Genetic determinants of mate recognition in Brachionus manjavacas(Rotifera)
© Snell et al; licensee BioMed Central Ltd. 2009
Received: 6 July 2009
Accepted: 9 September 2009
Published: 9 September 2009
Mate choice is of central importance to most animals, influencing population structure, speciation, and ultimately the survival of a species. Mating behavior of male brachionid rotifers is triggered by the product of a chemosensory gene, a glycoprotein on the body surface of females called the mate recognition pheromone. The mate recognition pheromone has been biochemically characterized, but little was known about the gene(s). We describe the isolation and characterization of the mate recognition pheromone gene through protein purification, N-terminal amino acid sequence determination, identification of the mate recognition pheromone gene from a cDNA library, sequencing, and RNAi knockdown to confirm the functional role of the mate recognition pheromone gene in rotifer mating.
A 29 kD protein capable of eliciting rotifer male circling was isolated by high-performance liquid chromatography. Two transcript types containing the N-terminal sequence were identified in a cDNA library; further characterization by screening a genomic library and by polymerase chain reaction revealed two genes belonging to each type. Each gene begins with a signal peptide region followed by nearly perfect repeats of an 87 to 92 codon motif with no codons between repeats and the final motif prematurely terminated by the stop codon. The two Type A genes contain four and seven repeats and the two Type B genes contain three and five repeats, respectively. Only the Type B gene with three repeats encodes a peptide with a molecular weight of 29 kD. Each repeat of the Type B gene products contains three asparagines as potential sites for N-glycosylation; there are no asparagines in the Type A genes. RNAi with Type A double-stranded RNA did not result in less circling than in the phosphate-buffered saline control, but transfection with Type B double-stranded RNA significantly reduced male circling by 17%. The very low divergence between repeat units, even at synonymous positions, suggests that the repeats are kept nearly identical through a process of concerted evolution. Information-rich molecules like surface glycoproteins are well adapted for chemical communication and aquatic animals may have evolved signaling systems based on these compounds, whereas insects use cuticular hydrocarbons.
Owing to its critical role in mating, the mate recognition pheromone gene will be a useful molecular marker for exploring the mechanisms and rates of selection and the evolution of reproductive isolation and speciation using rotifers as a model system. The phylogenetic variation in the mate recognition pheromone gene can now be studied in conjunction with the large amount of ecological and population genetic data being gathered for the Brachionus plicatilis species complex to understand better the evolutionary drivers of cryptic speciation.
One of the most important decisions in the life of animals is mate choice, which includes discrimination of conspecifics, avoidance of inbreeding, and identification of disease-free, high fitness genotypes. In many animals, pheromones are used to communicate information about the sex, age, and fitness of potential mates , and a wide variety of molecules are used to regulate life history transitions . Much is known about mate choice behavior, but little is known about the genes controlling this behavior . Recent work has elucidated some chemosensory genes involved in mate choice and how genetic variation in sensory genes influences mate discrimination. Comparative genomics has shown that genes involved in mate choice typically are highly polymorphic and arise through gene duplication, then are co-opted for mate choice . Reproductive genes often are rapidly evolving and exhibit the signature of positive selection .
Brachionid rotifers exhibit mate choice [5–8]. They are capable of both sexual (mixis) and asexual reproduction, where mixis is triggered by a quorum sensing process . If unfertilized, mictic females produce haploid males that are fast swimmers and do not feed . Upon encountering a female, a male exhibits a distinctive mating behavior that consists of tight circling around the female while maintaining contact with his corona and penis. The moment of male-female encounter is critical because this is when a male makes the key decision of whether or not to mate. Male mate recognition is based upon the mate recognition pheromone (MRP), a glycoprotein that is located on the body surface of females . Biochemical properties of the MRP from Brachionus manjavacas have been characterized [11–14]. The species B. manjavacas is a recently named clade  from the B. plicatilis species complex  and it was formerly known as B. plicatilis Russian. MRP can be removed by ethylenediaminetetraacetic acid (EDTA) treatment and transferred to the body surface of conspecific or heterospecific females . Males are thought to detect the MRP signal by contact chemoreception through a receptor in their corona. Males are capable of discriminating conspecifics by geographic origin, sex, and female age [5, 7, 10, 11, 18, 19], demonstrating a preference for young conspecific females [8, 19]. The basis of this choosy mating behavior of males is female MRP signal quality and strength, which in turn determines male mating persistence and circling intensity. Similar male mate choice has been observed in other rotifer species, including Brachionus calyciflorus [5, 6], Asplanchna brightwelli , and Epiphanes senta . Females also can exhibit mate choice by varying their resistance to male circling .
The MRP likely is a critically important molecule in the establishment and maintenance of reproductive isolation among rotifer species . Knowledge of the amount and pattern of variation at both the nucleotide and amino acid levels in natural populations would provide insight into the type and intensity of selection acting on mating barriers. The rate of evolution of reproductive proteins is known to be generally faster than for other proteins , but the reason for this is not clear. The evolutionary dynamics of mate recognition genes, their comparative divergence in sympatry and allopatry, and their divergence with local adaptation, remain a mystery. Explaining the abundance of cryptic species in rotifers and how morphologically similar species can coexist and maintain their species integrity also remains a challenge.
A key to answering these questions is the identification and characterization of the MRP gene and developing a hypothesis of its evolution. In this paper, we report on isolation of the MRP protein and determination of its N-terminal sequence. We used the N-terminal sequence to identify from genomic and cDNA a family of genes encoding extracellular matrix proteins, which have various numbers of MRP motif repeats. The predicted protein product of one of these genes, designated MMR-B3, has the biochemical properties expected of the MRP, and we used RNAi to validate the functional role of MMR-B3 as the MRP gene in rotifer mating.
The strain of Brachionus manjavacas  was originally collected from the Azov Sea  and formerly known as B. plicatilis Russian . It is part of the Manjavacas clade as defined by phylogenetic analysis of COI and ITS genes [7, 16, 23]. This strain has been cultured continuously in the laboratory since 1983, with periodic collection and storage of resting eggs. Methods for rotifer culture, isolation of males, and hatching of diapausing embryos (resting eggs) followed Snell et al. .
Purification and amino acid sequencing of MRP
Approximately 40 to 50 g wet-weight rotifer biomass was filtered from 250 liter mass cultures fed the green alga Tetraselmis suecica. Isolation of protein followed the EDTA extraction protocol of Snell and Stelzer . Extracted proteins were separated by SDS-PAGE, following the methods of Snell and Stelzer , and visualized with Sypro Orange protein gel stain (Molecular Probes, Eugene, OR, USA) according to the manufacturer's protocol. The detection limit for this stain is about 25 ng. Ion-exchange fractions capable of eliciting male mating reactions eluted from the column at 21 to 23 minutes , and were subjected to high-performance liquid chromatography (HPLC) chromatography as described in Snell et al. . Edman N-terminal amino acid sequence of the 29 kD protein band from the 17 minute HPLC fraction was determined by the Emory University microsequencing facility from a Western blot of a PAGE gel. Sequencing yielded a 0.7 pmol signal for the 29 kD protein.
Isolation and sequencing of the gene encoding MRP
Primers used in the study.
GenBank Accession Number
Primer sequence (5'-3')
MMR for RNAi (vector-specific)
MMR (vector-specific for double-stranded RNA synthesis)
GDP-mannose 4,6-dehydratase (MAN)
Amplifications for MMR-A were performed in 50 μl reactions containing 100 ng cDNA or 500 ng genomic DNA, 0.2 μM each primer (Table 1), 0.4 mM dNTPs, 0.75 M betaine, 2.5 U Taq polymerase (Qiagen), and 5 μl ×10 buffer (Qiagen). PCR conditions were 1 minute at 94°C followed by 35 cycles of 15 seconds at 94°C; 30 seconds at T a; 120 seconds at 72°C followed by 7 minutes at 72°C, where T a was 45°C, 48°C, 52°C or 55°C. Amplifications for MMR-B were performed in 25 μl reactions containing 1× TopTaq Reaction Buffer (Qiagen), 50 ng genomic DNA, 200 μM dNTPs, 0.4 μM each primer, and 1 U TopTaq (Qiagen). PCR conditions were 4 minutes at 94°C, followed by 27 cycles of 30 seconds at 50°C, 20 seconds at 50°C, and 2 minutes at 72°C, followed by 7 minutes at 72°C.
Amplification products were visualized on agarose gels, extracted using a MinElute Gel Extraction Kit (Qiagen), ligated into pCR4-TOPO using the TA-cloning reaction, and transformed into Escherichia coli Top10 cells following the supplier's protocol (Invitrogen). Plasmids were extracted from positive clones and sequenced in both directions using standard M13 forward and reverse primers with ABI Big Dye 3.1 chemistry on an ABI 3730 × l Genetic Analyzer. The sequences were edited and assembled using PHRED, CROSSMATCH, and PHRAP as previously described  or using Sequencher 4.6 (GeneCodes Corporation).
As two of the candidate MRP genes could not be reliably assembled from forward and reverse reads of the PCR products due to their length and repeat structure, we constructed a B. manjavacas fosmid genomic library using standard methods as previously described  and probed membranes representing about a 2-fold genome coverage using a 32P-labeled portion of MRP candidate gene MMR-A4 (4 indicating the number of repetitive motifs within the sequence). A positive fosmid was sub-cloned by shearing the fosmid DNA to about 3 kb fragments by nebulization, repairing the ends by extension with Taq polymerase, and ligating the repaired fragments into pCR4-TOPO. Plasmid clones were sequenced as described above. Due to the repetitive nature of the MRP candidate genes the sequences were assembled by hand using Sequencher 4.6. To resolve a large repeat region, the 2.1 kb insert of a sub-clone containing the entire region with 100 to 150 base pairs (bp) of flanking DNA was excised from pCR4-TOPO with EcoRI (which does not cut in the insert), eluted on an agarose gel and extracted with a MinElute column. The purified insert was partially digested with RsaI, a blunt cutter that cuts once in each repeat (500 ng purified DNA; 0.2 U RsaI; 1× New England Biolabs Buffer 1 for 5 minutes at 37°C). Digestion products were eluted on an agarose gel, extracted with a MinElute column, phosphatased using Shrimp Alkaline Phosphatase (USB, Cleveland, OH, USA) for 60 minutes at 37°C, and cloned into pCR4-TOPO as described above. The insert sizes of 96 clones were checked by digestion with EcoRI and 43 were sequenced and assembled using the non-repetitive flanking sequences as guides. This resulted in an unambiguous assembly with 28× average read coverage and no bp disagreements. Sequences were analyzed using the EMBOSS suite  and DnaSP . Signal peptides were detected using SIGNALP, which predicts the presence of an N-terminal peptide sequence that signals for secretion, and the peptidase I cleavage site that removes the peptide, using hidden Markov models and neural networks based on known signal peptide sequences.
Synthesis of double-stranded RNA
Plasmids containing MMR-A or MMR-B genes were used as templates for double-stranded (dsRNA) synthesis. The MMR-A4 plasmid insert originated from PCR amplification from fosmid DNA. The MMR-A7 insert was a sub-clone of a fosmid. The MMR-B3 insert was derived from a PCR product from genomic DNA. Due to the repetitive motifs in the putative MRP genes, primers were designed to the pCR4-TOPO vector (Table 1) and used to amplify a single product containing MMR-A7 or MMR-B3 flanked by around 80 vector nucleotides at the 5'- and 3'-ends. GDP-mannose 4,6-dehydratase isoform 1 (referred to as 'MAN' throughout) was amplified from genomic B. manjavacas DNA using primers from Snell et al. (Snell TW, Shearer TL, Smith HA: Exposure to dsRNA produces RNA interference in Brachionus manjavacas (Rotifera). BMC Genomics, submitted) (Table 1).
MRP and MAN PCR amplifications were performed in 10 μl volume solutions with 10 to 50 ng plasmid DNA (MMR-A7 and MMR-B3) or genomic DNA (MAN), 0.2 U Taq DNA polymerase and a final concentration of 0.2 mM of each dNTP, 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 0.001% gelatin, 2.5 mM MgCl2, and 0.2 μM vector-specific (MMR-A7 and MMR-B3) or gene-specific primers (Table 1). Thermal cycling conditions consisted of 95°C for 2 minutes followed by 40 cycles of 95°C for 30 seconds, 50°C for 90 seconds, 72°C for 90 seconds. PCR amplicons were gel-extracted using the QIAquick Gel Extraction Kit (Qiagen).
Gel-extracted PCR products were used as template (1 μl) for a subsequent PCR using vector-specific (MMR-A7 and MMR-B3) or gene-specific (MAN) primers with T7 bacterial promoter sequence at the 5'-end of each primer (Table 1). PCR amplification conditions were the same as above. Products were visualized on a 2% agarose gel, and gel-extracted using a QIAquick Gel Extraction Kit (Qiagen).
dsRNA of the candidate MRP and MAN genes and synthesized by in vivo transcription in 33 μl reactions following a modification of Piano et al. . Reactions consisted of 15 μl T7 PCR product (about 7 to 9 ng/μl), 1.21 mM NTPs, 4.53 mM DTT, 12.1 mM Tris (pH 7.9), 1.8 mM MgCl2, 0.6 mM spermidine, 3.0 mM NaCl, and 90 U T7 RNA polymerase (Promega, Madison, WI, USA), and were incubated at 37°C for 4 hours. The product was ethanol precipitated and purified dsRNA was re-suspended in 5 μl sterile, RNase-free water. To minimize degradation of the dsRNA, RNAi experiments were conducted within 48 hours of dsRNA production.
To confirm that the transcribed dsRNA was from MMR-A7 and MMR-B3, PCR products from the MMR-A7 and MMR-B3 PCR using the T7 primers were cloned using a TOPO TA Cloning kit (Invitrogen) and sequenced in both directions using M13 primers (Nevada Genomics, Reno, NV, USA). Sequences were aligned with MMR-A and MMR-B cDNA sequences using BioEdit v7.0.5 , confirming the dsRNA used in RNAi experiments was from the MMR genes (data not shown).
Diapausing embryos were decapsulated and transfected with dsRNA following the double transfection technique of Snell et al. (Snell TW, Shearer TL, Smith HA: Exposure to dsRNA produces RNA interference in Brachionus manjavacas (Rotifera). BMC Genomics, submitted). For the resting egg and the subsequent feeding transfection, 10 to 50 ng dsRNA was resuspended in 5 μl water. Mating bioassays were performed on maternal females approximately 24 hours following the feeding transfection.
Resting egg hatchlings are all females and were collected under a stereomicroscope at 10× magnification using a narrow bore glass micropipette. Males were filtered from a 5 to 7-day-old 200 ml culture using a 68 μm mesh pore size Nitex screen and re-suspended in ASW of the same salinity as the culture (12 to 16 ppt). Only vigorous, fast swimming males (ages unknown) were isolated and mated with the treated females. The mating bioassay was performed by placing seven males and one female into about 50 μl of ASW on a microscope slide printed with 10 spots. Mating behavior was observed for 3 minutes on a video monitor at 10× magnification using a CCD camera. The number of male-female encounters and the number of matings initiated (circlings) by males were recorded using 6 to 12 replicate females for each treatment. The number of encounters and circlings of males with control females, and with females exposed to a variety of HPLC fractions (described below) or dsRNA transfections of MMR-A7 and MMR-B3, were compared. To determine synergistic effects of MMR-A7 and MMR-B3 knockdown, a transfection was performed in which MMR-A7 and MMR-B3 at half dosages were together complexed with the lipofection reagent (Snell TW, Shearer TL, Smith HA: Exposure to dsRNA produces RNA interference in Brachionus manjavacas (Rotifera). BMC Genomics, submitted), for a total quantity of dsRNA similar to MMR-A7 or MMR-B3 alone.
A variation on the mating bioassay was used to test the ability of the HPLC fractions to elicit male mating reactions. Rotifer females were exposed to 50 mM EDTA then washed following Snell and Stelzer  to strip off their surface glycoproteins, making them receptive to binding of exogenous proteins. Females treated with EDTA were exposed to HPLC protein fractions to test the ability of each fraction to elicit male mating. Exposure of six to eight EDTA-treated females was performed using the protocol for treatment with ion-exchange fractions described in Snell and Stelzer ; control females were exposed to EDTA but not treated with protein fractions.
The null hypothesis tested in the mating bioassays was that the frequency of circling is independent of the treatments, which were an HPLC fraction or a dsRNA transfection. For RNAi assays, controls consisted of females transfected with phosphate-buffered saline (PBS) in place of dsRNA; PBS is a reagent in the transfection solution. A one-way analysis of variance (ANOVA) compared circling of control and treated females . The probability of male circling of 6 to 12 replicate females was recorded for each treatment. An ANOVA was performed on raw and arcsine transformed data. Since there was virtually no difference in the ANOVA values, we present results for the untransformed data.
HPLC fractions containing the MRP
SDS-PAGE of HPLC fractions
Identification and characterization of candidate MRP genes
The B. manjavacas cDNA library had two contigs with translated open reading frames containing close matches to the N-terminal sequence of the 29 kD band (Figure 3). The two contigs can be aligned with only minor gaps but differ by more than 50% of aligned nucleotides. As both contigs contained repeats of a conserved motif we designated the two sequences MRP Motif Repeat (MMR) types A and B. The first contig, designated MMR-A, appeared to be an incomplete transcript lacking the 5'-end: though sequenced in both directions from four clones none of the clones contained a sequence consistent with a 5'-end untranslated region (UTR) or with the 5'-end signal peptide region discussed below. The assembly of the reads, and thus the true length of the partial transcript, was ambiguous due to the near identity of the motif sequence across repeats. Sequencing a fosmid clone obtained from probing the B. manjavacas genomic library with MMR-A revealed two MMR-A genes with four and seven repeats, which we designated MMR-A4 and MMR-A7, respectively, and provided their full coding sequence. Other than differing in the number of repeats of the motif, the coding regions of MMR-A4 and MMR-A7 are nearly identical, differing at only 2% of aligned nucleotides. Using primers designed to the 5'-end of the coding sequence and to the junction of the 3'-end of the coding sequence and the 3'-end UTR, we amplified MMR-A4 from both cDNA and genomic DNA. There was no evidence of additional copies of MMR-A. Comparing the genomic and cDNA sequences of the two genes showed no evidence of introns or splicing for either gene.
The contig from the cDNA library representing sequence type MMR-B was made up of a single clone sequenced in both directions. As both reads spanned the entire insert, which contained an open reading frame flanked by 5'- and 3'-ends UTRs, the reads could be assembled unambiguously despite the presence of three repeats. We designed primers to the 5'- and 3'-ends of this transcript, MMR-B3, and used them to amplify cDNA and genomic DNA. The only two amplification products were MMR-B3 and a closely similar MMR-B gene with five repeats of the motif, MMR-B5. Comparing the genomic and cDNA sequences of the two genes showed no evidence of introns or splicing for either gene.
The conceptual translations of the MMR genes all share several features: each begins with a signal peptide region followed by nearly perfect repeats of a motif of 92 amino acids (MMR-A) or 87 amino acids (MMR-B); the terminator interrupts the final motif after amino acid 85 (MMR-A) or 81 (MMR-B) and there are no amino acids between repeats of the motif (Figure 3). The motif is 55% to 60% non-polar with a net negative charge, and is predicted to be composed of a series of alpha helices, producing a generally globular protein. The region corresponding to the N-terminal sequence obtained from the 29 kD band begins at the third amino acid of the first motif (MMR-A) or at the first amino acid of the first motif (MMR-B). The MMR-A genes lack codons for asparagine, and thus MMR-A protein products are unlikely to be N-glycosylated. Each motif in the product of the MMR-B genes contains three asparagines as potential sites for N-glycosylation, although they do not occur in the series N-X-T, which generally marks N-glycosylated asparagines. Two of the three asparagines are predicted to be on the hydrophilic sides of alpha helices, and therefore potentially exposed. None of the processed MMR peptides has significant similarity to any sequence in GenBank/EMBL/DDJ databases as judged by BLASTX and BLASTN searches, or to known protein motifs as judged by HMMPFAM searches of PFAM databases. After cleavage of the signal peptide the molecular weights of the peptides produced by MMR-A4, -A7, -B3, and -B5 are predicted to be 38.4 kD, 69.2 kD, 29.2 kD, and 48.2 kD respectively; each peptide is predicted to be negatively charged at pH 8.
MRP RNAi knockdown
Characteristics of MRP protein
This study is the culmination of extensive efforts to identify the MRP gene, beginning with studies of the role of glycoproteins as mating signals in rotifers. Nearly two decades ago the biochemical properties of surface glycoproteins involved in rotifer mate recognition were probed in binding studies with several fluorescently labeled lectins, which bind the carbohydrate moiety of the MRP protein [12, 33]. When these lectins were bound, females elicited significantly fewer mating responses from males. Cleavage of surface proteins by proteinase K also rendered females significantly less attractive to males , as did cleavage of N-linked oligosaccharides by the glycohydrolase N-glycanase . Treatment of B. manjavacas females with the detergent CHAPS and EDTA removed surface proteins and eliminated the male mating response [12, 17]. In the present study, binding of EDTA-extractable proteins to an anion exchange column suggests that the proteins possess a net negative charge at pH 8.0. Adsorption of the surface glycoproteins to a C3 reversed-phase HPLC column, eluting in 64% to 72% acetonitrile, further suggested the proteins contain hydrophobic domains.
While a putative MRP protein had been previously isolated, our study is the first to report its gene sequence and confirm its function. Snell et al.  described a 29 kD rotifer surface glycoprotein that likely represented the MRP. Yet, the amino acid sequence could not be determined from the small quantity of protein that had been purified, and there is no conclusive way to determine whether the glycoprotein described is identical to the MRP protein reported here. Nonetheless, in the current study, the HPLC fractions of EDTA-extractable proteins capable of eliciting male mating behavior had only two prominent proteins with 34 and 29 kD apparent mass. Given the earlier identification of the putative 29 kD MRP, here we focused on the 29 kD protein as the most likely candidate for the MRP. N-terminal amino acid sequencing of the 29 kD protein yielded enough sequence to perform a meaningful search of a rotifer cDNA library. Reduction of mating from RNAi knockdown of expression of the MMR-B3 gene as reported here validates the 29 kD protein as the MRP, and that this protein is critical for B. manjavacas mate recognition.
Characteristics of the MRP gene
Using the N-terminal sequence of the 29 kD protein, we identified two contigs from a cDNA library of B. manjavacas as candidates for the MRP gene. PCR of genomic DNA and cDNA, as well as sequencing a fosmid from a genomic library of B. manjavacas containing two candidate genes, revealed that the N-terminal sequence is a marker for a family of related genes. This gene family has no recognizable homology to sequences in public databases or to the protein patterns of PFAM but is characterized by a peptide signal sequence region followed by repeats of a motif that contains the N-terminal sequence at or near its 5'-end. There is a short spacer region between the signal peptide and the repeats but the repeats follow each other head-to-tail without interruption. The final motif is invariably truncated by four to seven codons. The presence of a signal peptide implies that the peptides are secreted via the endoplasmic reticulum (ER), consistent with being part of the extracellular matrix on the surface of the rotifer, and the negative charge and general hydrophobicity is consistent with previous biochemical studies of the MRP protein. The MMR-A and MMR-B genes differ at about 50% of aligned nucleotides (exact homology of the repeats cannot be reliably assigned), but the repeats of a motif within a gene are identical or differ by only a few nucleotides.
Data suggest MMR-B3 encodes a functional MRP protein. Despite considerable effort we have not identified a MMR-A type gene that would produce a 29 kD peptide. Also, although the MMR-A motif contains a single potential O-glycosylation site, it does not contain arginine, so is unlikely to be N-glycosylated. Finally, RNAi with the MMR-A motif did not decrease male mating behavior. In contrast, one of the MMR-B type genes, MMR-B3, is predicted to produce a 29 kD product, the MMR-B motif does contain arginines, which are predicted to be on the exposed surface of the protein. Finally, RNAi with the MMR-B3 gene significantly reduces mating behavior. Thus, we conclude that MMR-B3 encodes the 29 kD peptide originally identified as the MRP. We hypothesize that after cleavage of the signal region upon entry into the ER the peptide is further modified by the removal of the remaining residues before the first motif, consistent with N-terminal sequencing of the 29 kD band. Also, one or more arginines per motif may be N-glycosylated or another form of oligosaccharide moiety that mimics N-glycosylation in biochemical assays may be added.
That the MRP gene (MMR-B3) has no significant similarity to any deposited sequence may not be surprising, as at least 20% of the brachionid transcriptome appears to lack obvious homologs in sequence databases , and there is in general very little knowledge of the genes encoding extracellular proteins in non-ecdysozoan protostome phyla. The very low divergence between repeat units, even at synonymous positions, strongly suggests that the units are kept nearly identical through a process of concerted evolution analogous to the preservation of sequence homogeneity in rRNA clusters. Concerted evolution of repeats in turn suggests a mechanism by which a single non-synonymous mutation could rapidly spread to be present in multiple repeats, effecting a major change in the structure of the protein. Such saltational change of a protein involved in mate choice could lead to rapid speciation.
Comparison of the MRP to contact mate recognition pheromones of other animals
The critical role of a single gene in mate recognition of rotifers makes their signaling pathway for mate recognition particularly useful for studying the molecular mechanisms and evolution of mate recognition, and the role of mating genes in speciation. In contrast, in many other animals a variety of factors are influential, and thus assigning functional roles of genes is difficult. In B. manjavacas, glycoproteins from the female are necessary and sufficient to elicit male circling . Chemosensory mate recognition genes also have been described in other animal species . Perhaps most similar to rotifers is the mate recognition system of the fruit fly Drosophila. Males evaluate the approximately 20 cuticular hydrocarbons (CH) of females by contact chemoreception to discriminate mates and prefer to mate with same-strain females . Olfactory receptors on the antennae and gustatory receptors on the tarsa and proboscis of males detect immobilized surface heptacosadienes on females of 20 to 40 carbons in chain length . The neurobiological basis of mating decisions in Drosophila also is beginning to be revealed . It is now understood that the quantity and blends of various CHs contain information about species and geographic population and males use this information to determine whether or not to mate. Owing to the large number of CHs involved in mate recognition, the genetic basis of mating phenotypes is complex  and only a few genes have been characterized . In addition, temperature, food, and social experience can influence CH profiles. This has made it difficult to relate CHs directly to evolutionary processes like sexual selection and to determine explicitly their role in speciation. In contrast, in B. manjavacas the MRP is a glycoprotein immobilized on the surfaces of females . It is encoded by a single gene and the MRP phenotype apparently is subject to little environmental modification. The MRP therefore should be an excellent molecular marker for investigating selection and speciation processes in monogonont rotifers.
In comparison with other invertebrates, specifically insects, the composition of the MRP protein provides an indication of how differing selection pressures imposed by terrestrial versus aquatic habitats may have influenced evolution of mating systems. The use of CHs as recognition pheromones by insects is a general theme not limited to Drosophila [39–41]. This channel of communication is not surprising since CHs are the most prominent molecules on external surfaces of insects . CHs play a central role in minimizing desiccation , and thus their evolution may have resulted from strong selection for desiccation resistance that accompanied the colonization of land. The ability to decipher the information encoded in the molecular composition of cuticles is clearly a useful trait and eventually became the basis of mate recognition in many insects. Aquatic animals have very different cuticles since most have not experienced strong selection for desiccation resistance. For example, Alarie et al.  found substantial differences between CH profiles of an aquatic beetle compared with its terrestrial counterpart. We suggest that information-rich molecules like surface glycoproteins are well adapted for chemical communication and that aquatic animals may have evolved signaling systems based on these compounds. In this sense, the surfaces of aquatic animals are more like those of individual mammalian cells, which typically are highly decorated with information-rich glycoproteins (see, for example, Gahmberg and Tolvanen ). Glycoproteins and their oligosaccharide moieties have been implicated in a wide variety of cell-cell and cell-matrix recognition events [45, 46]. By providing evidence of a surface glycoprotein in brachionid rotifers involved in mate recognition of conspecifics, we provide one example of how such compounds are involved in chemical communication in aquatic environments.
Like rotifers, some copepods also use surface glycoproteins to make mating decisions . Mate-guarding in the harpacticoid copepod Tigriopus japonicus can be inhibited by binding lectins to surface glycoproteins of females [48, 49]. Before males initiate guarding behavior or transfer spermatophores, stroking behavior is engaged in which the male probes the female body surface, presumably in search of chemical cues. The carbohydrate moiety of glycoproteins on the female are thought to be detected by protein receptors on antennules of the male, enabling identification of appropriate partners for mating. Purification of female surface proteins revealed a 70 kD glycoprotein, with significant similarity to the sequence of α2-macroglobulin [49, 50].
Here we have isolated the rotifer MRP, identified MMR-B3 as the MRP gene, and demonstrated its essential role in male mate recognition by functional assays with RNAi. Our study provides an initial synthesis of behavioral, biochemical, and molecular assays to characterize the mating system of brachionid rotifers. This system may be an important model for comparative studies of mating signaling pathways in aquatic and terrestrial animals. The phylogenetic variation in the MRP gene can now be studied in conjunction with the large amount of ecological and population genetic data being gathered for the Brachionus plicatilis species complex to understand better the evolutionary drivers of cryptic speciation. Characterization of the MRP protein will facilitate identification and isolation of the second key element in the rotifer mate recognition signaling system, the male MRP receptor. Males have a variety of receptors in their coronas, and we are investigating a gene in a family of C-type lectins with putative roles in cell adhesion and pathogen recognition  as a potential target.
analysis of variance
high performance liquid chromatography
mate recognition pheromone
polymerase chain reaction
This work was supported by the National Science Foundation grant BE/GenEn MCB-0412674 to TWS, DBMW, and JK. We thank T. Abatzopoulos and an anonyomous reviewer for comments that improved this work.
- Wyatt TD: Pheromones and Animal Behavior. 2003, Cambridge: Cambridge University PressView ArticleGoogle Scholar
- Heyland A, Hodin J, Reitzel AM: Hormone signaling in evolution and development: a non-model system approach. BioEssays. 2004, 27: 64-75. 10.1002/bies.20136.View ArticleGoogle Scholar
- Horth L: Sensory genes and mate choice: Evidence that duplications, mutations, and adaptive evolution alter variation in mating cue genes and their receptors. Genomics. 2007, 90: 159-175. 10.1016/j.ygeno.2007.03.021.View ArticlePubMedGoogle Scholar
- Swanson WJ, Vacquier VD: The rapid evolution of reproductive proteins. Nat Genet. 2002, 3: 137-144.View ArticleGoogle Scholar
- Gilbert JJ: Contact chemoreception, mating behaviour, and sexual isolation in the rotifer genus Brachionus. J Exp Biol. 1963, 40: 625-641.Google Scholar
- Gilbert JJ, Walsh EJ: Brachionus calyciflorus is a species complex: Mating behavior and genetic differentiation among four geographically isolated strains. Hydrobiologia. 2005, 546: 257-265. 10.1007/s10750-005-4205-3.View ArticleGoogle Scholar
- Suatoni E, Vicario S, Rice S, Snell TW, Caccone A: Phylogenetic and biogeographic patterns in the salt water rotifer, Brachionus plicatilis. Mol Phylogenet Evol. 2006, 41: 86-98. 10.1016/j.ympev.2006.04.025.View ArticlePubMedGoogle Scholar
- Snell TW, Kim J, Zelaya E, Resop R: Mate choice and sexual conflict in Brachionus plicatilis (Rotifera). Hydrobiologia. 2007, 593: 151-157. 10.1007/s10750-007-9065-6.View ArticleGoogle Scholar
- Kubanek J, Snell TW: Quorum sensing in rotifers. Chemical Communication Among Microbes. Edited by: Winans SC, Bassler BL. 2008, Washington DC: ASM Press, 453-461.Google Scholar
- Snell TW: Systematics, reproductive isolation and species boundaries in rotifers. Hydrobiologia. 1989, 186/187: 299-310. 10.1007/BF00048925.View ArticleGoogle Scholar
- Snell TW, Rico-Martinez R, Kelly LN, Battle TE: Identification of a sex pheromone from a rotifer. Mar Biol. 1995, 123: 347-353. 10.1007/BF00353626.View ArticleGoogle Scholar
- Snell TW, Nacionales MA: Sex pheromones and mate recognition in rotifers. Comp Biochem Physiol. 1990, 97A: 211-216. 10.1016/0300-9629(90)90173-P.View ArticleGoogle Scholar
- Rico-Martinez R, Snell TW: Binding and blocking of mating in Brachionus plicatilis (Muller) and Brachionus rotundiformis (Tschugunoff) (Rotifera) females by an antibody against the mate recognition pheromone. Hydrobiologia. 1997, 358: 71-76. 10.1023/A:1003121531835.View ArticleGoogle Scholar
- Snell TW: Chemical ecology of rotifers. Hydrobiologia. 387/388: 267-276.
- Fontaneto D, Giordani I, Melone G, Serra M: Disentangling the morphological stasis in two rotifer species of the Brachionus plicatilis species complex. Hydrobiologia. 2007, 583: 297-307. 10.1007/s10750-007-0573-1.View ArticleGoogle Scholar
- Gómez A, Serra M, Carvalho GR, Lunt DH: Speciation in ancient cryptic species complexes: evidence from the molecular phylogeny of Brachionus plicatilis (Rotifera). Evolution. 2002, 56: 1431-1444.View ArticlePubMedGoogle Scholar
- Snell TW, Stelzer CP: Removal of surface glycoproteins and transfer among Brachionus species. Hydrobiologia. 2005, 546: 267-274. 10.1007/s10750-005-4207-1.View ArticleGoogle Scholar
- Snell TW, Hawkinson CA: Behavioral reproductive isolation among populations of the rotifer Brachionus plicatilis. Evolution. 1983, 37: 1294-1305. 10.2307/2408849.View ArticleGoogle Scholar
- Gómez A, Serra M: Mate choice in male Brachionus plicatilis rotifers. Funct Ecol. 1996, 10: 681-687. 10.2307/2390502.View ArticleGoogle Scholar
- Aloia RC, Moretti RL: Mating behavior and ultrastructural aspects of copulation in the rotifer Asplanchna brightwelli. Trans Am Microsc Soc. 1973, 92: 371-380. 10.2307/3225241.View ArticleGoogle Scholar
- Schröder T: Precopulatory mate guarding and mating behaviour in the rotifer Epiphanes senta (Monogononta, Rotifera). Proc R Soc Lond B Biol Sci. 2003, 270: 1965-1970. 10.1098/rspb.2003.2466.View ArticleGoogle Scholar
- Snell TW, Kubanek JM, Carter WE, Payne AB, Kim J, Hicks M, Stelzer CP: A protein signal triggers sexual reproduction in Brachionus plicatilis (Rotifera). Mar Biol. 2006, 149: 763-773. 10.1007/s00227-006-0251-2.View ArticleGoogle Scholar
- Derry AM, Hebert PDN, Prepas EE: Evolution of rotifers in saline and subsaline lakes: A molecular phylogenetic approach. Limnol Oceanogr. 2003, 48: 675-685.View ArticleGoogle Scholar
- Ewing B, Green P: Base-calling of automated sequencer traces using phred. II. Error probabilities. Genome Res. 1998, 8: 186-194.View ArticlePubMedGoogle Scholar
- Suga K, Mark Welch D, Tanaka Y, Sakakura Y, Hagiwara A: Analysis of expressed sequence tags of the cyclically parthenogenetic rotifer Brachionus plicatilis. PLoS One. 2007, 2: e671-10.1371/journal.pone.0000671. doi: 10.1371/journal.pone.0000671.PubMed CentralView ArticlePubMedGoogle Scholar
- Mark Welch DB, Mark Welch JL, Meselson M: Evidence for degenerate tetraploidy in bdelloid rotifers. Proc Natl Acad Sci USA. 2008, 105: 5145-5149. 10.1073/pnas.0800972105.PubMed CentralView ArticlePubMedGoogle Scholar
- Rice P, Longden I, Bleasby A: EMBOSS: The European Molecular Biology Open Software Suite. Trends Genet. 2000, 16: 276-277. 10.1016/S0168-9525(00)02024-2.View ArticlePubMedGoogle Scholar
- Rozas J, Sanchez-DelBarrio JC, Messeguer X, Rozas R: DnaSP, DNA polymorphism analyses by the coalescent and other methods. Bioinformatics. 2003, 19: 2496-2497. 10.1093/bioinformatics/btg359.View ArticlePubMedGoogle Scholar
- Emanuelsson O, Brunak S, von Heijne G, Nielsen H: Locating proteins in the cell using TargetP, SignalP, and related tools. Nat Protoc. 2007, 2: 953-971. 10.1038/nprot.2007.131.View ArticlePubMedGoogle Scholar
- Piano F, Schetter AJ, Morton DG, Gunsalus KC, Reinke V, Kim SK, Kemphues KJ: Gene clustering based on RNAi phenotypes of ovary-enriched genes in C. elegans. Curr Biol. 2002, 12: 1959-1964. 10.1016/S0960-9822(02)01301-5.View ArticlePubMedGoogle Scholar
- Hall TA: BioEdit: A user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symp Ser. 1999, 41: 95-98.Google Scholar
- Sokal RR, Rohlf FJ: Biometry. 1995, New York: W.H. FreemanGoogle Scholar
- Snell TW, Morris PD, Cecchine GA: Localization of the mate recognition pheromone in the marine rotifer Brachionus plicatilis. J Exp Mar Biol Ecol. 1993, 165: 225-235. 10.1016/0022-0981(93)90107-Y.View ArticleGoogle Scholar
- Snell TW, Childress M, Winkler BC: Characterization of the mate recognition factor of the rotifer Brachionus plicatilis. Comp Biochem Physiol. 1988, 89A: 481-485. 10.1016/0300-9629(88)91061-4.View ArticleGoogle Scholar
- Blows MW, Allan RA: Levels of mate recognition within and between two Drosophila species and their hybrids. Am Nat. 1998, 152: 826-837. 10.1086/286211.View ArticlePubMedGoogle Scholar
- Ferveur JF: Cuticular hydrocarbons: Their evolution and roles in Drosophila pheromonal communication. Behav Genet. 2005, 35: 279-295. 10.1007/s10519-005-3220-5.View ArticlePubMedGoogle Scholar
- Dickson BJ: Wired for sex: The neurobiology of Drosophila mating decisions. Science. 2008, 322: 904-909. 10.1126/science.1159276.View ArticlePubMedGoogle Scholar
- Gleason JM, Jallon JM, Rouault JD, Ritchie MG: Quantitative trait loci for cuticular hydrocarbons associated with sexual isolation between Drosophila simulans and D. sechellia. Genetics. 2005, 171: 1789-1798. 10.1534/genetics.104.037937.PubMed CentralView ArticlePubMedGoogle Scholar
- Dani FR, Jones GR, Destri S, Spencer SH, Turillazzi S: Deciphering the recognition signature within the cuticular chemical profile of paper wasps. Anim Behav. 2001, 62: 165-171. 10.1006/anbe.2001.1714.View ArticleGoogle Scholar
- Ginzel MD, Hanks LM: Contact pheromones as mate recognition cues of four species of longhorned beetles (Coleoptera: Cerambycidae). J Insect Behav. 2003, 16: 181-187. 10.1023/A:1023911701159.View ArticleGoogle Scholar
- Zhang A, Oliver JE, Chauhan K, Zhao B, Xia L, Xu Z: Evidence for contact sex recognition pheromone of the Asian longhorned beetle, Anoplophora glabripennis (Coleoptera: Cerambycidae). Naturwissenschaften. 2003, 90: 410-413. 10.1007/s00114-003-0452-1.View ArticlePubMedGoogle Scholar
- Singer TL: Roles of hydrocarbons in the recognition systems of insects. Am Zool. 1998, 38: 394-405.View ArticleGoogle Scholar
- Rouault JD, Marican C, Wicker-Thomas C, Jallon JM: Relations between cuticular hydrocarbon (HC) polymorphism, resistance against desiccation and breeding temperature; a model for HC evolution in D. melanogaster and D. simulans. Genetica. 2004, 120: 195-212. 10.1023/B:GENE.0000017641.75820.49.View ArticlePubMedGoogle Scholar
- Alarie Y, Joly H, Dennie D: Cuticular hydrocarbon analysis of the aquatic beetle Agabus anthracinus Mannerheim (Coleoptera: Dytiscidae). Can Entomol. 1998, 130: 615-629.View ArticleGoogle Scholar
- Gahmberg CG, Tolvanen M: Why mammalian cell surface proteins are glycoproteins. Trends Biochem Sci. 1996, 21: 308-311.View ArticlePubMedGoogle Scholar
- Bertozzi CR, Kiessling LL: Chemical glycobiology. Science. 2001, 291: 2357-2364. 10.1126/science.1059820.View ArticlePubMedGoogle Scholar
- Lonsdale DJ, Frey MA, Snell TW: The role of chemical signals in copepod reproduction. J Mar Syst. 1998, 15: 1-12. 10.1016/S0924-7963(97)00044-4.View ArticleGoogle Scholar
- Kelly LS, Snell TW: Chemical communication during mating of the copepod Tigriopus japonicus. Philos Trans R Soc Lond B. 1998, 353: 737-744. 10.1098/rstb.1998.0239.View ArticleGoogle Scholar
- Ting JH, Snell TW: Purification and sequencing of a mate recognition protein from the copepod Tigriopus japonicus. Mar Biol. 2003, 143: 1-8. 10.1007/s00227-003-1071-2.View ArticleGoogle Scholar
- Snell TW: Contact chemoreception and its role in zooplankton mate recognition. Chemical Communication in Crustaceans. Edited by: Thiel M, Breithaupt T. Springer, New York,
- East L, Isacke CM: The mannose receptor family. Biochim Biophys Acta. 2002, 1572: 364-386.View ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.