FS145, the first flea-derived disintegrin, inhibits angiogenesis through specifically binding integrin αvß3.

FS145, a protein containing a WGD motif, was previously described from the salivary transcriptome of the flea Xenopsylla cheopis. Nevertheless, its biological function and complete structure are still uncertain. Herein, FS145 was confirmed to adopt a common αßß structure with the WGD motif exposed on its surface and located right at the top of a loop composed of residues 72-81. Furthermore, FS145 dose-dependently inhibited the proliferation, adhesion, migration, and tube formation of HUVECs by not only binding to integrin αvß3 but also by subsequently inactivating the FAK/Src/MAPK pathway along with the reduction of the expression of MMP-2, MMP-9, VEGFA, bFGF, Ang2, Tie2, HIF-1α, and FAK. Moreover, FS145 also inhibited aortic vessel sprout and showed strong anti-angiogenic activities as assessed ex vivo, by employing the rat aortic ring assay, chick embryo chorioallantoic membrane, and zebrafish embryo models. Altogether, our results suggest that FS145 suppresses angiogenesis ex vivo and in vitro by blocking integrin αvß3. The current study reveals the first anti-angiogenesis disintegrin with WGD motif from invertebrates and provides a beneficial pharmacological activity to inhibit abnormal angiogenesis.

Different myrosinases activate sequestered glucosinolates in larvae and adults of the horseradish flea beetle.

ß-Glucosidases play an important role in the chemical defense of many insects by hydrolyzing and thereby activating glucosylated pro-toxins that are either synthesized de novo or sequestered from the insect's diet. The horseradish flea beetle, Phyllotreta armoraciae, sequesters pro-toxic glucosinolates from its brassicaceous host plants and possesses endogenous ß-thioglucosidase enzymes, known as myrosinases, for glucosinolate activation. Here, we identify three myrosinase genes in P. armoraciae (PaMyr) with distinct expression patterns during beetle ontogeny. By using RNA interference, we demonstrate that PaMyr1 is responsible for myrosinase activity in adults, whereas PaMyr2 is responsible for myrosinase activity in larvae. Compared to PaMyr1 and PaMyr2, PaMyr3 was only weakly expressed in our laboratory population, but may contribute to myrosinase activity in larvae. Silencing of PaMyr2 resulted in lower larval survival in a predation experiment and also reduced the breakdown of sequestered glucosinolates in uninjured larvae. This suggests that PaMyr2 is involved in both activated defense and the endogenous turnover of sequestered glucosinolates in P. armoraciae larvae. In activity assays with recombinant enzymes, PaMyr1 and PaMyr2 preferred different glucosinolates as substrates, which was consistent with the enzyme activities in crude protein extracts from adults and larvae, respectively. These differences were unexpected because larvae and adults sequester the same glucosinolates. Possible reasons for different myrosinase activities in Phyllotreta larvae and adults are discussed.

Yersinia pestis Δail Mutants Are Not Susceptible to Human Complement Bactericidal Activity in the Flea.

Ail confers serum resistance in humans and is a critical virulence factor of Y. pestis, the causative agent of plague. Here, the contribution of Ail for Y. pestis survival in the flea vector was examined. Rat or human but not mouse sera were bactericidal against a Y. pestis Δail mutant at 28°C in vitro. Complement components deposited rapidly on the Y. pestis surface as measured by immunofluorescent microscopy. Ail reduced the amount of active C3b on the Y. pestis surface. Human sera retained bactericidal activity against a Y. pestis Δail mutant in the presence of mouse sera. However, in the flea vector, the serum protective properties of Ail were not required. Flea colonization studies using murine sera and Y. pestis KIM6+ wild type, a Δail mutant, and the Δail/ail+ control showed no differences in bacterial prevalence or numbers during the early stage of flea colonization. Similarly, flea studies with human blood showed Ail was not required for serum resistance. Finally, a variant of Ail (AilF100V E108_S109insS) from a human serum-sensitive Y. pestis subsp. microtus bv. Caucasica 1146 conferred resistance to human complement when expressed in the Y. pestis KIM6+ Δail mutant. This indicated that Ail activity was somehow blocked, most likely by lipooligosaccharide, in this serum sensitive strain. IMPORTANCE This work contributes to our understanding of how highly virulent Y. pestis evolved from its innocuous enteric predecessor. Among identified virulence factors is the attachment invasion locus protein, Ail, that is required to protect Y. pestis from serum complement in all mammals tested except mice. Murine sera is not bactericidal. In this study, we asked, is bactericidal sera from humans active in Y. pestis colonized fleas? We found it was not. The importance of this observation is that it identifies a protective niche for the growth of serum sensitive and nonsensitive Y. pestis strains.

The Metagenomic Analysis of Potential Pathogenic Emerging Bacteria in Fleas.

<b>Background and Objective:</b> At present many pathogenic microbes that cause disease in humans are transmitted through animals. <i>Ctenocephalides felis</i>is specific ectoparasites in cats. Metagenomic research on the digestive tract and body surface of <i>C. felis</i>has been conducted. DNA genomics was extracted from the body surface and digestive tract of <i>C. felis</i>. <b>Materials and Methods:</b> Metagenomic analysis has used the 16S rRNA gene (V3-V4 region). Sequencing was carried out using New Generation Sequencing at the First BASE Laboratory, Singapore. <b>Results:</b> Wolbachia has the most significant bacterial composition in <i>C. felis</i> (94.4%), we were found bacteria with a composition >1% that have never been reported to be associated with <i>C. felis</i>. Also, there were 0.2% of bacteria whose taxonomic status cannot be determined. <b>Conclusion:</b> The results of this study become a vital reference pathogenic bacteria that can be transmitted to humans and animals through <i>C. felis</i>. It is necessary to study the resistance of bacteria isolated from<i>C. felis</i>to antibiotics in the future.

Oxidative stress responses in brackish water flea exposed to microcystin-LR and algal bloom waters from Nakdong River, Republic of Korea.

Microcystis blooms and the impact of their toxins, particularly microcystin (MC), in coastal ecosystems is an emerging threat, but the species-specific effects of MC and the potential for bioconcentration are not fully understood. We exposed the brackish water flea, Diaphanosoma celebensis, to MC-LR, which showed antioxidant responses measured at the molecular to enzyme levels but no acute toxicity. We extended our experimental investigation to measure the released MC and its uptake by D. celebensis exposed to river water. In a short-term exposure (48 h) experiment, D. celebensis exposed to water from an algal bloom (approximately 2 µg L-1 MC) assimilated more than 50 pg MC per individual. The significant increase of MCs suggests the potential for the species to accumulate MCs. The dose-dependent increase in the antioxidant response observed in the mRNA levels also showed that D. celebensis exposed to diluted algal bloom waters were affected by toxins from cyanobacteria.

Comparison of the transmission efficiency and plague progression dynamics associated with two mechanisms by which fleas transmit Yersinia pestis.

Yersinia pestis can be transmitted by fleas during the first week after an infectious blood meal, termed early-phase or mass transmission, and again after Y. pestis forms a cohesive biofilm in the flea foregut that blocks normal blood feeding. We compared the transmission efficiency and the progression of infection after transmission by Oropsylla montana fleas at both stages. Fleas were allowed to feed on mice three days after an infectious blood meal to evaluate early-phase transmission, or after they had developed complete proventricular blockage. Transmission was variable and rather inefficient by both modes, and the odds of early-phase transmission was positively associated with the number of infected fleas that fed. Disease progression in individual mice bitten by fleas infected with a bioluminescent strain of Y. pestis was tracked. An early prominent focus of infection at the intradermal flea bite site and dissemination to the draining lymph node(s) soon thereafter were common features, but unlike what has been observed in intradermal injection models, this did not invariably lead to further systemic spread and terminal disease. Several of these mice resolved the infection without progression to terminal sepsis and developed an immune response to Y. pestis, particularly those that received an intermediate number of early-phase flea bites. Furthermore, two distinct types of terminal disease were noted: the stereotypical rapid onset terminal disease within four days, or a prolonged onset preceded by an extended, fluctuating infection of the lymph nodes before eventual systemic dissemination. For both modes of transmission, bubonic plague rather than primary septicemic plague was the predominant disease outcome. The results will help to inform mathematical models of flea-borne plague dynamics used to predict the relative contribution of the two transmission modes to epizootic outbreaks that erupt periodically from the normal enzootic background state.

Structure and Function of FS50, a salivary protein from the flea Xenopsylla cheopis that blocks the sodium channel NaV1.5.

Naturally occurring toxins have been invaluable tools for the study of structural and functional relationships of voltage-gated sodium channels (VGSC). Few studies have been made of potential channel-modulating substances from blood-feeding arthropods. He we describe the characterization FS50, a salivary protein from the flea, Xenopsylla cheopis, that exhibits an inhibitory activity against the NaV1.5 channel with an IC50 of 1.58 µM. The pore-blocking mechanism of this toxin is evident from the kinetics of activation and inactivation suggesting that FS50 does not interfere with the voltage sensor of NaV1.5. FS50 exhibits high specificity for NaV1.5, since 10 µM FS50 had no discernable effect on voltage-gated Na+, K+ and Ca2+ channels in rat dorsal root ganglia or VGSC forms individually expressed in HEK 293T cells. Furthermore, intravenous injection of FS50 into rats and monkeys elicited recovery from arrhythmia induced by BaCl2, as would be expected from a blockade of NaV1.5. The crystal structure of FS50 revealed a ßαßß domain similar to that of scorpion ß toxin and a small N-terminal ßαß domain. Site-directed mutagenesis experiments have implicated a basic surface including the side chains of Arg 6, His 11 and Lys 32 as potentially important in the FS50 NaV1.5 interaction.

Gene expression analysis of Xenopsylla cheopis (Siphonaptera: Pulicidae) suggests a role for reactive oxygen species in response to Yersinia pestis infection.

Fleas are vectors for a number of pathogens including Yersinia pestis, yet factors that govern interactions between fleas and Y. pestis are not well understood. Examining gene expression changes in infected fleas could reveal pathways that affect Y. pestis survival in fleas and subsequent transmission. We used suppression subtractive hybridization to identify genes that are induced in Xenopsylla cheopis (Rothschild) (Siphonaptera: Pulicidae) in response to oral or hemocoel infection with Y. pestis. Overall, the transcriptional changes we detected were very limited. We identified several genes that are likely involved in the production or removal of reactive oxygen species (ROS). Midgut ROS levels were higher in infected fleas and antioxidant treatment before infection reduced ROS levels and resulted in higher bacterial loads. An ROS-sensitive mutant strain of Y. pestis lacking the OxyR transcriptional regulator showed reduced growth early after infection. Our results indicate that ROS may limit Y. pestis early colonization of fleas and that bacterial strategies to overcome ROS may enhance transmission.

Do fleas affect energy expenditure of their free-living hosts?

BACKGROUND: Parasites can cause energetically costly behavioural and immunological responses which potentially can reduce host fitness. However, although most laboratory studies indicate that the metabolic rate of the host increases with parasite infestation, this has never been shown in free-living host populations. In fact, studies thus far have shown no effect of parasitism on field metabolic rate (FMR). METHODOLOGY AND RESULTS: We tested the effect of parasites on the energy expenditure of a host by measuring FMR using doubly-labelled water in free-living Baluchistan gerbils (Gerbillus nanus) infested by naturally occurring fleas during winter, spring and summer. We showed for the first time that FMR of free-living G. nanus was significantly and positively correlated with parasite load in spring when parasite load was highest; this relationship approached significance in summer when parasite load was lowest but was insignificant in winter. Among seasons, winter FMRs were highest and summer FMRs were lowest in G. nanus. DISCUSSION: The lack of parasite effect on FMR in winter could be related to the fact that FMR rates were highest among seasons. In this season, thermoregulatory costs are high which may indicate that less energy could be allocated to defend against parasites or to compensate for other costly activities. The question about the cost of parasitism in nature is now one of the major themes in ecological physiology. Our study supports the hypothesis that parasites can elevate FMR of their hosts, at least under certain conditions. However, the effect is complex and factors such as season and parasite load are involved.

Identification and characterization of two arginine kinases from the parasitic insect Ctenocephalides felis.

Arginine kinase (ATP:l-arginine omega-N-phosphotransferase, EC2.7.3.3.; AK) is an enzyme crucial for the energy metabolism of insects and other invertebrates, that has known allergenic potential in humans and that has been proposed as a pesticidal drug target. Here we report the identification, cDNA cloning, genomic gene structure and functional expression of AK genes from Ctenocephalides (C.) felis (cat flea). In contrast to other insect species investigated so far, C. felis possesses two AK genes, cfak1 and cfak2, encoding the functional enzymes CfAK1 and CfAK2 that can be distinguished by their guanidino substrate specificity and the kinetic parameters for their natural substrates. Molecular modelling on CfAK1 and CfAK2 based on the Limulus polyphemus AK X-ray structure (Zhou et al., 1998) and substrate docking studies provide a potential rational for the observed specificities. Evidence is provided that adult fleas express predominantly CfAK1 as an abundant soluble protein, and that in vivo in C. felis, the AK metabolites are present in concentration ranges relevant for this enzyme.

Identification and characterisation of the dopamine receptor II from the cat flea Ctenocephalides felis (CfDopRII).

G protein-coupled receptors (GPCRs) represent a protein family with a wide range of functions. Approximately 30% of human drug targets are GPCRs, illustrating their pharmaceutical relevance. In contrast, the knowledge about invertebrate GPCRs is limited and is mainly restricted to model organisms like Drosophila melanogaster and Caenorhabditis elegans. Especially in ectoparasites like ticks and fleas, only few GPCRs are characterised. From the cat flea Ctenocephalides felis, a relevant parasite of cats and dogs, no GPCRs are known so far. Thus, we performed a bioinformatic analysis of available insect GPCR sequences from the honeybee Apis mellifera, the mosquito Anopheles gambiae, the fruit fly Drosophila melanogaster and genomic sequences from insect species. Aim of this analysis was the identification of highly conserved GPCRs in order to clone orthologs of these candidates from Ctenocephalides felis. It was found that the dopamine receptor family revealed highest conservation levels and thus was chosen for further characterisation. In this work, the identification, full-length cloning and functional expression of the first GPCR from Ctenocephalides felis, the dopamine receptor II (CfDopRII), are described.

Molecular characterisation of nicotinic acetylcholine receptor subunits from the cat flea, Ctenocephalides felis (Siphonaptera: Pulicidae).

As part of a program to monitor the susceptibility of cat flea populations to the insecticide imidacloprid we have examined the cat flea nicotinic acetylcholine receptor, the target site protein of the neonicotinoid group of insecticides. Seven nAChR subunits (six alpha-type and one beta-type) were identified in cat flea using a degenerate PCR-based strategy. Five of these were expressed in vitro by creating chimeras containing the N-terminal ligand-binding domain of the cat flea subunits and the C-terminal region of the Drosophila Dalpha2 (SAD) subunit. Two of the five chimeric subunits, Cfalpha1/Dalpha2 and Cfalpha3/Dalpha2, when co-expressed with rat beta2 in Drosophila S2 cells, showed high-affinity binding of both epibatidine (Kd=1.6+/-0.6 and 0.13+/-0.06nM, respectively), and imidacloprid (Ki=142+/-34 and 28.7+/-2.4nM, respectively). It is likely therefore that Cfalpha1 and Cfalpha3 contribute to nAChR populations in vivo that are sensitive to imidacloprid. The identification of cat flea nAChR subunits that have a high affinity for imidacloprid presents candidate genes in which to look for resistance-associated mutations if target-site resistance to imidacloprid arises in domestic pet flea populations.

Energy costs of blood digestion in a host-specific haematophagous parasite.

Fleas consume and digest blood from their hosts. We hypothesized that the energy costs of digestion of blood by fleas is dependent on the host species. To test this hypothesis, we studied CO2 emission, a measure of energy expenditure, during digestion of a blood meal taken by Parapulex chephrenis from a preferred (Acomys cahirinus) and a non-preferred (Gerbillus dasyurus) host. We predicted that the energy cost of digestion would be lower for A. cahirinus blood than that for G. dasyurus. Male and female fleas consumed similar amounts of blood per unit body mass, independent of host species. Our prediction was supported in that fleas expended significantly more energy digesting blood of G. dasyurus than blood of A. cahirinus. We also found CO2 emission rates of fed fleas were higher than those of unfed fleas and differed significantly among stages of blood digestion when a flea fed on G. dasyurus but not when it fed on A. cahirinus. When fed on G. dasyurus, fleas spent less energy during earlier than later stages of digestion.

Disposition of 3H-selamectin and 3H-ivermectin in the brain of the cat flea Ctenocephalides felis felis using micro-image analysis.

In addition to intrinsic potency and metabolic stability, the disposition of an antiparasitic drug within the target parasite plays a major role in determining drug activity. A novel technique that allows the disposition of radiolabelled drugs to be visualised within the body of the cat flea (Ctenocephalides felis felis) is described. The concentrations of two macrocyclic lactones, (3)H-selamectin and (3)H-ivermectin, within the supra- and sub-oesophageal ganglia of the flea brain following in vitro feeding of fleas on different doses of drug solubilised in calf blood have been measured. Drug disposition was visualised in cryostat sections of fleas using a micro-image analysis (MIA). A relationship between the concentration of radioactivity in the ganglia and the dose of drug in the blood meal was obtained. The concentration of selamectin in the ganglia was significantly higher than ivermectin at all doses investigated. The enhanced concentration of selamectin, at a site rich in glutamate-gated chloride channels may, in part, explain the higher potency of selamectin against fleas compared to ivermectin.

Efficacy of fipronil/(S)-methoprene combination spot-on for dogs against shed eggs, emerging and existing adult cat fleas (Ctenocephalides felis, Bouché).

The inhibitory activities of fipronil (10% (w/v) solution), (S)-methoprene (9% (w/v) solution), and fipronil/(S)-methoprene (10 and 9% (w/v) solution, respectively) combination against eggs and emerging adult cat fleas (Ctenocephalides felis) and adulticidal activity were tested on experimentally infested dogs. Thirty-two Beagle dogs were selected for this study and eight replicates of four animals were formed based on body weight within sex. One dog in each replicate was randomly allocated to treatment with: (1) untreated control; (2) fipronil 10% (w/v) solution, (3) (S)-methoprene 9% (w/v) solution, and (4) fipronil 10% (w/v) and (S)-methoprene 9% (w/v) combination solution. Treatments were applied once topically on Day 0 at the rate of 0.067 ml/kg. On Days -12, -1, 21, and weekly to Day 84 each dog was infested with approximately 200 fleas and comb counted approximately 24h later, or 2 days (our 48 h) after in the case of Day -1 infestation. On Days -11, 1, 22, and weekly to Day 85 each dog was again infested with approximately 200 fleas. Flea eggs were collected over approximately 24 h beginning 3 days after infestation. Fleas were combed off of the dogs and counted at the end of the egg collection period (approximately 96 h count). One aliquot of up to about 100 eggs, if available, from each animal at each infestation time was incubated for approximately 72 h to determine larval hatch and the other for 35 days to determine the number of adults that developed. The 10% (w/v) fipronil spot-on provided excellent control (>95%) of adult fleas on dogs for 5 weeks. Similarly, the combination spot-on of 10% (w/v) fipronil and 9% (w/v) (S)-methoprene provided excellent control of adult fleas, i.e., >95% for 5 weeks. From week 6 post-treatment onward, the relatively low inhibition of adult flea emergence substantiated the lack of significant ovicidal/larvicidal activity in the fipronil (10%, w/v) treatment group. However, the combination product provided excellent (>90%) ovicidal activity for 8 weeks and high (91.4%) inhibition of adult flea emergence for 12 weeks. In addition, a synergistic effect of the two compounds in combination was demonstrated with fipronil enhancing the ovicidal and inhibition of adult flea emergence activity of (S)-methoprene against cat flea eggs. When all stages of the life cycle of the cat flea are considered, the combination spot-on product provided a high level of total flea control yielding a curative effect against adult fleas and inhibition of flea development stages with little to no potential reinfestation pressure on the animal or in the environment for 12 weeks.

Identification of two cDNAs encoding synaptic vesicle protein 2 (SV2)-like proteins from epithelial tissues in the cat flea, Ctenocephalides felis.

Two distinct cDNAs that appear to encode proteins in the synaptic vesicle-2 (SV2) family were identified as expressed sequence tags from a Ctenocephalides felis hindgut and Malpighian tubule (HMT) cDNA library. To date, SV2 proteins have been described only in vertebrates, and have been detected only in synaptic vesicles in neuronal and endocrine tissues, where they are thought to regulate synaptic vesicle exocytosis. The cDNAs for the C. felis SV2-like proteins SVLP-1 and SVLP-2 encode predicted full-length proteins of 530 and 726 amino acids, respectively. Of characterized proteins, the SVLP protein sequences were most similar to rat SV2B. Northern blot analysis revealed that both mRNAs were up-regulated in larval stages that feed and in adults after feeding, and were expressed primarily or exclusively in the HMT tissues in adult fleas. These results suggest that the flea SVLP-1 and SVLP-2 gene products may have roles that are specific for the HMT tissues, and may differ in function from vertebrate SV2 proteins.

Metabolic rate and jump performance in seven species of desert fleas.

We hypothesized that sexual and interspecific differences in jumping performance of fleas found in our previous study are correlated with differences in resting metabolic rate (RMR) between sexes and among species. To test this hypothesis, we measured RMR of seven flea species (Xenopsylla conformis mycerini, Xenopsylla ramesis, Xenopsylla dipodilli, Parapulex chephrenis, Synosternus cleopatrae pyramidis, Nosopsyllus iranus theodori and Stenoponia tripectinata medialis). We compared RMR between sexes and among species and examined whether there is intra- and interspecific correlation between RMR and jumping ability. Both mass-specific and mass-independent RMR were the highest in female S. t. medialis, whereas mass-specific RMR was the lowest in male X. dipodilli and mass-independent RMR was the lowest in three Xenopsylla species and P. chephrenis. Mass-specific and mass-independent RMR were significantly higher in females than in males in all fleas except S. t. medialis. Differences in jumping ability between males and females were found to be correlated with sexual differences in mass-specific or mass-independent RMR. Interspecific comparison showed that the length of jump in both male and female fleas was strongly affected by their mass-specific and mass-independent RMR.

Isolation, characterization, and recombinant expression of multiple serpins from the cat flea, Ctenocephalides felis.

Several clones encoding serine protease inhibitors were isolated from larval and adult flea cDNA expression libraries by immunoscreening and PCR amplification. Each cDNA contained an open reading frame encoding a protein of approximately 45 kDa, which had significant sequence similarity with the serpin family of serine protease inhibitors. The thirteen cDNA clones isolated to date encode serpin proteins, which share a primary structure that includes a nearly identical constant region of about 360 amino acids, followed by a C-terminal variable region of about 40-60 amino acids. The variable C-terminal sequences encode most of the reactive site loop (RSL) and are generated by mutually exclusive alternative exon splicing, which may confer unique protease selectivity to each serpin. Utilization of an alternative exon splicing mechanism has been verified by sequence analysis of a flea serpin genomic clone and adjacent genomic sequences. RNA expression patterns of the cloned genes have been examined by Northern blot analysis using variable region-specific probes. Several putative serpins have been overexpressed using the cDNA clones in Escherichia coli and baculovirus expression systems. Two purified baculovirus-expressed recombinant proteins have N-terminal amino acid sequences identical to the respective purified native mature flea serpins indicating that appropriate N-terminal processing occurred in the virus-infected insect cells.

Water uptake in the cat flea Ctenocephalides felis (Pulicidae: Siphonaptera).

To counteract water loss due to excretion, cuticular transpiration and respiration, various groups of arthropods have developed mechanisms for active uptake of water vapor from unsaturated air. In this study, active uptake capabilities and water loss rates were examined in the various developmental stages of the cat flea, Ctenocephalides felis. To determine critical equilibrium humidity, the lowest relative humidity at which active water uptake can occur, pre-desiccated immature and adult fleas were placed in a series of humidity regimes ranging from 44 to 93% RH. Active uptake occurred in larval stages at relative humidities above 53% and in pre-pupae at 75-93% RH. Pupae and adults did not demonstrate active uptake at any humidity. Optimal uptake for larvae occurred between 20 and 30 degrees C. When placed over Drierite (<10% RH), larval and adult stages demonstrated a higher rate of water loss than pre-pupal and pupal stages. Active water uptake is necessary to ensure proper development of the larvae of C. felis. Active uptake ceases after the larval-pupal ecdysis and it appears that adults have lost the ability to actively uptake water.

Cloning and characterization of five cDNAs encoding peritrophin-A domains from the cat flea, Ctenocephalides felis.

Five cDNAs encoding peritrophin-A domains were identified as expressed sequence tags (ESTs) from flea hindgut and Malpighian tubule (HMT) cDNA libraries. The full-length cDNAs for each were subsequently isolated and sequenced. Three of the encoded proteins were similar to published peritrophin sequences, and thus were called "peritrophin-like", or PL1, PL2, and PL3. The other two sequences had similarity to both mucin and peritrophin proteins, and were called "mucin/peritrophin-like", or MPL1 and MPL2. The predicted protein sequences encoded by these cDNAs all contained a signal sequence and one or more peritrophin-A domains, which have been shown in other proteins to bind chitin. Aside from the peritrophin-A domains, the sequences shared little or no similarity to each other or to other proteins in the GenBank non-redundant database. The predicted protein sequences were variable in size, ranging in length from 81 to 453 amino acids. The two MPL proteins contained putative N-linked and O-linked glycosylation sites, including a region of seven nearly perfect tandem repeats in the MPL1 protein sequence. Northern blot analysis of different flea lifestages and fed adult timepoints showed distinct mRNA expression patterns for each gene, although all five transcripts were primarily or exclusively detected in the HMT tissues in adults. The PL1 protein was detected by immuno-blot in soluble and insoluble protein extracts from unfed and fed adult fleas. The PL1 protein from the insoluble fractions appeared to be approximately 1 kDa larger than the PL1 protein from the soluble protein fractions. Immunohistochemistry performed on flea thin sections revealed that the PL1 protein was detected in the Malpighian tubules, hindgut, rectum, and trachea. Unpurified native PL1 protein from both soluble and insoluble protein fractions was tested for chitin-binding activity but did not bind to chitin under the conditions tested. These results show that the flea peritrophin-like proteins may have biological functions that are distinct from the peritrophic matrix and from the binding of chitin.