|
|
||||||||
Papers In Press, published online ahead of print March 1, 2005
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||

* Department of Biochemical Physiology, Faculty of Biology and Institute of Biomembranes, Utrecht University, Utrecht, The Netherlands
Department of Chemistry, Coastal Carolina University, Conway, SC 29528
Published, JLR Papers in Press, December 16, 2004. DOI 10.1194/jlr.M400374-JLR200
The online version of this article (available at http://www.jlr.org) contains an additional three figures. ![]()
1 To whom correspondence should be addressed. e-mail: k.w.rodenburg{at}bio.uu.nl
| ABSTRACT |
|---|
|
|
|---|
These results show that apoLp-II/I is posttranslationally cleaved by an insect furin and that biosynthesis and secretion of HDLp can occur independent of this processing step. Structure modeling indicates that the cleavage of apoLp-II/I represents a molecular adaptation in homologous apolipoprotein structures. We propose that cleavage enables specific features of insect lipoproteins, such as low density lipoprotein formation, endocytic recycling, and involvement in coagulation.
Abbreviations: apoB, apolipoprotein B; apoLp, apolipophorin; decRVKRcmk, decanoyl-Arg-Val-Lys-Arg-chloromethyl ketone; HDLp, high density lipophorin; LDLp, low density lipophorin; LDLR, low density lipoprotein receptor; LLT, large lipid transfer; MTP, microsomal triglyceride transfer protein; PC, proprotein convertase
Supplementary key words apolipoprotein B homology modeling low density lipophorin insect lipoprotein receptor lipovitellin precursor proprotein convertase vitellogenin
| INTRODUCTION |
|---|
|
|
|---|
Cleavage of apoLp-II/I can be related to the activity of furin, a member of the proprotein convertase (PC) family of subtilisin-like serine endoproteases that is mainly active in the trans-Golgi network (7). The preferred consensus substrate sequence for furin, R-X-K/R-R, is present in all apoLp-II/I sequences characterized to date (811). In agreement with the activity of furin, Locusta migratoria apoLp-II/I appears to be cleaved immediately C terminal of its furin substrate sequence, RQKR720, as indicated by the N-terminal sequence of apoLp-I (10).
The predicted furin cleavage site in each insect apoLp-II/I is located in the large lipid transfer (LLT) domain, which constitutes the N-terminal region of apoLp-II/I that has sequence homology to that of apoB, the microsomal triglyceride transfer protein (MTP) large subunit, and vitellogenin (3, 4). In apoB, this domain is essential for lipoprotein biosynthesis. The interaction between the LLT domain of apoB and that of the MTP large subunit enables the assembly of apoB-containing lipoproteins (1, 2). The homology between apoB and apoLp-II/I, as well as the presence of an MTP large subunit in insects (12), suggest that the LLT domain of apoLp-II/I enables lipoprotein biosynthesis in insects as well. Therefore, we hypothesized that the cleavage of apoLp-II/I in the LLT domain functions in the biosynthesis and secretion of insect lipoprotein.
In the present report, we characterize the involvement of insect furin in the cleavage of L. migratoria apoLp-II/I and investigate the importance of this posttranslational modification to insect lipoprotein biosynthesis and secretion. To this end, apoLp-II/I cleavage was investigated in a recombinant insect expression system for truncated apoLp-II/I as well as in the locust fat body, the insect tissue that expresses apoLp-II/I and secretes its cleavage products apoLp-I and -II together in the form of a high density lipophorin (HDLp) particle (6, 13). The results indicate that L. migratoria apoLp-II/I is cleaved by an insect furin. Uncleaved apoLp-II/I could be secreted and formed a high density lipoprotein. We conclude that cleavage of apoLp-II/I is not required for the biosynthesis and secretion of mature lipophorin. Rather, this posttranslational modification may represent a molecular adaptation enabling specific features of insect lipoproteins. Modeling of apoLp-II/I to the available homologous structures, the lipovitellin crystal structure and an apoB model, indicated the position of the cleavage site in the apoLp-II/I structure, and putative functions for apoLp-II/I cleavage are discussed accordingly.
| MATERIALS AND METHODS |
|---|
|
|
|---|
|
Stable expression of apoLp-II/I-38 constructs in the Sf9 cell line
Spodoptera frugiperda Sf9 cells were maintained in adherent culture in serum-free Insect-Xpress medium (Cambrex) in polystyrene flasks (Greiner) at 27°C in a humidified atmosphere and passed twice each week. Transfections with the wild-type and mutant apoLp-II/I-38 constructs were performed using Cellfectin reagent (Invitrogen) according to the manufacturer's instructions. Stable transformants were selected using 400 µg/ml zeocin (Cayla) and maintained at 100 µg/ml zeocin. Experiments were performed in 25 cm2 flasks with cells grown to 80% confluence. Incubations with the furin inhibitor decanoyl-Arg-Val-Lys-Arg-chloromethyl ketone (decRVKRcmk; Bachem) were performed by including either the inhibitor or its solvent methanol in fresh culture medium for 6 h. Culture media were collected and cleared from whole cells and cellular debris by centrifugation (1,000 g, 10 min; 22,000 g, 10 min, 4°C), and analyzed for apoLp-II/I-38 and cleavage products by immunoblotting.
In vitro incubation of L. migratoria fat body tissue
L. migratoria were reared in cages under crowded conditions at 30°C, 40% relative humidity, a light-dark cycle of 12 h/12 h, and a diet of reed grass supplemented with rolled oats. Fat bodies were dissected from day 9 adult males, separated in two halves along the length, and rinsed in saline buffer (10 mM HEPES, 150 mM NaCl, 10 mM KCl, 4 mM CaCl2, 2 mM MgCl2, pH 7.0). Subsequent incubations were performed in a shaking water bath at 32°C. After washing for 1 h at 32°C in 4 ml of saline buffer, halves of six fat bodies were transferred to 250 µl of fresh saline buffer with 30 µCi of [35S]Met/Cys (Promix; Amersham) supplemented with either 100 µM decRVKRcmk or an equivalent volume of its solvent methanol. Proteins in the incubation medium were analyzed for apoLp-II/I and cleavage products by immunoblotting.
Density gradient analysis
The buoyant density of secreted apoLp-II/I and cleavage products was compared by subjecting the incubation media of decRVKRcmk-treated and control fat body tissue to KBr density gradient ultracentrifugation (13). After gravimetric analysis of density, fractions were assessed for the presence of apoLp-II/I and cleavage products by immunoblotting. Radiolabeled proteins on the blot were visualized by phosphorimaging on a Molecular Imager FX system with Quantity One software (Bio-Rad).
Immunoblot analysis of apoLp-II/I and cleavage products
Proteins were precipitated from incubation media by the addition of TCA to a concentration of 5%. Pellets were resuspended and heated (10 min at 95°C) in Laemmli sample buffer that was modified by the addition of 5% (v/v) 1 M Tris to circumvent acidification by residual TCA. Protein samples were separated using SDS-PAGE (9% slab for transfected cell culture media; 410% slab for fat body tissue incubation media). Precision Plus Protein Dual Color Prestained Standards (Bio-Rad) were used as a protein molecular mass marker, and isolated wild-type L. migratoria HDLp (6) was used as a positive control. Proteins were transferred to a polyvinylidene difluoride membrane (Millipore), and subsequent immunostaining was performed as described (16). Primary antibodies (1:10,000 dilution) were monoclonal
-V5 (Invitrogen), polyclonal
-II or
-I (17), or polyclonal
-IIC and
-IN (raised in rabbits against the peptides CKSLYNRITERFEKTFRQKR and SVSKDAVDNIRQQAYKSLLC, respectively, which correspond to the C terminus of apoLp-II and the N terminus of apoLp-I, excluding the terminal cysteines). Alkaline phosphatase-conjugated goat anti-rabbit or goat anti-mouse IgG (Jackson ImmunoResearch Laboratories) was used as a secondary antibody (1:10,000 dilution).
Modeling of L. migratoria apoLp-II/I
The L. migratoria apoLp-II/I sequence was used to identify homologous proteins and conserved domains using the web-based BLASTp program from the National Center for Biotechnology (18). The default settings were used for the BLASTp search. The most significant matches were to sequences from apoLp-II/I, apoB, and vitellogenin, including silver lamprey lipovitellin. Sequence homologies were located within the first 1,000 amino acid residues.
The alignment program CLUSTAL W (19) was used to align the homologous N-terminal regions of silver lamprey lipovitellin, human apoB-100, and L. migratoria apoLp-II/I. This alignment was used to generate the structural model of the first 1,009 amino acids (excluding the signal peptide) of locust apoLp-II/I using the program Modeller6 (20), based on the crystal structure of silver lamprey lipovitellin (PDB 1LSH) (21) and an all atom model for apoB (22). To eliminate steric problems and to optimize bond lengths and angles, the apoLp-II/I model was subjected to 250 steps of steepest descent energy minimization using the DISCOVER program package from INSIGHT2 (Accelerys, Inc.). A graphic representation of the model was generated by the Swiss-PdbViewer (23).
| RESULTS |
|---|
|
|
|---|
Stable transfection of Sf9 cells with this apoLp-II/I-38 construct resulted in the secretion of apoLp-II and apoLp-I567/V5 cleavage products into the incubation medium, as demonstrated by immunoblot analysis (Fig. 2A)
. Secretion of apoLp-I567/V5 is indicated by the single reactive band obtained with antibodies directed against apoLp-I (
-IN and
-I; Fig. 2A, lanes 4 and 5) as well as the V5 epitope (
-V5; Fig. 2A, lane 6) at
66 kDa, similar to the estimated molecular mass of apoLp-I567/V5. The weak immunoreactivity of polyclonal
-I (Fig. 2A, lane 5) likely reflects the limited representation of its epitopes in apoLp-I567/V5. Secretion of recombinant apoLp-II is indicated by immunoblotting with antibodies directed against apoLp-II (
-II and
-IIC; Fig. 2A, lanes 2 and 3), which results in a single immunoreactive band at
72 kDa. This recombinant protein appears to migrate identically to the apoLp-II from purified L. migratoria HDLp (Fig. 2A, lane 1 vs. lane 2). This similar size indicates that recombinant apoLp-II is similarly glycosylated as wild-type apoLp-II. Indeed, deglycosylation with endoglycosidase H resulted in a single additional apoLp-II immunoreactive band with a decreased molecular mass of ±3 kDa (see supplementary Fig. I), which is similar to that reported for wild-type locust apoLp-II (6). Moreover, the similar migration behavior of recombinant and wild-type apoLp-II indicates that the cleavage of apoLp-II/I-38 proceeds identical to that in fat body (6, 10). Expression of a construct expressing the N-terminal 33% of apoLp-II/I in Sf9 cells, and apoLp-II/I-38 in Drosophila melanogaster S2 cells, also resulted in the secretion of the two expected apoLp-II/I cleavage products (data not shown).
|
-IIC and
-IN antibodies, raised against synthetic oligopeptides of the sequences that flank the cleavage site N- and C-terminally, specifically recognize recombinant apoLp-II and apoLp-I567/V5, respectively (Fig. 2A, lanes 3 and 4). In addition, mutation of the three basic residues in the putative furin cleavage site (yielding the mutant designated QQAQ720; Fig. 1B) results in the secretion of uncleaved apoLp-II/I-38 (Fig. 2B), as shown by the single immunoreactive band obtained with
-II,
-IIC,
-I,
-IN, and
-V5 antibodies at
145 kDa (Fig. 2B, lanes 15, respectively), similar to the combined molecular mass of apoLp-II and apoLp-I567/V5. The limited immunoreactivity of
-IIC antibody against mutant apoLp-II/I (Fig. 2B, lane 2) likely relates to the introduced mutations. In further experiments, the degree of apoLp-II/I-38 cleavage was quantified using
-V5 antibody. Thus, Sf9 cells cleave recombinant apoLp-II/I-38 into apoLp-II and apoLp-I567/V5 at the consensus substrate sequence for furin, similar to apoLp-II/I in locust fat body. These results validate the use of this expression system to characterize apoLp-II/I cleavage. The secreted recombinant apoLp-II/I-38 cleavage products apoLp-II and apoLp-I567/V5 were further characterized for complex formation and lipidation. The coelution of apoLp-II with apoLp-I567/V5 after affinity chromatography suggests that both cleavage products can form a complex (see supplementary Fig. II). Upon density gradient ultracentrifugation, the cleavage products apoLp-II and apoLp-I567/V5 were found together in the fractions with densities between 1.20 and 1.25 g/ml, as was uncleaved mutant apoLp-II/I-38 (QQAQ720) (see supplementary Fig. III) . These results indicate that the secreted recombinant apoLp-II/I-38 products are poorly lipidated. Therefore, the hypothesized role for apoLp-II/I cleavage in lipidation was assessed in L. migratoria fat body tissue, whereas the present expression system was used to characterize apoLp-II/I cleavage.
|
-V5 (Fig. 3) and
-II (data not shown). Based on the amount of secreted protein, cleavage appeared to be prevented for up to 90% at a concentration of 10 µM decRVKRcmk. The sensitivity to decRVKRcmk indicates the involvement of a PC in the cleavage of apoLp-II/I.
A consensus substrate sequence for furin is required to efficiently process recombinant apoLp-II/I
To provide further evidence for the involvement of PCs in apoLp-II/I-38 cleavage, a negatively charged amino acid was introduced at position P4 (Fig. 1B), which is expected to disturb binding by PCs via charge repulsion at the S4 binding pocket (2628). As demonstrated by immunoblot analysis of media from stably transfected Sf9 cells using
-V5 antibody, introduction of Asp at the P4 position (R to D; DQKR720) completely prevented cleavage (Fig. 4A)
, implicating the involvement of a PC in the cleavage of apoLp-II/I.
|
Uncleaved native apoLp-II/I is secreted and forms a stable lipoprotein
To validate the involvement of an insect furin in lipophorin biosynthesis, and to establish the importance of cleavage for (apo)lipoprotein secretion in a physiological situation, fat body tissue from adult male locusts was treated in vitro with decRVKRcmk. Experiments showed the release of large amounts of HDLp that had adhered to fat body tissue. Therefore, [35S]Met/Cys was included in incubation media to label newly biosynthesized proteins.
To identify secreted apoLp-II/I, incubation media were submitted to density gradient ultracentrifugation, and the resulting fractions were analyzed for radiolabeled proteins by phosphorimaging. Compared with control-treated fat body tissue, incubation with 100 µM decRVKRcmk resulted in the appearance of an additional radiolabeled band of high molecular mass and a concomitant decrease in the putative apoLp-I, in fractions with an average buoyant density of 1.12 ± 0.01 g/ml (Fig. 5A)
, identical to the density previously reported for wild-type HDLp (13). The identity of the radiolabeled proteins in these fractions as apoLp-II/I and apoLp-I was confirmed by immunoblotting with
-I (Fig. 5B). Although apoLp-II could not readily be identified among other labeled proteins, immunoblotting demonstrated its presence in the same fractions as apoLp-II/I and apoLp-I (data not shown). The sensitivity of apoLp-II/I cleavage to decRVKRcmk incubation (
60% with 100 µM inhibitor; Fig. 5A) was reduced in the fat body, compared with the recombinant expression system (
90% with 10 µM inhibitor; Fig. 3). This difference may reflect reduced delivery of decRVKRcmk within the fat body, as the hydrophobic decanoyl group of this inhibitor may cause it to accumulate at the surface of the lipid droplets in this tissue. Thus, fat body tissue can secrete uncleaved apoLp-II/I with a buoyant density identical to wild-type HDLp. Gel filtration chromatography indicates that this secreted apoLp-II/I forms particles with a molecular size identical to wild-type HDLp (data not shown). Together, these results demonstrate that fat body can secrete uncleaved apoLp-II/I that has been integrated in a high density lipoprotein similar to wild-type HDLp. Apparently, the biosynthesis as well as secretion of insect lipoprotein can occur independently of apoLp-II/I cleavage.
|
-helical structure (Fig. 6)
. The ßA and ßB sheets face each other to form a putative lipid pocket and are partly surrounded by the
-helical region. The ßC sheet is curved, resulting in a barrel-like structure, which is situated above the putative lipid pocket. The apoLp-II/I cleavage site, located between residues 720 and 721, was situated in an extended loop region (residues 669748) that connects two ß-strands of the ßA sheet at the base of the putative lipid pocket. This region could not be modeled confidently. The corresponding sequence was not resolved in the crystal structure of silver lamprey lipovitellin either (21), which is suggestive of structural flexibility in this region of lipovitellin and possibly also in the region of apoLp-II/I cleavage. According to the present model, apoLp-II includes the ßC sheet, the
-helical region, and most of the ßA sheet, whereas apoLp-I consists of two distal ß-strands from the ßA sheet, the full ßB sheet, and further unmodeled C-terminal sequences.
|
| DISCUSSION |
|---|
|
|
|---|
The involvement of an insect furin in the cleavage of L. migratoria apoLp-II/I is indicated by three results. First, apoLp-II/I cleavage in both the Sf9 recombinant expression system and locust fat body can be inhibited by decRVKRcmk, which specifically inhibits PCs (25), including Sfurin, the furin homolog characterized from the insect Sf9 cell line (24). Second, cleavage is completely prevented by mutation of the basic amino acid residue (Arg) at the P4 position to a negatively charged amino acid residue (Asp). Based on the presence of a negatively charged residue at the S4 substrate binding pocket of PCs, this mutation is predicted to impair the substrate interaction via charge repulsion (2628). Third, complete proteolytic processing requires an intact furin consensus substrate sequence with basic amino acid residues at P1, P2, and P4. Mutation of any of these sites into amino acids neutral of charge results in the secretion of uncleaved apoLp-II/I, yet some cleavage does occur in the P2 (Lys
Ala) and P4 (Arg
Gln) mutants. Mammalian furin is known to cleave, with reduced activity, at sequences that partially match the preferred R-X-K/R-R substrate sequence (26, 29). Likewise, the partial cleavage in these P2 and P4 mutants may result from limited activity of furin at nonideal sites. Together, these results indicate cleavage of apoLp-II/I by an insect furin.
Several furin homologs have been identified in insects. In D. melanogaster, cleavage activity and homology have been demonstrated for the dfurin1 and dfurin2 gene products (3032). Furthermore, a furin was cloned from the insect Sf9 cell line (24). Moreover, a PC with furin-like activity is present in the fat body of Aedes aegypti (33). However, no putative furin has yet been identified in L. migratoria fat body.
In transfected Sf9 cells as well as locust fat body, uncleaved apoLp-II/I could be secreted in similar amounts as its cleavage products. The recombinant apoLp-II/I-38 cleavage products apoLp-II and apoLp-I567/V5 appeared to form a complex, yet were not integrated in a high density lipoprotein, in contrast to apoLp-II/I products secreted by locust fat body, as they were recovered in the very high density range upon density gradient ultracentrifugation (see supplementary Figs. II, III). The poor lipidation of the apoLp-II/I-38 truncation products is in agreement with the decreased lipidation observed for apoB truncations (1, 22) but may also reflect the absence of cofactors for lipidation in the present expression system. Therefore, the possible role of cleavage in the lipidation and secretion of insect lipoprotein was investigated in locust fat body. Here, decRVKRcmk treatment resulted in the secretion of uncleaved apoLp-II/I that formed a stable high density lipoprotein, as indicated by its density and molecular mass identical to wild-type HDLp. Consequently, the uncleaved precursor apoLp-II/I can function as a single apolipoprotein in the formation of lipoprotein, like its mammalian homolog apoB. Thus, apoLp-II/I can be lipidated to form a high density lipoprotein, irrespective of its cleavage.
The lipidation of apoB starts cotranslationally in the rough endoplasmic reticulum and is completed posttranslationally in the smooth endoplasmic reticulum and/or cis-Golgi network, possibly by fusion with an intralumenal lipid droplet (1, 2). The lipidation process starts with the lipidation of the lipid pocket in the apoB LLT domain and requires the lipid transfer activity of MTP (1, 2). Based on the homology between apoB and apoLp-II/I (3, 4) as well as the discovery of an insect MTP (12), insect lipoprotein assembly may also occur early in the secretory pathway. However, cleavage by furin homologs is performed late in the secretory pathway, mainly in the trans-Golgi network (34). Therefore, we propose that insect lipoprotein biosynthesis by the fat body involves lipidation of apoLp-II/I to a lipoprotein first, followed by cleavage of apoLp-II/I into apoLp-I and -II. The occurrence of cleavage before any lipidation might result in the parting of apoLp-I and apoLp-II, and hence impairment of lipoprotein biosynthesis. The uncleaved LLT domain in apoLp-II/I may be essential to enable the first steps in lipidation, as in apoB.
Homology modeling shows that the LLT domain of apoLp-II/I can form a putative lipid pocket. Remarkably, the cleavage site is located in an unresolved loop region between two ß-strands in the ßA sheet at the base of this pocket (Fig. 6). In apoB, the corresponding region is proposed to function in early lipidation events, as it may form a hairpin structure that temporarily closes the basal opening of the lipid pocket by connecting to the ßB sheet via salt bridges (22). When the lipid pocket of apoB reaches a certain lipid content during lipoprotein assembly, these salt bridges are proposed to dissociate. This would allow for widening of the V-shaped lipid pocket formed by ßA and ßB sheets and the progression of apoB lipidation (22, 35). The cleavage of apoLp-II/I might represent an alternative structural solution to unlock the lipid pocket and increase the flexibility of the nonexchangeable apolipoprotein, hence allowing for further lipidation of the lipoprotein particle. Interestingly, insects can lipidate circulating HDLp to a low density lipophorin (LDLp) during conditions that require enhanced lipid transport (e.g., long-term flight and vitellogenesis) (36, 37). Therefore, apoLp-II/I cleavage may function to enable the formation of LDLp from HDLp.
The apparent conservation of apoLp-II/I cleavage in all insects characterized to date reveals the importance of this processing step. Besides LDLp formation, however, apoLp-II/I cleavage may enable other unique insect lipoprotein characteristics, such as the ability of lipophorin to function as a reusable lipid transporter (36) and to be recycled after endocytic uptake by an insect member of the low density lipoprotein receptor (LDLR) family (16). Vitellogenin, another ligand of this receptor family, is homologous to apoB and apoLp-II/I, and is also cleaved at a furin consensus substrate sequence in the LLT domain during biosynthesis in most insect species, but not vertebrates (38). Lipophorin, vitellogenin, and LDLR family members are involved in insect vitellogenesis, the transfer of nutrients to the developing oocyte (38), and perhaps the posttranslational cleavage of apoLp-II/I and vitellogenin facilitates this process (e.g., by enabling receptor binding). In addition, apoLp-II/I cleavage might enable a function for lipophorin in coagulation. Lipophorin has been implicated in this process by its abundant presence in clots (3942), capacity to aggregate (43, 44), and interactions with other insect hemostasis factors (45). Moreover, the disappearance of apoLp-I but not apoLp-II from plasma during coagulation (46) suggests that apoLp-II/I cleavage enables distinct roles for apoLp-I and apoLp-II in coagulation.
Cleavage of the insect apoB homolog apoLp-II/I appears to be a molecular adaptation within homologous structures and may have a significant impact on insect lipoprotein function. The possibility of obtaining apoLp-II/I lipoprotein from fat body using a furin inhibitor may aid in establishing the physiological role of apoLp-II/I cleavage. Furthermore, the recombinant expression of truncated apoLp-II/I may be used to explore aspects of insect lipoprotein assembly and structure, such as lipid binding regions in apoLp-I and -II, the role of protein-protein and lipid-protein interactions in apoLp-I:apoLp-II complex formation and lipoprotein solubility (47), as well as the cofactors involved in insect lipoprotein biosynthesis. For example, apoB truncations with a length similar to apoLp-II/I-38 can form high density lipoprotein particles upon recombinant expression, but only when cells are supplied with exogenous lipids and (co)express MTP (12, 48). Likewise, the present recombinant expression system may be used to investigate putative roles for lipid availability and insect MTP homologs in the biosynthesis of insect lipoprotein.
| ACKNOWLEDGMENTS |
|---|
-IIC and
-IN antibodies as well as purified insect lipoprotein, Jan van Doorn for technical assistance, and Jana Kerver for help in obtaining the apoLp-II/I construct. The authors thank Jere Segrest (Department of Medicine, University of Alabama at Birmingham Medical Center) for his initiating role in the modeling work. Manuscript received September 29, 2004 and in revised form November 24, 2004.
| REFERENCES |
|---|
|
|
|---|
ali, A., and T. L. Blundell. 1993. Comparative protein modelling by satisfaction of spatial restraints. J. Mol. Biol. 234: 779815.[CrossRef][Medline]
1-Antitrypsin Portland, a bioengineered serpin highly selective for furin: application as an antipathogenic agent. Proc. Natl. Acad. Sci. USA. 95: 72937298.This article has been cited by other articles:
![]() |
M. M. W. Smolenaars, A. de Morree, J. Kerver, D. J. Van der Horst, and K. W. Rodenburg Insect lipoprotein biogenesis depends on an amphipathic {beta} cluster in apolipophorin II/I and is stimulated by microsomal triglyceride transfer protein J. Lipid Res., September 1, 2007; 48(9): 1955 - 1965. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. M. W. Smolenaars, O. Madsen, K. W. Rodenburg, and D. J. Van der Horst Molecular diversity and evolution of the large lipid transfer protein superfamily J. Lipid Res., March 1, 2007; 48(3): 489 - 502. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. A. Sellers, L. Hou, D. R. Schoenberg, S. R. Batistuzzo de Medeiros, W. Wahli, and G. S. Shelness Microsomal Triglyceride Transfer Protein Promotes the Secretion of Xenopus laevis Vitellogenin A1 J. Biol. Chem., April 8, 2005; 280(14): 13902 - 13905. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||