| Glycobiology | Pages |
Stable expression of mammalian [beta]1,4-galactosyltransferase extends the N-glycosylation pathway in insect cells
Introduction
Results
Discussion
Materials and methods
Acknowledgments
Abbreviations
References
Stable expression of mammalian [beta]1,4-galactosyltransferase extends the N-glycosylation pathway in insect cells
An established lepidopteran insect cell line (Sf9) was cotransfected with expression plasmids encoding neomycin phosphotransferase and bovine [beta]1,4-galactosyltransferase. Neomycin-resistant transformants were selected, assayed for [beta]1,4-galactosyltransferase activity, and the transformant with the highest level of enzymatic activity was characterized. Southern blots indicated that this transformed Sf9 cell derivative contained multiple copies of the galactosyltransferase-encoding expression plasmid integrated at a single site in its genome. One-step growth curves showed that these cells supported normal levels of baculovirus replication. Baculovirus infection of the transformed cells stimulated [beta]1,4-galactosyltransferase activity almost 5-fold by 12 h postinfection. This was followed by a gradual decline in activity, but the infected cells still had about as much activity as uninfected controls as late as 48 h after infection and they were able to produce a [beta]1,4-galactosylated virion glycoprotein during infection. Infection of the transformed cells with a conventional recombinant baculovirus expression vector encoding human tissue plasminogen activator also resulted in the production of a galactosylated end-product. These results demonstrate that stable transformation can be used to add a functional mammalian glycosyltransferase to lepidopteran insect cells and extend their N-glycosylation pathway. Furthermore, stably-transformed insect cells can be used as modified hosts for conventional baculovirus expression vectors to produce foreign glycoproteins with 'mammalianized" glycans which more closely resemble those produced by higher eucaryotes.
Introduction
Most glycoproteins produced by insect cells lack complex N-linked oligosaccharides containing penultimate galactose and terminal sialic acid (März et al., 1995; Jarvis, 1997). One reason for this is that insect cells lack galactosyltransferase and sialyltransferase or have insufficient levels of these terminal glycosyltransferases to convert most N-linked side chains to complex forms (Butters et al., 1981). A second reason is that some insect cell lines have a membrane-bound N-acetylglucosaminidase (Licari et al., 1993; Altmann et al., 1995; Wagner et al., 1996) which processes GlcNAcMan3GlcNAc2 to Man3GlcNAc2 and precludes terminal extension of the former by galactosyltransferases and sialyltransferases. Thus, the most highly processed N-linked oligosaccharide side chain typically found on insect cell glycoproteins is Man3GlcNAc2(± Fuc).
We have shown previously that the N-glycosylation pathway in lepidopteran insect cells can be extended by introducing a functional mammalian processing enzyme (Jarvis and Finn, 1996). Our approach was to construct a novel recombinant baculovirus vector that included a bovine [beta]1,4-galactosyltransferase cDNA under the transcriptional control of an immediate early viral promoter. This recombinant virus expressed enzymatically active [beta]1,4-galactosyltransferase early in infection, which enabled the cells to [beta]-galactosylate N-linked oligosaccharides on a virion glycoprotein that was expressed later in infection. This result showed that the newly introduced mammalian [beta]1,4-galactosyltransferase could function in the environment of a baculovirus-infected lepidopteran insect cell and that it could effectively compete with the N-acetylglucosaminidase activity in these cells.
The purpose of the present study was to determine whether we could directly extend the N-glycosylation pathway of lepidopteran insect cells by modifying the cells instead of the viral vector. We used a cellular transformation procedure to isolate a stable insect cell derivative that contains a mammalian [beta]1,4-galactosyltransferase cDNA and constitutively expresses [beta]1,4-galactosyltransferase activity without viral infection. Unlike the parental cells, this transformed derivative was able to add galactose to two different foreign glycoproteins. Thus, this study demonstrates that genetic transformation can be used to produce a lepidopteran insect cell line which has de novo mammalian [beta]1,4-galactosyltransferase activity and extended N-linked oligosaccharide processing capabilities.
Results
Isolation of stably transformed insect cells that express [beta]1,4-galactosyltransferase activity
A bovine [beta]1,4-galactosyltransferase cDNA (Shaper et al., 1986; Russo et al., 1990; Shaper et al., 1997) was subcloned into an immediate early expression plasmid (Jarvis et al., 1996) to produce pIE1HRGalT, in which the complete [beta]1,4-galactosyltransferase coding sequence was positioned under the transcriptional control of a baculovirus immediate early (ie1) promoter (Guarino and Summers, 1987). pIE1HRGalT was used to cotransfect Sf9 cells together with pIE1Neo, which encodes a neomycin phosphotransferase gene under the control of the ie1 promoter (Jarvis et al., 1990). Neomycin-resistant colonies were isolated and amplified, and the resulting clones were assayed for [beta]1,4-galactosyltransferase activity as described in Materials and methods. The results showed that each of 10 independent clones had [beta]1,4-galactosyltransferase activity, while the parental, untransformed Sf9 cells had none (Figure 1). Different amounts of [beta]1,4-galactosyltransferase activity were observed in different clones and there was nearly a 10-fold difference between the clones with the lowest (clone 5) and highest (clone 9) amounts of activity. Clone 9 was designated Sf[beta]4GalT and used for the remainder of this study.
Figure 1. Expression of [beta]1,4-galactosyltransferase enzymatic activity by neomycin-resistant Sf9 cell derivatives. Sf9 cells were cotransfected with pAcP(+)IE1GalT plus pIE1Neo and 10 independent clonal derivatives were isolated by G418 selection and limiting dilution, as described in Materials and methods. Lysates were prepared from each derivative (labeled 1-10 in the figure) and assayed for [beta]1,4-galactosyltransferase enzymatic activity as described in Materials and methods. Lysates from uninfected (mock) and wild-type baculovirus-infected Sf9 cells were used as controls.
Analysis of genomic DNA in Sf[beta]4GalT cells
Genomic DNA was extracted from Sf[beta]4GalT or Sf9 cells, digested with various restriction endonucleases, and examined by Southern blotting (Southern, 1975) with a [beta]1,4-galactosyltransferase cDNA probe. This probe did not hybridize with genomic DNA from untransformed Sf9 cells, but it did hybridize with genomic DNA from Sf[beta]4GalT cells (Figure 2A). Two distinct bands were detected when Sf[beta]4GalT genomic DNA was cut with HincII or EcoRV, each of which has a single recognition site within pIE1HRGalT. The lower, more intense band comigrated with the linearized pIE1HRGalT positive control at about 6.5 Kb, while the upper, less intense bands were larger. These results suggested that multiple copies of pIE1HRGalT are integrated at a single site in the Sf[beta]4GalT genome. According to this model, the intense 6.5 kb band would arise from multiple unit-length fragments derived from the tandem array and the less intense, larger bands would arise from the single downstream junction fragment, as shown diagramatically in Figure 2B. The upstream junction fragment would not be detected by the probe used in this experiment.
Figure 2. Genomic DNA structure in Sf[beta]4GalT cells. (A) Southern blotting analyses. Genomic DNA was isolated from Sf[beta]4GalT (lanes A) or Sf9 (lanes B) cells and equal aliquots were digested with the restriction endonucleases indicated above each pair of lanes, as described in Materials and methods. The lane marked (+) was a positive control containing 10 ng of pAcP(+)IEHRGalT linearized with EcoRV. All digests were resolved by agarose gel electrophoresis and analyzed by Southern blotting with a 1095 bp fragment of the bovine [beta]1,4-galactosyltransferase cDNA which included nearly the entire coding sequence, as described in Materials and methods. The numbers and arrows to the left of the figure indicate the sizes (in kb) and positions of standard lambda DNA fragments obtained by HindIII digestion. (B)and (C), Transgene structure in Sf[beta]4GalT cells. The drawing shows that there are multiple copies of pAcP(+)IE1GalT (open boxes) inserted as a tandem array at a single site in the insect cell genome (solid boxes).
Further support for this model was obtained by Southern blotting analysis of Sf[beta]4GalT DNA digested with BglII or SalI, each of which has no recognition site in pIE1HRGalT (Figure 2A). In both cases, a single band was detected, as expected if pIE1HRGalT is integrated at a single site in the Sf[beta]4GalT genome (Figure 2C). Slight differences in the sizes of the fragments obtained after digestion with these two different enzymes argue against the possibility that a concatemerized form of the plasmid might exist as a stable episome in Sf[beta]4GalT cells. We also addressed this possibility by Southern blotting analysis of Hirt fractions (Hirt, 1967), which revealed intense hybridization with DNA from Hirt pellets and minor, if any, hybridization with DNA from Hirt supernatants (data not shown). Thus, all our Southern blotting results indicated that multiple copies of pIE1HRGalT are integrated into a single site in the Sf[beta]4GalT genome, as depicted in Figure 2B,C.
Baculovirus infection of Sf[beta]4GalT cells
One-step growth curves were done to compare baculovirus replication in Sf9 and Sf[beta]4GalT cells. The results showed that there were no differences in the kinetics of viral replication or amounts of progeny produced (Figure 3). Thus, expression of [beta]1,4-galactosyltransferase activity had no beneficial or adverse effect on the ability of Sf[beta]4GalT cells to serve as hosts for baculovirus infection.
Figure 3. Baculovirus replication in Sf[beta]4GalT cells. One-step growth curves were done to compare the replication of wild-type baculovirus in Sf9 or Sf[beta]4GalT cells, as described in Materials and methods. The open circles show the results obtained with Sf9 cells, and the solid circles show the results obtained with Sf[beta]4GalT cells.
Previous analyses of two other stably-transformed Sf9 cell subclones that express E.coli [beta]-galactosidase or human tissue plasminogen activator (t-PA) revealed that baculovirus infection initially stimulates, then eliminates expression of the integrated transgene at the transcriptional level (Jarvis, 1993). This prompted us to determine how viral infection would influence the intrinsic level of [beta]1,4-galactosyltransferase activity in Sf[beta]4GalT cells. We found that this activity was initially stimulated and that it gradually declined later in infection (Figure 4). However, the level of [beta]1,4-galactosyltransferase activity observed at 48 h postinfection was similar to the level found in uninfected Sf[beta]4GalT cells. Thus, the potential baculovirus-mediated shutdown of transgene expression was not a problem, as [beta]1,4-galactosyltransferase activity persisted until very late times after infection.
Figure 4. Influence of baculovirus infection on [beta]1,4-galactosyltransferase activity in Sf[beta]4GalT cells. Sf[beta]4GalT cells were mock-infected (open circles) or infected (solid circles) with wild type baculovirus exactly as described for the one-step growth curves in Materials and methods. After the viral adsorption period, the cells were washed three times with TN-MFH medium containing 10% heat-inactivated fetal bovine serum then seeded at a density of 0.5 × 106 cells/ml into 50 ml spinner flasks. Samples containing equal numbers of cells were removed from the spinners at various times after infection, detergent-extracted, and the extracts were assayed for [beta]1,4-galactosyltransferase activity as described in Materials and methods.
Galactosylation of foreign glycoproteins expressed in Sf[beta]4GalT cells
gp64 is a major envelope glycoprotein of progeny virions produced during baculovirus infection. gp64 has four N-linked oligosaccharide side chains and at least one is converted to an endo-[beta]-D-N-acetylglucosaminidase H-resistant structure (Jarvis and Garcia, 1994). However, gp64 contains no detectable [beta]-linked galactose, [alpha]2,3-linked sialic acid, or [alpha]2,6-linked sialic acid when analyzed by lectin blotting (Jarvis and Finn, 1995). Thus, gp64 can be conveniently used to monitor N-linked oligosaccharide processing in baculovirus-infected insect cells. Progeny virions were partially purified from baculovirus-infected Sf9 or Sf[beta]4GalT cells and, as a positive control, progeny also were purified from Sf9 cells infected with AcP(+)IE1GalT, the recombinant virus that expresses a [beta]1,4-galactosyltransferase cDNA early in infection (Jarvis and Finn, 1996). gp64 was extracted from the partially-purified virus preparations, immunoprecipitated, and the immunoprecipitates were analyzed by lectin blotting as described in Materials and methods. The results showed that concanavalin A (Con A),which is specific for [alpha]-linked mannose, bound to gp64 from wild type virus grown in either Sf9 or Sf[beta]4GalT cells (Figure 5A). By contrast, Ricinus communis agglutinin (RCA), which is specific for [beta]-linked galactose, bound only to gp64 from wild type virus grown in Sf[beta]4GalT cells or from recombinant virus grown in Sf9 cells. In every case, lectin binding was carbohydrate-specific, as binding was eliminated or greatly reduced by preincubating the lectin with competing sugars (Figure 5B). Furthermore, every lectin blot included a mammalian glycoprotein, IgG, as an internal positive control for lectin binding. Finally, pretreatment of gp64 with peptide:N-glycosidase F (Figure 6A) or Diplococcus pneumoniae [beta]-galactosidase (Figure 6B) precluded RCA binding. These results indicated that RCA specifically binds to galactose in a terminal [beta]1,4-linked position of an N-linked oligosaccharide side chain on gp64. Together, these results demonstrate that Sf[beta]4GalT cells, unlike the parental Sf9 cells, can terminally [beta]1,4-galactosylate gp64 during baculovirus infection.
Figure 5. [beta]1,4-galactosylation of gp64 by Sf[beta]4GalT cells. Progeny budded virions were partially purified from Sf9 or Sf[beta]4GalT cells infected with wild type baculovirus (wt) or AcP(+)IE1GalT (rec) as detailed in Materials and methods. gp64 was extracted, immunoprecipitated, resolved by SDS-PAGE, and transferred to Immobilon. After blocking, the blots were cut into strips and probed with rabbit anti-gp64 (Ab), Con A, or RCA. Each lectin was preincubated in buffer alone (A) or in buffer containing excess competing sugar (B) prior to being used to probe the filters. Lectin or antibody binding was detected with alkaline phosphatase-conjugated secondary antibodies and a standard color reaction as described in Materials and methods. The arrows on the right mark the positions of gp64 and IgG heavy chain, which served as an internal positive control for lectin binding. The arrows and numbers on the left indicate the positions and sizes (in kDa) of protein standards.
Figure 6. Effects of glycosidases on RCA binding to gp64. gp64 was extracted and immunoprecipitated from partially-purified progeny virions from Sf9 or Sf[beta]4GalT cells infected with wild-type baculovirus (wt) or AcP(+)IE1GalT (rec). gp64 was then recovered and treated with buffer alone (-), peptide:N-glycosidase F (lanes marked + in A) or Diplococcus pneumoniae [beta]-galactosidase (lanes marked + in B) as described in Materials and methods. The reaction products were resolved by SDS-PAGE, transferred to Immobilon, and the filters were cut into strips and probed with either anti-gp64 (Ab) or RCA (RCA). The arrows on the right mark the positions of gp64 and IgG heavy chain, which served as an internal positive control for lectin binding, while the arrows and numbers on the left indicate the positions and sizes (in kDa) of protein standards.
Our final goal was to determine if Sf[beta]4GalT cells could produce a [beta]1,4-galactosylated end-product when infected by a conventional recombinant baculovirus expression vector. With this type of vector, transcription of the foreign gene is controlled by the polyhedrin promoter and the foreign protein is expressed even later than gp64. While gp64 is expressed during the late phase of baculovirus infection, with a peak around 24 hr, polyhedrin-mediated expression occurs during the very late phase, which begins around 20 h and peaks around 36 h postinfection. A recombinant virus encoding t-PA, a secreted human glycoprotein, was chosen for this analysis. Sf[beta]4GalT or Sf9 cells were mock-infected, infected with wild-type virus, or infected with the recombinant virus (NTPA). The cells were radiolabeled, t-PA was immunoprecipitated from the extracellular medium, and the immunoprecipitates were analyzed by lectin blotting. The t-PA band was identified by its absence in mock- and wild type virus-infected controls and in control immunoprecipitations done with normal goat serum (Figure 7A). t-PA was present in every lane derived from recombinant virus-infected cells, as shown by direct autoradiography of the dried lectin blots (Figure 7C-D). However, only the t-PA produced by Sf[beta]4GalT cells was recognized by RCA (Figure 7A). There was no detectable RCA binding when the lectin was preincubated with excess competing galactose (Figure 7B).
Figure 7. [beta]1,4-Galactosylation of recombinant human t-PA by Sf[beta]4GalT cells. Sf9 or Sf[beta]4GalT cells were mock-infected (mock), infected with wild type baculovirus (wt), or infected with NTPA (rec), a conventional baculovirus vector encoding human t-PA under the control of the polyhedrin promoter. The cells were radiolabeled and the extracellular fractions were collected and immunoprecipitated with goat anti-t-PA ([alpha]-t-PA) or normal goat serum (NGS) as described in Materials and methods. The immunoprecipitates were analyzed by lectin blotting using RCA preincubated in buffer alone (A and C) or buffer plus excess galactose (B and D) as described in the Figure 5 caption. (A) and (B) show the results of the alkaline phosphatase color reactions, while (C) and (D) show autoradiograms of the same blots. The arrows on the right mark the positions of t-PA and IgG heavy chain. The arrows and numbers on the left indicate the positions and sizes (in kDa) of protein standards.
Discussion
The main purpose of this study was to determine whether the N-glycosylation pathway in lepidopteran insect cells could be directly extended by stable genetic transformation. An expression plasmid was constructed in which a bovine [beta]1,4-galactosyltransferase cDNA was placed under the transcriptional control of an immediate early baculovirus promoter. This plasmid was used to produce a stably transformed Sf9 cell subclone that contains multiple copies of the plasmid integrated at a single site in its genome. Unlike the parental Sf9 cell line, these transformed cells constitutively expressed [beta]1,4-galactosyltransferase activity. Furthermore, this activity contributed to the N-glycosylation pathway in these cells, which were able to add [beta]1,4-galactose to N-linked glycans on two different foreign glycoproteins expressed during baculovirus infection. These results demonstrate that stable genetic transformation can, indeed, be used to add a mammalian glycosyltransferase to lepidopteran insect cells and to directly extend their N-glycosylation pathway.
Most investigators recognize lepidopteran insect cells as hosts for recombinant baculovirus expression vectors (Summers and Smith, 1987; O'Reilly et al., 1992; Jarvis, 1997). Lepidopteran insect cells are extremely important in this context, but insect cells, in general, also provide interesting and unique model systems for studying protein glycosylation. The N-linked oligosaccharide processing pathway in insect cells probably differs from, and is surely not as well-defined as, the corresponding pathway in mammalian cells. Structural studies have shown that the N-linked glycans from most insect cell-derived glycoproteins lack penultimate galactose and terminal sialic acid (März et al., 1995; Jarvis, 1997). These data, together with the negligible levels of galactosyltransferase and sialyltransferase activities found in insect cell extracts (Butters et al., 1981), suggest that insect cells lack functional levels of these terminal N-linked oligosaccharide processing enzymes. Moreover, it has been found that some insect cells have a membrane-bound N-acetylglucosaminidase which converts processing intermediates to trimmed structures that cannot be extended by galactosyltransferases or sialyltransferases (Licari et al., 1993; Altmann et al., 1995). Yet, structural studies of a few glycoproteins have shown that insect cells can sometimes produce trimmed and extended N-linked glycans similar or identical to those produced by mammalian cells (Davidson et al., 1990; Ogonah et al., 1996). Taken together, these findings make it difficult to understand the precise nature of the N-glycosylation pathway in insect cells. They also make it difficult to predict with absolute certainty what kind of N-linked carbohydrate structure one should expect to find on any specific recombinant glycoprotein produced in the baculovirus-insect cell system.
The results of the present study provide new information concerning the nature of the N-glycosylation pathway in insect cells. We have shown that an exogenous mammalian glycosyltransferase cDNA can be introduced and constitutively expressed in uninfected lepidopteran insect cells, resulting in the production of a functional enzyme that can contribute to the N-glycan processing pathway in these cells. Previous work had shown that functional mammalian N-acetylglucosaminyltransferase I (Wagner et al., 1996) or [beta]1,4-galactosyltransferase (Jarvis and Finn, 1996) could be introduced by infecting insect cells with baculovirus expression vectors engineered to express these enzymes. But, this study is the first to show that a transferase can be added to uninfected lepidopteran insect cells by stable genetic transformation in the absence of viral infection. The ability of [beta]1,4-galactosyltransferase to function as part of the native insect cell N-linked oligosaccharide processing pathway indicates that these cells must be able to properly localize the heterologous enzyme, synthesize and transport the nucleotide sugar needed to transfer galactose to an acceptor molecule, and produce the N-linked oligosaccharide acceptor. This raises several interesting questions. What feature of the heterologous enzyme signals its proper localization, and precisely where is this enzyme localized in these insect cells? Why would insect cells, which usually fail to add galactose to newly-synthesized glycoproteins, have the machinery and expend the energy needed to produce and transport the nucleotide sugar? What is the spatial and temporal relationship between the N-acetylglucosaminidase and [beta]1,4-galactosyltransferase activities in these cells, each of which produces structures that cannot be processed by the other? Perhaps our new data provide an additional hint that Sf9 cells have the genotypic potential to extend N-linked intermediates, but, for unknown reasons, are usually unable to fulfill this potential. Finally, one might ask about the phenotypic effects of extending the insect cell N-glycosylation pathway. Interestingly, we found that the addition of [beta]1,4-galactosyltransferase to Sf9 cells had no measurable effect on their growth rate, morphology, or ability to serve as hosts for baculovirus infection.
From a more practical perspective, the results of this study provide a new tool for foreign glycoprotein production by baculovirus expression vectors. None of the lepidopteran insect cell lines that are commonly used as hosts for baculovirus expression vectors can routinely produce glycoproteins with trimmed and extended N-linked oligosaccharide side chains. By contrast, the Sf[beta]4GalT cells described in this study were able to add [beta]1,4-galactose to N-linked glycans on two different foreign glycoproteins, including one human secretory glycoprotein encoded by a conventional baculovirus expression vector. We have maintained Sf[beta]4GalT cells continuously for over 100 passages in our lab and, while we have not quantitatively monitored the amounts of [beta]1,4-galactosyltransferase activity in these cells at various passage levels, we know that they can still produce [beta]1,4-galactosylated gp64 (J.R. Hollister and D.L. Jarvis, unpublished observations). Thus, Sf[beta]4GalT appears to be a genetically stable insect cell line that should be able to routinely produce more extensively processed foreign glycoproteins. Sf[beta]4GalT and Sf9 cells could be used to produce galactosylated and nongalactosylated recombinant glycoproteins, respectively, for interesting functional comparisons. This is important because there are no specific inhibitors of [beta]1,4-galactosyltransferase that can be used to study the function of this late step in N-linked oligosaccharide processing. Perhaps more importantly, it should be possible to use baculovirus vectors or stable transformation methods to introduce additional mammalian processing enzymes, namely sialyltransferase(s), into Sf[beta]4GalT cells. The resulting cells might be able to produce more completely 'mammalianized" foreign glycoproteins with complex N-linked glycans containing penultimate galactose and terminal sialic acid. We anticipate that this goal can be achieved by further metabolic engineering of the baculovirus-insect cell system.
Materials and methods
Plasmids
pIE1HRGalT is an immediate-early expression plasmid designed to express [beta]1,4-galactosyltransferase under the transcriptional control of the Autographa californica multicapsid nuclear polyhedrosis virus (AcMNPV) ie1 promoter. pIE1HRGalT was constructed by subcloning the 1.5 kb BamHI fragment of a bovine [beta]1,4-galactosyltransferase cDNA (Shaper et al., 1986) into the unique BglII site of pIE1HR3 (Jarvis et al., 1996). This fragment includes the entire coding sequence for the short form of bovine [beta]1,4-galactosyltransferase (Russo et al., 1990) plus 3 bp of upstream and 300 bp of downstream noncoding sequences. The construction of pIE1Neo, which encodes a neomycin phosphotransferase gene under the control of the ie1 promoter, has been described previously (Jarvis et al., 1990).
Cells and viruses
Sf9 cells, a subclone of IPLB-SF21-AE (Vaughn et al., 1977), were routinely maintained as adherent cultures in antibiotic-free TN-MFH medium supplemented with 10% (v/v) heat-inactivated fetal bovine serum (Sigma Chemical Co., St. Louis, MO) as described previously (Summers and Smith, 1987). The Sf[beta]4GalT derivative was produced by stable transformation of Sf9 cells, using a modification of an established procedure (Jarvis et al., 1990; Jarvis and Guarino, 1995). Briefly, Sf9 cells were cotransfected with a mixture of pIE1HRGalT plus pIE1Neo using a calcium phosphate method (Summers and Smith, 1987). The cells were allowed to recover, selected in 1 mg/ml G418 (GIBCO-BRL Life Sciences; Grand Island, NY) for 2 weeks, and cloned by limiting dilution in 96-well plates. After stepwise amplification into larger cultures, neomycin-resistant clones were screened for [beta]1,4-galactosyltransferase enzymatic activity, as described below. The clone with the highest activity was designated Sf[beta]4GalT and used for the remainder of this study. Sf[beta]4GalT cells were maintained as adherent cultures in TN-MFH medium supplemented with 10% (v/v) heat-inactivated fetal bovine serum, 1.25 µg of amphotericin B per ml (Sigma), and 25 µg of gentamicin per ml (Sigma).
AcMNPV, strain E2, was used as the wild-type baculovirus in this study (Summers and Smith, 1987). AcP(+)IE1GalT is an immediate early baculovirus vector that expresses [beta]1,4-galactosyltransferase under ie1 control (Jarvis and Finn, 1996), and NTPA is a conventional baculovirus vector that expresses human tissue plasminogen activator under polyhedrin control.
[beta]1,4-galactosyltransferase enzymatic activity assays
Extracts were prepared from uninfected or infected Sf9 or Sf[beta]4GalT cells as described previously (Jarvis and Finn, 1996) and total protein concentrations were determined using the bicinchoninic acid method with BSA as the standard (Smith et al., 1985; Pierce Chemical Company; Rockford, IL). Triplicate samples, each containing 100 µg of solubilized protein, were assayed for [beta]1,4-galactosyltransferase activity in a total volume of 110 µl of 10 mM HEPES (pH 7.4), 140 mM NaCl, 5 mM MnCl2, and 0.5% (v/v) NP40 containing 0.3 µCi of uridine diphosphate [6-3H] galactose (15 Ci/mmol; American Radiolabeled Chemicals, Inc., St. Louis, MO) and 450 µg/ml of ovalbumin (Grade V, Sigma). After 1 h at 37°C, the reaction mixtures were quenched with 0.4 ml of cold assay buffer and spotted onto glass fiber filters (Whatman GF/D; Hillsboro, OR). The filters were dried and washed once with cold 10% (w/v) trichloroacetic acid, twice with cold 5% (w/v) trichloroacetic acid, once with cold 95% (v/v) ethanol, and once with absolute ethanol. The filters were then redried and placed in vials containing liquid scintillation cocktail (Packard UltimaGold F; Meriden, CT), and radioactivity was measured in a liquid scintillation counter (Beckman model LS6000-IC; Fullerton, CA). [beta]1,4-galactosyltransferase enzymatic activity was presented as µmol galactose transferred per µg of protein (or per million cells) per min under the reaction conditions described above.
Genomic Southern blotting
Genomic DNA was prepared from Sf9 or Sf[beta]4GalT cells by using a standard procedure (Sambrook et al., 1989), and 12.5 µg aliquots were digested with HincII, EcoRV, SalI, or BglII under standard conditions (Sambrook et al., 1989). The digests were resolved on a 0.6% (w/v) agarose gel, depurinated by soaking for 10 min in 250 mM HCl, and transferred to a positively charged nylon membrane under alkaline conditions, as recommended by the manufacturer (Bio-Rad Laboratories, Hercules, CA). The membrane was prehybridized for 1 h at 43°C in 0.12 M Na2HPO4 (pH 7.2) containing 250 mM NaCl, 7% (w/v) SDS, 1 mM EDTA, and 50% (v/v) formamide. An overnight hybridization was performed under the same conditions with the addition of a 1095 bp BsaI fragment of pIE1HRGalT, which had been gel purified twice and uniformly labeled by the random primer method (Feinberg and Vogelstein, 1983). After hybridization, the filters were washed with a series of solutions recommended by the manufacturer, blotted dry, sealed in baggies, and exposed to Kodak X-OMAT AR x-ray film.
One-step growth curves
Sf9 or Sf[beta]4GalT cells were pelleted in a 50 ml polypropylene centrifuge tube, and then gently resuspended with a small volume of TN-MFH medium containing wild type baculovirus to initiate infections at a multiplicity of about 10 plaque-forming units per cell. The virus and cells were incubated together on a rocking platform for 1 h at 28°C, and then the cells were repelleted, washed with TN-MFH medium three times, and dispensed in 3.0 ml aliquots into 6-well culture plates at a density of 1.0 × 106 cells/well. Growth media were harvested from triplicate plates at various times postinfection and pooled, and progeny virus production was quantified by plaque assay on Sf9 cells, as described previously (Summers and Smith, 1987).
Analysis of gp64 glycosylation
Sf9 or Sf[beta]4GalT cells were grown to a density of 1.0 × 106 cells/ml in a 250 ml spinner flask, and then wild type baculovirus or AcP(+)IE1GalT was added at a low multiplicity of infection (<1 plaque-forming unit/cell). When over 50% of the cells contained viral occlusions, the cultures were harvested, clarified by low speed centrifugation, and budded virus was pelleted from the cell-free medium by centrifugation in a Beckman Type 35 rotor for 30 min at 30,000 r.p.m. at 4°C. The budded virus pellets were gently resuspended with 0.1× TE (1 mM Tris-HCl, pH 8.0; 0.1 mM EDTA), and then carefully layered onto 25-56% (w/v) linear sucrose gradients. The gradients were centrifuged in a Beckman SW41 rotor for 90 min at 24,000 r.p.m. at 4°C. The budded virus band was harvested, diluted in 0.1× TE, and reconcentrated by centrifugation in a Beckman SW41 rotor for 30 min at 24,000 r.p.m. at 4°C. Finally, the budded virus pellets were resuspended with 0.1× TE and gp64 was extracted with 50 mM Tris-HCl (pH 8.0) containing 140 mM NaCl and 1% (v/v) NP40. gp64 was immunoprecipitated with a mouse monoclonal antibody against gp64, as described previously (Hohmann and Faulkner, 1983; Jarvis and Garcia, 1994). The immunoprecipitates were resolved by discontinuous SDS-PAGE (Laemmli, 1970) and transferred to Immobilon membranes (Millipore Corp., Bedford, MA) by using an electrophoretic transfer method (Towbin et al., 1979). The resulting blots were cut into strips and blocked by soaking overnight at 4°C in 50 mM Tris-HCl (pH 7.5) containing 150 mM NaCl and 0.5% (v/v) Tween 20 (Sigma). The strips were then probed with various digoxigenylated lectins (Boehringer-Mannheim Corp., Indianapolis, IN) or a rabbit antibody against gp64. The lectins were preincubated for 2 h at room temperature with buffer alone, 0.7 M methyl [alpha]-mannopyranoside (for Con A), or 0.7 M galactose (for RCA). Lectin or antibody binding was detected by a secondary reaction with alkaline phosphatase-conjugated antibodies against digoxigenin (Boehringer-Mannheim) or rabbit IgG (Promega, Inc., Madison, WI), respectively, and the secondary probes were detected by a standard color reaction (Blake et al., 1984). Pretreatments of gp64 with peptide:N-glycosidase F (Tarentino et al., 1985; Boehringer Mannheim) or Diplococcus pneumoniae [beta]-galactosidase (Jacob and Scudder, 1994; Boehringer Mannheim) prior to lectin blotting were done as described previously (Jarvis and Finn, 1996).
Analysis of t-PA glycosylation
Sf9 or Sf[beta]4GalT cells were seeded at a density of 2.5 × 106 cells/flask in 25 cm2 culture flasks and allowed to attach to the plastic for 1 h. The medium was drained and the cells were mock-infected or infected with wild-type baculovirus or NTPA at a multiplicity of infection of about 10 plaque-forming units per cell. After a 1 h adsorption period, the inoculum was removed from each flask and the cells were fed with Grace's medium containing 0.5% heat-inactivated fetal bovine serum. At 20 h postinfection, the medium was removed and replaced by Grace's medium containing 0.5% serum and one-tenth the normal concentration of methionine plus 100 µCi per ml Translabel (ICN Radiochemicals, Irvine, CA), which is about 20% [35S]cysteine and 80% [35S]methionine. The growth medium was harvested at 48 h postinfection, clarified by low speed centrifugation, adjusted to 1% (v/v) NP40, and immunoprecipitated with normal goat serum or goat anti-t-PA (American Diagnostica, New York). The washed immunoprecipitates were resolved by gel electrophoresis and analyzed by lectin blotting as described above for gp64. In this case, however, autoradiography was used to identify the position of t-PA in the blots because a second antibody against human t-PA was unavailable.
Acknowledgments
This work was supported by grants from the National Institutes of Health (GM49734 to D.L.J. and GM45799 to J.H.S.).
Abbreviations
AcMNPV, Autographa californica multicapsid nuclear polyhedrosis virus; AcP(+)IE1GalT, immediate early recombinant baculovirus encoding bovine [beta]1,4-galactosyltransferase; Con A, concanavalin A; ie1, AcMNPV immediate early gene 1; IgG, immunoglobulin G; NP40, Nonidet-P40; NTPA, recombinant baculovirus encoding native human tissue plasminogen activator; PAGE, polyacrylamide gel electrophoresis; RCA, Ricinus communis agglutinin; SDS, sodium dodecyl sulfate; Sf9, Spodoptera frugiperda clone 9 insect cell line; Sf[beta]4GalT, Stably-transformed Sf9 derivative that constitutively expresses bovine [beta]1,4-galactosyltransferase; t-PA, tissue plasminogen activator.
References
3Present address: Department of Molecular Biology, University of Wyoming, Laramie, WY 82071, USA
4To 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: 2 Apr 1998
Copyright© Oxford University Press, 1998.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  H. Kajiura, H. Koiwa, Y. Nakazawa, A. Okazawa, A. Kobayashi, T. Seki, and K. Fujiyama Two Arabidopsis thaliana Golgi {alpha}-mannosidase I enzymes are responsible for plant N-glycan maturation Glycobiology, February 1, 2010; 20(2): 235 - 247. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Leonard, D. Rendic, C. Rabouille, I. B. H. Wilson, T. Preat, and F. Altmann The Drosophila fused lobes Gene Encodes an N-Acetylglucosaminidase Involved in N-Glycan Processing J. Biol. Chem., February 24, 2006; 281(8): 4867 - 4875. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Vadaie and D. L. Jarvis Molecular Cloning and Functional Characterization of a Lepidopteran Insect {beta}4-N-Acetylgalactosaminyltransferase with Broad Substrate Specificity, a Functional Role in Glycoprotein Biosynthesis, and a Potential Functional Role in Glycolipid Biosynthesis J. Biol. Chem., August 6, 2004; 279(32): 33501 - 33518. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Koles, K. D. Irvine, and V. M. Panin Functional Characterization of Drosophila Sialyltransferase J. Biol. Chem., February 6, 2004; 279(6): 4346 - 4357. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. Choi, N. Tomiya, J. H. Kim, J. M. Slavicek, M. J. Betenbaugh, and Y. C. Lee N-glycan structures of human transferrin produced by Lymantria dispar (gypsy moth) cells using the LdMNPV expression system Glycobiology, July 1, 2003; 13(7): 539 - 548. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. J. Aumiller, J. R. Hollister, and D. L. Jarvis A transgenic insect cell line engineered to produce CMP-sialic acid and sialylated glycoproteins Glycobiology, June 1, 2003; 13(6): 497 - 507. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Hollister, H. Conradt, and D. L. Jarvis Evidence for a sialic acid salvaging pathway in lepidopteran insect cells Glycobiology, June 1, 2003; 13(6): 487 - 495. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. van Remoortere, C. M.C. Bank, A. K. Nyame, R. D. Cummings, A. M. Deelder, and I. van Die Schistosoma mansoni-infected mice produce antibodies that cross-react with plant, insect, and mammalian glycoproteins and recognize the truncated biantennaryN-glycan Man3GlcNAc2-R Glycobiology, March 1, 2003; 13(3): 217 - 225. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Tomiya, D. Howe, J. J. Aumiller, M. Pathak, J. Park, K. B. Palter, D. L. Jarvis, M. J. Betenbaugh, and Y. C. Lee Complex-type biantennary N-glycans of recombinant human transferrin from Trichoplusia ni insect cells expressing mammalian {beta}-1,4-galactosyltransferase and {beta}-1,2-N-acetylglucosaminyltransferase II Glycobiology, January 1, 2003; 13(1): 23 - 34. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Guo, T. Sato, K. Shirane, and K. Furukawa Galactosylation of N-linked oligosaccharides by human {beta}-1,4-galactosyltransferases I, II, III, IV, V, and VI expressed in Sf-9 cells Glycobiology, October 1, 2001; 11(10): 813 - 820. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Marchal, M. Cerutti, A.-M. Mir, S. Juliant, G. Devauchelle, R. Cacan, and A. Verbert Expression of a membrane-bound form of Trypanosoma cruzi trans-sialidase in baculovirus-infected insect cells: a potential tool for sialylation of glycoproteins produced in the baculovirus-insect cells system Glycobiology, July 1, 2001; 11(7): 593 - 603. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Bakker, M. Bardor, J. W. Molthoff, V. Gomord, I. Elbers, L. H. Stevens, W. Jordi, A. Lommen, L. Faye, P. Lerouge, et al. Galactose-extended glycans of antibodies produced by transgenic plants PNAS, February 27, 2001; 98(5): 2899 - 2904. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. R. Hollister and D. L. Jarvis Engineering lepidopteran insect cells for sialoglycoprotein production by genetic transformation with mammalian {{beta}}1,4-galactosyltransferase and {{alpha}}2,6-sialyltransferase genes Glycobiology, January 1, 2001; 11(1): 1 - 9. [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]
|
|
|
|
 |
|
 D. L. Jarvis, L. Wills, G. Burow, and D. A. Bohlmeyer Mutational Analysis of the N-Linked Glycans on Autographa californica Nucleopolyhedrovirus gp64 J. Virol., December 1, 1998; 72(12): 9459 - 9469. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||