OUP user menu

Characterization of the AeaHP Gene and its Expression in the Mosquito Aedes aegypti (Diptera: Culicidae)

Travis H. Stracker, Stacie Thompson, Genelle L. Grossman, Michael A. Riehle, Mark R. Brown
DOI: http://dx.doi.org/10.1603/0022-2585-39.2.331 331-342 First published online: 1 March 2002


The sequence and tissue expression of the gene encoding a peptide hormone Aea-HP-I, known to inhibit host-seeking behavior, has been characterized for the yellowfever mosquito, Aedes aegypti (L.). The open reading frame reveals a prepropeptide that would be processed into three identical peptides. The gene contains four short introns and exists as a single genomic copy. Transcripts of the gene were present in the brain, terminal ganglion, and midgut of adults, and in females, its expression profile differed for each tissue before and during a reproductive cycle. Peptides resulting from this expression were identified in the female tissues by immunoassays. Numerous neurosecretory cells and neurons in the nervous system were immunostained by an Aea-HP-I antiserum. Hundreds of endocrine cells were stained similarly in the midgut, thus contributing to the 10 times greater amount of immunoreactive peptide in an abdomen than in a head, as determined with an Aea-HP-I radioimmunoassay. Based on these results, neurosecretory cells and midgut endocrine cells are likely sources of Aea-HPs shown to reach highest hemolymph titer at the same time as host seeking is inhibited in female Ae. aegypti during a reproductive cycle.

  • Aedes aegypti
  • behavior
  • peptide hormone
  • nervous system
  • midgut endocrine system

Female mosquitoes exhibit behavioral responses to light, temperature, and chemical emanations that direct them to take a blood meal from vertebrate hosts (Klowden 1995). Ingestion of a blood meal triggers egg maturation in the yellowfever mosquito, Aedes aegypti (L.), and thereafter, females show a marked decrease in responses to host emanations that initially stimulated host seeking behavior. This inhibition of host seeking is known to have two phases (Klowden and Lea 1979). The first phase is stimulated by distention of the abdomen by the blood meal and continues for ≈30 h until the meal is digested. The second phase is thought to be regulated by a neuropeptide, Ae. aegypti head peptide I (Aea-HP-I; pERPhPSLKTRF-NH2) (Brown et al. 1994). This peptide was isolated from an extract of female Ae. aegypti heads based on its immunoreactivity in an FMRF-NH2 radioimmunoassay (Matsumoto et al. 1989). Another peptide, which is not hydroxylated at Pro4, Aea-HP-III, and truncated forms were identified in the same head extract (Matsumoto et al. 1989) and an extract of adult Ae. aegypti abdomens (Veenstra 1999). With an Aea-HP-I radioimmunoassay, the hemolymph titer of Aea-HP-I/III (Aea-HPs) was found to peak at 36 h after blood meal and decline to before blood meal levels by 48 h after a blood meal (Brown et al. 1994). This period coincides with the absence of host seeking by females receiving only a small blood meal (Klowden and Lea 1979). A functional basis for this coincidence was demonstrated by the failure of sugar-fed females injected with synthetic Aea-HP-I to respond to host cues, whereas females injected with Aea-HP-III and analogues did respond to the cues (Brown et al. 1994).

Because host seeking is a fundamental aspect of mosquito and vertebrate host interactions, our studies focused on the identification of the gene encoding Aea-HPs to gain a better understanding of the expression of this gene in female Ae. aegypti. Expression of the gene was examined at two levels. First, transcript presence in the nervous system and midgut was determined for females before and after a blood meal. Second, presence of Aea-HPs in female tissues was demonstrated by quantification of the peptides in different body regions of females with a specific radioimmunoassay (RIA) and by immunocytochemistry. Larvae and males also were found to express the Aea-HP gene, thus suggesting that these peptides, not only regulate host seeking in blood-fed female Ae. aegypti, but also other stage and sex-specific processes.

Materials and Methods


Ae. aegypti larvae were fed a mixture of brewer’s yeast, lactalbumin, and finely ground rat chow (1:1:1). Adults were maintained at 27°C, a photoperiod of 16:8 (L:D) h, and had access to a 10% sucrose solution for the first 2 d after eclosion, and thereafter only water. Only 3- to 5-d-old adults were used for the described procedures, and for blood feeding, females were given access to anesthetized rats.

Amplification of Nucleotide Sequences Encoding Aea-HPs.

To obtain cDNA sequences encoding Aea-HPs, two amplification steps by polymerase chain reaction (PCR) were required (Fig. 1A). Products were sequenced at the Molecular Genetics and Instrumentation Facility (University of Georgia), and Integrated DNA Technologies (Coralville, IA) synthesized the primers. For the first step, a degenerate nucleotide primer for Aea-HP-I (AHP 5; 5′-CAR CGN CCN CCI TCN YTI AAR ACN MGI TTY-3′; I = inosine) and a T7 primer (5′-TAATACGACTCACTATA-3′) were used in PCR (40 cycles in an Idaho Technologies Rapidcycler: 94°C for 10 s; 45°C, 15 s; 72°C, 10 s; primers (5 ng/μl) and Taq polymerase (2.5 U, Promega, Madison, WI) in 50 mM TRIS buffer with 250 μg/ml BSA, 4% creosol red, 0.2 mM dNTPs, and 4 mM MgCl2; 20 μl total volume) to amplify products from a λZAPII cDNA library (Stratagene, La Jolla, CA) constructed from mosquito heads (Graf et al. 1997). After the reaction mixture was separated on an agarose gel, products of the expected size were purified (Sephaglas Kit, Amersham Pharmacia Biotech, Piscataway, NJ), cloned into a pCR2.1 cloning vector (Invitrogen, Carlsbad, CA), and sequenced. A 211 bp product was identified that encoded an Aea-HP. For the second step (Fig. 1A), a specific primer for the 211 bp product (HP51R2; 5′-CAAACTGCATCATCCAGGTCCG-3′) and the T3 primer were used to amplify products with PCR (same conditions as above) from the head cDNA library. This reaction yielded a 433 bp product with additional 5′ sequence including the start codon, putative signal peptide, and two more encoded Aea-HPs.

Fig. 1

Identification of an AeaHP cDNA. (A) Anchored PCR strategy: AeaHP cDNA is shown flanked by the T7 and T3 promoters (striped boxes) of the pBluescript cloning vector, and Aea-HP-encoding repeats are indicated by stippled areas. Products from the PCR steps were obtained with anchored primers to the T7 and T3 promoters and specific forward and reverse primers. (B) Nucleotide sequence of AeaHP cDNA (GenBank AF155738) with the predicted amino acid sequence of the prepropeptide: putative signal peptide in bold capitals, Aea-HP sequences underlined, proteolytic processing sites boxed, and polyadenylation consensus sequences in bold lowercase and underlined.

For another approach, genomic DNA template was prepared from mosquitoes (Collins et al. 1987), and a 619 bp product was amplified from this template by PCR (94°C for 10 s; 55°C, 15 s; 72°C, 10 s; 30 cycles) with the primers, HP51F2 (5′-CAT GGG CCG TCT ATT GCG AG-3′) and HP51R2. It was found to encode the Aea-HPs but was interrupted by introns (Fig. 2A).

Fig. 2

Structure of the AeaHP gene. (A) Genomic DNA sequence of the AeaHP gene aligned with the AeaHP cDNA sequence—intron sequences underlined and coding sequences for Aea-HPs in bold. Primer sequences used to obtain the genomic DNA sequence by PCR amplification are boxed. (B) Five exons in the AeaHP gene are separated by four introns of varying length. Numbering begins with the start codon and ends with the stop codon. Regions coding for Aea-HPs are hatched. Untranslated sequence is indicated at the 3′ end followed by the first putative polyadenylation signal.

cDNA Library Screening.

A 278 bp product was amplified as above from the head cDNA library with a specific reverse primer for the first PCR product (5′-CCG TCA CTT TCT GGA TTG GC-3′) and a T3 primer (5′-AAT TAA CCC TCA CTA AAG GG-3′) and labeled with digoxigenin (DIG; Genius 2 DNA labeling kit, Roche, Indianapolis, IN). This probe was used to screen the head cDNA library (≈50,000 pfu/150 by 15-mm plate, after overnight growth, 37°C). Duplicate lifts were taken from each plate with nylon membranes (Magnacharge, Micron Separations, Westboro, MA). The DNA on the membranes was denatured (1.5 M NaCl and 0.5 M NaOH for 2 min), neutralized (twice for 2 min; 1.5 M NaCl and 0.5 M TRIS, pH 8.0), and UV crosslinked (twice at 0.2 Joules/cm2). The membranes were then treated with a prehybridization solution (5 × SSC, 0.1% N-lauroylsarcosine, 0.02% SDS, and 1% blocking reagent (Roche), overnight at 65°C), followed by the probe (2 ng/μl in the prehybridization solution; overnight at 65°C). After hybridization, membranes were washed (3 × 20 min in 0.5 × SSC with 0.1% SDS at 65°C), blocked (1 × blocking reagent, 1 h), incubated with anti-DIG antibody-alkaline phosphate (Roche; 1:20,000 in 1 × blocking reagent) for 30 min, washed (3 × 20 min in 0.1 M maleic acid buffer with 0.15 M NaCl and 0.3% Tween 20, pH 7.5), treated with chemiluminescent substrate (CDP-star at 1:200 dilution; Tropix, Foster City, CA), and exposed to XLS film (Kodak, Rochester, NY). Positive plaques were subjected to secondary and tertiary screens to isolate DNA from individual positive phage clones by standard methods (Sambrook et al. 1989).

Genomic DNA Blot.

Genomic DNA (10 μg) prepared as above was digested with EcoRI and separated on a 0.8% agarose gel, after which it was denatured (2 × 15 min in 1.5 M NaCl and 0.5 M NaOH), and neutralized (2 × 15 min in 1.5 M NaCl and 0.5 M TRIS, pH 8.0). The DNA was transferred to a nylon membrane (Magnacharge) overnight and UV crosslinked, as above. The membrane was then treated with a prehybridization solution (50 mM sodium phosphate buffer (pH 7.0), containing 7% SDS, 50% formamide, 5 × SSC, 2% blocking reagent, and 0.1% N-lauroylsarcosine; overnight at 45°C) followed by the DIG labeled 619 bp genomic DNA in the prehybridization solution (overnight at 45°C). Bands on the membrane to which the probe hybridized were detected as above.

Reverse Transcription-PCR and Southern Blots.

Before mRNA isolation, females were blood-fed on rats, and after ≈20 min, engorged females were divided into groups of 20. At specific times after a blood meal, heads, midguts, and the last two abdominal sections were dissected, pooled separately in RNAlater (20 tissues/50 μl; Ambion, Austin, TX), and stored at 4°C. Females had deposited eggs by 72 h after a blood meal. Tissues from fourth instars and 3- to 5-d-old males and females (nonblood-fed) were dissected and treated similarly.

Messenger RNA was isolated from tissue samples with the mRNA DIRECT Micro Kit (DYNAL, Lake Success, NY). After removal of the RNAlater solution, tissues were homogenized in 250 μl of lyses/binding buffer (100 mM TRIS-HCl [pH 7.5] containing 500 mM LiCl, 10 mM EDTA p[pH 8.0], 1% LiDS, and 5 mM DTT) and centrifuged for 2 min. Supernatant solutions were transferred to tubes containing 10 μl of prewashed Dynabeads Oligo (dT) and gently rocked for 3 to 4 min at room temperature. Tubes containing this solution were placed in a magnetic stand for 2 min to pellet the beads, and after the grinding/grinding buffer was removed, the beads were washed twice with 10 mM TRIS (pH 7.5) containing 0.15 M LiCl, 1 mM EDTA, and 0.1% LiDS and once with the same solution lacking LiDS. Beads were resuspended in 8 μl of ice cold 10 mM TRIS (pH 7.5) and held at 65°C for 8 min to free the bound mRNA. After the beads were pelleted, mRNA solutions were transferred to new tubes and incubated with an enzyme to remove DNA (DNA-free kit; Ambion, Austin, TX). After centrifugation, supernatant solutions with mRNA were transferred to a new tube.

Tissue mRNA (8 μl) was transcribed (1 h, 37°C) into cDNA with M-MLV reverse transcriptase using the NotI-d(T) 18 primer (First-Strand cDNA Synthesis Kit; Amersham Pharmacia) and stored at 4°C. Integrity of tissue cDNA (1.0 μl) was evaluated by amplification of a 550 bp nucleotide product encoding a portion of the actin gene in Ae. aegypti (forward primer, 5′ GCG ATC CCG ACT ACC TGA TG 3′, and reverse primer, 5′ CCA GAT TCA TCG TAC TCC TGC 3′) with PCR (35 cycles: 94°C for 30 s; 50°C, 30 s; 72°C, 1 min; Eppendorf gradient cycler). Only those tissue cDNA samples yielding an actin PCR product were used to amplify (40 cycles, same as above) a 337 bp product encoding Aea-HPs (forward primer, 5′ CAT GGG CCG TCT ATT GCG AG 3′ and reverse primer, 5′ CAA ACT GCA TCA TCC AGG TCC G 3′). Products were separated on a 1.5% agarose gel containing ethidium bromide and visualized. Reactions with both primers and no tissue cDNA, as a negative control, and both primers and Ae. aegypti genomic DNA, as a positive control (larger product with introns), were included on the gel.

After gel electrophoresis, PCR products were denatured (0.5 M NaOH and 1.5 M NaCl for 45 min), neutralized (2 × 15 min in 1.5 M NaCl and 0.5 M TRIS, pH 8.0), capillary transferred (10 × SSC) overnight to nylon membranes (Nytran SuperCharge; Schleicher & Schuell, Keene, NH), and UV cross-linked for 30 s. Membranes were prehybridized as above at 50°C for 2–4 h. To detect AeaHP products, a biotinylated degenerate nucleotide probe (5′ CAR MGN CCW CCW TCN YTI AAR ACA CGI TTY GG 3′) was added to the buffer (30 ng/ml, 10 ml total volume) for hybridization to membranes at 50°C overnight. A specific biotinylated actin probe (5′ GAT TTC CTT CTG CAT ACG ATC AGC AAT ACC 3′) was used on membranes with actin PCR products. Thereafter, membranes were washed in low stringency buffer (2 × 15 min in 2 × SSC and 0.1% SDS) and high stringency buffer (2 × 15 min in 0.5% SSC and 0.1% SDS, 50°C), blocked (SuperBlock, Pierce, Rockford, IL, at room temperature, 1 h), and treated with streptavidin-horseradish peroxidase (1:1200; Pierce) for 15 min. After washing (6 × 15 min at room temperature in 0.025 M TRIS (pH 7.5), 0.15 M NaCl, and 1% Tween 20), membranes were covered with reagents from a Renaissance Chemiluminescence Kit (NEN, Boston, MA) and exposed to BioMax LS film for 5 to 30 min. These procedures were repeated at least four times using tissues from different cohorts of larvae, males, and females before and at different times after a blood meal.

Peptide Extraction.

Body parts of nonblood fed females were separated for peptide extraction (Matsumoto et al. 1989). One gram each of heads (≈11,000) thoraces (≈1,300), and joined abdomens and thoraces (≈1,000) were homogenized in 12 ml of 0.2 M acetic acid solution containing 0.1% thiodiglycol, 1 mM EDTA, and 0.2 mM phenylmethylsulfonyl fluoride. After centrifugation (17,200 × g, 20 min) and removal of supernatant solutions, pellets were reextracted (3 ml) by sonication (30 s), and centrifuged. For each body part sample, supernatant solutions were combined for lipid extraction (3 × 3 ml of hexane), and the aqueous portion lyophilized. The samples were rehydrated (5.0 ml of aqueous 0.1% TFA solution, 1 h on ice), centrifuged, and subjected to solid-phase extraction (SPE) by loading onto C18 cartridges (3 ml, Analytichem, Harbor City, CA), preconditioned with acetonitrile (CH3CN) followed by the TFA solution. Each cartridge was eluted sequentially with 4.0 ml each of TFA solution alone and then 10, 40, and 80% CH3CN/TFA solutions, which were lyophilized.

Aea-HP-I Radioimmunoassay.

The SPE eluants of the body parts were rehydrated with distilled water and then diluted with 0.05 M Tris-buffer (pH 7.2) containing 0.1% bovine serum albumin and 0.02% sodium azide for the Aea-HP-I radioimmunoassay (RIA; Brown et al. 1994). Serial dilutions in triplicate of the body parts were incubated overnight (4°C) with Aea-HP-I antiserum (71D; 1/100 K-150 K final dilution in 1.0 ml; bound/free ratio = 1.0 in the absence of unlabeled peptide) and 125I-labeled Aea-HP-I (3,000-3,500 cpm). Bound and free-labeled peptides in the sample tubes were separated with a charcoal-dextran-serum mixture, and the pellets counted. For each RIA, a standard curve was plotted from the bound/free ratios and log values of synthetic peptide and the body part dilutions. The amount of immunoreactive peptide in body part extracts was calculated from a regression equation for the linear portion of the Aea-HP-I standard curve.


Whole mosquitoes were fixed and embedded in Epon/Araldite plastic before sectioning (Brown and Lea 1988). To observe co-immunostaining of cells with the Aea-HP-I antiserum (71D) (Brown et al. 1994) and FMRFamide antiserum (lot #155, donated by G. J. Dockray, Physiological Laboratory, University of Liverpool, UK), adjacent serial sections (3–4 μm thick) of embedded bodies were mounted on two or three different slides in sets (2–3 sets/female). A slide from each set was treated with an antiserum (1/500-1,000 dilution), preimmune serum, or an antiserum preabsorbed with synthetic Aea-HP-I, YGGFMRF-NH2, or pERPPSLKTRC (Aea-HP-Cys10; Peninsula Laboratories, Belmont, CA; 5–10 μg peptide/ml diluted antiserum, 24 h, 4°C). Afterward, sections were subjected to avidin-biotin immunocytochemistry (Vectastain Elite ABC Kit, Vector Laboratories, Burlingame, CA). Sections from three or more females were similarly treated and observed for each antiserum or preabsorbed antiserum.


Identification of an AeaHP cDNA.

Two steps of anchored PCR using degenerate and specific primers and an Ae. aegypti head cDNA library as template yielded products with overlapping sequences encoding Aea-HPs (Fig. 1A). For the first step, a degenerate primer to the amino acid sequence of Aea-HPs was used to amplify a product to the 3′ end of a putative cDNA. The nucleotide sequence of this product encoded an Aea-HP followed by Gly-Arg, which are required for proteolytic cleavage and amidation. Second, a specific primer was used to amplify a 433 bp product that began with a start codon at the 5′ end and encoded a putative signal peptide and three Aea-HPs. In addition, the head cDNA library was screened for positive clones, but they varied in length and lacked the start codon.

Sequences from the PCR products and library clones were compiled into a nucleotide sequence for the "Ae. aegypti Head Peptide cDNA" (AeaHP cDNA, Fig. 1B; GenBank AF155738). The open reading frame (ORF) encodes a prepropeptide of 128 amino acids with a signal peptide of 22 residues and three repeats of QRPPSLKTRFG, which after posttranslational processing, would yield Aea-HP-I or III. The signal peptide was identified with the SignalP program (Nielsen et al. 1997). The first Aea-HP begins immediately after the signal peptide. The second and third copies are preceded by a sequence of MEKRSA, and all end with Arg-Ser.

Genomic DNA Sequence and Gene Number.

To further characterize the gene for Aea-HPs, a genomic DNA sequence encoding the Aea-HPs was obtained, and its copy number determined by analysis of a Southern blot of genomic DNA. With specific primers, a 619 bp product was amplified by PCR from genomic DNA, and the above ORF was found to contain four introns with no significant sequence homology (Fig. 2 A and B). The introns begin and end with the nucleotides GT and AG, respectively as expected, and range in size from 57 to 69 bp (Fig. 2 A and B). The first three introns occur between the fourth and fifth nucleotides encoding each of the Aea-HP repeats. The fourth intron is in the 3′ region of the ORF after the last Aea-HP repeat. The existence of a single gene encoding Aea-HPs was indicated by hybridization of a labeled genomic DNA PCR product to a single 9 kb band of EcoRI-digested genomic DNA and a single 7 kb band of ApaI-digested genomic DNA (Fig. 3).

Fig. 3

Southern blot of Ae. aegypti genomic DNA digested with EcoRI or ApaI after hybridization to a digoxygenin-labeled AeaHP probe. Only a single 9 kb band in the EcoRI lane and a 7 kb band in the ApaI lane are positive, thus indicating that the AeaHP gene exists as a single copy.

RT-PCR and Detection of AeaHP mRNA.

The qualitative assessment of AeaHP expression in the brain (heads), midgut, and terminal ganglion (last two abdominal segments) of females during a reproductive cycle was based on the results from Southern blots of RT-PCR products. The same tissues from males and larva heads were included to determine the ubiquity of gene expression. As shown with representative Southern blots (Fig. 4), the AeaHP product (370 bp) was amplified and detected as a single positive band in lanes of separated PCR reactions from adult tissues and larva heads. Because detection of this band varied among trials using different cohorts of larvae and adults, AeaHP mRNA (transcript) was judged to be present (+, Fig. 4) in a tissue only when the band was detected in half or more of the trials given beneath each lane in Fig. 4.

Fig. 4

Representative Southern blots of RT-PCR products amplified with specific AeaHP primers from mRNA of brains (heads), midgut, and terminal ganglion (last two abdominal segments) of fourth instars, males, and females before and at different times after a blood meal. In the lanes for each tissue blot, the AeaHP product is indicated by the presence of single positive band of 370 bp. Below each lane, detection of AeaHP product is summarized for the trials using tissues from different cohorts of larvae, males, and females. AeaHP mRNA was judged to be present, +, in a tissue only when the band is detected in half or more of the trials. A larger product (≈600 bp) was evident after PCR amplification with genomic DNA (gDNA; first lane in the Terminal Ganglion panel). All experiments included a no template (NT) lane for RT-PCR with primers but no template.

Companion blots for the detection of the actin product were made for all trials with the tissues of larvae and adults, and in all instances, actin expression was evident (blots not shown). These results indicated not only that the expression of the actin gene in tissues of larvae and adults was constitutive, but more importantly, that the absence of AeaHP transcript observed in female tissues at times after the blood meal was not due to degradation or lack of tissue mRNA. Of note, a larger 600 bp product was detected after amplification by PCR with AeaHP primers and genomic DNA (gDNA; Fig. 4, first lane in the "terminal ganglion" panel). This product includes the introns described above and was not amplified in the tissue cDNA lanes. In addition, all experiments included a lane for "no template" (NT) RT-PCR mixtures with the primers, and no AeaHP positive band was detected in these lanes (Fig. 3), thus showing the absence of contaminating template.

The presence of AeaHP mRNA was demonstrated in the brain, midgut, and terminal ganglion of both sexes and the larva brain (Fig. 4). More specifically, these transcripts are presumed to occur in specific neurons or neurosecretory cells in the brain or terminal ganglion and endocrine cells in the midgut. For females, each of the tissues had a different temporal profile of AeaHP expression before and during a reproductive cycle (Fig. 4), which is initiated by a blood meal and completed 48 to 72 h after a blood meal with egg deposition. Before a blood meal, transcripts were present in the brain and terminal ganglion of females, but not in the midgut. After a blood meal, transcripts were present in brains at 2 h after a blood meal, in the intervening period from 12 to 24 h after a blood meal, and at 48 h after a blood meal, whereas in the midgut, they were evident from 2 to 24 h after a blood meal. Notably, transcripts were detected most consistently in the terminal ganglion throughout the reproductive cycle, with the exception of the 12 h after a blood meal time point.

Quantification of Aea-HPs in Female Body Parts.

Extracts of body parts from nonblood fed females were enriched for peptides by solid-phase extraction (SPE) before quantifying Aea-HP content with the Aea-HP-I RIA. Because the 40% CH3CN step eluant from the SPE of the body part extracts contained >90% of the immunoreactive peptide, only these eluants were diluted serially for the RIA. The regression line for Aea-HP-I standards (Fig. 5A) appeared parallel to those obtained for dilutions of the body part extractions within their respective ranges: 0.05-0.15 heads, 5–20 thoraces, and 0.005-0.0125 thoraces-abdomens (Fig. 5B). This parallelism indicates that the same antigenic Aea-HPs were recognized in the RIA of the body part extracts, thus the Aea-HP content/body part was estimated to be 512 fmol/head, 5 fmol/thorax, and 5,577 fmol/thorax-abdomen. Because a thorax contained a thousand times less peptide than a thorax-abdomen, the results presented in Fig. 5B for "abdomens" are values obtained from dilutions of the thorax-abdomen extract.

Fig. 5

Parallel displacement of radiolabeled Tyr0-Aea-HP-I by (A) unlabelled Aea-HP-I, 10–150 fmol, and (B) dilutions of female body parts, as indicated by regression lines.


The Aea-HP-I antiserum was used to determine the tissue distribution and number of cells in nonblood fed females containing the endogenous peptide and to compare these results to those obtained with a heterologous antiserum to FMRF-NH2. The same tissue preparation and immunocytochemical techniques were used as in a previous study of FMRF-NH2-like immunostaining in nonblood fed females (Brown and Lea 1988), but this time, adjacent sections of tissues were treated separately with each of the antisera to determine whether the same or different cells were immunostained. As recorded from several sectioned females treated with the Aea-HP-I antiserum, ≈120 neurons and neurosecretory cells in the brain (Fig. 6A), optic lobes, and subesophageal ganglion were stained in the head. In the thorax, 7–13 cells in the fused ganglia, the ventricular ganglia (Fig. 6B) on the cardia, and axons along the anterior midgut were stained. In the abdomen, a pair of cells in each ganglion, a large cell associated with the spermatheca, and several hundred endocrine cells in the posterior midgut (Fig. 6C) were stained. Detailed observations of adjacent sections treated with the antisera revealed that, in general, the same cells were immunostained with both antisera.

Fig. 6

Immunocytochemical localization of Aea-HPs in female tissues. (A) Immunostained lateral and medial neurosecretory cells (arrows) in the brain (dorsal half; anterior, top). CE, compound eyes; NP, neuropil (bar = 100 μm). (B) Ventricular ganglia (arrows) on each side of the cardiac valve (CV) at the junction of the foregut (FG) and midgut are immunostained (bar = 50 μm). Dark staining in the cells and lumen of the cardiac valve is nonspecific. (C) Numerous endocrine cells (arrows) are immunostained in the posterior midgut (bar = 100 μm).

To characterize the epitopes recognized predominantly by the Aea-HP-I antiserum, adjacent serial sections of tissues were treated with the antiserum or by antiserum preabsorbed with different peptides of interest. As shown on sections of a representative tissue (female brain and subesophageal ganglion), preabsorption of the Aea-HP-I antiserum with Aea-HP-I blocked immunostaining (Fig. 7 A versus B), whereas, preabsorption with Aea-HP-(Cys10), which lacks only the C-terminal Phe-NH2, did not block staining (Fig. 7C), thus showing that the antiserum binds the Phe-NH2 epitope. More importantly for specificity, preabsorption of the antiserum with YGGFMRF-NH2, which has only the Phe-NH2 in common with Aea-HP-I, did not block immunostaining (Fig. 7D) and demonstrates that other epitopes of Aea-HP-I were recognized by the antiserum. In contrast, FMRF-NH2 antiserum (Fig. 7E) recognized only the Phe-NH2 epitope, as shown by the absence of immunostaining when it was preabsorbed with Aea-HP-I (Fig. 7F). For negative controls, no immunostaining was observed in tissue sections treated with preimmune serum from the rabbit immunized with the Aea-HP-I antigen (not shown) and FMRF-NH2 antiserum preabsorbed with FMRF-NH2 (not shown).

Fig. 7

Immunocytochemistry with Aea-HP-I and FMRF-NH2 antisera alone and preabsorbed with related peptides (dorsal, top; anterior, left; bar = 100 μm). (A-D) Adjacent sagittal sections of brain treated with Aea-HP-I antiserum alone or preabsorbed with peptides. (A) Aea-HP-I antiserum alone, immunostained cells indicated by arrows. (B) Absence of immunostaining by antiserum preabsorbed with Aea-HP-I. (C) Antiserum preabsorbed with Aea-HP-[Cys10]—same cells immunostained (arrows) in A and D. (D) Antiserum preabsorbed with YGGFMRF-NH2, same cells immunostained (arrows) in A and C. (E, F) Adjacent sagittal sections of brain and subesophageal ganglion (dorsal, top; anterior, left): (E) treated with FMRF-NH2 antiserum—immunostained cells (arrows), and (F) same serum preabsorbed with Aea-HP-I—immunostaining not in cells (arrows) indicated in E.


Originally, Aea-HP-I and other related peptides were isolated from adult Ae. aegypti based on their immunoreactivity in an FMRF-NH2 RIA, and only later was one of the peptides, Aea-HP-I, shown to affect host seeking by females of this species (Brown et al. 1994). Now, the gene has been identified that encodes three Aea-HPs (Fig. 1B). In females, expression of the gene, as demonstrated by the presence of AeaHP mRNA, was shown to occur in the brain, terminal ganglion, and midgut. For each of these tissues, the temporal expression profile appeared to differ during a reproductive cycle. Not surprisingly, the gene was expressed in the same tissues in males and in larva brains. Results from immunocytochemistry and RIA using the Aea-HP-I antiserum support the evidence for AeaHP gene expression in these female tissues and further indicate that cells in these tissues secrete Aea-HP-I into the hemolymph, thus enabling its modulatory effect on female behavior.

The prepropeptide encoded by the AeaHP gene contains three repeats of the amino acid sequence, QRPPSLKTRFG, that are enclosed by protease cleavage sites (Fig. 1B). For Aea-HPs, the amino-terminal Gln presumably is converted to pyroglutamic acid by a glutamine peptide cyclase (Busby et al. 1987), after the signal peptide is cleaved from the first Aea-HP copy. For the other two copies, this conversion would follow proteolytic cleavage at the dibasic residues in the sequence, KRSA, by a prohormone convertase (Veenstra 2000) and removal of the remaining residues by an aminopeptidase. The Phe of all repeats is followed by Gly-Arg-Ser, and is amidated presumably after cleavage between the Arg and Ser by a convertase, removal of the Arg by a carboxypeptidase, and conversion of Gly into an amide by a mono-oxygenase and a lyase (Kolhekar et al. 1997). A carboxy-terminal peptide of 38 residues follows the processing site of the last Aea-HP, and its fate or function is unknown.

Further characterization of the AeaHP gene revealed introns and its copy number in the genome. Sequences of PCR products from genomic DNA contained introns of 57–69 bp following the fourth nucleotide in each region encoding an Aea-HP repeat, and a fourth intron was found in the C-terminal peptide (Fig. 2 A and B). These short introns must facilitate recognition and splicing after the mRNA is transcribed, and in Drosophila, such short introns are common (Kennedy and Berget 1997). The identical placement and similar size of the introns within the Aea-HP encoding repeats (Fig. 2 A and B) suggest that the peptide coding region was amplified in the AeaHP gene by unequal crossing over of the chromosomes. Southern blots showed that the labeled AeaHP DNA probe hybridized strongly to a single band of genomic DNA digested with ApaI or EcoRI, indicating that the Aea-HP-I gene occurs as a single copy in the Ae. aegypti genome.

Expression of the AeaHP gene was examined in females at two levels. First, tissues of females known to contain Aea-HP-like immunoreactivity, based on the immunocytochemical survey, were analyzed for the presence of AeaHP products obtained by RT-PCR on Southern blots. Second, the relative quantity of Aea-HPs and likely cell sources were determined for these same tissues with immunoassays based on an Aea-HP I antiserum, which specifically recognizes both Aea-HP-I and -III but not unrelated short peptides having a C-terminal Phe-NH2 in an RIA (Brown et al. 1994).

Detection of AeaHP products from RT-PCR on Southern blots was surprisingly variable for the tissues from different cohorts (trials) of females (Fig. 4). These products, however, were present in almost all trials made simultaneously with male tissues, thus indicating that the molecular techniques were consistently applied. Detection of actin RT-PCR products was required for all tissues in all trials to assure mRNA integrity. Amplification of AeaHP RT-PCR products from these tissues depends on the presence of AeaHP mRNA in specific peptidergic cells that can store or release Aea-HPs in response to specific stimuli. Although the presence of AeaHP mRNA generally indicates active translation and subsequent processing of the encoded peptides, it does not indicate directly whether the peptides are being stored or released by the cells. Presumably, storage of peptides would turn off transcription, and with the release of stored peptide, transcription would be initiated again. It is likely that the observed variability in the presence of AeaHP RT-PCR products in female tissues at a particular time is due to differences in the dynamics of peptide storage and release in pooled tissues compounded by the ability of PCR to amplify many thousands of products from few templates.

Given this variability, there were notable trends in the tissue-specific expression of the AeaHP gene in females before and during a reproductive cycle. The absence of AeaHP RT-PCR products in midguts of nonblood fed females (Fig. 4) may reflect storage of Aea-HPs in midgut endocrine cells, and with the release of Aea-HPs from such cells after a blood meal, transcription would be activated, as indicated by its detection at 2 h after a blood meal. Generally, AeaHP transcripts in brains appeared to have a 24 h periodicity after a blood meal, and in midguts, its presence from 2 to 24 h after a blood meal coincides with blood digestion by this tissue (Noriega and Wells 1999). The terminal ganglion showed the most consistent AeaHP expression throughout the reproductive cycle, including the period of 30 to 48 h after a blood meal when the Aea-HP hemolymph titer is at its highest and host seeking is inhibited (Brown et al. 1994). Because the terminal ganglion alone appears to express AeaHP transcripts during this period, neurosecretory cells in the ganglion, as described below, may be the principal source of circulating Aea-HPs during this time.

The Aea-HP-I RIA was used to quantify immunoreactive peptide content of the main body parts of a nonblood fed female. An abdomen contained approximately five times more immunoreactivity than a head by weight (29 fmol/μg of abdomen versus 6 fmol/μg of head), whereas a thorax contained an insignificant amount. This distribution of immunoreactive peptide in the body parts was consistent with the tissue localization and number of immunostained cells observed in sectioned nonblood fed females after immunocytochemistry with the Aea-HP-I antiserum. Approximately 120 cells in the brain (head), fewer than 20 cells in the thoracic ganglia, and a few hundred cells in the posterior midgut, along with ≈20 cells in the abdominal ganglia, (abdomen) were immunostained with this antiserum. Thus, the greater amount of Aea-HPs in an abdomen relative to the head is probably due to the considerably more numerous cells immunostained in the midgut than in the brain. Veenstra (1999) also noted the surprisingly large quantity of Aea-HPs isolated from extracted thoraces and abdomens of adult Ae. aegypti.

The overall pattern of Aea-HP-I immunostaining observed and recorded for female Ae. aegypti in this study is essentially the same as that reported for an FMRF-NH2 antiserum (Brown and Lea 1988). Because the Aea-HPs were isolated based on their immunoreactivity to an FMRF-NH2, the cross-reactivity to both antisera is thought to reflect the presence of Aea-HPs, although other types of peptides with the carboxyl terminus of Arg-Phe-NH2 may exist in mosquitoes. The distribution of Aea-HP-I immunostained cells, as well, supports that obtained for AeaHP gene expression and indicates likely sources of circulating Aea-HPs in females. These sources include the medial and lateral neurosecretory cells in the brain with axons to a neurohemal organ, the corpus cardiacum, and midgut endocrine cells. Specific cells in the terminal ganglion of females are another source and may release the peptides from axons extending along the hindgut to an axonal ring around the pyloric valve at the junction of the hindgut and midgut, as indicated by immunostaining with Arg-Phe-NH2 antisera (Veenstra et al. 1995). In addition, Aea-HPs may function as neurotransmitters or neuromodulators within the nervous system of females, as suggested by the immunostaining of neurons. In this mode, Aea-HP-I does not affect muscle contractions of oviducts and hindguts isolated from female Ae. aegypti that are responsive to other peptides and neurotransmitters (Messer and Brown 1995), but it may block peripheral sensory cues affecting host-seeking behavior.

The Aea-HPs belong to a growing family of "head peptides" in arthropods (Veenstra 1999), as shown by the sequence SL(R/K)(L/T)RF-NH2 common to peptides isolated from heads of the horseshoe crab, Limulus polyphemus (L.) (Gaus et al. 1993), midguts of the cockroach, Periplaneta americana (L.) (Veenstra and Lambrou 1995), and brains of the Colorado potato beetle, Leptinotarsa decemlineata (Say) (Spittaels et al. 1996). These peptides were purified based on their immunoreactivity to different antisera recognizing Arg-Phe-NH2, and except for Aea-HP-I, no function has been ascribed to these peptides. The evolutionary relationship of the arthropod "head peptides" is best ascertained from comparisons of the encoding genes or cDNAs, yet only one such gene from the fruitfly Drosophila melanogaster (Meigen) has been identified (Broeck 2000). This putative peptide gene (CG13968) only encodes two "head peptides," AQRSPSLRLRF-NH2 and WFGDVNQKPIRSPSLRLRF-NH2, with the requisite proteolytic processing sites, whereas the AeaHP gene encodes three such peptides, as described above. Characterization of the AeaHP gene and its expression in female Ae. aegypti will aid the identification of related peptides and their encoding genes in other insects and arthropods, thus supporting the existence of a distinct "head peptide" family. In time, these peptides may be shown to affect host-seeking behavior in other medically important insects and other arthropods, given the great extent to which structure and function of peptides are conserved across the phyla of higher animals.


We thank Joe W. Crim for preparing the radiolabeled peptide used in the RIA. This work was funded by a grant (AI33108) from the National Institutes of Health to M.R.B.


  • In conducting the research described in this report, the investigators adhered to the "Guide for the Care and Use of Laboratory Animals" as promulgated by the Committee on Care and Use of Laboratory Animals of the Institute of Laboratory Animal Resources, National Research Council. The facilities are fully accredited by the American Association of Laboratory Animal Care.

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