Vitellogenin of the parasitoid wasp, Encarsia formosa (Hymenoptera: Aphelinidae): gene organization and differential use by members of the genus

Insect Biochemistry and Molecular Biology 34 (2004) 951–961 www.elsevier.com/locate/ibmb

Vitellogenin of the parasitoid wasp, Encarsia formosa (Hymenoptera: Aphelinidae): gene organization and differential use by members of the genus5,55 David M. Donnell
Interdisciplinary Program in Insect Science, University of Arizona, Tucson, AZ 85721, USA Received 28 May 2004; accepted 15 June 2004

Abstract The vitellogenin (Vg) gene of the parasitoid wasp, Encarsia formosa (Hymenoptera: Aphelinidae), has been cloned and sequenced. The gene codes for a protein consisting of 1814 amino acids in seven exons. The position of the six introns in the E. formosa gene align with those inferred for the Vg gene of the honeybee, Apis mellifera. The position of two introns in the hymenopteran sequences are shared with every full-length insect Vg gene characterized to date. The deduced amino acid sequence of the E. formosa Vg gene most closely resembles that of the ichneumonid parasitoid, Pimpla nipponica (38% identity). The gene product, less the putative signal peptide, contains large quantities of serine (11.3% of total residues) but lacks the extensive polyserine tracts found in the Vgs of insects outside the apocritan Hymenoptera. The gene also codes for the highest level of lysine (9.5%), and lowest levels of phenylalanine (2.6%) and tyrosine (2.3%), observed in any insect Vg characterized to date. The mature gene product retains 12 cysteine residues in positions conserved in other insect Vgs. Ovary homogenates suggest that processed Vg is stored in the egg as an uncleaved molecule of approximately 200 kDa. Vg expression was examined in three additional Encarsia species. The protein was found in female E. sophia and E. luteola, but not in male E. luteola or female E. pergandiella. Despite extensive screening of a phage library prepared from E. pergandiella genomic DNA, a Vg gene was not detected in this species. # 2004 Elsevier Ltd. All rights reserved. Keywords: Parasitoids; Aphelinidae; Encarsia; Vitellogenin; Vitellin; Egg provisioning

1. Introduction Vitellogenin (Vg) genes code for the precursors of the egg storage protein, vitellin, used by many oviparous organisms (Kunkel and Nordin, 1985). Insect Vgs are large molecules (~200 kDa) synthesized in the fat body in a process that involves substantial structural modifications of the nascent protein to facilitate the transport of carbohydrates, lipids and other nutrients through the hemolymph to the ovaries (Kunkel and Nordin, 1985; Dhadialla and Raikhel, 1990). In groups 5

GenBank accession number: AY553878. Supplementary data (Figs. A.1 and A.2) associated with this article can be found at doi: 10.1016/j.ibmb.2004.06.011.
Present address: Department of Entomology, University of Georgia, Biological Sciences, Room 420, Athens, GA 30602, USA. Tel.: +1-706-583-8238; fax: +1-706-542-2279. E-mail address: [email protected] (D.M. Donnell). 55

0965-1748/$ - see front matter # 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.ibmb.2004.06.011

outside the apocritan Hymenoptera, Vgs are routinely cleaved into multiple subunits (Harnish and White, 1982; Sappington et al., 2002). Vgs are secreted from the fat body and travel as large (>550 kDa) multimeric structures to the ovaries where they are taken up by oocytes via receptor-mediated endocytosis (Chen et al., 1978; Harnish and White, 1982; Byrne et al., 1989; Raikhel and Dhadialla, 1992). The primary role of Vg as a storage protein has allowed substantial changes in the order of amino acids within the protein (Hagedorn and Kunkel, 1979). Nevertheless, the nutrient demands of embryos have necessitated that the balance of amino acids in insect Vgs and the position of certain residues essential for the post-translational modification of the protein remain conserved (Wahli, 1988; Byrne et al., 1989). Comparisons of coding sequences have demonstrated that sufficient information remains in the primary

952

D.M. Donnell / Insect Biochemistry and Molecular Biology 34 (2004) 951–961

structure of Vg genes to permit reconstruction of the phylogenetic relationship between insect Vgs and those of their nematode and vertebrate homologs (Nardelli et al., 1987; Spieth et al., 1991; Trewitt et al., 1992; Chen et al., 1997; Hagedorn et al., 1998). Early analyses of gene organization suggested that introns in vertebrate Vg genes are more stable than those in the Vg genes of invertebrates (Nardelli et al., 1987; Spieth et al., 1991). At the present time, the intron organization of insect Vg genes has been determined for the boll weevil, Anthonomous grandis (Trewitt et al., 1992), the moths, Bombyx mori and Lymantria dispar (Yano et al., 1994b; Hiremath and Lehtoma, 1997), and the mosquitoes, Aedes aegypti and Anopheles gambiae (Romans et al., 1995). Alignments of these sequences revealed that all of the genes retain the position of two introns (Romans et al., 1995; Chen et al., 1997). The significance of these introns has not been determined. It is also unknown if these two intron positions are retained in the Vg genes of insects in other orders. Vgs are supplemented or supplanted by other egg storage proteins in many insect species (Bownes and Hames, 1977; Fourney et al., 1982; Wang et al., 1988; Chinzei et al., 1992; Telfer, 2002). These proteins often vary in structure and composition and appear to have a diversity of origins although Sappington (2002) recently noted that the egg-specific proteins (ESPs) that constitute a minority of the yolk in lepidopteran eggs are homologs of the yolk proteins (YPs) that have replaced Vg entirely in the eggs of higher Diptera. In extreme circumstances, YPs are entirely absent. Yolkless eggs have been described in numerous species of endoparasitic wasps (Flanders, 1942; Fisher, 1971; King et al., 1971; Ivanova-Kasas, 1972; Le Ralec, 1995; Quicke, 1997). Flanders (1942) termed such eggs ‘‘hydropic’’ because they swell upon oviposition and were assumed to absorb the nutrients necessary for embryonic development from the host hemolymph. Ferkovich and Dillard (1986) demonstrated the capacity of the hydropic eggs of the braconid parasitoid, Microplitis croceipes, to absorb radiolabeled amino acids, but not proteins, from the hemolymph of their hosts. Wasps within the genus Encarsia (Hymenoptera: Aphelinidae) are endoparasitoids of whiteflies and scale insects. This group of wasps varies markedly in the extent to which they provision their eggs. The eggs of E. pergandiella, for example, appear yolkless (Donnell and Hunter, 2002) and swell upon oviposition in the host (Gerling, 1966). Those of the congener, Encarsia formosa, are much larger and clearly possess yolk despite the fact that both wasp species develop within the same hosts. This paper details the isolation and organization of the Vg gene of E. formosa and provides information regarding the differential reliance upon this protein by members of the genus.

2. Materials and methods 2.1. Cultures All parasitoids used in this study were reared on the sweet-potato whitefly, Bemisia tabaci, or the greenhouse whitefly, Trialeurodes vaporariorum, feeding on green bean plants, Phaseolus vulgaris cv. Landmark. The E. pergandiella culture was established in 1994 using insects obtained from the USDA/APHIS Mission Biological Control Laboratory, TX (Quarantine No. M94055). This population of E. pergandiella was originally collected in Brazil on B. tabaci. The E. formosa culture was established in 1997 from insects obtained from a commercial insectary (CIBA Bunting, Colchester, UK). The E. formosa and E. pergandiella populations are both thelytokous parthenogens. The bisexual wasps E. luteola and E. sophia were also used in this study. The E. luteola culture was established in 1997 using wasps collected in agricultural fields in Brawley, CA. The E. sophia culture was established in 1996 using wasps originally collected in Murcia, Spain, and cultured at the USDA/APHIS Mission Biological Control Laboratory. Cultures and experiments were conducted in an environmental chamber maintained at v 27  1 C, 35  5% r:h:, under a 16L:8D photoperiod regime. 2.2. Preparation of a Vg probe RNA was prepared from approximately 50 mg each of newly eclosed E. formosa and E. pergandiella wasps using a Bio 101 RNAid kit (Qbiogene, Carlsbad, CA). The kit utilizes the guanidinium isothiocyanate RNA isolation procedure (Chomczynski and Sacchi, 1987). Messenger RNA was isolated using a poly(A)+ RNA binding matrix under high salt conditions then removed from the matrix using DEPC-treated water. This mRNA was used to generate cDNA using Superscript II reverse transcriptase (Life Technologies) and an oligonucleotide with adaptor primer (AP) and poly(dT) nucleotide sequences (APdT; 50 -CGGAATTCTCTAGACTCGAG(T)18-30 ). A degenerate genespecific oligonucleotide primer (GSP) was designed to anneal to a conserved region near the 30 end of hymenopteran Vg genes (50 -TGYTGGCAYGTNGTNATGA-30 ). This primer was used in conjunction with the APdT oligonucleotide in polymerase chain reactions (30 -RACE) using both E. formosa and E. pergandiella cDNAs. This procedure yielded a product 1641 base pairs in length from E. formosa. The product was cloned into a plasmid designed to accept PCR products (pCR2.1, Invitrogen) and sequenced. The product showed significant homology to the C-termini of Vg gene sequences in the GenBank database. The same

D.M. Donnell / Insect Biochemistry and Molecular Biology 34 (2004) 951–961

primers were used in numerous PCR reactions with E. pergandiella cDNA but failed to yield a product. 2.3. Genomic library construction and clone isolation Genomic DNA was extracted from approximately 400 mg each of newly eclosed E. formosa and E. pergandiella adults by homogenizing the wasps in chilled buffer (100 mM Tris–Cl (pH 8.0), 80 mM EDTA (pH 8.0), 160 mM sucrose, 0.5% SDS). The extracts were treated sequentially with RNase, Proteinase K and phenol/chloroform (25:24). The DNA was precipitated using 0.2 volumes of 10 M ammonium acetate and 2.5 volumes of 100% ethanol. The DNA was resuspended in a 10 mM Tris–Cl, 1 mM EDTA (pH 8.0) buffer and 5 lg partially digested using the restriction enzyme Sau3a1 (Life Technologies). The digest products were layered on a linear 10–40% sucrose gradient and centrifuged at 30,000 rpm in an SW41 rotor for 24 h in a Beckman H-class ultracentrifuge. Following centrifugation, the sucrose gradient was collected as a series of fractions and aliquots of each fraction were analyzed on a 0.7% agarose gel. DNA was precipitated in the fractions containing fragments ranging in size from 5 to 7 kb then resuspended as above and the DNA cloned into kZAP vectors using a kZAP cloning kit with Gigapack III Gold packaging extract (Stratagene). The resulting phage libraries were plated at 50,000 plaque forming units (pfus) per plate using XL1-BLU cells (Stratagene). Plaque lifts were performed using nylon membranes (Boehringer Mannheim). The 30 -RACE product obtained from E. formosa was labeled with digoxigenin (DIG)-conjugated nucleotides (Boehringer Mannheim) by conventional PCR using the same AP and GSP primer sequences as in the initial amplification. The DIG-labeled product was used to screen a minimum of nine membranes (e4:5  105 phage plaques) from each of the E. formosa and E. pergandiella libraries using the buffers and procedures described in the Genius detection system (Boehringer Mannheim). Positive plaques were identified using NBT-BCIP colorimetric detection reagents supplied with the Genius kit. These plaques were cored from the original agar plates and the vector containing putative genomic sequence excised and grown in XLOLR cells (Stratagene). The vector inserts were digested with a series of restriction enzymes. The products of these digests were subcloned into a pBluescript SK (+) phagemid vector (Stratagene) and sequenced using a combination of priming sites from within the multiple cloning site of the vector and primer walking with primers designed from sequence data provided from initial sequencing operations. All sequencing was conducted at the sequencing facility at the University of Arizona using an automated sequencer (Model 373, Applied Biosystems, Foster City, CA).

953

2.4. Analysis of Vg gene sequence The gene sequence obtained from the E. formosa clones was assembled into a contiguous unit in the Faktory sequence assembly program (Miller and Myers, 1999; http://www.cs.arizona.edu/research/reports.html). The contiguous sequence was analyzed for structural characteristics including a transcriptional start site, open reading frame and the presence of introns. Putative intron positions in the gene were identified by scrutinizing regions of the gene containing unusually high levels of adenine and thymine residues and those where discontinuities in alignments with cDNA sequences of other wasps were observed. Intron positions were confirmed by RT-PCR using RNA isolated from newly emerged adults and oligonucleotide primers designed to anneal to regions on either side of the putative intron sites. The cDNA fragments generated with this procedure were analyzed on a 0.8% agarose gel opposite PCR products generated using the same oligonucleotide primers in the presence of genomic DNA. The RT-PCR products were excised from the gel and purified using a Sephaglas BandPrep kit (Pharmacia) then sequenced to verify intron splice sites. The positions of introns in the E. formosa gene were compared to those characterized in the full-length Vg genes of other insects and the African clawed frog, Xenopus laevis (see accession numbers below). The Vg cDNA sequence of the honeybee, Apis mellifera, identified by Piulachs et al. (2003) was used as a virtual probe in a BLAST search of the latest release of sequence data from the Honeybee Genome Sequencing Project (Baylor College of Medicine). A contiguous sequence assembly containing what appears to be the full-length Vg gene of A. mellifera was identified (gnl|Amel_1.1|Contig2354). The positions of introns in this gene were inferred from discontinuities in the alignment between the probe and genomic sequences and included in the analysis. Two alignments were prepared between the deduced amino acid sequence of the E. formosa Vg gene and those obtained from both insect and non-insect sources (see below). The first alignment was constructed with the full-length gene sequences, less their signal peptides, in the ClustalW program (Thompson et al., 1994) with a gap penalty of 10.0 and a gap extension penalty of 0.20. The second alignment was constructed by hand using only the more conserved regions of the same sequences. These regions were defined by Sappington et al. (2002) in a recent alignment of Vg and Vg-related genes. Phylogenetic trees were constructed from both alignments using a neighbor-joining algorithm and support for the branches determined by 1000 iterations of a bootstrap routine (DNASTAR, Inc.). Branches that were not supported by at least 50% of the bootstrap

954

D.M. Donnell / Insect Biochemistry and Molecular Biology 34 (2004) 951–961

iterations were collapsed into a polytomy. The E. formosa Vg gene sequence was aligned with the following insect sources (reference and accession numbers in parentheses): Aedes aegypti (Chen et al., 1994; U02548; Romans et al., 1995; genomic sequence: L41842), Anopheles gambiae (AF281078), Antheraea pernyi (AB049631), Antheraea yamamai (AB055843), Anthonomous grandis (Trewitt et al., 1992; M72980), Apis mellifera (Piulachs et al., 2003; AJ517411), Athalia rosae (Kageyama et al., 1994; Nose et al., 1997; AB007850), Blattella germanica (Comas et al., 2000; AJ005115), Bombyx mandarina (AB055845), Bombyx mori (Yano et al., 1994a, b; D13160; genomic sequence: D30732, D30733), Graptopsaltria nigrofuscata (Lee et al., 2000b; AB026848), Leucophaea madera (Tufail and Takeda, 2002; AB052640), Lymantria dispar (Hiremath and Lehtoma, 1997; U90756), Periplaneta americana (Tufail et al., 2000, 2001; Vg 1: AB034804, Vg 2: AB047401), Pimpla nipponica (Nose et al., 1997; AF026789), Plautia stali (Lee et al., 2000a; Vg 1: AB033498; Vg 2: AB033499; Vg 3: AB033500), Riptortus clavatus (Hirai et al., 1998; U97277), Samia cynthia (AB055844), Solenopsis invicta (AF512520). The following non-insect sources of Vg gene sequences were included in these alignments (identification, reference and accession numbers in parentheses): Caenorhabditis elegans (nematode; Spieth et al., 1985; Vg 5: X03044), Cherax quadricarinatus (crayfish, Abdu et al., 2002; AF306784), Ichthyomyzon unicuspus (lamprey; Sharrock et al., 1992; M88749), Metapenaeus ensis (shrimp, Tsang et al., 2003; AF548364), Penaeus japonicus (prawn, Tsutsui et al., 2000; BAB01568), Xenopus laevis (frog; Gerber-Huber et al., 1987; VgA2: Y00354). The position of the signal peptide in the E. formosa Vg was predicted using the SignalIP computer program (Nielsen et al., 1997). The mass, less the signal peptide, and hydrophilicity profile of the protein were determined from the amino acid sequence using the Protean program (DNASTAR, Inc.). 2.5. SDS-PAGE Sodium dodecyl sulfate-polyacrylimide gel electrophoresis (SDS-PAGE) was used to analyze Vg use in a number of parasitoid species within the genus Encarsia. Whole body homogenates were prepared from 10 newly eclosed adults representing each of the following groups: female E. formosa, E. pergandiella, E. sophia, E. luteola and male E. luteola. Ovary homogenates were prepared from newly eclosed E. formosa and E. pergandiella females. Earlier investigations suggested E. pergandiella ovaries contained markedly lower levels of protein than E. formosa ovaries. Thus, ovaries were dissected from 60 E. pergandiella females and 20 E. formosa females. Ovaries were isolated in Tris-buffered saline (20 mM Tris, 150 mM NaCl, 5 mM EDTA, pH

7.5) containing a mixture of protease inhibitors (leupeptin, antipain, chymostatin and aprotinin at 17 lg/ml each, pepstatin A at 1.7 lg/ml and 1 mM 4-(2-aminoethyl) benzenesulfonyl fluoride (AEBSF)), then frozen v in liquid nitrogen and stored at 20 C until analysis. Just prior to analysis, the adult wasps and ovary sets were homogenized using Kontes pellet pestles (VWR Scientific) in 1.5 ml microcentrifuge tubes containing sample loading buffer. Samples were centrifuged at v 12; 000g for 10 min at 4 C and the supernatants then transferred to new tubes. Samples were boiled for 3 min then loaded onto 6–15% gradient slab gels containing SDS. High molecular weight marker standards (BioRad) were also loaded onto the gels. SDS-PAGE was carried out according to the method of Laemmli (1970) with modification for the use of gradient gels. Gels were stained with Coomasie blue and destained in a series of methanol/acetic acid solutions (Wheeler and Buck, 1995) before being photographed. An egg from the ant, Camponotus festinatus, was processed in the same manner as the adult wasps and analyzed alongside the whole-body homogenates to provide an example of an uncleaved, apocritan vitellin. The Vg of C. festinatus was identified by Martinez and Wheeler (1991).

3. Results and discussion 3.1. Isolation of a Vg gene from E. formosa A cDNA product consisting of 1641 nucleotides was isolated from E. formosa mRNA using the 30 -RACE technique with a poly(dT) oligonucleotide and degenerate oligonucleotides designed to anneal to a conserved region near the 30 end of hymenopteran Vg genes. The product was subcloned and tentatively identified as Vg gene sequence following a comparison with Vg sequences available in GenBank. The product was DIGlabeled and used to screen k phage libraries constructed from E. formosa and E. pergandiella genomic DNA. An estimated 4:5  105 phage plaques of each species were exposed to the probe during an initial screening (5–7 kb average insert size). The probe hybridized with three clones in the E. formosa library. These clones were bidirectionally sequenced and the resulting data compiled into a single contiguous sequence after the overlapping regions of the clones were found to align with 100% homology. The contiguous sequence consisted of 6594 nucleotides (Fig. 1). A BLAST search (Altschul et al., 1997) indicated that the sequence contained a complete Vg gene. The 30 -RACE technique failed to yield a product from E. pergandiella cDNA preparations. Further, the E. formosa DIG probe failed to hybridize with any clones from the E. pergandiella genomic library despite repeated screenings (approximately 1:2  106 phage plaques screened in total).

D.M. Donnell / Insect Biochemistry and Molecular Biology 34 (2004) 951–961

955

Fig. 1. Genomic sequence containing the Vg gene of E. formosa. Nucleotide positions are shown to the right of each line using the first base of the putative transcription initiation site (arrow) as the starting point. Residue positions in the deduced amino acid sequence of the coding region are shown to the left of each line using the inferred translation initiation methionine as the starting point. Noncoding sequence is shown in lower case and the introns are numbered. The TATA, CAAT and initiation sites are underlined as well as the polyadenylation signal. Subtilisin-like convertase recognition motifs (R-X-X-R) are boxed.

956

D.M. Donnell / Insect Biochemistry and Molecular Biology 34 (2004) 951–961

Fig. 2. Intron alignments for Vg genes of the African clawed frog, Xenopus laevis (Xla), and seven insect species for which complete genomic sequences are available: the moths, Bombyx mori (Bmo) and Lymantria dispar (Ldi), the mosquitoes, Aedes aegypti (Aae) and Anopheles gambiae (Aga), the boll weevil, Anthonomous grandis (Agr), the honeybee, Apis mellifera (Ame), and the parasitoid wasp, Encarsia formosa (Efo) (references and accession numbers are provided in Section 2). The intron positions in the A. mellifera Vg gene were inferred from an alignment of the Vg cDNA identified for this species (Piulachs et al., 2003), with sequence obtained from the honeybee genome sequencing project (see Section 2). Numerals to the left of each sequence indicate the position of the first residue in the deduced amino acid sequence relative to the initiator methionine. Intron positions are underlined. In instances where two residues are separated by an intron, both are underlined. Intron numbers are indicated in brackets while a dash [–] indicates that an intron is absent in the region shown. When an intron is present in a region its length is indicated in parentheses at the end of each sequence. Bold type indicates positions in which all species use the same residue. An asterisk (
) under the alignment indicates a position in which the majority of the insect species use the same residue. A cross (+) below the alignment indicates a position in which X. laevis and at least three of the insect species use the same residue.

These findings suggest that the E. formosa genome contains a single Vg gene and that the E. pergandiella genome no longer contains a gene for this protein. 3.2. Organization of the E. formosa Vg gene The E. formosa sequence includes a putative transcription initiation site, Vg gene and polyadenylation signal (Fig. 1). The transcript initiation site was inferred using the consensus sequence for transcription initiation by RNA polymerase II (RNAPII; Breathnach and Chambon, 1981; Bucher, 1990). A TATA box begins 32 bases upstream of the putative transcription site and a CAAT site (Bucher, 1990) begins 68 bases further upstream. Unusually high levels of adenine and thymine residues in parts of the genomic sequence were used to infer the position of six introns in the gene. The intron positions were confirmed by

RT-PCR using RNA obtained from whole animal homogenates. The introns ranged in size from 69 to 91 nucleotides (Figs. 1 and 2). The consensus donor and acceptor intron splice junctions adhere to the ‘‘GTAG’’ rule (50 -GT-AG-30 ), characteristic of genes transcribed by RNAPII (Breathnach and Chambon, 1981). Following the identification of introns, the gene was found to code for a Vg consisting of 1814 amino acids. The E. formosa Vg contains four R-X-X-R amino acid motifs identified by Barr (1991) as recognition sites for subtilisin-like convertases (Fig. 1). The first of these sites lies in a hydrophilic region that in other insect Vgs is thought to be exposed on the surface of the folded protein (Chen et al., 1997). Most of the insect Vgs studied to date are cleaved following convertase recognition sites in this region (Sappington et al., 2002). Denaturing gel electrophoresis suggests, however, that the E. formosa Vg is stored uncleaved

D.M. Donnell / Insect Biochemistry and Molecular Biology 34 (2004) 951–961

(Fig. 4, see Section 3.4). Indeed, Vg cleavage appears rare within the apocritan Hymenoptera, having been confirmed only within the Ponerinae (Wheeler et al., 1999). The Vg of the ichneumonid, P. nipponica, is thought to be refractory to cleavage because the R-XX-R site in the hydrophilic region is followed by an aliphatic residue (Sappington et al., 2002). The A. mellifera gene lacks a dibasic motif in this region (see Fig. A.1) and is also uncleaved (Wheeler and Kawooya, 1990). Multiple Vg bands have been observed on denaturing gel studies of S. invicta extracts, but these are thought to represent the products of multiple Vg genes rather than subunits of a single, cleaved protein (Lewis et al., 2001). An obvious characteristic of most non-apocritan insect Vgs is the presence of extensive polyserine tracts proximal to the cleavage site in the hydrophilic region (see Fig. A.1). Although most apocritans contain higher levels of serine in this region than elsewhere in their Vgs, polyserine tracts exceeding the five contiguous residues observed in the E. formosa sequence are absent. These tracts were once thought to facilitate cleavage of Vgs by convertases (Chen et al., 1997), but more recent analyses suggest their role in cleavage is unknown (Sappington and Raikhel, 1998; Sappington et al., 2002). With this study, there are now seven complete genomic sequences for insect Vg genes: the moths, B. mori and L. dispar (Yano et al., 1994b; Hiremath and Lehtoma, 1997), the mosquitoes, Ae. aegypti and An. gambiae (Romans et al., 1995), the weevil, A. grandis (Trewitt et al., 1992), the bee, A. mellifera, and the parasitoid, E. formosa. Each of the genes contains six introns except for those of the mosquitoes, which contain only two. The intron positions in the Vg gene of E. formosa were compared to those in the Vg genes of other insects and a Vg gene (VgA2) of the frog, X. laevis (Fig. 2). The intron positions in the A. mellifera sequence used in this analysis were inferred following a comparison of the Vg cDNA sequence identified for this species by Piulachs et al. (2003) with the latest release of genomic sequence from the Honeybee Genome Sequencing Project (see Section 2). All of the intron positions in the E. formosa gene correspond to those inferred in the A. mellifera gene. The first and fifth intron positions in the hymenopteran genes are shared by all of the full-length insect Vg genes characterized thus far and an additional intron position is shared with the gene of A. grandis (Fig. 2). The moths and A. grandis also share an intron that is not observed in the hymenopteran or mosquito sequences (alignment not shown). The significance of the two intron positions shared by all insect Vg genes has not been determined. The length of the conserved introns varies substantially (Fig. 2). The first intron varies from 70 nucleotides in Ae. aegypti to more than 4400 nucleo-

957

tides in B. mori. The length of the second conserved intron varies from 57 nucleotides in Ae. aegypti to 3016 nucleotides in L. dispar. Each of the introns observed in the hymenopteran Vg genes occur in positions proximal to or corresponding to introns in the X. laevis gene (Fig. 2). A previous comparison of intron positions from the Vg genes of A. grandis, C. elegans and X. laevis led Trewitt et al. (1992) to suggest that while substantial divergence was apparent between the sequences, conserved residues supported their alignment and indicated that a number of the introns in the vertebrate and invertebrate sequences shared a common ancestry. The significance of such comparisons has been called into question in the past because the large number of introns (34) observed in many vertebrate Vg genes increases the likelihood of chance intron alignments with invertebrate sequences (Spieth et al., 1991; Hagedorn et al., 1998). The pattern of intron loss and/or gain between insect orders illustrated here contrasts with the stability of introns observed in vertebrate Vg genes. Nardelli et al. (1987) noted that the positions of the 34 introns observed in the VgA2 gene of X. laevis aligned with those identified in the Vg of the chicken, Gallus gallus. More recently, Mouchel et al. (1997) noted the loss of a single intron position in a Vg gene from the rainbow trout, Oncorhynchus mykiss, relative to that of other vertebrate sequences. The Vg of E. formosa is similar to those of most other insects in that it contains approximately 50% essential amino acids and very little cysteine. Of the 13 cysteine residues coded for in the E. formosa Vg gene, one residue occurs in the putative signal peptide and the remainder are found in conserved positions near the C-terminus (see Figs. A.1 and A.2). The E. formosa Vg contains high levels of serine (11.3 %) despite the absence of the extensive polyserine tracts observed in the Vgs of most insects outside the Apocrita. The gene also codes for the highest level of lysine (9.5%) and lowest levels of phenylalanine (2.6%) and tyrosine (2.3%) observed in any insect Vg characterized to date. 3.3. Phylogenetic analysis A ClustalW alignment (Thompson et al., 1994) was conducted between the E. formosa Vg sequence and those of other insects as well as a number of non-insect organisms (see Fig. A.1). The E. formosa sequence is most similar to those of other hymenopterans: P. nipponica (38% identity), A. rosae and S. invicta (37%), and A. mellifera (36%). Discontinuities within the hymenopteran alignments exceeding five bases occur in regions where the A. rosae sequence contains extensive polyserine tracts and at the point where the S. invicta sequence terminates (the S. invicta sequence used in

958

D.M. Donnell / Insect Biochemistry and Molecular Biology 34 (2004) 951–961

Fig. 3. Vg phylogenies constructed from coding sequence alignments using all insect Vg genes characterized to date and those of a number of non-insect organisms (full species names and accession numbers are provided in Section 2). Both trees were constructed by the neighbor-joining method. The numbers to the right or above a node indicate the percentage of 1000 bootstrap iterations supporting the branch. Branch lengths are proportional to the number of changes assigned to each branch. (A) Tree constructed from an alignment of full-length gene sequences, less the signal peptides, using the ClustalW computer program (Thompson et al., 1994). The alignment is shown in Fig. A.1. (B) Tree constructed from an alignment by hand of only the more conserved regions of the Vg genes using the alignments of Sappington et al. (2002) as a guide. The alignment is shown in Fig. A.2.

this analysis consists of only 1641 amino acid residues). A phylogenetic tree was constructed from the aligned sequences using the neighbor-joining method (Fig. 3A).

Bootstrap support for most of the branches was generally approaching 100%. The analysis placed E. formosa opposite P. nipponica in a distinct clade within the

Fig. 4. Results of SDS-PAGE analysis. The first lane in each panel contains molecular weight markers (M) (45, 66.2, 97.4, 116.2 and 200 kDa). (A) Gel showing egg extracts from the ant, Camponotus festinatus (Cf) and whole-body homogenates from female Encarsia sophia (Es), female and male Encarsia luteola (El(m) and El(f), respectively), E. formosa (Ef) and E. pergandiella (Ep) wasps. The most abundant protein in the extracts of Cf (~185 kDa) is that of vitellin (Martinez and Wheeler, 1991). (B) Gel showing ovary homogenates of E. pergandiella (Ep) and E. formosa (Ef).

D.M. Donnell / Insect Biochemistry and Molecular Biology 34 (2004) 951–961

Hymenoptera. The tree does not show the distinction between the Holometabola and Hemimetabola obtained by Comas et al. (2000) but the difference could be due to the larger number of sequences used in the current analysis. The tree in Fig. 3A is similar to that recently proposed by Sappington et al. (2002) in that the Lepidoptera appear the most derived among the insects but differs in that the current phylogeny places the Dictyoptera between the Lepidoptera and remaining insect orders. It is clear from an analysis of the ClustalW alignment (Fig. A.1) that the highly variable regions of the Vg genes, such as those containing the polyserine tracts, are difficult to align. This led Sappington et al. (2002) to construct phylogenies based only on the use of the more conserved regions of the Vg sequences. Thus, a second alignment was prepared using the same organisms as in the ClustalW alignment but with highly variable regions of the genes removed (Fig. A.2). This alignment was prepared by hand using the alignment of Sappington et al. (2002) as a guide and then used to construct a neighbor-joining tree. This tree was similar to that of the first in that the Lepidoptera appear the most derived within insects. In keeping with the findings of Sappington et al. (2002), however, while bootstrap support was sufficient to define intraordinal relationships, it was not sufficient (bootstrap support <50%) to reveal the branching sequence of most of the non-lepidopteran orders. This tree also resulted in a restructuring of the Hymenoptera, with P. nipponica and A. rosae forming one clade and E. formosa and the remaining apocritans forming another (Fig. 3B).

959

the eggs of the ant, C. festinatus, by Martinez and Wheeler (1991) (Fig. 4A). Analysis of the whole body and ovary extracts suggests that E. pergandiella does not use Vg in its provisioning strategy (Fig. 4). This finding is in keeping with the failure to isolate clones containing Vg gene sequence during the screening of the genomic library for this species (Section 3.1), and with an earlier microscopy study in which it was noted that the clearly yolk-rich eggs of E. formosa were approximately 6 times larger by volume than the transluscent eggs of E. pergandiella (Donnell and Hunter, 2002). The eggs of E. pergandiella are similar to other yolkless eggs in that they swell upon oviposition (Gerling, 1966). The capacity of E. pergandiella eggs to absorb large quantities of nutrients from the hemolymph of their host is evidenced by the considerably larger size of their larvae relative to those of E. formosa at the time of hatching (unpublished data). Despite the apparent adaptiveness of the egg provisioning strategy of E. pergandiella, numerous aphelinid parasitoids, including E. formosa, E. sophia, and E. luteola, continue to utilize yolk in their eggs. In the case of E. sophia, Vg is necessary because although females of this species are endoparasitic primary parasitoids, males develop as hyperparasitic ectoparasitoids on conspecific females or other primary parasitoids (Hunter and Kelly, 1998). Ectoparasitism of any sort is unusual in Encarsia (Hunter and Woolley, 2001), however, and is not observed in E. formosa or E. luteola. Thus, other factors are likely involved in the retention of yolk in the egg provisioning strategies of these parasitoids.

3.4. SDS-PAGE SDS-PAGE analysis of whole-body extracts from four species within the Encarsia genus revealed that the protein profiles of the wasps differ noticeably with respect to the presence or absence of a protein that is approximately 200 kDa in size (Fig. 4A). This protein is believed to be the product of the Vg gene. The protein is present in the extracts from female E. formosa, E. sophia and E. luteola and absent in those from female E. pergandiella and male E. luteola (Fig. 4A). The protein is clearly present in large quantities in the ovaries of E. formosa (Fig. 4B). Attempts to chemically determine the N-terminal amino acid sequence of this protein were unsuccessful suggesting that the N-terminus may be modified. This has not been confirmed nor have attempts yet been made to determine the identity of the protein by other means. The mass of this protein does, however, approximate that expected for the E. formosa Vg (203 kDa) in the absence of extensive modification of the amino acid residues. This is only slightly larger than the uncleaved vitellin identified in

Acknowledgements The author would like to thank Norm Buck, Jun Isoe and Kugao Oishi for sharing technical advice, Suzanne Kelly for maintaining the wasp cultures and Henry Hagedorn for sharing facilities and equipment. This work also benefited from helpful discussions with Reg Chapman, Henry Hagedorn, and Martha Hunter. The author was supported by funds from the Center for Insect Science and a NSF Research Training Grant to the University of Arizona Department of Ecology and Evolutionary Biology.

References Abdu, U., Davis, C., Khalaila, I., Sagi, A., 2002. The vitellogenin cDNA of Cherax quadricarinatus encodes a lipoprotein with calcium binding ability, and its expression is induced following the removal of the androgenic gland in a sexually plastic system. Gen. Comp. Endocrinol. 127 (3), 263–272.

960

D.M. Donnell / Insect Biochemistry and Molecular Biology 34 (2004) 951–961

Altschul, S.F., Madden, T.L., Scha¨ffer, A.A., Zhang, J., Zhang, Z., Miller, W., Lipman, D.J., 1997. Gapped BLAST and PSIBLAST: a new generation of protein database search programs. Nucleic Acids Res. 25, 3389–3402. Barr, P.J., 1991. Mammalian subtilisins: the long-sought dibasic processing endoproteases. Cell 66, 1–3. Bownes, M., Hames, B.D., 1977. Accumulation and degradation of 3 major yolk proteins in Drosophila melanogaster. J. Exp. Zool. 200 (1), 149–156. Breathnach, R., Chambon, P., 1981. Organization and expression of eucaryotic split genes coding for proteins. Annu. Rev. Biochem. 50, 349–383. Bucher, P., 1990. Weight matrix descriptions of four eukaryotic RNA polymerase II promoter elements derived from 502 unrelated promoter sequences. J. Mol. Biol. 212, 563–578. Byrne, B.M., Gruber, M., Ab, G., 1989. The evolution of egg yolk proteins. Prog. Biophys. Mol. Biol. 53 (1), 33–69. Chen, J.S., Cho, W.L., Raikhel, A.S., 1994. Analysis of mosquito vitellogenin cDNA. Similarity with vertebrate phosvitins and arthropod serum proteins. J. Mol. Biol. 237, 641–647. Chen, J.S., Sappington, T.W., Raikhel, A.S., 1997. Extensive sequence conservation among insect, nematode, and vertebrate vitellogenins reveals ancient common ancestry. J. Mol. Evol. 44, 440–451. Chen, T.T., Strahlendorf, P.W., Wyatt, G.R., 1978. Vitellin and vitellogenin from locusts (Loscusta migratoria)—properties and posttranslational modification in fat-body. J. Biol. Chem. 253 (15), 5325–5331. Chinzei, Y., Miura, K., Kobayashi, L., Shinoda, T., Numata, H., 1992. Cyanoprotein—developmental stage, sex and diapausedependent expression and synthesis regulation by juvenile hormone in the bean bug, Riptortus clavatus. Arch. Insect Biochem. Physiol. 21, 61–73. Chomczynski, P., Sacchi, N., 1987. Single-step method of RNA isolation by acid guanidinium thiocyanate phenol chloroform extraction. Anal. Biochem. 162 (1), 156–159. Comas, D., Piulachs, M., Belles, X., 2000. Vitellogenin of Blatella germanica (L.) (Dictyoptera, Blatellidae): nucleotide sequence of the cDNA and analysis of the protein primary structure. Arch. Insect Biochem. Physiol. 45, 1–11. Dhadialla, T.S., Raikhel, A.S., 1990. Biosynthesis of mosquito vitellogenin. J. Biol. Chem. 265 (17), 9924–9933. Donnell, D.M., Hunter, M.S., 2002. Developmental rates of two congeneric parasitoids, Encarsia formosa and E. pergandiella (Hymenoptera: Aphelinidae), utilizing different egg provisioning strategies. J. Insect Physiol. 48, 487–493. Ferkovich, S.M., Dillard, C.R., 1986. A study of uptake of radiolabled host proteins and protein synthesis during development of eggs of the endoparasitoid, Microplitis croceipes (Cresson) (Braconidae). Insect Biochem. 16 (2), 337–345. Fisher, R.C., 1971. Aspects of the physiology of endoparasitic Hymenoptera. Biol. Rev. 46, 243–278. Flanders, S.E., 1942. Oosorption and ovulation in relation to oviposition in the parasitic Hymenoptera. Ann. Entomol. Soc. Am. 35, 251–266. Fourney, R.M., Pratt, G.F., Harnish, D.G., Wyatt, G.R., White, B.N., 1982. Structure and synthesis of vitellogenin and vitellin from Calliphora erythrocephala. Insect Biochem. 12 (3), 311–321. Gerber-Huber, S., Nardelli, D., Haefliger, J.A., Cooper, D.N., Givel, F., Germond, J.E., Engel, J., Green, N.M., Wahli, W., 1987. Precursor-product relationship between vitellogenin and the yolk proteins as derived from the complete sequence of a Xenopus vitellogenin gene. Nucleic Acids Res. 15 (12), 4737– 4760. Gerling, D., 1966. Studies with whitefly parasites of Southern California. I. Encarsia pergandiella Howard (Hymenoptera: Aphelinidae). Can. Entomol. 98, 707–724.

Hagedorn, H.H., Kunkel, J.G., 1979. Vitellogenin and vitellin in insects. Annu. Rev. Entomol. 24, 475–505. Hagedorn, H.H., Maddison, D.R., Tu, Z., 1998. The evolution of vitellogenins, cyclorrhaphan yolk proteins and related molecules. Adv. Insect Physiol. 27, 235–284. Harnish, D., White, B., 1982. An evolutionary model for the insect vitellins. J. Mol. Evol. 18, 405–413. Hirai, M., Watanabe, D., Kiyota, A., Chinzei, Y., 1998. Nucleotide sequence of vitellogenin mRNA in the bean bug, Riptortus clavatus: analysis of processing in the fat body and ovary. Insect Biochem. Mol. Biol. 28, 537–547. Hiremath, S., Lehtoma, K., 1997. Structure of the gypsy moth vitellogenin gene. Arch. Insect Biochem. Physiol. 36, 151–164. Hunter, M.S., Kelly, S.E., 1998. Hyperparasitism by an exotic autoparasitoid: secondary host selection and the window of vulnerability of conspecific and native heterospecific hosts. Entomol. Exp. Appl. 89 (3), 249–259. Hunter, M.S., Woolley, J.B., 2001. Evolution and behavioral ecology of heteronomous aphelinid parasitoids. Annu. Rev. Entomol. 46, 251–290. Ivanova-Kasas, O.M., 1972. Polyembryony in insects. In: Counce, S.J., Waddington, C.H. (Eds.), Developmental Systems: Insects, vol. 1. Academic Press, New York, pp. 243–271. Kageyama, Y., Kinoshita, T., Umesono, Y., Hatekeyama, M., Oishi, K., 1994. Cloning of cDNA for vitellogenin of Athalia rosae (Hymenoptera) and characterization of the vitellogenin gene expression. Insect Biochem. Mol. Biol. 24, 599–605. King, P.E., Ratcliffe, N.A., Fordy, M.R., 1971. Oogenesis in a braconid, Apanteles glomeratus (L.) possessing an hydropic type of egg. Z. Zellforsch. Mik. Ana. 119, 43–57. Kunkel, J.G., Nordin, J.H., 1985. Yolk proteins. In: Kerkut, G.A., Gilbert, L.I. (Eds.), Comprehensive Insect Physiology, Biochemistry and Pharmacology, vol. 1. Pergamon Press, New York, pp. 83–111. Laemmli, U.K., 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680–685. Le Ralec, A., 1995. Egg contents in relation to host-feeding in some parasitic Hymenoptera. Entomophaga 40 (1), 87–93. Lee, J.M., Hatakeyama, M., Oishi, K., 2000a. A simple and rapid method for cloning insect vitellogenin cDNAs. Insect Biochem. Mol. Biol. 30 (3), 189–194. Lee, J.M., Nishimori, Y., Hatakeyama, M., Bae, T., Oishi, K., 2000b. Vitellogenin of the cicada Graptosaltria nigrofuscata (Homoptera): analysis of its primary structure. Insect Biochem. Mol. Biol. 30, 1–7. Lewis, D.K., Campbell, J.Q., Sowa, S.M., Chen, M., Vinson, S.B., Keeley, L.L., 2001. Characterization of vitellogenin in the red imported fire ant, Solenopsis invicta (Hymenoptera: Apocrita: Formicidae). J. Insect Physiol. 47, 543–551. Martinez, T., Wheeler, D., 1991. Identification of vitellogenin in the ant, Camponotus festinatus. changes in hemolymph proteins and fat body development in workers. Arch. Insect Biochem. Physiol. 17, 143–155. Miller, S.J., Myers, E., 1999. The FAKtory DNA Sequence Fragment Assembly System. TR99-03. Department of Computer Science, University of Arizona, Tucson, AZ, USA. Mouchel, N., Trichet, V., Naimi, B.Y., Le Pennec, J., Wolff, J., 1997. Structure of a fish (Oncorhynchus mykiss) vitellogenin gene and its evolutionary implication. Gene 197, 147–152. Nardelli, D., Gerber-Huber, S., van het Schip, F.D., Gruber, M., Ab, G., Wahli, W., 1987. Vertebrate and nematode genes coding for yolk proteins are derived from a common ancestor. Biochemistry 26, 6397–6402. Nielsen, H., Engelbrecht, J., Brunak, S., von Heijne, G., 1997. Identification of prokaryotic and eukaryotic signal peptides and prediction of their cleavage sites. Protein Eng. 10 (1), 1–6.

D.M. Donnell / Insect Biochemistry and Molecular Biology 34 (2004) 951–961 Nose, Y., Lee, J.M., Ueno, T., Hatakeyama, M., Oishi, K., 1997. Cloning of cDNA for vitellogenin of the parasitoid wasp, Pimpla nipponica (Hymenoptera: Apocrita: Ichneumonidae): vitellogenin primary structure and evolutionary considerations. Insect Biochem. Mol. Biol. 27 (12), 1047–1056. Piulachs, M.D., Guidugli, K.R., Barchuk, A.R., Cruz, J., Simoes, Z.L.P., Belles, X., 2003. The vitellogenin of the honey bee, Apis mellifera: structural analysis of the cDNA and expression studies. Insect Biochem. Mol. Biol. 33, 459–465. Quicke, D.L.J., 1997. Parasitic Wasps. Cambridge University Press, London. Raikhel, A.S., Dhadialla, T.S., 1992. Accumulation of yolk proteins in insect oocytes. Annu. Rev. Entomol. 37, 217–251. Romans, P., Zu, Z., Ke, Z., Hagedorn, H.H., 1995. Analysis of a vitellogenin gene of the mosquito, Aedes aegypti and comparisons to vitellogenins from other organisms. Insect Biochem. Mol. Biol. 25, 939–958. Sappington, T.W., 2002. The major yolk proteins of higher Diptera are homologs of a class of minor yolk proteins in Lepidoptera. J. Mol. Evol. 55, 470–475. Sappington, T.W., Oishi, K., Raikhel, A.S., 2002. Structural characteristics of insect vitellogenins. In: Adiyodi, K.G., Adiyodi, R.G. (Eds.), Reproductive Biology of Invertebrates, vol. 12, Part A, pp. 69–101. Sappington, T.W., Raikhel, A.S., 1998. Molecular characteristics of insect vitellogenins and vitellogenin receptors. Insect Biochem. Mol. Biol. 28, 277–300. Sharrock, W.J., Rosenwasser, T.A., Gould, J., Knott, J., Hussey, D., Gordon, J.I., Banaszak, L., 1992. Sequence of lamprey vitellogenin. Implications for the lipovitellin crystal structure. J. Mol. Biol. 226 (3), 903–907. Spieth, J., Denison, K., Zucker, E., Blumenthal, T., 1985. The nucleotide sequence of a nematode vitellogenin gene. Nucleic Acids Res. 13 (19), 7129–7138. Spieth, J., Nettleton, M., Zucker-Aprison, E., Lea, K., Blumenthal, T., 1991. Vitellogenin motifs conserved in nematodes and vertebrates. J. Mol. Evol. 32, 429–438. Telfer, W.H., 2002. Insect yolk proteins: a progress report. In: Adiyodi, K.G., Adiyodi, R.G. (Eds.), Reproductive Biology of Invertebrates, vol. 12, Part A, pp. 29–67. Thompson, J.D., Higgins, D.G., Gibson, T.J., 1994. ClustalW: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22, 4673–4680.

961

Trewitt, P.M., Heilmann, L.J., Degrugillier, S.S., Kumaran, A.K., 1992. The boll weevil vitellogenin gene: nucleotide sequence, structure, and evolutionary relationship to nematode and vertebrate vitellogenin genes. J. Mol. Evol. 34, 478–492. Tsang, W.S., Quackenbush, L.S., Chow, B.K., Tiu, S.H., He, J.G., Chan, S.M., 2003. Organization of the shrimp vitellogenin gene: evidence of multiple genes and tissue specific expression by the ovary and hepatopancreas. Gene 303 (1–2), 99–109. Tsutsui, N., Kawazoe, I., Ohira, T., Jasmani, S., Yang, W.J., Wilder, M.N., Aida, K., 2000. Molecular characterization of a cDNA encoding vitellogenin and its expression in the hepatopancreas and ovary during vitellogenesis in the kuruma prawn, Penaeus japonicus. Zool. Sci. 17, 651–660. Tufail, M., Takeda, M., 2002. Vitellogenin of the cockroach, Leucophaea maderae: nucleotide sequence, structure and analysis of processing in the fat body and oocytes. Insect Biochem. Mol. Biol. 32, 1469–1476. Tufail, M., Lee, J.M., Hatakeyama, M., Oishi, K., Takeda, M., 2000. Cloning of vitellogenin cDNA of the American cockroach, Periplaneta americana (Dictyoptera), and its structural and expression analyses. Arch. Insect Biochem. Physiol. 45, 37–46. Tufail, M., Hatakeyama, M., Takeda, M., 2001. Molecular evidence for two vitellogenin genes and processing of vitellogenins in the American cockroach, Periplaneta americana. Arch. Insect Biochem. Physiol. 48, 72–80. Wahli, W., 1988. Evolution and expression of vitellogenin genes. Trends Genet. 4 (8), 227–232. Wang, X., Cole, K.D., Law, J.H., 1988. cDNA cloning and deduced amino acid sequence of microvitellogenin, a female specific hemolymph and egg protein from the tobacco hornworm, Manduca sexta. J. Biol. Chem. 263 (18), 8851–8855. Wheeler, D.E., Buck, N.A., 1995. Storage proteins in ants during development and colony founding. J. Insect Physiol. 41, 885–894. Wheeler, D.E., Kawooya, J.K., 1990. Purification and characterization of honey bee vitellogenin. Arch. Insect Biochem. Physiol. 14, 253–267. Wheeler, D., Liebig, J., Holldobler, B., 1999. Atypical vitellins in ponerine ants (Formicidae: Hymenoptera). J. Insect Physiol. 45, 287–293. Yano, K., Sakurai, M.T., Watabe, S., Izumi, S., Tomino, S., 1994a. Structure and expression of mRNA for vitellogenin in Bombyx mori. Biochim. Biophys. Acta 1218 (1), 1–10. Yano, K., Sakurai, M.T., Izumi, S., Tomino, S., 1994b. Vitellogenin gene of the silkworm, Bombyx mori: structure and sex-dependent expression. FEBS Lett. 356 (2–3), 207–211.