| Glycobiology | Pages |
N-Linked glycosylation of a baculovirus-expressed recombinant glycoprotein in insect larvae and tissue culture cells
Introduction
Results
Discussion
Materials and methods
Acknowledgments
Abbreviations
References
N-Linked glycosylation of a baculovirus-expressed recombinant glycoprotein in insect larvae and tissue culture cells
The potential of insect cell cultures and larvae infected with recombinant baculoviruses to produce authentic recombinant glycoproteins cloned from mammalian sources was investigated. A comparison was made of the N-linked glycans attached to secreted alkaline phosphatase (SEAP) produced in four species of insect larvae and their derived cell lines plus one additional insect cell line and larvae of one additional species. These data survey N-linked oligosaccharides produced in four families and six genera of the order Lepidoptera. Recombinant SEAP expressed by recombinant isolates of Autographa californica and Bombyx mori nucleopolyhedroviruses was purified from cell culture medium, larval hemolymph or larval homogenates by phosphate affinity chromatography. The N-linked oligosaccharides were released with PNGase-F, labeled with 8-aminonaphthalene-1-3-6-trisulfonic acid, fractionated by polyacrylamide gel electrophoresis, and analyzed by fluorescence imaging. The oligosaccharide structures were confirmed with exoglycosidase digestions. Recombinant SEAP produced in cell lines of Lymantria dispar (IPLB-LdEIta), Heliothis virescens (IPLB-HvT1), and Bombyx mori (BmN) and larvae of Spodoptera frugiperda, Trichoplusia ni, H.virescens, B.mori, and Danaus plexippus contained oligosaccharides that were structurally identical to the 10 oligosaccharides attached to SEAP produced in T.ni cell lines. The oligosaccharide structures were all mannose-terminated. Structures containing two or three mannose residues, with and without core fucosylation, constituted more than 75% of the oligosaccharides from the cell culture and larval samples.
Introduction
Insect cells infected with recombinant baculoviruses are useful for expressing recombinant proteins particularly when high production levels, secretion and posttranslational modifications of the protein are desired (Luckow, 1995; Shuler et al., 1995). There is special interest in the potential of insect cells to produce recombinant vertebrate glycoproteins with oligosaccharides similar to the original proteins, a feature lacking with many other expression vector technologies. Information available to date suggests that the processing of N-linked oligosaccharides on recombinant proteins produced in insect cells differs from that of the same protein produced in the vertebrate cells of their origin or in vertebrate cell culture expression systems (März et al., 1995). Most mammalian recombinant glycoproteins produced by baculovirus expression vectors in insect cell culture are produced with mannose-terminated oligosaccharides or otherwise less complete oligosaccharides compared to the native protein produced in the original tissue (reviewed in Luckow, 1995; März et al., 1995). There are notable exceptions to this in a few reports of recombinant proteins produced with small amounts of oligosaccharides terminated with hexosamine and galactose, such as plasminogen (Davidson et al., 1990; Davidson and Castellino, 1991a,b), interferon (Ogonah et al., 1996) and IgG (Hsu et al., 1997). Sialic acid residues on N-linked oligosaccharides have been reported only for recombinant plasminogen produced in insect cell culture (Davidson et al., 1990, 1991; Davidson and Castellino, 1991a,b) and SEAP produced in insect larvae (Davis and Wood, 1995).
Although a large number of recombinant glycoproteins have been produced with the baculovirus expression system, variation in the glycoprotein processing by different insect cell lines has not been studied systematically. Since both the intrinsic structure of glycoprotein and the cell type in which it is expressed can have a profound influence on the final oligosaccharide attached, a more systematic approach will be required to predict which insect cell lines will be most useful for the production of glycoproteins. Lepidopteran larvae are also useful for baculovirus-driven expression of recombinant proteins and have been reported to carry out recombinant protein modifications that are not performed in cell culture (Luckow, 1995). Davis and Wood (1995) reported that Trichoplusia ni larvae produce recombinant SEAP with sialic acid residues, but there has been no comparison of susceptible insect species for their capacity to glycosylate recombinant proteins.
In a previous publication (Kulakosky et al., 1998), we compared N-linked glycosylation of SEAP (Davis et al. 1992) by three insect cell lines. SEAP produced in mammalian cells normally has oligosaccharides that are sialylated (Takami et al., 1988; Chuang, 1989). However, SEAP produced by the T.ni (BTI-Tn-5b1-4 and TN368) and the Spodoptera frugiperda (Sf21) cell lines only had oligosaccharides with terminal mannose residues. Both T.ni cell lines produced large amounts of oligosaccharides with two or three mannose residues, lacking terminal [alpha]1,3-linked mannose. However, the oligosaccharides attached by Sf21 cells do not contain the two or three mannose residue structures which lacked [alpha]1,3 terminal linkages. In this article we have used SEAP as a model glycoprotein to investigate the potential of several additional insect cell lines as well as insect larvae to glycosylate recombinant proteins. By comparing samples produced in cultured cells and larvae of the same species, we explored whether the variability in glycosylation of a recombinant protein with different cells arises because of tissue diversity within an insect, or from differences between insects in their basic glycoprotein processing machinery. In addition, we compared SEAP oligosaccharides produced in tissue culture cells and larvae infected with a recombinant BmNPV with samples produced using a recombinant AcMNPV.
Results
We analyzed the oligosaccharides attached to SEAP produced in several insect cell lines in order to evaluate the range of differences in glycoprotein processing and to compare the data with previously observed differences in N-linked oligosaccharides on SEAP produced in S.frugiperda and T.ni cell cultures. The oligosaccharides on SEAP produced in Sf21 and BTI-Tn-5b1-4 cell lines are shown in Figure
Figure 1. SEAP oligosaccharide profile and structures produced in BTI-Tn-5b1-4 (T.ni) and IPLB-SF-21A (Sf21) tissue culture cells. Asterisk denotes unbranched structures. Arrows on the right indicate positions of glucose polymer standards. Previous studies (Kulakosky et al., 1998) showed that Sf21 cells produce the largest proportion of SEAP oligosaccharides larger than trimannose (27%). The BmN and LdEIta cells produced the lowest amount of these oligosaccharides (12%). The BmN, HvT1 and LdEIta cells produced SEAP with 49%, 55%, and 62% core-fucosylated oligosaccharides, respectively. The TN368, Tn5-b1-4, and Sf21 cells produce 2%, 19%, and 49% core fucosylated oligosaccharides. All the cell lines produced more SEAP oligosaccharides containing six mannose residues than five or seven mannose residues. We were able to obtain and infect several lepidopteran species reported to be susceptible to AcMNPV or BmNPV. The level of SEAP from the perfused hemolymph of these larvae varied greatly (Table I). Although susceptible to AcMNPV, the perfusates from Helicoverpa zea, L.dispar, Pseudaletia unipuncta, and Manduca sexta larvae contained only small amounts of recombinant SEAP and were not further analyzed. SEAP samples harvested from T.ni, S.frugiperda, B.mori, Danaus plexippus, and H.virescens larvae (Figure Figure 3. Oligosaccharide profile of SEAP produced in Trichoplusia ni (Tn), Spodoptera frugiperda (Sf), Heliothis virescens (Hv), Danaus plexippus (Dp) and Bombyx mori (Bm) larvae. The Bm larvae were infected with recombinant BmNPV, and the remaining larvae with recombinant AcMNPV expressing the SEAP gene. Arrows indicate positions of glucose polymer standards. Table I. The profile of SEAP oligosaccharides from B.mori larvae was distinct, having a high proportion of very small oligosaccharides. The only oligosaccharides produced in significant quantities were trimannose, dimannose, and dimannose fucosylated structures. These oligosaccharides were devoid of terminal [alpha]1,3-linked mannose and differed from the oligosaccharides produced in BmN cells (Figure Except for B.mori samples, all the larval samples contained large amounts of fucosylated oligosaccharides (T.ni, 42%; D.plexippus, 50%; S.frugiperda, 55%; H.virescens, 60%). The B.mori larval samples contained only 17% fucosylated oligosaccharides. Samples produced in the L.dispar, H.virescens, and B.mori cell lines contained 62%, 55%, and 49% fucosylated oligosaccharides, respectively. The result with T.ni larval SEAP oligosaccharides (Figure 
Figure 2. Oligosaccharide profile from SEAP produced in Bombyx mori (BmN), Heliothis virescens (HvT1), and Lymantria dispar (LdEIta) tissue culture cells. The HvT1 and LdEIta cells were infected with the recombinant AcMNPV, and the BmN cells were infected with recombinant BmNPV expressing the SEAP gene. Arrows indicate positions of glucose polymer standards.
Species
No. of larvae
Mean µg/larva
Danaus plexippus
29
106.4 ± 93.9
Pseudaletia unipuncta
45
2.3 ± 2.5
Heliothis virescens
37
105 ± 97.1
Helicoverpa zea
15
0.5 ± 0.5
Trichoplusia ni
20
87.5 ± 45
Lymantria dispar
21
2.4 ± 2.5
Bombyx mori
25
67.7 ± 23.3
Manduca sexta
11
14.6 ± 12.2
Spodoptera frugiperda
18
173 (pooled sample)
Discussion
There was a remarkable similarity in the composition of SEAP oligosaccharides produced in cell lines and larvae from six genera in four families. The cell lines that we investigated were isolated from a variety of tissue types-embryonic (BTI-Tn-5b1-4, Granados et al., 1994; IPLB-LdEIta, Lynn et al., 1988), pupal ovarian (TN368, Hink, 1970), unspecified ovarian (BmN, Maeda, 1984), imaginal disks (Sf21, Vaughn et al., 1977), and larval testicular sheath (IPLB-HvT1, Lynn et al., 1988). These cell lines are composed of dedifferentiated cells. Most baculovirus replication in larvae occurs in the fat body cells, and it is therefore assumed that most of the SEAP was produced in this differentiated cell type. Clearly, phylogenetic divergence, tissue origin, and state of cellular differentiation did not result in major differences in the glycosylation of SEAP. The constant nature of the oligosaccharide structures suggests that (1) insect cells from widely divergent insect species and their derived cell cultures have similar potential for processing recombinant proteins, (2) the recombinant protein itself is the overriding determinant in processing events, and/or (3) baculovirus infection alters the secretory pathway in a way that limits the glycosylation potential of insect cells for N-linked glycosylation.
The processing of N-linked oligosaccharides of SEAP by Sf21 cells (Figure
The other notable difference between samples from larvae and their derived cell lines is that of SEAP produced in B.mori larvae (Figure
This laboratory previously reported (Davis and Wood, 1995) that SEAP oligosaccharides produced in T.ni larvae contained sialic acid residues. These findings were based on positive Sambucus nigra agglutinin lectin blots of a 64 kDa, immunoprecipitated protein separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The results of the current study revealed no evidence of sialic acid-terminated oligosaccharides from SEAP purified from T.ni larvae. This discrepancy could arise in at least two ways. First, the immunoprecipitation may have precipitated another sialylated glycoprotein with a molecular weight of 64 kDa. Secondly, lectin blot detection is not quantitative, and a very minor sialylated oligosaccharide component of SEAP could be responsible for the positive reaction. Such a minor component could have been produced in tissues other than the fat body cells that produce most of the SEAP in larvae. Because of the quantitative properties of FACE analyses, we now know that, if sialylated SEAP oligosaccharides are produced in insect cells, they occur in very low proportions, a fact not discernible from the lectin blotting data.
Based on this survey of SEAP glycosylation in a wide range of lepidopteran larvae and tissue culture cells, the potential for lepidopteran cells to process oligosaccharides in a manner similar to mammalian cells is questionable. However, the limited variation in SEAP processing may be a result of the properties of SEAP rather than cellular determinants. To investigate this question, an evaluation of the oligosaccharide processing of additional glycoproteins in several lepidipoteran larvae and tissue culture cells will be required.
Materials and methods
Insect cell culture
BmN cells (Maeda, 1984) were obtained from Susumu Maeda (Department of Entomology, UC-Davis). The HvT1 cells (Lynn et al., 1988) were supplied by Dwight Lynn (USDA ARS, Beltsville, MD). The LdEIta (Lynn et al., 1988) were obtained from R. Granados (Boyce Thompson Institute). All cell lines were maintained in TNM-FH medium (Graces insect cell medium, Gibco, Grand Island, NY) containing 10% fetal bovine serum. Healthy cells were maintained at 27°C and infected cells at 28°C except for HvT1 cells, which were maintained 23°C.
Insect larvae
T.ni and D.plexippus were from colonies at the Boyce Thompson Institute. P.unipuncta were a gift from R. Granados (Boyce Thompson Institute). The H.virescens and H.zea were reared from eggs purchased from USDA-ARS (Stoneville, MS), and S.frugiperda were reared from eggs supplied by W. Deryck Perkins (USDA ARS Insect Biology Lab, Tifton, GA). M.sexta caterpillars were a gift from Ronald Booker (Department of Neurobiology & Behavior, Cornell University), and L.dispar larvae were a gift from Ann Hajek (Dept. of Entomology, Cornell Univ.). B.mori mulberry diet and eggs of the Kinshu×Showa race were kindly provided by Yoshifumi Hashimoto (Kyoto Institute of Technology, Japan). All insect larvae were maintained on wheat germ diet (Bell et al., 1979) except B.mori, D.plexippus, and P.unipuncta, which were fed on commercial Morus spp. leaf-based diet, Asclepias curassavica, and Avena sativa seedlings, respectively.
SEAP production in larvae and cells
Early fifth instar H.virescens, S.frugiperda, T.ni, D.plexippus and B.mori larvae were injected with the budded form of the recombinant viruses produced in tissue culture cells. Viral injections were performed with both fourth- and fifth-instar larvae of L.dispar, H.zea, M.sexta, and P.unipuncta. Hemocoel injections of AcMNPV were 106 PFU for T.ni and S.frugiperda, 2 × 106 PFU for P.unipuncta, 4.3 × 106 PFU for H.zea and H.virescens, and 1 × 107 PFU for L.dispar, D.plexippus, and M.sexta larvae. The B.mori larvae were injected with between 6.4 × 105 and 6.4 × 106 PFU of BmNPV.
The SEAP was harvested from the hemocoel of larvae by perfusion in order to minimize the release of cellular hydrolases during the sampling from larvae. When the larvae appeared moribund (minimal response to touching with a forceps or unable to right themselves when inverted), they were chilled on ice and perfused with a lepidopteran saline solution (Anderson and Telfer, 1969) containing 5 mM phenylthiourea (recrystallized from ethanol). The perfusion was carried out with cold saline from a syringe fitted with a suitable gauge (18-27) needle for a particular species. The chilled larvae were bled by one of two procedures. In the first procedure, they were injected posteriorly or anteriorly with the saline solution before cutting a proleg and then perfused until the perfusate was colorless. Alternatively, the larvae were injected at the anterior end with a small amount of saline and then perfused from the posterior end until the solution was colorless. The perfusate was collected in conical tubes containing a small amount of saline on ice. In specified experiments with T.ni larval samples, a sialidase inhibitor, 1 mM 2,3-dehydro-2-deoxy-N-acetylneuraminic acid (Boehringer Mannheim, Indianapolis, IN), was added to the perfusion solution. The perfusate was centrifuged for 5 min at 500 × g. An aliquot of the supernatant was removed for enzyme assay and the remainder of the supernatant was stored at -70°C. For purification, the SEAP samples were thawed, centrifuged at 70,000 × g for 2 h or 250,000 × g for 30 min., and the supernatant was exchanged into affinity column buffer (20 mM Tris, 1 mM MgCl2, pH 8.0) using Biogel P-6, P-10, or P-30 gel permeation columns (Bio-Rad Hercules, CA). The SEAP was then purified with affinity chromatography (Kulakosky et al., 1998).
Homogenates of infected T.ni larvae were prepared in the lepidopteran saline solution containing 2 mM dithiothreitol and 2 mM 2,3-dehydro-2-deoxy-N-acetylneuraminic acid. Twenty-five larvae were homogenized with a Heat Systems homogenizer (Heat Systems, Plainview, NY) in 5 ml buffer. The homogenate was centrifuged at 500 × g for 5 min in conical tubes and the supernatant solution was then centrifuged at 2000 × g for 10 min. The resulting supernatant solution was stored at -70°C. For purification, the samples were thawed and centrifuged at 165,000 × g for 2.5 h in a fixed-angle rotor. The supernatant solution was filtered through a Millex-PI 0.8 µm syringe filter (Millipore, Bedford, MA) before desalting on a Biogel P-30 column equilibrated in the affinity chromatography buffer. The SEAP was then purified by affinity chromatography (Kulakosky et al., 1998).
Tissue culture cells in midexponential phase were infected with recombinant AcMNPV or BmNPV at a multiplicity of infection of 10 PFU per cell. SEAP expressed with the recombinant BmNPV (Huang et al. , 1997) was produced in BmN cells, and the remaining cell lines were used to produce SEAP by infection with recombinant AcMNPV (Davis et al., 1992). At 4-7 days postinoculation, the cell culture medium containing SEAP was harvested, and the cellular debris was pelleted at 500 × g for 10 min. The virus in the cell culture medium was pelleted at 70,000 × g for 2 h in a fixed-angle rotor. The supernatant was concentrated 2- to 3-fold by dialysis against polyethylene glycol (MW 15,000-20,000, Sigma, St. Louis, MO). The samples were then dialyzed against column buffer, 20 mM Tris, 1 mM MgCl2, pH 8.0, and purified by phosphate affinity column chromatography (Kulakosky et al., 1998).
Oligosaccharide analysis
Characterization of the oligosaccharides attached to SEAP was conducted according to Kulakosky et al. (1998). Briefly, the oligosaccharides were released from the protein with PNGase-F and labeled with 8-aminonaphthalene-1-3-6-trisulfonic acid (Molecular Probes, Eugene, OR). Samples were subjected to single or multiple digestion reactions containing jack bean mannosidase (Oxford GlycoSystems, Rosedale, NY) and bovine epididymus [alpha]l-fucosidase (Oxford GlycoSystems, Rosedale, NY) as previously described (Kulakosky et al., 1998). Digestions with neuraminidase (Oxford GlycoSystems, Rosedale, NY) were performed according to the manufacturer's recommendations. All digestions were carried out at 37°C for 24 h. Fractionation of the oligosaccharides was done on a Glyko (Novato, CA) apparatus and analyzed using a Glyko SE1000 FACE imaging system (CCD camera and image analysis software).
Acknowledgments
We acknowledge the outstanding technical assistance provided by Molly Ingersoll during this project. This research was partially supported by the National Science Foundation, Grant Number BES-9421381, and the National Aeronautics and Space Administration, Grant Number NAG8-1384.
Abbreviations
AcMNPV, Autographa californica nucleopolyhedrovirus; BmNPV, Bombyx mori nucleopolyhedrovirus; FACE, fluorophore-assisted carbohydrate electrophoresis; HvT1, IPLB-HvT1; LdEIta, IPLB-LdEIta; PFU, plaque-forming units; SEAP, secreted human placental alkaline phosphatase; Sf21, IPLB-SF-21AE.
References
1To whom correspondence should be addressed
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 10 Jun 1998
Copyright©Oxford University Press, 1998.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  S. M. Van Patten, H. Hughes, M. R. Huff, P. A. Piepenhagen, J. Waire, H. Qiu, C. Ganesa, D. Reczek, P. V. Ward, J. P. Kutzko, et al. Effect of mannose chain length on targeting of glucocerebrosidase for enzyme replacement therapy of Gaucher disease Glycobiology, May 1, 2007; 17(5): 467 - 478. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Jayachandran, C. M. Radcliffe, L. Royle, D. J. Harvey, R. A. Dwek, P. M. Rudd, and A. A. Karande Oligosaccharides modulate the apoptotic activity of glycodelin Glycobiology, November 1, 2006; 16(11): 1052 - 1063. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K.-Y. Lee, H. G. Kim, M. R. Hwang, J. I. Chae, J. M. Yang, Y. C. Lee, Y. K. Choo, Y. I. Lee, S.-S. Lee, and S.-I. Do The Hexapeptide Inhibitor of Galbeta 1,3GalNAc-specific alpha 2,3-Sialyltransferase as a Generic Inhibitor of Sialyltransferases J. Biol. Chem., December 13, 2002; 277(51): 49341 - 49351. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Ailor, N. Takahashi, Y. Tsukamoto, K. Masuda, B. A. Rahman, D. L. Jarvis, Y. C. Lee, and M. J. Betenbaugh N-glycan patterns of human transferrin produced in Trichoplusia ni insect cells: effects of mammalian galactosyltransferase Glycobiology, August 1, 2000; 10(8): 837 - 847. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H Zieler, J. Nawrocki, and M Shahabuddin Plasmodium gallinaceum ookinetes adhere specifically to the midgut epithelium of Aedes aegypti by interaction with a carbohydrate ligand J. Exp. Biol., January 3, 1999; 202(5): 485 - 495. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||