larvae: χ2 = 0.01, df = 1, P > 0.5; females: χ2 = 0.97, df = 1, P > 0.25). For each species total lipids were greatest in individuals with the highest dry weight.

Thin-layer chromatography separated the neutral lipids into four components: two major components of free fatty acids and triglyceride, and two minor components of diglyceride and phospholipid (Fig. 3). The percentage composition of free fatty acids and triglycerides in Cx p. molestus was >80% of the total neutral lipids obtained from larvae and females. Culex p. molestus had a significantly higher percentage composition of triglyceride than An. stephensi (larvae: χ2 = 27.39, df = 1, P < 0.0001; females: χ2 = 3.97, df = 1, P < 0.05). Free fatty acid composition of females was similar (χ2 = 2.22, df = 1, P < 0.25), but fourth-instar larvae was significantly lower (χ2 = 7.75, df = 1, P < 0.01) than that in An. stephensi. The two minor components generally were lower in Cx. p. molestus than in An. stephesi for both larvae (phospholipid: χ2 = 2.85, df = 1, P < 0.1; diglyceride: χ2 = 7.71, df = 1, P < 0.01) and females (phospholipid: χ2 = 8.31, df = 1, P < 0.01; diglyceride: χ2 = 2.51, df = 1, P < 0.25).

Fig. 3

Percentage composition of main neutral lipids of An. stephensi (left) and Cx. p. molestus (right) for fourth instars and 1-d-old females. The main neutral lipids were phospholipid (PL), diglyceride (DG), triglyceride (TG), and free fatty acids (FFA). Fifty individuals of each stage were used for TLC analysis as described in Materials and Methods.

Fatty Acid Analysis.

The fatty acid components of the two major neutral lipids (free fatty acids and triglyceride), determined by GLC and GLC-MS analyses, contained saturated and unsaturated carbon chains of 14–23 carbon length (Fig. 4). The main fatty acids obtained from both An. stephensi and Cx. p. molestus in larval and adult stages were palmitic acid (C16:0), palmitoleic acid (C16:1), stearic acid (C18:0), oleic acid (C18:1), linoleic acid (C18:2), and arachidonic acid (C20:4). Triglycerides of An. stephensi had a higher percentage composition of stearic acid in both larvae (χ2 = 9.55, df = 1, P < 0.005) and females (χ2 = 6.19, df = 1, P < 0.05) than in these developmental stages of Cx. p. molestus. However, palmitoleic acid was lower than in Cx. p. molestus at both developmental stages of An. stephensi (larvae: χ2 = 11.86, df = 1, P < 0.001; females: χ2 = 5.02, df = 1, P < 0.05); linoleic acid also was lower in females (χ2 = 7.07, df = 1, P < 0.01). Free fatty acid content was almost the same as triglycerides (Fig. 4). Significantly higher percentage composition of palmitoleic acid and linoleic acid were obtained from Cx. p. molestus larvae (χ2 = 9.22, df = 1, P < 0.005) and females (χ2 = 8.09, df = 1, P < 0.005) than in An. stephensi, respectively. There were no significant differences for oleic acid or arachidonic acid between species and developmental stages.

Fig. 4

Fatty acids composition of triglycerides (TG) and free fatty acids (FFA) isolated from An. stephensi and Cx. p. molestus at fourth-instars and 1-d-old females. The main fatty acids were palmitic acid (C16:0), palmitoleic acid (C16:1), stearic acid (C18:0), oleic acid (C18:1), linoleic acid (C18:2), and arachidonic acid (C20:4).

Table 2 shows the ratio between saturated and unsaturated fatty acids among >20 components detected by GLC and GLC-MS analysis. At all stages in the two major neutral lipids examined, relative proportions of unsaturated fatty acids were higher than those of saturated ones (1.02–2.33), except for triglycerides in An. stephensi. The ratios of unsaturated fatty acids obtained from Cx. p. molestus females were greater than in An. stephensi (triglycerides: χ2 = 8.57, df = 1, P < 0.005; free fatty acids: χ2 = 7.94, df = 1, P < 0.005).

Table 1

View this table:
Table 1
View this table:
Table 2


The autogenous mosquito Cx. p. molestus initiates vitellogenesis in the absence of blood feeding during the first ovarian cycle. In this study, we compared lipid utilization by two anautogenous mosquitoes, Cx. p. pallens and An. stephensi, with Cx. p. molestus. Culex p. pallens showed no ovarian development even in glucose-fed females, but the infusion of amino acids through the hemolymph induced egg development (Uchida et al. 1992). By taking in an additional supply of nutrition, such as amino acids, after adult emergence, these females may be induced to develop eggs. Glucose-fed females of An. stephensi also showed no ovarian development, and egg development was not induced previously by the infusion of any amounts of amino acids (K. Uchida, personal communication). An. stephensi is not only anautogenous like Cx. p. pallens, but it also seems to have developed enzymatic activities associated with lipid synthesis that are different from Cx. pipiens species (Tsukamoto 1982).

The decline in neutral lipids at days 1–2 after emergence followed by an increase was unique for starved Cx. p. molestus females; neutral lipids in anautogenous An. stephensi and Cx. p. pallens only decreased. Starved females of An. stephensi and Cx. p. pallens died within 2–3 d after emergence. No great differences in the amounts of neutral lipids were found between these two Cx. pipiens species before emergence. Therefore, as a first step toward understanding differences of lipid utilization between autogeny and anautogeny, we compared Cx. p. molestus and An. stephensi for analysis of fatty acid composition. Information on lipids from Cx. p. pallens would be a useful next step in our research.

Yolk deposition was completed in Cx. p. molestus 2 d after emergence without either sugar feeding or a blood meal, but not in Cx. p. pallens and An. stephensi. With no exogenous supply of nutrients, Cx. p. molestus must have carried over sufficient reserves for vitellogenin production from the larval stage. Larval nutrition plays an important role in the expression of autogeny. Poor larval nutrition reduces autogeny in facultatively autogenous mosquitoes, such as Cx. pipiens (Spielman 1957), Aedes togoi (Theobald) (Laurence 1964), and Wyeomyia smithii (Coquillett) (Lounibos et al. 1982). Autogenous mosquito larval development is longer than anautogenous larvae (Clements 1992). Newly emerged autogenous females are usually heavier, and contain more lipids, glycogen, protein and nitrogen content than newly emerged anautogenous females (Clements 1956, Lang 1963, Briegel 1969). In our study, the same diet was supplied to larvae of all three species, but no great differences were found in larval-pupal duration among them (unpublished data). Female Cx. p. molestus had a larger body size and greater amounts of neutral lipids than An. stephensi. We suggest that female Cx. p. molestus may use lipids not only for adult longevity but also for ovarian development. These females transfer the lipids from the larval–pupal stage to adult females to be used as yolk lipids in the oocytes.

The ratios of triglycerides and free fatty acids in Cx. p. molestus were usually higher than in An. stephensi. The chemical structure of these two neutral lipids made them more useful as energy sources than other lipid components such as diglycerides and phospholipids. In Cx. p. molestus, the ratio of unsaturated compared with saturated fatty acids also was greater than in An. stephensi. The melting point for unsaturated fatty acids is lower than that for saturated fatty acids, and permeability through the cell membrane is increased with a high ratio of unsaturated fatty acids. Unsaturated fatty acids may be more efficacious as an energy source than saturated fatty acids. These results indicate that Cx. p. molestus probably accumulates lipid reserves to be used as an energy source for adult longevity as well as a source for yolk lipids.

Newly emerged females of Cx. p. molestus had the highest percentage composition of linoleic acid (C18:2). Linoleic acid, a precursor of arachidonic acid, and arachidonic acid (C20:4) are regarded as essential fatty acids for many insects (Dadd 1983). Arachidonic acids act as specific inhibitors of prostaglandin syntheses in mosquitoes (Dadd and Kleinjan 1984). One of the arachidonic acids (C20:4n-6) metabolites releases gonadotropin in the common goldfish (Chang et al. 1989). In autogenous females of Ae. detritus (Haliday), ecdysteroid and juvenile hormone titers peaked between 35 and 40 h, respectively, at 43 h postemergence (Guilvard et al. 1984). These authors reported that the follicles entered stage IIIa at the time of the ecdysteroid peak, and reached stage IIIb immediately after the juvenile hormone peak. Linoleic acid and arachidonic acid may act as hormones in autogenous mosquitoes. Further studies of arachidonic acid are required to examine the possibility that some lipid components, such as linoleic acid and arachidonic acid, are associated with vitellogenesis as a releaser of reproductive hormones or act directly on these hormones as shown by Guilvard et al. (1984).

Our results indicate that Cx. p. molestus is able to use the linoleic acid-rich lipid accumulated during larval-pupal stage for egg development. Lipid reserves are as essential for vitellogenesis as protein. More comparative studies using other autogenous species, such as Cx. tarsalis Coquillett and Ae. togoi, are needed to clarify whether or not the lipid metabolism as shown in Cx. p. molestus is typical for all autogenous mosquitoes.


We thank S. Tojo for his technical suggestions and encouragement in the TLC analysis, and Lisa Filippi for correcting the English of this manuscript. This study was supported in part by a grant-in-Aid from the Ministry of Science, Education and Culture, Japan (Grant No. 05770180).

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0/), which permits non-commercial reuse, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions{at}oup.com

References Cited

View Abstract