|
Advertisement | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
J. Biol. Chem., Vol. 278, Issue 43, 41954-41962, October 24, 2003
Tissue- and Stage-specific Expression of Two Lipophorin Receptor Variants with Seven and Eight Ligand-binding Repeats in the Adult Mosquito*![]() ![]() **![]() ![]()
From the
Received for publication, July 28, 2003 , and in revised form, August 13, 2003.
We identified two splice variants of lipophorin receptor (LpR) gene products specific to the mosquito fat body (AaLpRfb) and ovary (AaLpRov) with respective molecular masses of 99.3 and 128.9 kDa. Each LpR variant encodes a member of the low density lipoprotein receptor family with five characteristic domains: 1) ligand recognition, 2) epidermal growth factor precursor, 3) putative O-linked sugar, 4) single membrane-spanning domains, and 5) the cytoplasmic tail with a highly conserved internalization signal FDNPVY. Proposed phylogenetic relationships among low density lipoprotein receptor superfamily members suggest that the LpRs of insects are more closely related to vertebrate low density lipoprotein receptors and very low density lipoprotein receptor/vitellogenin receptor than to insect vitellogenin receptor/yolk protein receptors. Two mosquito LpR isoforms differ in their amino termini, the ligand-binding domains, and O-linked sugar domains, which are generated by differential splicing. Polymerase chain reaction and Southern blot hybridization analyses show that these two transcripts originated from a single gene. Significantly, the putative ligand-binding domain consists of seven and eight complement-type, cysteine-rich repeats in AaLpRfb and AaLRov, respectively. Seven cysteine-rich repeats in AaLpRfb are identical to the second through eighth repeats of AaLpRov. Previous analyses (1) have indicated that the AaLpRov transcript is present exclusively in ovarian germ-line cells, nurse cells, and oocytes throughout the previtellogenic and vitellogenic stages, with the peak at 2430 h after blood meal, coincident with the peak of yolk protein uptake. In contrast, the fat body-specific AaLpRfb transcript expression is restricted to the postvitellogenic period, during which yolk protein production is terminated and the fat body is transformed to a storage depot of lipid, carbohydrate, and protein.
In insects, lipophorin (Lp)1 is the main transport vehicle, delivering lipids through the hemolymph to various organs. A characteristic feature of Lp is the selective mechanism by which it shuttles its lipid cargo between cells without concomitant degradation of the protein matrix of the Lp particle (24). The fat body, which is an insect analog of vertebrate liver and adipose tissue combined, plays a key role in lipid metabolism by being the site of both lipid storage and mobilization. The loading and unloading of lipids into and from fat body cells is accomplished by a shuttle mechanism involving Lp and a multi-protein complex, called a lipid transfer particle (24). Despite the fact that lipid transfer occurs on the fat body cell surface without apparent internalization of transfer proteins, specific Lp receptors have been described in the larval fat body of several insects (59). The larval fat body cells have been shown to internalize high density lipophorin (HDLp) by means of receptor-mediated endocytosis, suggesting HDLp turnover between the fat body and the hemolymph (6, 8). Moreover, the locust lipophorin receptor (LpR), which has recently been cloned (10), has been proposed to be the endocytotic receptor for HDLp in the Locusta fat body (11). However, it is unclear whether or not the LpR plays any role in the lipid shuttling mechanism (3, 4). Structurally, the Locusta LpR (LmLpR) is a homologue of the vertebrate very low density lipoprotein receptor (VLDLR) and the chicken vitellogenin receptor (VgR), containing the putative ligand recognition domain with eight complement-type, cysteine-rich repeats (10).
Lp plays a dual role in insect vitellogenesis, shuttling precursors from the fat body to the ovaries for the deposition of lipid yolk droplets and, in some species, becoming one of the constituents of the protein yolk bodies (1214). In insects, most yolk protein precursors are synthesized in and secreted from the fat body, and are transported as hemolymph proteins to developing oocytes where they are internalized via receptor-mediated endocytosis (15, 16). Receptor-mediated endocytosis, an essential process in all eukaryotes, is required for general cellular functions, including uptake of large molecules and recycling of membranes and membrane proteins (17). Uptake of yolk protein precursors by developing oocytes is a dramatic example of the receptor-mediated endocytosis pathway in oviparous animals (18). Recently, we cloned and characterized an ovarian LpR from the yellow fever mosquito, Aedes aegypti (AaLpRov) (1). It is homologous to the LmLpR and vertebrate VLDLR. The AaLpRov transcript is present exclusively in ovarian germ-line cells, nurse cells, and oocytes throughout the previtellogenic and vitellogenic stages, with the peak at 2430 h post blood meal (PBM), coincident with the peak of yolk protein uptake (1). In this paper, we describe the cloning and characterization of the new isoform of LpR (AaLpRfb) from the fat body of the adult female mosquito. For the most part, the fat body AaLpRfb is identical to the ovarian AaLpR (AaLpRov), reported previously (1). Surprisingly, the ligand-binding domain of the AaLpRfb is composed of seven tandem copies of a 40-amino acid, cysteine-rich repeat, similar to the mammalian low density lipoprotein receptor (LDLR). AaLpRfb is also different from AaLpRov in having a short O-linked sugar domain. Moreover, the two isoforms of AaLpR that arise via combinations of tissue-specific 5'-exon splicing and in-frame deletion show entirely different expression patterns during the vitellogenesis of an adult female mosquito.
InsectsThe yellow fever mosquitoes, A. aegypti, were maintained in laboratory culture as described elsewhere (19). Adults were provided with water and a 10% sucrose solution. Vitellogenesis was initiated by feeding females, 35 days after eclosion, with a blood meal on rats.
Isolation of Mosquito LpR cDNA and Genomic ClonesA cDNA fragment of the mosquito LpR was first amplified from the Reverse Transcription (RT)- and 5'-Rapid Amplification of cDNA Ends (RACE) PCRTotal RNA was extracted from the female fat body and ovary 36 h PBM using RNeasy mini kits (Qiagen) according to the instructions from the manufacturer. RT-PCR was performed using the Titan One-Step RT-PCR kit (Roche Molecular Biochemicals) with samples of 0.2 µg of total RNA as templates. Tubes containing RNA and RNase inhibitor (1 unit/µl, Roche Molecular Biochemicals) were incubated for 30 min at 50 °C for RT reaction. Amplification conditions included rapid heating to 94 °C for 2 min, followed by 2530 cycles of 94 °C for 45 s, 55 °C for 30 s, and 68 °C for 1min [PDB] . To obtain the start region of the mosquito LpRfb open reading frame, 5'-RACE PCR was performed as described by the manufacturer (Invitrogen). Two gene-specific primers (A and B) were synthesized as antisense primers: A, 5'-TTCCGAATCCTCGTCGGAACC-3' (6773 amino acids); B, 5'-GTTGGCGCAGGTAAACTCATC-3' (123129 amino acids).
From 5 µg of fat body total RNA at 36 h PBM, single-stranded cDNA was synthesized using SuperScript reverse transcriptase (Invitrogen) and gene-specific primer B. After cDNA synthesis, the product was purified using a Glass MAX Spin Cartridge and tailed with dCTP and TdT. The tailed cDNA was amplified with the abridged anchor primer and nested gene-specific primer A, then re-amplified using the same primers. The re-amplified product was separated by means of agarose gel electrophoresis, and the Inverse PCRTo clone the 5'-flanking regions of the AaLpR gene, amplification by inverse PCR (28) was performed. Primer Inv1 (5'GAGGTTCACTTCATCCCGAATCG-3') and primer Inv2 (5'-ATGTTCGGAGGACACCTTGTG-3') were designed on the basis of the nucleotide sequence encoding the NH2-terminal end of AaLpRfb. As a template, genomic DNA from A. aegypti was completely digested with MspI and then circularized with T4 DNA ligase. Amplification by inverse PCR was performed, i.e. 35 cycles of denaturation (94 °C for 30 s), annealing (60 °C for 30 s), and extension (72 °C for 3 min) with primers Inv1 and Inv2. The inverse PCR products were ligated into the pCRII-TOPO vector (Invitrogen) and sequenced. Genomic DNA Isolation and Southern BlotGenomic DNA from ten mosquitoes was purified using the DNeasy tissue kit (Qiagen, Chatsworth, CA). Two-microgram aliquots of DNA were digested with the corresponding endonuclease; the DNA fragments were separated by electrophoresis in a 0.8% agarose gel, transferred to nitrocellulose filters, and hybridized with a DNA probe. Northern BlotTotal RNA was isolated from the mosquito fat body and ovary of different stages using the RNeasy mini kit (Qiagen) according to the instructions from the manufacturer. Total RNA (10 µg) was then subjected to agarose gel electrophoresis in the presence of formaldehyde. The RNA from the gel was capillary blotted onto an Immobilon Ny+ membrane (Millipore Corp., Bedford, MA), UV cross-linked, and hybridized as described by Sambrook et al. (29) using a 1.2-kb cDNA fragment of the LpRfb and other probes. Probes were labeled with 32P by random priming. The hybridization was carried out at 65 °C for 16 h with the labeled probe (2 x 106 cpm/ml) using Hybrisol II (Oncor, Gaithersburg, MD) containing 0.5 mg/ml denatured salmon sperm DNA. The blots were then washed twice in 1x SSC and 0.1% SDS for 30 min at room temperature, once in 0.1x SSC and 0.1% SDS for 10 min at 65 °C, then exposed to Kodak film. Co-immunoprecipitation Analysis of Ligand Binding of the in Vitro Expressed AaLpRThe full-length cDNAs of AaLpRov and AaLpRfb were subcloned into pBluescript II vectors (Stratagene) and expressed in the TNT-coupled reticulocyte lysate system according to the instructions from the manufacturer (Promega). Protein labeling was performed by incorporation of [35S]methionine during the translation. The binding activities of the expressed receptors with the mosquito Lp were analyzed using co-immunoprecipitation, as described previously (1).
The LpR from the Female Mosquito Fat Body Is a Seven-repeat LDLR-like ReceptorCloning of the LpR variant (LpRfb) from the female mosquito fat body was accomplished using a combination of PCR-based and cDNA library screening. The missing 5' end of the coding sequence was obtained using 5'-RACE PCR. The resultant cDNA clone of AaLpRfb comprised 900 bp in the 5'-untranslated region, a 2673-bp open reading frame encoding a protein of 891 amino acids with a calculated molecular mass of 99,303 Da, and 2108 bp in the 3'-untranslated region (Fig. 1). Restriction mapping and sequence analyses of the LpRfb (accession no. AY348869 [GenBank] ) and LpRov (accession no. AF355595 [GenBank] ) cDNAs indicated that they contained a large common region. In both cDNAs, the putative start codons (ATG) were preceded by several in-frame stop codons, indicating that the open reading frame was full-length in each clone. It seems likely that a single AaLpR gene encodes the two mosquito isoforms (Fig. 1). Although the entire insert of AaLpRfb was longer than that of AaLpRov, the open reading frame for AaLpRfb (2673 bp, 891 residues) was shorter than that of AaLpRov (3468 bp, 1156 residues) (Fig. 1).
According to the deduced amino acid sequences, both AaLpRs encoded by full-length cDNAs were members of the LDLR family (Figs. 1 and 2,2). In both proteins, there were five domains, characteristic of the low density lipoprotein-type receptors (3032). The two variant cDNAs in the mosquito shared identical sequences in the EGF homology, transmembrane, and cytoplasmic domains, but differed at their NH2 termini, in the number of ligand-binding repeats, and in the length of the O-linked sugar domain (Figs. 1 and 2,2).
Remarkably, the first domain encoding the putative ligand-binding domain consisted of seven cysteine-rich repeats in AaLpRfb in contrast to eight in AaLpRov. The number of cysteine-rich repeats in their putative ligand-binding domains was the most striking structural difference between AaLpRfb and AaLpRov. Indeed the AaLpRfb with only seven ligand-binding repeats (LR7) was similar to the human LDLR, whereas AaLpRov contained eight repeats (LR8) like LpRs of other insects (9, 10) and the vertebrate LR8 receptors (2025, 34). Although the two mosquito variants differed in the number of repeats in the ligand-binding domain, when the first repeat of AaLpRfb was aligned with the second of AaLpRov, the sequences in the rest of the domain were identical (Figs. 1 and 2,2). This suggests utilization of alternative 5'-regions in the transcripts encoding these mosquito LpR variants.
The second EGF-precursor domain was identical in both receptors and contained three EGF-precursor repeats and five of copies of the characteristic YWXD sequence, spaced by The third domain was located between the EGF precursor and the transmembrane domains and was rich in serine and threonine, providing multiple potential sites for O-linked carbohydrates. The AaLpRfb was also quite different from AaLpRov with respect to the length of the putative O-linked sugar domain: only 63 residues in AaLpRfb relative to 257 residues in AaLpRov (Figs. 1 and 2,2). To confirm the difference in the O-linked sugar domains of LpR variants, RT-PCR was performed using total RNA from the ovarian and fat body tissues with specific primers (Fig. 2,Fig. 2, Pr-4 and Pr-5) annealing to identical regions flanking these domains. The RT-PCR indeed resulted in amplified fragments of two different sizes of 189 and 771 bp for AaLpRfb and AaLpRov, respectively (Fig. 4B). The precise function of the O-linked sugar domain in LDLR family members is not known; however, it is the most divergent of the five domains, with many deletions and insertions having occurred in the LDLR (35). Variant isoforms of LDLR family members with or without the O-linked sugar domain have been reported in vertebrates (34, 36, 37). In laying hens, the LR8 expressed by the oocytes lacks the O-linked sugar domain (34), but somatic tissue cells express a non-spliced isoform containing this domain (37). The two variants show unequal distribution among various organs. For example, the LR8 isoform lacking the O-linked sugar domain is predominant in the ovary and testis (34, 37), whereas its counterpart is present in muscle and heart (38). Different ratios of two variants of human VLDLR mRNA have been found in various cell types and tissues in vitro (39).
The fourth domain was a single membrane-spanning hydrophobic stretch of 22 amino acids, which are also poorly conserved in locust LpR; only 41% of the amino acids in this domain of the mosquito LpR are identical to those in the locust LpR (data not shown). In general, transmembrane domains that serve only as anchors vary in length and show little sequence homology other than an absence of amino acids with charged side chains (3033). The cytoplasmic domain of the mosquito LpR, consisting of 58 amino acids, contained a well conserved signal (FDNPVY) responsible for targeting the LDLR to coated pits on the plasma membrane (40). This internalization signal has been conserved for 350 million years among vertebrate LDLRs (18), and it is apparent that this motif shared a common ancestor of vertebrates and insects. Interestingly, this internalization motif, which is present in both fat body and ovarian-specific variants of the mosquito LpR, is absent in insect ovarian VgRs/YPRs, which possess an alternative well conserved leucine internalization signal (41, 42). Thus, it appears that differences in internalization signals are not cell- or tissue-specific. Tissue-specific 5' Exon Splicing and In-frame Deletion Are Involved in Generation of Two AaLpR Transcripts from a Single GeneSouthern blot analysis was performed using genomic DNA from 10 mosquitoes and two tissue-specific probes: NH2 termini of AaLpRfb (Pr1-Pr3 region) and AaLpRov (Pr2-Pr3 region). The hybridization results with either probe showed a single band in a digest with different restriction enzymes (Fig. 3). This suggested that a single genomic copy of the LpR gene existed in A. aegypti. In the human, two variant forms of the LDLR (LDLR, VLDLR; ref. 43) were identified. The exon-intron organization of the gene was almost the same as that of the LDLR gene, except for an extra exon that encodes an additional repeat in the ligand-binding domain of the VLDLR. Although the structure and organization of the VLDLR gene is highly similar to the LDLR gene, the two genes were located on different chromosomes (43).
Two AaLpR isoforms seem to be generated by alternative splicing from a single gene. Probes designed from both isoform-specific sequences were used to screen the genomic library, and the AaLpR gene was isolated and partially sequenced. This genomic region contained the full portion of AaLpRov, except for the 5'-UTR, NH2 terminus, and 3'-UTR of AaLpRfb. To obtain the fat body-specific 5' exon, amplification by inverse PCR was performed. The sequences of inverse PCR product were connected with the known nucleotide sequences of AaLpRfb 5'-UTR and NH2 terminus. The inverse PCR product contained exon-intron junction sequences that revealed splice site. In Fig. 4, we show the structure of the AaLpR gene from partial sequence analysis of the 5' exon boundary and the O-linked domain that is alternatively spliced. Because these two alternative 5' exons are not expressed in single tissues, the fat body-specific exon was named exon 1f (E1f) and encoded the signal sequence, and the ovary-specific exon was named exon 1o (E1o) and encoded the signal sequence and extra cysteinerich repeat of the ligand-binding domain. These were found to reside in a single E1f of 1-kb nucleotides that lies 5' of the E1o (Fig. 4A). The size of the intron separating E1f and E1o failed to reveal it, because we focused on a specific splicing site. The transcript start site has not yet been determined for these two isoforms. Nevertheless, it is clear that transcription of the two AaLpR mRNA variants is regulated by two alternative promoters, which we designated distal (for AaLpRfb) and proximal (for AaLpRov). Data base mining with annotated Drosophila gene (CG31094, GenBankTM accession no. AE003753 [GenBank] ) and adjacent two annotated Anopheles gambiae genes (ebiP6309, accession no. EAA03795 [GenBank] ; agCP14677, accession no. EAA03681 [GenBank] ) revealed a high degree of amino acid homology and similar 5' first exon splicing pattern with AaLpR two isoforms. The AaLpRov was similar to the translated sequence of Anopheles gambiae (ebiP6309) in the structure of the first exon, which has an extra cysteine-rich repeat in the ligand-binding domain. The AaLpRfb was similar to the Drosophila gene in the structure of the first exon, which has a short repeat in the ligand-binding domain (Fig. 5). However, because LpR genes of these insects have not been analyzed in detail, existence of splice variants with eight and seven cysteine-rich repeats in Drosophila and Anopheles, respectively, cannot be ruled out.
Tissue specificity of alternative 5' end exon splicing and the possible correspondence of the two AaLpR isoforms to the fat body and ovary transcripts were examined using RT-PCR of RNA fractions from the fat body and ovary 30 h PBM. Three PCR primers were used: the reverse primer Pr3 (containing a sequence from exon 2 common to both the AaLpRfb and AaLpRov isoforms) and two forward primers, Pr1 (specific to the AaLpRfb isoform, derived from the E1f sequence) and Pr2 (specific to the AaLpRov isoform, derived from the E1o sequence). The hybridizing fragments of different expected sizes were detectable in the PCR amplification products obtained from the fat body and ovary. We showed that both E1f and E1o were expressed and that E1f and E1o were alternatively spliced to exon 2 (Fig. 4B). These results strongly suggest that there is tissue-specific usage of E1f and E1o. This is similar to the case with the mosquito clathrin heavy chain gene, in which each sex has been found to use a different 5' exon (44). In the course of these experiments, we also identified an additional splice acceptor site within exon 8 (E8). The PCR products using Pr4/Pr5 located to E8 revealed that the O-linked domain of AaLpRfb was generated by an in-frame deletion within this exon (Fig. 4B). The transcript of two isoforms had different sizes of 3'-UTR. The 3'-UTR of ovarian transcript was shorter than that of the fat body transcript. The sequences from the 3' end of E8 corresponded to common 3'-UTRs of two isoforms (Fig. 4A). It remains to be determined whether these differences in size of 3'-UTR were generated by alternative polyadenylation or alternative splicing of exons in the 3'-region. However, according to the sequence of the 3'-UTR of the two transcripts, there were two alternative polyadenylation signals and several putative consensus sites (ATTTA, which are thought to increase instability of mRNA; Refs. 45 and 46) located in this region. Eukaryotic gene expression is partly controlled by the rate of mRNA degradation, which is generally a function of regulatory sequences in the 3'-UTR (47). The longer 3'-UTR of the AaLpRfb transcript contains more potent consensus sequences (46, 48) that promote mRNA decay than does the shorter 3'-UTR of the AaLpRov transcript (data not shown). These data suggest that the ovarian AaLpR transcript is more stable than that of the fat body. Additionally, these data are in agreement with Northern blot analysis (Fig. 8), showing that the AaLpRov transcript was expressed and resistant to decay during all developmental stages.
Phylogenetic Relationship of AaLpR Variants with Other Members of the LDLR SuperfamilyTo identify the phylogenetic relationship of the newly discovered mosquito LDLR-like LpRfb with other members of the LDLR superfamily, we conducted distance-based amino acid analysis, limiting it to smaller receptors of the family with LR5 to LR13. In general, the obtained dendrogram revealed that there were three early divergent lineages, with the worm Caenorhabditis elegans REM-2 (LR5 receptor) being the earliest and most divergent. This dendrogram tree also indicated an early split between the insect VgR/YPR lineage (LR13 receptors) and that of insect LpRs and vertebrate LDLR/VLDLR (LR7 and LR8 receptors). Indeed, the high divergence between the mosquito LpRs and the mosquito VgR (18.3% (AaLpRov) and 22.0% (AaLpRfb) identity, respectively) suggests a distant common ancestor (Table I). In contrast, the insect LpRs shared a much more recent common ancestor with vertebrate LDLRs and VLDL/VgRs. The mosquito LpR bears 3338% amino acid identity with its vertebrate counterparts. Interestingly, AaLpRfb isolated from the adult female fat body was structurally less similar to the locust fat body LpR (10) than the mosquito ovarian variant, AaLpRov, in that the two latter receptors had eight ligand-binding repeats (Fig. 2,Fig. 2). The AaLpRfb also shared lower overall identity (62.3%) with LmLpR than AaLpRov (64.3%).
Thus, the insect LpRs likely share ancestral lineage with the vertebrate LDLR/VLDLRs. Based on amino acid fingerprint analyses, Sappington and Raikhel (31) previously proposed that seven-repeat LDLRs arose from an eight-repeat ancestor through the loss of the first module of the latter. This conclusion is supported by the direct observations of the mosquito LpR variants, as well as reconstruction of LpR phylogeny presented in this study (Figs. 1 and 4). Both Mosquito LpR Variants, with Either Seven or Eight Ligand-binding Repeats, Bind LpThe receptor proteins were expressed using their cDNAs in the TNT-coupled reticulocyte lysate system. The sizes of both AaLpRov and AaLpRfb isoforms were similar to those estimated from the deduced amino acid sequences of both receptor cDNAs: 140 and 95 kDa, respectively (Fig. 6, lanes 1 and 4).
Expressed and labeled receptor proteins were first incubated with Lp from vitellogenic fat bodies (14), and then reacted with anti-Lp-specific polyclonal antibodies. The resulting immuno-complexes were precipitated using agarose-conjugated protein A and separated by means of SDS-PAGE. The results indicated that AaLpRfb bound to mosquito Lp, similar to AaLpRov (Fig. 6, lanes 2 and 5). In the controls, in which Lp was omitted, no precipitated band was found after incubation of anti-Lp antibodies with either LpR, prepared as in experimental treatments (Fig. 6, lanes 3 and 6). Thus, the AaLpRfb variant had a similar ability to bind to Lp as its ovarian counterpart, despite the differences in their ligand-binding domains. It is well established that the binding specificity of receptors belonging to the LDLR family varies greatly, with some receptors binding several structurally dissimilar ligands (22, 30, 32, 33). For example, in addition to VLDL and vitellogenin, chicken VLR8 specifically binds a variety of other ligands, possibly because of having eight ligand-binding repeats instead of seven like the less promiscuous LDLR (18, 49, 50). The molecular structures of AaLpRov and AaLpRfb are very similar to those of VLDLR and LDLR, respectively. It remains to be determined whether or not the differences in the ligand-binding domains of two mosquito LpR variants reflect particular properties of the Lp-binding kinetics or the scope of ligand specificity beyond Lp binding. Developmental Expression of Mosquito LpR mRNAs in the Fat Body and OvaryFirst, we measured the transcript size of two mosquito LpR variants, which were detected using either the common or the variant-specific cDNA probes in mRNA isolated from either the fat body or ovaries at 30 h PBM. In both cases, transcripts yielded expected sizes of 5.7 kb in the fat body and 4.5 kb in the ovary (Figs. 7 and 8). Each tissue exhibited a single specific variant of the LpR.
To determine the levels of the LpRfb transcript expression levels in the fat body during the vitellogenic cycle of the female mosquito, Northern hybridization was performed over a time course. Equal samples of total RNA collected from the fat body of pre- and vitellogenic females (FB) throughout the vitellogenic cycle were hybridized with the AaLpR and Lp genes (Fig. 8). As a control, ovarian samples (OV) collected at the same time points were hybridized with the AaLpR and VgR genes, the expression of which have been reported previously (1, 42). The Northern blot analyses demonstrated a striking difference in expression time of the LpR gene transcript variants. The AaLpRfb transcript was absent in the fat body of previtellogenic females as well as during the first 24 h PBM, at the time when major events of vitellogenesis occur; the expression of yolk protein precursor genes is initiated by a blood meal and reaches maximal levels by 24 h PBM (Fig. 8). The level of Lp transcript in the fat body on the same blot increased dramatically from 12 h PBM, peaking by 24 h PBM. The Lp transcript level began to decrease by 30 h PBM, when the expression of yolk protein precursor genes declines. However, although the AaLpRfb transcript appeared 24 h PBM, its level increased considerably by 30 h PBM, persisting through the next 18 h. Thereafter, the AaLpRfb transcript disappeared from the postvitellogenic fat body, whereas expression of the Lp gene continued at the lower level (Fig. 8). The AaLpRov transcript was clearly detectable during all vitellogenic stages. It was present in the previtellogenic stage and increased further after the onset of vitellogenesis, peaking by 24 h PBM. This is the time when yolk protein gene transcription nears its maximum (26) under the control of a rising titer of 20E and when the endocytotic activity of oocytes is at its highest point. AaLpRov mRNA was present in the ovary until 48 h PBM, the time of termination of vitellogenic events in the female mosquito. The pattern of AaLpR mRNA expression in the ovaries is similar to that of AaVgR, but AaVgR mRNA was expressed at higher levels, both in previtellogenic and postvitellogenic ovaries. The developmental expression of both transcripts in the ovary supports their putative functions in coding for receptors that mediate Lp or vitellogenin uptake. The appearance of the fat body-specific LpR variant coincides with the termination of vitellogenesis and of accumulation of yolk protein precursors. During this time, the mosquito fat body undergoes remodeling from a protein-synthesizing tissue to a storage depot for lipid, protein and carbohydrate reserves. Previously, we have shown that the Lp hemolymph titer reaches its peak at the termination time (Ref. 14 and Fig. 8). The presence of AaLpRfb also correlates with a considerable increase in the Lp amount in the fat body (Ref. 14 and Fig. 8). Taken together, these findings suggest that the seven-ligand-binding repeat LpRfb represents a specialized fat body receptor for re-absorption of Lp during the postvitellogenic period. In the mosquito fat body, regulation of expression of most genes involved in vitellogenesis is governed via a blood meal-driven hormonal cascade, with the terminal signal being a steroid, 20-hydroxyecdysone (20E) (51). This hormone controls most genes, which are maximally expressed at the peak of vitellogenesis, such as yolk protein precursors and Lp (14, 26, 52, 53). Lysosomal enzymes are overproduced for cellular remodeling in the fat body at the termination stage (54). Examination of the cathepsin D-like lysosomal aspartic protease has revealed, however, that mRNA is maximally expressed at the peak of vitellogenesis despite the fact that its protein peak is 12 h later (26). Recently, we have demonstrated that 20E is involved in translational inhibition of this enzyme mRNA, and the falling titer of 20E permits translation to proceed (55). Unlike lysosomal enzymes, the AaLpRfb mRNA is elevated only during the termination stage 30 h PBM. Therefore, the signal activating this gene is likely different from that of the lysosomal enzyme genes. The rising Lp titer (14) or falling titer of 20E (51) could serve as a signal activating the transcription of this gene. Alternatively, the oostatic peptide hormone that is presumably released by the ovary at the time of termination of vitellogenesis (56, 57) is another candidate for being a signal activating the AaLpRfb transcript. Future studies should reveal the precise nature of this late postvitellogenic gene activation.
The nucleotide sequence(s) reported in this paper has been submitted to the GenBankTM/EBI Data Bank with accession number(s) AY348869 [GenBank]
* This work was supported by National Institutes of Health Grants AI024716
[GenBank]
and AI36959 (to A. S. R.), a grant from the Brain Korea 21 program (to S.-J. S.) and by Advanced Basic Research Laboratory Grant R 14-2002-056-01001-0 from the Korea Science Engineering Foundation (to S. J. S.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 The abbreviations used are: Lp, lipophorin; HDLp, high density lipoprotein; LDLR, low density lipoprotein receptor; VLDLR, very low density lipoprotein receptor; LpR, lipophorin receptor; VgR, vitellogenin receptor; YPR, yolk protein receptor; EGF, epidermal growth factor; PBM, post-blood meal; LR, ligand-binding repeat; 20E, 20-hydroxyecdysone; E, exon; UTR, untranslated region; RT, reverse transcription; RACE, rapid amplification of cDNA ends.
This article has been cited by other articles:
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Advertisement | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||