Antimicrobial activity of alcohols from Musca domestica

The house fly, Musca domestica, is a well-known cosmopolitan pest, both on farms and in homes. This species is always found in association with humans or human activity. The house fly has a complete metamorphosis with distinct egg, larvae/maggot, pupal and adult stages. Warm summer conditions are generally optimal for the development of the house fly. The complete life cycle of M. domestica lasts from 7 to 10 days; however, under suboptimal conditions the life cycle may require up to 2 months.

The control of M. domestica is vital to human health and comfort in many areas of the world. The most important impacts of M. domestica are damages caused by the potential transmission of pathogens (viruses, bacteria, fungi, protozoans and nematodes). Pathogenic organisms are picked up by flies from garbage, sewage and other sources of filth, and then transferred on their mouthparts, through their vomitus, feces and contaminated external body parts to human and animal food.

The more commonly used control measures for house flies are traps and insecticides, but in some instances integrated fly control has been implemented. The use of biological control in fly management is still at a relatively early stage. The sex pheromone (Z)-9-tricosene is formulated with sugar as a commercially available fly bait for local population suppression, as well as an enhancement for population monitoring. Ultraviolet light traps can be used to assess population levels, but also serve as a non-chemical control technique that can be applied in both agricultural and non-agricultural confined spaces. Natural biological suppression of the house fly results primarily from the actions of certain chalcidoid wasps (Hymenoptera: Pteromalidae), many species of which are associated with the house fly around the world. Among the most important are Muscidifurax and Sphalangia spp. Ichneumonids and other parasitoids, as well as some predatory insects [especially histerids (Coleoptera: Histeridae) and staphylinids (Coleoptera: Staphylinidae)], also contribute to fly mortality.

The potential application of fungal pathogens as control agents against insect pests is also well documented (Roberts and Hajek, 1992; Clarkson and Charnley, 1996; Khachatourians, 1996; Shah and Pell, 2003; Faria and Wraight, 2007). Unlike other insect-pathogenic microorganisms, which must be ingested to initiate disease (e.g. viruses, bacteria, nematodes and protozoans), entomopathogenic fungi such as Metarhizium anisopliae, Conidiobolus coronatus and Beauveria bassiana mostly invade target hosts by penetrating through their cuticle. Susceptibility or resistance of various insect species to fungal invasion may be caused by several factors, including differences in the structure and composition of the exoskeleton and the presence of antifungal compounds in the cuticle (Vilcinskas and Götz, 1999). It is believed that the epicuticular lipid profile of insect hosts may be one of the pivotal factors determining insect susceptibilities or resistance to fungal attack (Gillespie et al., 2000).

The outermost insect surface or epicuticle is covered by a lipid-rich layer, usually composed of hydrocarbons (Gołębiowski et al., 2012a) together with variable amounts of alcohols (Buckner, 1993), fatty acids (Gołębiowski et al., 2007) and wax esters (Nelson et al., 1998). Entomopathogenic fungi have the ability to degrade insect cuticular lipids with hydrocarbons, the preferred components capable of supporting fungal growth (Napolitano and Juárez, 1997). Although several aspects of the interaction between entomopathogenic fungi and insect host cuticular hydrocarbons and fatty acids have been examined (Pedrini et al., 2007; Boguś et al., 2010), there is no information concerning the other surface compounds.

Free fatty alcohols are not commonly found in epicuticular lipids of insects, although high molecular weight alcohols were identified in honeybees (Blomquist et al., 1987). It was found that honeybees produced alcohols of 17–22 carbon atoms (Donze et al., 1998). In the epicuticle of the blood-sucking bug Triatoma infestans, two female-specific fatty alcohols, docosanol (C22) and eicosanol (C20), were found (Cocchiararo-Bastias et al., 2011). Long-chain alcohols have also been reported in the defensive secretions of the scale insects Bemisia tabaci and Trialeurodes vaporariorum (Byrne and Hadley, 1988).

Information on both stimulatory and inhibitory effects of cuticular alcohols on growth and virulence of insecticidal fungi is unavailable. This paper describes the content of cuticular and internal alcohols in the body of housefly larvae, pupae, males and females. The surface lipids of flies were separated into classes of compounds using high performance liquid chromatography equipped with a laser light scattering detector (HPLC-LLSD). Qualitative and quantitative analysis was achieved by gas chromatography (GC) and gas chromatography combined with mass spectrometry (GC-MS).

Isolated compounds were tested according to their potential antimicrobial activity. We have selected three representative strains each of gram-positive bacteria, gram-negative bacteria and five strains of fungi. Activity against reference strains was determined for individual alcohols as well as for their mixtures found in cuticular lipids of larvae, pupae, males and females of M. domestica. Antimicrobial activity of alcohol mixtures was considered according to their composition.

Musca domestica Linnaeus 1758 raised from eggs laid on diet by adult flies were reared at 28°C with 70% relative humidity and a 12 h:12 h light:dark photoperiod. The maternal generation was maintained under the same conditions. House flies were reared on a mixture 1:1 of standard rodent chow (LSM and LSK, Fodder Manufacture, Motycz, Poland) dissolved in water. For experiments, 1-day-old larvae, freshly emerged pupae and 6-day-old sexually mature adults were used. The insects were either used for lipid extractions or exposed for 18 h to fully grown and sporulating colonies of the entomopathogenic fungus Conidiobolus coronatus. In a Petri dish containing a C. coronatus colony, 10 flies (males and females separately), 10 pupae or 10 larvae were maintained. The insects exposed for 18 h to sterile uninoculated Sabouraud agar medium (SAM) served as controls. After exposure, the adults were transferred to clean Petri dishes with sugar and water, whereas larvae were transferred to dishes with fresh diet. Pupae were transferred to clean Petri dishes without food. All insects exposed to C. coronatus were kept at proper growing conditions for 10 days and their condition was monitored daily. Exposure of tested insects to the C. coronatus colony for 18 h was found to be the most efficient method of resembling the natural infection process (Wieloch and Boguś, 2005). To avoid pseudoreplication, all assays of fungi versus insects were performed with flies from different stocks incubated in three different chambers.

A culture of the wax moth Galleria mellonella was maintained and reared in temperature and humidity controlled chambers (30°C, 70% relative humidity) in constant darkness on an artificial diet (Sehnal, 1966). Fully grown larvae were collected before pupation, surface sterilized, homogenized and used as supplement in fungal cultures.

Conidiobolus coronatus (Zygomycetes), isolate number 3491, originally isolated from Dendrolaelaps spp., was obtained from the collection of Prof. Bałazy (Polish Academy of Sciences, Research Center for Agricultural and Forest Environment, Poznań, Poland), and was maintained in 90 mm Petri dishes at 20°C with cyclic changes of light (12 h:12 h light:dark) on SAM with the addition of homogenized G. mellonella larvae to a final concentration of 10% wet mass. Addition of homogenized G. mellonella larvae has been shown to enhance SAM cultures of C. coronatus. The levels of mycelial growth, conidia production and virulence were good in hundreds of successive transfers (Wieloch, 2006), suggesting a stimulatory effect of insect proteins, carbohydrates and lipids on C. coronatus growth and insecticidal properties. Seven-day-old cultures were briefly washed with sterile water to harvest conidia. The number of harvested conidia was determined under a microscope with the use of a hemocytometer. One-hundred-microliter portions of suspension each containing ~50 conidia were used for inoculations.

Bacterial (Bacillus subtilis ATCC 6633, Rhodococcus equi ATCC 6939, Staphylococcus aureus ATCC 25923, Escherichia coli ATCC 25922, Klebsiella pneumoniae ATCC 13882 and Proteus vulgaris ATCC 13315) and fungal (Aspergillus niger ATCC 16404, Candida albicans ATCC 10231, Candida lipolytica PCM 2680 and Candida tropicalis PCM 2681) reference strains were obtained from The Polish Collection of Microorganisms (Polish Academy of Sciences, Wrocław, Poland). Reference strains of bacteria were inoculated in Mueller–Hinton II broth 24 h before performing the minimal inhibitory concentration (MIC) test, and incubated at 37°C with 150 r.p.m. shaking. Fungal strains were cultured at the temperature 25°C with 150 r.p.m. shaking for 48 h by the time the MIC was tested.

The cuticular lipids of M. domestica were obtained by immersing insects in 40 ml of petroleum ether for 10 s (extract I) and in 40 ml of dichloromethane for 5 min (extract II). The same insects were transferred to dichloromethane for 10 days (extract III). Extracts I and II contained cuticular lipids and extract III contained internal lipids. The extracts were evaporated on a rotary evaporator at a temperature not exceeding 40°C. Table 1 lists the number of insects, as well as the masses of the extracts. Extraction was carried out according to previously published methods (Gołębiowski et al., 2008a).

The lipid fractions obtained by HPLC were silylized with 100 μl of a mixture of 99% bis(trimethylsilyl)acetamide and 1%chlorotrimethylsilane for 1 h at 100°C. Silylized samples were analyzed by GC-MS.

The lipid extracts were separated using HPLC-LLSD. The detector evaporation temperature was 45°C, and the carbon dioxide pressure was 0.1–0.2 MPa. The separation in the normal phase was performed on a silica gel column (Econosil Silica 5 Micron, 25 cm × 4.6 mm, Alltech, Nicholasville, KY, USA). The mobile phase consisted of hexane (Solvent A) and dichloromethane with the addition of 15% acetone (Solvent B). The gradient was programmed linearly from 100% A to 100% B within 30 min. Total flow was maintained at 0.8 ml min⁻¹.

Electron-impact (EI; 70 eV) mass spectra were obtained on an HP 5890 gas chromatograph (Hewlett-Packard, Palo Alto, CA, USA) coupled with an SSQ-7100 mass spectrometer (San Jose, CA, USA). GC analysis was performed on an Rtx-5MS column (30 m length × 0.25 mm i.d. × 0.25 μm film thickness; Restek, Warsaw, Poland). The temperature was programmed from 80°C (held for 8 min) to 320°C (held for 5 min) at a rate of 10°C min⁻¹. Each extract (1 μl) was injected (injector temperature, 320°C), using helium as the carrier gas. MS conditions were as follows: interface temperature, 320°C; ion source temperature, 220°C; mass range, 40–650 a.m.u.

Antimicrobial activity was determined for individual alcohols and mixtures of alcohols found in larvae, pupae, males and females of M. domestica. The MIC was determined using a microbroth dilution method recommended by the Clinical and Laboratory Standards Institute (CLSI). For bacterial strains, the Mueller Hinton Broth II (MHB II) (Fluka Analytical, Sigma-Aldrich, Steinheim, Germany) and initial inoculums of 5×10⁵ CFU ml⁻¹ were applied, whereas Sabouraud glucose 2% broth (Carl Roth, Karlsruhe, Germany) and initial inoculums 10³ CFU ml⁻¹ were used for tested fungi. Polypropylene 96-well plates (Becton Dickinson, Le Point de Claix, France) were incubated for 18 h at 37°C (bacteria) or 48 h at 25°C (fungi). The MIC was taken as the lowest alcohol concentration at which a noticeable growth was inhibited. The experiments were performed in triplicate.

The antifungal potential of crude extracts obtained from M. domestica against the entomopathogenic fungus C. coronatus was screened using the disc diffusion method recommended by the CLSI. Different concentrations of each tested extract dissolved in DMSO (five dilutions for each extract; concentration range 1 μg–2.6 mg per disc) were applied onto sterile discs (5 mm diameter, Whatman filter paper no. 42) presoaked in DMSO (Sigma-Aldrich). Discs were placed onto Petri dishes containing SAM supplemented with G. mellonella homogenate and inoculated with C. coronataus as described above. Inoculated plates were incubated at 20°C for 7 days, after which plates were checked for the presence of inhibition zones. Each test was performed in triplicate.

Exposure of M. domestica adults to the sporulating C. coronatus resulted in prompt death of nearly all adults: 93±3% of tested males and 87±6% of females died around the termination of the 18 h exposition to fungal culture. The lack of resistance of M. domestica males and females to C. coronatus infection was confirmed using a pairwise t-test: we did not notice any significant differences in the susceptibility of males and females. In contrast, house fly pupae and larvae were totally resistant to fungal infection (Table 2).

In order to obtain total amounts of the cuticular lipids, two short extractions with petroleum ether and dichloromethane were applied. The cuticular lipids of larvae, pupae, males and females amounted to 1.25, 0.16, 8.78 and 3.97 mg g⁻¹ of the insect body, respectively. The amount of internal lipids obtained by long dichloromethane extraction was considerable (48.73, 33.81, 8.70 and 17.38 mg g⁻¹ of the insect body, respectively; Table 1).

The cuticular alcohols of larvae contained eight saturated even-numbered carbon-chain alcohols ranging from C_12:0 to C_30:0 (Fig. 1). The major free alcohol in M. domestica larvae was C_12:0 (70.4%). The alcohols occurring in smaller quantities (from 1 to 15%) were C_14:0 (8.9%), C_16:0 (3.5%), C_18:0 (14.2%) and C_30:0 (1.3%). The cuticular lipids also contained three alcohols present in concentrations <1%: C_20:0 (trace), C_24:0 (0.9%) and C_28:0 (0.9%). The total cuticular alcohol content in the cuticular lipids (petroleum ether and dichloromethane extracts) was 4.59 μg g⁻¹ of the insect body. Internal alcohols of M. domestica larvae were not found. Table 3 lists the alcohol contents calculated per gram of insect body in larvae.

Among cuticular alcohols of pupae, C_12:0 (31.0%) was the most abundant compound (Table 4, Fig. 2). Other alcohols were detected in comparable amounts from 3.7% (C_14:0) to 14.4% (C_16:0). Only C_22:0 was present in trace amounts. In the internal lipids of pupae, only five alcohols were identified, and these occurred in trace amounts.

Eight alcohols were present in the cuticular lipids of males. The most abundant alcohol was C_24:0 (57.5%) (Table 5, Fig. 3). The percentage content of cuticular C_24:0 in males and females (57.5 and 36.5%, respectively) was significantly higher than that in cuticular lipids in larvae and pupae (0.9 and 5.6%, respectively). The major alcohol found in larvae and pupae (C_12:0) was also present in the cuticular lipids of males, and had a high percentage content (27.8%). Three alcohols of males present in smaller quantities were C_18:0 (10.1%), C_20:0 (3.0%) and C_16:0 (1.5%). Three alcohols were also present in trace amounts (C_14:0, C_26:0 and C_28:0). Only two alcohols were in the internal lipids of males (C_18:0 and C_20:0, in trace amounts).

Similar profiles of alcohols were identified in the cuticular and internal lipids in males and females. The most abundant cuticular alcohols in females were C_24:0 (36.5%) and C_12:0 (26.8%) (Fig. 4). Other alcohols were present in smaller quantities from 1.5% (C_14:0) to 11.9% (C_22:0) and two alcohols were present in trace amounts (C_26:0 and C_28:0). Only two alcohols (C_18:0 and C_20:0) were detected in comparable amounts in the internal lipids of females (3.61±0.32 and 5.01±0.42 μg g⁻¹ of the insect body, respectively). Table 6 lists the alcohol contents calculated per gram of insect body in females.

The total cuticular alcohols in females, males and larvae of M. domestica were detected in comparable amounts (4.03, 3.95 and 4.59 μg g⁻¹ of the insect body, respectively). The total cuticular alcohols occurred in smaller quantities in pupae (2.16 μg g⁻¹ of the insect body). No internal alcohols were found in larvae, and these only occurred in trace amounts in males and pupae. Only in females were internal alcohols detected in high amounts (8.62 μg g⁻¹ of the insect body). The cuticular alcohols of pupae contained 10 saturated even-numbered carbon chain alcohols ranging from C_12:0 to C_30:0, whereas C_30:0 in females, C_22:0 and C_30:0 in males, and C_22:0 and C_26:0 in larvae were absent.

Seven and eight alcohols were present in the petroleum ether extract (I) and in the dichloromethane extract (II) of larvae, respectively. The dichloromethane extract generally had more alcohols than the petroleum ether extract. The respective quantities were 2.92 μg g⁻¹ and 1.67 μg g⁻¹ of the insect body. C_20:0 (alcohol) was present only in the dichloromethane extract. The following alcohols were qualitatively predominant in the petroleum ether extract: C_12:0 (70.7%), C_18:0 (13.8%) and C_16:0 (9.6%). In the dichloromethane extract, the following alcohols were the most abundant: C_12:0 (70.2%), C_18:0 (14.4%) and C_14:0 (10.6%). In the petroleum ether extract, three alcohols were presented in trace amounts (C_24:0, C_28:0 and C_30:0). In the dichloromethane extract, two alcohols were presented in trace amounts (C_16:0 and C_20:0).

The most abundant alcohol in both pupae extracts (petroleum and dichloromethane) was C_12:0. The respective contents were 37.9 and 26.4%. In the petroleum ether and dichloromethane extracts of pupae, the relative contents of C_14:0 (4.6 and 3.1%), C_18:0 (11.5 and 8.5%), C_20:0 (6.9 and 5.4%), C_22:0 (both in trace amounts), C_24:0 (6.9 and 4.7%), C_26:0 (14.9 and 10.9%) and C_28:0 (17.2 and 9.3%) alcohols were similar. The relative contents of C_16:0 and C_30:0 in the petroleum ether extract were less than in the dichloromethane extract (trace versus 24.0% for C_16:0 and trace versus 7.8% for C_30:0).

The alcohol occurring in the highest concentrations in petroleum ether extract of males was C_12:0 (75.5%). The relative content of this compound in dichloromethane extract was 13.0%. There was a significant difference in the relative content of C_24:0 in both extracts. This alcohol was present in the highest concentrations in dichloromethane extract (75.4%), but in the petroleum ether extract occurred in trace amounts. In both cuticular extracts of males, three alcohols were present in trace amounts (C_14:0, C_26:0 and C_28:0). The C_20:0 alcohol present in the dichloromethane extract (4.0%) was absent in the petroleum ether extract. The percentage contents of other alcohols in petroleum ether and dichloromethane extracts were very distinct: C_16:0 (trace versus 2.0%), C_18:0 (24.5 versus 5.6%).

The most abundant cuticular alcohols in females were C_12:0 in petroleum extract (51.8%) and C_24:0 in dichloromethane extract (50.2%). Four alcohols present in the dichloromethane extract (C_22:0, C_24:0, C_26:0 and C_28:0) were absent in the petroleum ether extract. In contrast, C_16:0 occurred only in petroleum ether extract (16.4%). Other alcohols were present in both extracts: C_14:0 (trace versus 2.0%), C_18:0 (18.2 versus 6.1%) and C_20:0 (13.6 versus 7.8%).

The activity of individual alcohols varied significantly depending on the tested strain (Table 7). In general, long-chain alcohols were less active in comparison to shorter-chain compounds. The alcohol 1-tetradecanol (C_14:0) turned out to be very effective against gram-positive bacteria (B. subtilis, R. equi and S. areus). This compound also presented antifungal activity in tests with the use of C. albicans, C. tropicalis, C. lipolytica and A. niger. Its activity against gram-negative bacteria (E. coli, K. pneumoniae and P. vulgaris) was minor, similar to the activity of all remaining alcohols. Tested alcohols presented approximately equal activity against reference strains of fungi. C_10:0 and C_12:0 were the most active antifungal compounds and presented some antibacterial potency against gram-positive bacteria. The remaining alcohols exhibited minor antibacterial activity. Rhodococcus equi turned out to be the most susceptible bacterial strain – most tested alcohols (with the exception of C_20:0, C_26:0 and C_30:0) presented certain activity against these bacteria. Gram-negative bacteria were resistant to all alcohols at tested concentrations (data not shown).

Antimicrobial activity of alcohol mixtures was determined for all representative strains of fungi and R. equi as the most susceptible bacterial strain. A mixture of alcohols found in the cuticular lipids of females presented the highest activity against R. equi and C. lipolytica. Cuticular alcohols of larvae acted the strongest against the above-mentioned strains and C. tropicalis, whereas cuticular alcohols of males were the most efficient against C. tropicalis and C. lipolytica. Candida albicans and A. niger were the more resistant fungal strains. All extracts acted on C. albicans with the same strength, whereas A. niger was slightly more susceptible to female and larval lipids (Table 8).

In contrast to data presented in Tables 7 and 8, diffusion disc tests performed with the use of lipids extracted from larvae, pupae and adults of M. domestica (extracts I, II and III, as depicted in Table 1) showed no antifungal activity against the entomopathogenic fungus C. coronatus. No inhibition zones were seen around discs containing both cuticular and internal lipids. In addition, mycelia grew over not only substrate but also discs (data not shown).

Thin layer chromatography, together with HPLC and column chromatography, are standard methods in lipid class analysis. GC and GC-MS are often the methods used for identification of insect lipids (Lockey, 1988; Buckner, 1993; Gołębiowski et al., 2011). In the present study, the use of HPLC-LLSD and subsequent GC-MS for lipid analysis allowed us to obtain more accurate results than direct analysis by GC-MS. It was found that under optimized chromatographic conditions, closely eluting peaks did not interfere with the alcohols. LLSD is often used for plant lipid class analysis (Gołębiowski et al., 2010a) and can also be used for detection of insect lipids (Gołębiowski et al., 2008b; Gołębiowski et al., 2010b). In other words, LLSD is the universal detector for all substances less volatile than mobile phase, which is used in HPLC. In the present study, the GC analysis for quantities of underivatized n-alcohols was not reliable because of peak broadening. Therefore, the lipid samples reacted with a mixture of 99% bis(trimethylsilyl) acetamide and 1% chlorotrimethylsilane and were heated for 1 h at 100°C, and the resulting alcohol trimethylsilyl ether was analyzed by GC-MS. Alcohols were identified on the basis of silyl derivative ions [M-15]⁺ and m/z 103 (Evershed, 1992).

Early analyses of cuticular lipids of insect species indicated that cuticular alcohols vary in qualitative and quantitative composition. In insect cuticular lipids, the following alcohols have been identified: C_22:0 to C_32:0 in Melanoplus sanguinipes and Melanoplus packardii adults (Soliday et al., 1974), C_28:0 to C_34:0 in Bemisia argentifolii nymphs (Buckner et al., 1999), C_24:0 to C_34:0 in Heliothis virescens pupae and Aleyrodes singularis exuviae (Buckner et al., 1996; Nelson et al., 1998), C_26:0 to C_34:0 in Helicoverpa zea pupae (Buckner et al., 1996), C_22:0 to C_34:0 in A. singularis adults (Nelson et al., 1998), C_8:0 to C_22:0 in Acyrthosiphon pisum (Brey et al., 1985), C_32:0 and C_34:0 in Bemisia tabaci adults (Buckner et al., 1994), C_30:0 and C_32:0 in Trialeurodes vaporariorum adults (Buckner et al., 1994), C_12:0 to C_34:0 in Locusta migratoria migratoriodes adults (Oraha and Lockey, 1990) and C_12:0 to C_32:0 in Schistocerca gregaria adults (Oraha and Lockey, 1990) (Table 9). In the present study, the cuticular lipids of M. domestica males and females contained alcohols from C_12:0 to C_28:0 and those of larvae and pupae contained alcohols from C_12:0 to C_30:0.

The major free alcohols that have been found in insects are: C_14:0 (16.0%) in L. m. migratoriodes adults (Ohara and Lockey, 1990), C_15:0 (18.8%) in S. gregaria adults (Ohara and Lockey, 1990), C_16:0 (35%) in A. pisum (Brey et al., 1985), C_18:0 (55.2 and 48.8%) in Lucilia sericata males and females, respectively (Gołębiowški et al., 2012b), 14.6% in L. m. migratoriodes adults and 12.6% in S. gregaria adults (Ohara and Lockey, 1990), C_22:0 (32%) in A. pisum (Brey et al., 1985), C_24:0 (34%) in M. sanguinipes adults and 48% in M. packardii adults (Soliday et al., 1974), C_26:0 (36%) in M. sanguinipes and M. packardii adults (Soliday et al., 1974), 73% in H. virescens pupae (Buckner et al., 1996), <60% in A. singularis exuviae and <20% in A. singularis adults (Nelson et al., 1998), C_28:0 (16%) in H. virescens and H. zea pupae (Buckner et al., 1996), C_30:0 (74%) in H. zea pupae (Buckner et al., 1996), C_32:0 (36.9%) in B. argentifolii nymphs (Buckner et al., 1999), <35% in A. singularis exuviae and <70% in A. singularis adults (Nelson et al., 1998) and C_34:0 (36.1%) in B. argentifolii nymphs (Buckner et al., 1999). All major alcohols had an even number of carbon atoms, except C_15:0, which was present in S. gregaria adults (Ohara and Lockey, 1990). In the present study, the major free alcohols were: C_12:0 (70.4%) and C_18:0 (14.2%) in M. domestica larvae, C_12:0 (31.0%) and C_16:0 (14.4%) in pupae, C_12:0 (27.8%) and C_24:0 (57.5%) in males, and C_12:0 (26.8%) and C_24:0 (36.5%) in females.

Alcohols found in insects' lipids mainly contain even-numbered carbon chains. For example, only even-numbered alcohols occurred in the lipids of B. argentifolii nymphs (Buckner et al., 1999), Tenodera sinensis females, T. angustipennis males, Stagmomantis carolina females (Jones et al., 1997), and Bemisia tabaci and T. vaporariorum adults (Buckner et al., 1994). Likewise, in the present study, only even-numbered alcohols were identified in the cuticular and internal lipids of larvae, pupae, and males and females of M. domestica. In some insect species, even-numbered alcohols predominate and alcohols with an odd number of carbon atoms are present in smaller quantities. For example, cuticular lipids of H. virescens pupae contained only one odd-numbered alcohol (C_29:0) in trace amounts (Buckner et al., 1999). And in the cuticular lipids of H. zea pupae, only two alcohols are present: C_29:0 (2%) and C_31:0 in trace amounts (Buckner et al., 1999). A similar relative content of the alcohol C_29:0 has been found in the cuticular lipids of adult M. sanguinipes and M. packardii (5 and 2%, respectively) (Soliday et al., 1974). Moreover, extracts from M. sanguinipes have been shown to contain traces of two odd-numbered saturated alcohols (C_25:0 and C_27:0) and M. packardii has been shown to contain only one alcohol in traces (C_27:0) (Soliday et al., 1974). In the cuticular lipids of A. pisum (Brey et al., 1985) and L. m. migratoriodes and S. gregaria adults (Oraha and Lockey, 1990), odd-numbered alcohols shorter than C_20:0 were observed. The relative content of the alcohol C_13:0 was 11% in A. pisum (Brey et al., 1985). Locusta migratoria migratoriodes and S. gregaria adults contained three shorter odd-numbered alcohols: C_15:0 (6.3 and 18.8%, respectively), C_17:0 (5.0 and 4.4%, respectively) and C_19:0 (0.8 and 0.6%, respectively) (Ohara and Lockey, 1990).

In the present study, the alcohols C_12:0 to C_30:0 were identified. However, the alcohols C_31:0 to C_34:0 are also present in most insects. Alcohols with molecular weight >C30 occurred in cuticular lipids of B. argentifolii nymphs (Buckner et al., 1999), H. virescens and H. zea pupae (Buckner et al., 1996), A. singularis exuviae and adults (Nelson et al., 1998), M. sanguinipes and M. packardii adults (Soliday et al., 1974), B. tabaci adults, T. vaporariorum adult (Buckner et al., 1994), and L. m. migratoriodes and S. gregaria adults (Oraha and Lockey, 1990). In B. argentifolii nymphs, alcohols with a molecular weight >C30 are the major alcohols. The relative contents of C_32:0 and C_34:0 alcohols were 36.9 and 36.1%, respectively. In contrast, the cuticular lipids of H. virescens pupae and M. sanguinipes and M. packardii adults contained alcohols with molecular weight >C30 in trace amounts.

Alcohols from C_22:0 to C_34:0 are typically found in many insect species, but the presence of alcohols with molecular weight <C22 are seldom met. Alcohols with molecular weight <C22 are present in cuticular lipids of L. m. migratoriodes adults, S. gregaria adults (Oraha and Lockey, 1990) and A. pisum (Brey et al., 1985). Among alcohols with low molecular weight, the following were identified: C_8:0 to C_16:0 in A. pisum (Brey et al., 1985), C_12:0 to C_21:0 in L. migratoria migratoriodes and S. gregaria adults (Ohara and Lockey, 1990), and C_12:0 to C_20:0 in larvae, pupae, males and females of M. domestica (present study).

Most insects contain a small number of alcohols. For example, in the cuticular lipids of Tenodera sinensis (females), T. angustipennis (males) and Stagmomantis carolina (females), only one alcohol (C_30:0) was identified (Jones et al., 1997). Two alcohols were present in the cuticular lipids of B. tabaci adults (C_30:0 and C_32:0) and T. vaporariorum adults (C_30:0 and C_32:0) (Buckner et al., 1994). And the following insect species contained from four to nine alcohols: B. argentifolii nymphs, H. virescens pupae, H. zea pupae, A. singularis exuviae and adults, M. sanguinipes adults, M. packardii adults and A. pisum (Brey et al., 1985). However, greater numbers of alcohols were identified in the lipids of L. m. migratoriodes adults (19 alcohols) and S. gregaria adults (19 alcohols) (Ohara and Lockey, 1990). In our work, eight, 10, eight and nine alcohols were present in the cuticular lipids of larvae, pupae, males and females, respectively.

The antimicrobial activity of components of cuticular lipids isolated from insects has been previously reported. It was found that cuticular fatty amides may contribute to Liposcelis bostrychophila's tolerance to entomopathogenic fungi (Lord and Howard, 2004). Caprylic acid, found on the surface of Heliothis zea and Spodoptera frugiperda larvae, possesses the ability to inhibit the germination of B. bassiana. Other free fatty acids present on the surface of H. zea and S. frugiperda also present mycostatic activity towards fungi (Smith and Grula, 1982). The current state of knowledge on diverse, species-specific effects of free fatty acids on the growth and virulence of insecticidal fungi is reviewed by Boguś et al. (Boguś et al., 2010).

Mixtures of alcohols present in cuticular lipids of M. domestica turned out to be effective against tested microbial strains. The gram-positive bacteria, R. equi, was the most susceptible to cuticular lipids found in larvae. This extract contained a high amount of C_14:0, which was the most active individual alcohol against all tested gram-positive bacteria. The least active compound against R. equi was the extract isolated from males, where C_14:0 was present only in trace amounts. The compound 1-tetradecanol seems to play an important role in the effectiveness against gram-positive bacteria. However, the activity of alcohols in females was slightly stronger in comparison to that in pupal extract, despite the lower amount of C_14:0. Females' alcohols contained considerably higher amounts of alcohols C_18:0 to C_24:0, and perhaps some synergistic action of these weak antimicrobials is responsible for the stronger activity. According to the literature, linoleic and oleic fatty acids inhibit the growth of gram-positive bacteria and fungi, whereas gram-negative bacteria are resistant to these compounds (Dilika et al., 2000). Similarly, alcohols didn't show any activity against gram-negative bacteria. Antifungal activity was presented by all tested compounds and extracts. In most cases, the activity of mixtures was higher in comparison to that of individual alcohols, which suggests some synergy in action. A synergistic effect between oleic and linoleic acid was confirmed for Micrococcus kristinae and Staphylococcus aureus (Dilika et al., 2000). Our results suggest synergistic action between components of cuticular lipids against C. lipolytica. The MICs obtained for females, males and larvae alcohols were lower than any MIC for individual compounds. A different situation was observed for C. albicans: the activity of mixtures was equal or even slightly lower compared with alcohols. The activity against C. tropicalis was determined mainly by the content of C_12:0 and/or C_24:0, which were the most potent alcohols against this fungus. Correspondingly, the activity against A. niger of alcohols isolated from larvae was greater than that from males and pupae extract. Alcohols extracted from females presented activity comparable with larval mixtures. However, female extracts did not contain high amounts of either C_12:0 or C_24:0. In this case, C_18:0–C_22:0 seem to play crucial role in antimicrobial effectiveness as they were the most abundant alcohols.

Primary aliphatic alcohols from C₆ to C₁₃ tested for their antifungal activity against Saccharomyces cerevisiae showed high fungicidal potential of undecanol and decanol (Kubo et al., 2003). The antifungal action of amphipathic medium-chain alcohols is mediated by biophysical processes and comes mainly from their ability as nonionic surfactants to disrupt the native membrane associated function of the integral proteins. This inhibition pattern in yeast is not specific to only alkanols, but also applies to alkenals and fatty acids (Kubo et al., 2003). However, it should be pointed out that none of the reference bacteria strains and medically important fungi we have tested here are pathogenic for M. domestica. A variety of microorganisms, including S. aureus, Pseudomonas sp., Proteus sp., Escherichia sp., Klebsiella sp., A. niger and A. flavus, which are highly pathogenic for humans, have been previously isolated from healthy house flies (Omalu et al., 2011; Banjo et al., 2005; Nazni et al., 2005). It seems that antimicrobial properties of alcohols and other cuticular compounds, together with the innate immunity mechanisms, efficiently protect M. domestica from the infection by common pathogens inhabiting similar habitats as those of the house fly. In contrast, the same compounds fail to protect adult flies against the entomopathogenic fungus C. coronatus ‘equipped’ with effective weaponry composed of proteolytic, chitinolytic and lipolytic enzymes that degrade the insect cuticle, as well as toxic metabolites that disable the victim's immune system (Wieloch et al., 2011). Similarly, Kontoyiannis and Lewis (Kontoyiannis and Lewis, 2010) reported high mortality of fruit flies infected with other Zygomycetes representatives – Rhizopus, Mucor and Cunninghamella species – contrasting with the resistance of Drosophila to Aspergillus and Candida species. In studying Drosophila pathogenesis caused by various fungal species, the authors learned that an intact Toll signaling pathway, the crucial element of innate immunity, is not sufficient to protect fruit flies against Zygomycetes. This finding contrasts with Aspergillus and Candida infections and shows complexity of insect victim–fungal pathogen relationships.

More experiments are needed to decipher the reason for the high susceptibility of adult flies versus the total resistance of larvae and pupae to the C. coronatus infection. Comparing developmental fluctuations in the concentration of cuticular alcohols, the contribution of these compounds seems doubtful. Assuming that adult flies are licking all substrates, it seems probable that during 18 h of exposition to fungal colonies, insects ingest fungal spores, which quickly germinate inside the gut. Growing hyphae release toxic metabolites (Samborski, 2000; Boguś and Scheller, 2002; Wieloch et al., 2011), which may promptly kill the flies. Whether larvae burrowing in the fungal colony during the exposure ingest it as well remains unknown. In contrast, fungus can invade pupae only via the cuticle.

The role of cuticular fatty acids in the resistance of insects to fungal infection has been widely discussed. According to our results, the composition of cuticular alcohols seems to be no less important. It would be worthwhile to continue the research and focus on the synergistic action between components of cuticular lipids against various microbial strains including multidrug-resistant bacteria.

We would like to express our gratitude to Katarzyna Kowalewska for her assistance.