MUC1 in Insect Cells 471
471
39
Expression of MUC1 in Insect Cells
Using Recombinant Baculovirus
Pawel Ciborowski and Olivera J. Finn
1. Introduction
MUC1 mucin undergoes multistep posttranslational modifications before it is
finally expressed on the apical surface of mammalian ductal epithelial cells. Two early
precursor proteins are both N-glycosylated and differ in molecular weight owing to a
proteolytic cleavage of a 20-kDa fragment. Proteolytically modified form is trans-
ported to the Golgi, where it undergoes extensive, although not complete, O-gly-
cosylation on serine and threonine residues within the tandem repeat (TR) region.
MUC1 is then transported to the cell surface. For additional glycosylation and
sialylation, surface MUC1 is internalized and directed to trans-Golgi compartments.
Mature form is again transported to the cell surface (1).
MUC1 expressed by malignant epithelial cells such as breast and pancreatic adeno-
carcinomas is underglycosylated (aberrantly glycosylated), which makes it structur-
ally and antigenically distinct from that expressed by normal cells (2). As such, it may
be an excellent target for immunotherapy. One of the ways to utilize tumor-specific
forms of this molecule is as immunogens. Purifying these forms from tumor cells is
not feasible because it is a labor-intensive process that gives low yields. A much more
desirable approach is purification of a recombinant molecule from an appropriate
expression system. Recombinant MUC1 expressed in a convenient prokaryotic sys-
tem that does not glycosylate proteins, such as Escherichia coli, undergoes rapid and
random proteolytic degradation. To obtain underglycosylated recombinant tumorlike
forms of MUC1 in mammalian cells through expression vectors such as vaccinia virus,
retroviral vectors, and plasmid vectors requires a prolonged treatment of infected or
transfected cells with toxic and expensive inhibitors of O-linked glycosylation (3,4).
Furthermore, vaccinia and retroviral constructs spontaneously recombine out most TRs
that characterize the major portion and the most immunogenic portion of MUC1 (5).
We explored the baculovirus system that allows expression of MUC1 mucin in
Spodoptera frugiperda Clone 9 (Sf-9) insect cells. We found that these cells, when
From:
Methods in Molecular Biology, Vol. 125: Glycoprotein Methods and Protocols: The Mucins
Edited by: A. Corfield © Humana Press Inc., Totowa, NJ
472 Ciborowski and Finn
infected with a MUC1 recombinant baculovirus, produce fully glycosylated, full-size
(no deletion of TRs) molecules that are expressed on the cell surface (6). Moreover,
under specific starvation growth conditions that we determined empirically, Sf-9 cells
can also produce underglycosylated MUC1, similar to the MUC1 produced by tumor
cells. The state of glycosylation of various forms can be evaluated by their migration
in sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gels and
reactivity with different anti-MUC1 antibodies in Western blot analysis (6,7). In this
chapter, we present the techniques of expression of MUC1 mucin using three
baculoviral vectors: pBlueBacIII, pFastBac, and pIE1-4. Additional vectors are com-
mercially available and, as one can expect, more will emerge on the market in the
future. In our opinion, they provide an ideal expression system to study different forms
of MUC1 protein, their function, and utility.
2. Materials
2.1. Cloning Reagents
1. Vectors: pBlueBacIII was purchased from Invitrogen, San Diego, CA (see Note 1);
pFastBac was purchased from Gibco, Life Technologies, Grand Island, NY; and pIE1-4
was purchased from Novagen, Madison, WI.
2. Competent E. coli cells such as MAX Efficiency DH5α™ Competent Cells and MAX
Efficiency DH10Bac™ Competent Cells were obtained from Gibco-BRL.
3. Restriction enzymes, agarose, ligase, and other reagents for cloning may be obtained from
any supplier of molecular biology reagents. Wizard™ Minipreps and Wizard™ Megapreps
were obtained from Promega, Madison, WI. Cationic liposomes InsecticinPlus™ were
obtained from Invitrogen, but can be also obtained from other commercial sources.
BluoGal and isopropyl-β-
D
-thiogalactopyranoside (IPTG) were purchased from Sigma,
St. Louis, MO; X-gal was purchased from Boehringer Mannheim, Indianapolis, IN; and
SeaPlaque agarose was purchased from FMC BioProducts, Rockland, ME.
2.2. Cells, Media, and Antibodies
1. The insect cell line Sf-9 can be obtained from American Type Culture Collection
(Rockville, MD) or from other suppliers such as Invitrogen, San Diego, CA.
2. Hink’s TNM-FH Insect Medium can be obtained from several sources such as JRH Bio-
sciences, Lenexa, KS. Penicillin, streptomycin, fungizone, and geneticin can be obtained
from Gibco. Fetal bovine serum (FBS) was from Gibco-BRL.
3. Anti-MUC1 antibodies used in this study are not commercially available. Monoclonal
antibodies (MAbs) used for Western blot and flow cytometry analysis are listed in
Table 1. The TD-4 MUC1 Workshop (see ref. 7, pp. 1–152) provides the most up-to-date
list of anti-MUC1 antibodies, their specific reactivities, and their sources.
4. Tissue culture flasks, plates, roller bottles, and disposable plastic tubes of various sizes
can be obtained from various sources, e.g., Sarsted, Falcon, etc. Any 27°C incubator can
be used, although one with a water jacket is recommended.
2.3. Western Blot
All reagents and equipment for PAGE and Western blot, except nitrocellulose, were
purchased from Bio-Rad, Hercules, CA. Other suppliers can also be used. Nitrocellu-
lose BioBlot-NC was purchased from Corning Costar, Corning, NY. Chemilumines-
MUC1 in Insect Cells 473
cence Western blotting detection kit was purchased from Amersham, Buckinghamshire,
England.
3. Methods
3.1. Vector Construction
The cDNAs coding for MUC1 of various lengths owing to various numbers of TRs
were obtained from previously made plasmid constructs. Plasmid expression vectors
encoding MUC1 with 22 repeats (22TRMUC1) and two repeats (2TRMUC1) were
made in our laboratory (3). The cDNAs can be isolated from the plasmid vectors as
HindIII cassettes (Fig. 1). Plasmid expression vectors containing MUC1 cDNA with
42 TRs (42TRMUC1) and MUC1 cDNA without TRs (TR
–
MUC1), both BamHI cas-
settes, were obtained from Dr. A. Hollingsworth, University of Nebraska, Omaha.
3.1.1. Cloning into pBlue
Bac
III Transfer Vector
An example we will use for cloning of the 3.2-kbp cDNA MUC1 with 22 TRs
(22TRMUC1) and the 1.8-kbp cDNA MUC1 with 2 TRs (2TRMUC1). The resulting
pBlueBacIII-22TR-MUC1 recombinant transfer plasmid is used for inserting MUC1
cDNA into the genome of the wild-type Autographa californica Multiple Nuclear
Polyhedrosis Virus (wtAcMNPV), as described under Subheading 3.3.1.
1. Digest pBlueBacIII transfer vector with HindIII or BamHI and treat with calf intestine
phosphatase (CIP) to protect against self-ligation using standard methodology (see Note
1 and 2).
2. Prepare MUC1 cDNA cassette by HindIII digestion.
3. Purify fragments by electrophoresis in 0.7% agarose.
4. Ligate the cDNA cassette into the pBlueBacIII vector using T4 DNA ligase at 16°C over-
night.
Table 1
MUC1 Specific Antibodies
a
Antibody Isotype Specificity
SM-3 IgG1 APDTRP
b
, underglycosylated
c
VU-4-H5 IgG1 PDTRPAP, underglycosylated
c
VU-3-C6 IgG1 PDTRPAP, all forms
BC-3 IgM APDTR, all forms
BC-2 IgG1 APDTR, all form
232A1 IgG Proteolytic cleavage site
a
For more details about antibodies see ISOBM TD-4 International Workshop
on Monoclonal Antibodies against MUC1, Tumor Biology, 1998, 19 (Suppl. 1),
1–152.
b
Single letter code for amino acids. A, alanine; P, proline; T, threonine; D,
glutamic acid; R, arginine.
c
Tumor specific, recognizing underglycosylated but not fully glycosylated
MUC1.
474 Ciborowski and Finn
5. Transform the ligated construct into E. coli MAX Efficiency DH5α Competent Cells fol-
lowing the protocol provided by the manufacturer.
6. Select recombinants using Luria agar with 10 µg/mL of ampicillin (8).
7. Amplify ampicillin-resistant clones in 5-mL Luria broth/ampicillin (8) cultures. Use 1.5–
3.0 mL of the culture to isolate plasmid DNA using Wizard
Minipreps.
8. Analyze recombinant DNA by restriction enzyme digestion for orientation of the insert.
3.1.2. Cloning into pFast
Bac
Transfer Vector
As an example, we will use cloning of 4.6-kbp cDNA with 42 TRs (42TRMUC1).
Fragment of MUC1 cDNA coding for transmembrane and cytoplasmic domains was
replaced with a sequence linking the outer membrane portion of MUC1 with
glycosylphosphatidylinositol (GPI) anchor of human decay accelerating factor. This
new construct (42TRMUC1-GPI) was made in our laboratory and remains as a BamHI
cassette (Alter, M., unpublished data). The resulting pFastBac-42TRMUC1-GPI
recombinant transfer plasmid is used for inserting MUC1 cDNA into the genome of
the wtAcMNPV, as described under Subheading 3.3.2.
1. Linearize pFastBac transfer vector with BamHI digestion and protect it with CIP against
self-ligation using standard methodology.
2. Cut out the 42TRMUC1-GPI cDNA cassette by BamHI digestion.
3. Purify a fragment of the correct size by electrophoresis in 0.7% agarose.
4. Ligate the cDNA cassette into the vector using T4 DNA ligase at 16°C for overnight.
5. Transform E. coli MAX Efficiency DH5α
Competent Cells with the ligated construct.
6. Select recombinants using Luria agar with 10 µg/mL of ampicillin.
7. Select ampicillin-resistant clones, and amplify and purify plasmid DNA using Wizard Minipreps.
8. Analyze recombinant DNA by restriction enzyme digestion for orientation of the insert.
Fig. 1. MUC1 cDNA expression plasmid. The MUC1 cDNA is downstream from transla-
tional start codon. Constructs with 2 or 22 TRs that were made in our laboratory are contained
in the HindIII.
MUC1 in Insect Cells 475
3.1.3. Cloning into the Episomal Transfer Vector pIE1-4
The vector pIEI-4 is used to provide stable expression of a cloned gene from the
baculovirus ie1 promoter. Cells are cotransfected with pIE1-neo providing neomycin
selection marker expressed from ie1 promoter. As an example, we will use cloning of
1.4-kbp MUC1 cDNA that lacks TRs (TR
–
MUC1). The resulting pIE1-4TR
–
MUC1-
GPI recombinant transfer plasmid is used for cotransfection of Sf-9 cells with pIE-
neo, as described under Subheading 3.5.
1. Linearize the pIE1-4 transfer vector with BamHI and protect it with CIP against self-
ligation using standard methodology (see Note 3).
2. Prepare TR
–
MUC1 cDNA cassette by BamHI digestion.
3. Purify the desired fragment by electrophoresis in 0.7% agarose.
4. Ligate the cDNA cassette into the vector using T4 DNA ligase at 16°C for overnight.
5. Transform E. coli MAX Efficiency DH5α Competent Cells with the ligated construct.
6. Select recombinants using Luria agar with 10 µg/mL of ampicillin. Amplify ampicillin-
resistant clones and purify plasmid DNA using Wizard
Minipreps.
7. Analyze recombinant DNA by restriction enzyme digestion for orientation of the insert.
3.2. Conditions for Culturing the Sf-9 Cells
Sf-9 insect cells are cultured in Hink’s TNM-FH Insect Medium supplemented with
5 or 10% FBS and penicillin/streptomycin/fungizone at the concentrations of 100
U/mL, 100 µg/mL, and 2.5 µg/mL, respectively. Cells are grown as a monolayer at
27°C. For small scale growth, 75-cm
2
vented flasks are used (Costar, Cambridge, MA).
Typically 5 × 10
5
cells and 20 mL of medium are used to start the culture of this size.
For larger-scale growth, roller bottles are used. Cultures are usually started at the cell
density of 10
6
cells/mL. During the logarithmic phase of growth, cells typically double
every 24 h. Therefore, equal amounts of fresh medium are added each day to the roller
bottle for up to 300 mL total volume. Figure 2 shows the kinetics of growth in a
typical roller bottle culture (see Note 4).
3.3. Production of Recombinant Virus by Cotransfection with Viral
and Recombinant Transfer Vector DNAs
The wtAcMNPV viral DNA and the recombinant transfer vector DNA are shuttled
into Sf-9 cells by cationic liposomes. Within the cells, transfer vector DNA and viral
DNAs recombine, incorporating the gene of interest into the viral genome. Depending
on the transfer vectors different protocols can be used to make recombinant virus. Two
protocols are given next.
3.3.1. Using pBlue
Bac
III Vector
When using pBlueBacIII vectors, recombination leads to the replacement of the
viral polyhedrin gene (phenotypically occ
+
) with part of the transfer vector containing
lacZ gene and gene of interest. Therefore, the selection is based on the phenotypic
observation—lack of occlusion bodies (occ
-
) and expression of β-galactosidase
(lacZ
+
).
1. Seed Sf-9 cells in a 6-well plate (10
6
cells/well) prior to the cotransfection, and rock them
gently side-to-side for 1 h at room temperature to evenly distribute and attach the cells.
476 Ciborowski and Finn
2. Remove nonattached cells and medium, gently wash the adherent monolayer once with
serum-free medium, cover with 2 mL of serum free medium, and incubate for 30 min at
room temperature.
3. Prepare five independent transfection mixtures. Mix 100, 200, 500, and 750 ng, or 1 µg of
the recombinant pBlueBacIII transfer plasmid, respectively, with 500 ng of linearized
AcMNPV DNA, 40 mL Insectin-Plus Liposomes, and 1 ml of Hink’s TNM-FH Insect
Medium.
4. Vortex transfection mixtures vigorously for 10 s and incubate at room temperature for 30 min.
5. Remove serum-free medium from the cells, cover cell monolayer with one of the transfec-
tion mixtures, swirl to mix, and incubate for 4 h at room temperature with slow rocking.
6. Add 2 mL of complete Hink’s TNM-FH Insect Medium (containing 10% FBS) to each
well, wrap plates with clear plastic wrap, and incubate at 27°C for 48 h.
7. Take 100 µL of the culture supernatant that contains viruses produced by the transfected cells
from each well and screen by plaque assay for the presence of double recombinants (occ
-
,
lacZ
+
). Transfer the remaining medium to sterile microcentrifuge tubes and store at 4°C.
3.3.2. Using pFast
Bac
Transfer Vector
pFastBac transfer vector is a part of the Bac-To-Bac™ Baculovirus Expression
System developed by Gibco. In the first step, competent MAX Efficiency DH10Bac
E. coli cells are transformed with pFastBac donor plasmid with a gene of interest. The
competent DH10Bac E. coli cells contain baculovirus shuttle vector (bacmid) and a
Fig. 2. Growth of SF-9 cells in a typical roller bottle culture. Infection was on d 4 and no
new medium was added afterward. On d 5 all cells were expressing β-galactosidase (see Sub-
heading 3.6.1. for details). Cells were usually harvested after 72 h.
MUC1 in Insect Cells 477
helper plasmid. Bacmid propagates in E. coli and, besides viral DNA, contains several
other elements such as attachment sites for transposon Tn7 and an open reading frame
of lacZ
α
peptide. Recombination between pFastBac transfer vector and bacmid occurs
within bacterial cells by transposition, with the aid of helper plasmid. Another feature
of this system is that insertion of the gene of interest into the viral genome causes
disruption of lacZ gene and E. coli containing recombinant bacmid grow as white
colonies in the presence of Bluo-gal and IPTG. This feature makes the system easy to
use and eliminates posttransfection isolation of recombinant viruses.
1. Transform E. coli MAX Efficiency DH10Bac Competent Cells that contain Bacmid DNA
and helper plasmid. After transformation with recombinant pFastBac-MUC1-42TR-GPI,
the transposition occurs inside E. coli cells disrupting LacZ gene within Bacmid DNA.
2. Select white growing colonies from Luria agar supplemented with kanamycin (50 µg/mL),
tetracycline (10 µg/mL), gentamicin (7 µg/mL), BluoGal (100 µg/mL), and IPTG (40 µg/mL)
to make a larger amount of the recombinant Bacmid DNA.
3. Amplify selected clones and purify Bacmid DNA using Wizard Minipreps.
4. Seed 10
6
Sf-9 cells/well in a 6-well plate immediately prior to transfection, and rock them
gently side-to-side for 1 h at room temperature to evenly distribute and attach the cells.
Alternatively, seed 2.5 × 10
5
/well and grow them until desired density, usually 1 to 2 d.
5. Remove nonattached cells and medium, wash gently the cell monolayer once with serum
free medium, cover with 2 mL of serum-free medium, and incubate for 30 min at room
temperature.
6. Prepare three independent transfection mixtures: 100 to 200 ng of Bacmid DNA, mixed
with 20, 40, or 60 µL Insectin-Plus
Liposomes, and 1 mL of Hink’s TNM-FH Insect
Medium.
7. Vortex transfection mixtures vigorously for 10 s and incubate at room temperature for 30 min.
8. Remove serum-free medium from the wells, cover cell monolayers with the transfection
mixture, swirl to mix, and incubate for 4 h at room temperature with slow rocking.
9. Add 2 mL of complete Hink’s TNM-FH Insect Medium (containing 10% of FBS) to each
well, wrap with clear plastic wrap, and incubated at 27°C for 48 h.
10. Transfer the culture supernatants that contain viruses produced by the transfected cells to
sterile microtubes and store at 4°C.
3.4. Isolation of Recombinant Baculovirus
The culture supernatants from cells cotransfected with the pBlueBacIII as a transfer
vector contain a mixture of recombinant and wild-type viruses. To isolate recombi-
nant viruses, a plaque assay is performed followed by an end-point dilution round of
purification.
3.4.1. Plaque Assay
1. Seed Sf-9 cells in 6-cm dishes with 2 × 10
6
cells/dish, and rock them gently side-to-side
for 1 h at room temperature to evenly distribute and attach the cells.
2. Grow cells at 27°C to approx 80% confluency. Alternatively, seed more cells, and after 2
to 3 h during which the cells attach, the plates are ready for plaque assay.
3. Make a serial 10-fold dilution of the medium harvested from transfected cells, in full
Hink’s TNM-FH Insect Medium. Dilutions should range from 10
–1
to 10
–5
.
4. Remove the medium from the wells and add 1 mL of the culture supernatant containing a
mixture of wild-type and recombinant viruses (viral inoculum) to the side of the dish, and
478 Ciborowski and Finn
tilt the dish slowly to cover evenly all cells. It is important to do this gently in order not to
disturb the attached cells. Incubate at room temperature for 1 h.
5. Melt the required volume of 3% SeaPlaque agarose, cool down to 45°C, and keep in a
water bath. Prewarm to 37°C an equal volume of Hink’s TNM-FH Insect Medium to
which was added 120 µg/mL of X-gal.
6. Remove the viral inoculum by tilting the plate and aspirating from the edge.
7. Mix warm agar with medium and overlay dishes with 5 ml of this medium. Leave leveled
until agarose sets.
8. Incubate at 27°C until blue plaques develop, usually 5–7 d.
9. Using a sterile Pasteur pipet, pick isolated plaques with the recombinant virus (blue plaques
without occlusion bodies), transfer to tubes containing 2 to 3 mL of Hink’s TNM-FH Insect
Medium, and vortex for 30 s. Allow viral particles to diffuse from agar for another hour at
room temperature. This could be used for screening and further purification.
3.5. Production of an Episomal Recombinant Vector
for Stable Expression
3.5.1. Using pIE1-4 Vector
1. Seed Sf-9 cells in a 6-well plate (10
6
cells/well) prior to the cotransfection, and rock them
gently side-to-side for 1 h at room temperature to evenly distribute and attach the cells.
2. Remove nonattached cells and medium, gently wash the adherent monolayer once with
serum-free medium, cover with 2 mL of serum free medium, and incubate for 30 min at
room temperature.
3. Prepare five independent transfection mixtures. Mix 3 µg of recombinant transfer plas-
mid pIE1-4-TR
–
-MUC1; 400 ng of pIE-neo plasmid DNA, and 20, 40, 60, 80, or 100 µL
of Insectin-Plus liposomes. For mocktransfection, use 20 µL of Insectin-Plus liposomes.
4. Vortex transfection mixtures vigorously for 10 s and incubate at room temperature for 30 min.
5. Remove serum-free medium from the wells, cover cell monolayer with the transfection
mixture, swirl to mix, and incubate for 4 h at room temperature with slow rocking.
6. Add 2 mL of complete Hink’s TNM-FH Insect Medium (containing 10% FBS) to each
well, wrap the plates with clear plastic wrap, and incubate at 27°C for 48 h.
7. Replace medium after 48 h with new medium containing 600 µg/mL of neomycin and
grow cells for another 7 d.
8. After 7 d cells can be tested for MUC1 expression by flow cytometry and Western blot
(see Notes 5 and 6).
3.6. Infection of Sf-9 Cells and MUC1 Production
3.6.1. Flask Cultures
1. Grow cells to approx 100% confluency.
2. Aspirate all medium and cover the cell monolayer with the minimal volume of a viral
stock at 3 multiplicity of infection (see Note 7). This is usually 4 to 5 mL/75-cm
2
flask.
3. Rock the flask for 1 h at room temperature.
4. Add 20 mL of serum-free medium and incubate at 27°C for a desired time. To control
yield of infection with recombinant virus carrying lacZ gene, aliquot a small sample of
the culture (cells and supernatant) into a 1.5-mL microtube containing 1 µL of X-Gal at a
concentration of 40 mg/mL, and incubate for 30 to 60 min at room temperature. After that
time, infected cells should exhibit blue color when observed microscopically owing to
β-galactosidase expression, and the culture supernatant should turn blue. If recombinant
virus does not carry and express lacZ gene, other signs of infection such as swollen nuclei
MUC1 in Insect Cells 479
can be used to assess the efficiency of infection. Typically more than 80% of cells are
infected within the first day.
3.6.2. Roller Bottle Cultures
1. Start a roller bottle culture with Sf-9 cells in 20 mL of in Hink’s TNM-FH Insect Medium,
at a density of 10
6
cells/mL. The cell number usually doubles once every 24 h. Expand
cells by adding every day an equal amount of fresh medium: 20 mL on d 2, 40 mL on d 3,
80 mL on d 4, and so forth, up to half the desired final volume of the culture (see Note 8).
2. Infect cells with the viral stock. There are two procedures for infection of a roller bottle
culture. In the first, the culture is infected by adding the viral stock directly to the bottle
and supplementing the culture with an equal amount of fresh medium 2 to 3 h after infec-
tion. The second method is to collect cells by centrifugation, resuspend in minimal vol-
ume of medium, usually one-tenth of the original, add the viral stock, and rock for 1 h at room
temperature. Transfer infected cells back to the bottle, and add fresh medium in the amount
equal to that prior to infection. We found both methods worked equally well. The only differ-
ence is that in the first method, in order to accomplish complete infection of all insect cells
within 24 h, cultures should be infected at 5 m.o.i. or higher. In the second method, although
more laborious, less viral stock is used and infection at 3 m.o.i. is sufficient.
3.7. Starving Sf-9 Cells in Culture
to Obtain Underglycosylayted Forms of MUC1
3.7.1. Flask Cultures
1. Grow cells to 100% confluency, and prolong the time of culture for another day or two
without changing the medium in order to deplete most of the nutrients. At this time, the
number of dead cells slightly increases.
2. Aspirate all medium, clear by centrifugation and filtration through a 0.22-µm filter, and save.
3. Cover the cell monolayer with a minimal volume of the viral stock, usually 4 to 5 mL/75-
cm
2
flask.
4. Rock the flask for 1 h at room temperature.
5. Add the saved nutrients-depleted medium and incubate the culture at 27°C for a desired
time. To control efficiency of infection with recombinant virus carrying lacZ gene, fol-
low the procedure described under Subheading 3.6.2.
3.7.2. Roller Bottle Cultures
1. Start and expand a roller bottle culture as described under Subheading 3.6.
2. Starve cells by growing at high density for 48 h without supplementing with fresh
medium.
3. Infect cells with a viral stock as described under Subheading 3.6. In starved cultures, if
cells are harvested for infection, nutrient-depleted medium should be collected, cleared
by centrifugation and filtration through a 0.22-µm filter, and added back to the bottle.
3.8. Analysis of Recombinant MUC1 Produced by Sf-9 Cells
The quickest method to test the expression of MUC1 in insect cells is by
immunostaining with specific MAbs. There is a large number of well characterized
MAbs against different peptide and sugar epitopes on MUC1. Western blot is the
method of choice for testing total expression, and flow cytometry is used for testing
cell surface expression. The pattern of reactivity of baculovirus expressed MUC1 with
480 Ciborowski and Finn
various anti-MUC1 MAbs does not reflect the degree of glycosylation of the recombi-
nant product. Although we do not provide a detailed protocol for Western blot and
flow cytometry in this chapter, some technical details could be found in the figure
legends (see Note 9).
MUC1 is the only transmembrane mucin known to date. It is transported to the
apical surface of the ductal epithelial cells and anchored in the cell membrane via its
transmembrane domain. MUC1 is removed from the cell surface by proteolytic cleav-
age in the membrane proximal domain, or, to lesser extent, it is internalized and
degraded in the phagolysosomes. In insect cells, MUC1 also undergoes complex pro-
cess of posttranslational modification as in mammalian cells. It is also transported to
the surface of Sf-9 cells (6).
A single, large band with an apparent molecular weight above the 221 kDa, repre-
sents the only form of MUC1 expressed by insect cells when they are grown in fully
supported medium containing 10% FBS. However, when cells are starved for at least
two days prior to infection and then grown in nutrient-depleated medium, the majority
Fig. 3. Different forms of MUC1 expressed in Sf-9 cells using recombinant baculovirus.
Form designated as (A) is a protein precursor which undergoes a proteolytic modifiction
yeilding a 20 kDa smaller protein (B). Form (C) has not been characterized yet, and form (D) is
MUC1 released from the cell surface.
MUC1 in Insect Cells 481
of the MUC1 product is of low molecular weight, as shown in Fig. 3, lane B (72 h post
infection), and lane C (96 h postinfection). Another important fact can be derived from
Fig. 3: The baculovirus expression system is capable of producing the full-length
mucin polypeptide core (approx 106 kDa corresponding to 3.2 kbp cDNA) without the
tandem repeat region undergoing deletions due to homologous recombination, a prob-
lem previously encountered with other expression systems (5).
Based on previous studies of MUC1 and processing (9,10) three out of four major
immunoblotted products shown in lane A of Fig. 3 can be identified. Protein precursor
form (a) undergoes a proteolytic modification yielding a 20 kDa smaller protein desig-
nated as form (b). Form (c) has not been characterized yet, and form (d) is MUC1-
released from the cell surface (Fig. 4, lane 72 h/s). It is important to note that MUC1 is
mostly cell associated and only a very small fraction of the low-molecular-weight
form is found in the supernatant. It must be noted, however, that in this experiment the
culture supernatant was not cleared by short ultracentrifugation (120,000g/1 h), and
therefore, the form of MUC1 found in the supernatant may have also been associated
with fragments of cell membrane, rather than being a bona fide secreated form.
We have determined that MUC1 is also inserted into the insect cell membrane
(Fig. 5). Surface expression was tested by flow cytometry analysis of immunostained
cells. Panels A, B, and C represent cells stained respectively with IgG-control, BC-3,
or VU-4H5 primary antibody followed with fluorescein isothiocyanate (FITC)-labeled
secondary antibody. Antibody BC-2, like antibody BC-3, recognizes MUC1 indepen-
dent of its state of glycosylation, while antibody VU-4H5 recognizes underglycosylated,
tumor-specific MUC1 similarly to SM-3. It has been recently reported by Wright (12)
and us (submitted for publication) that neither of these forms is glycosylated. It is not
clear why insect cells do not O-glycosylate this recombinant product. GalNAc trans-
ferases purified from insect cells are able to glycosylate in vitro a synthetic peptide
corresponding to one tandem repeat but not the recombinant product.
3.9. Large-Scale Purification of Recombinant MUC1
The most commonly used method for purification of a protein from complex mix-
tures is affinity chromatography using specific antibodies as ligands. This method is
not easy to use for MUC1 purification because most MAbs are against epitopes within
the TR region. High density of repeating epitopes leads to high-avidity binding to the
affinity columns, which makes it almost impossible to elute bound MUC1 from IgG-
Sepharose without boiling in sodium dodecyl sulfate (SDS). Insect cells appear to
retain most of the MUC1 on the cell surface, and, thus, the cells rather than the super-
natant are the primary source for MUC1 purification.
Purification of recombinant MUC1 containing highly hydrophobic transmembrane
domain is a challenge. Fractionation of whole insect cell lysate by ultracentrifugation
in cesium chloride gradient does not separate recombinant mucin effectively. Recom-
binant MUC1 is found in the top layer containing lipids. The form of MUC1 that is
shed from the cell surface and does not have transmembrane and cytoplasmic domains
migrates to the middle of the CsCl gradient (approx 1.4 g/mL). MUC1 binds
nonspecifically to various resins such as glass (high-performance liquid chromatogra-
482 Ciborowski and Finn
phy precolumn from Toso-Haas, Stuttgart, Germany), silica (HPLC size-exclusion
column from Shodex, Showa Denko K.K., Tokyo, Japan), and Sephadex (CM-
Sephadex from Pharmacia, Uppsala, Sweden), significantly decreasing yield of sepa-
Fig. 4. Time course of MUC1 expression. Cell lysates (C) and supernatants (S) were ana-
lyzed by SDS-PAGE in 7% gel under reducing conditions. After transblotting, one membrane
was probed with BC-3 MAb (A), and the other was probed with SM-3 MAb (B). The arrows on
the left indicate the migration distances of protein molecular weight standards. (Reprinted with
permission from ref. 8).
MUC1 in Insect Cells 483
Fig. 5. Fluorescence-activated cell sorter analysis of surface expression of MUC1 on infected
Sf-9 cells. Pairs of infected (bold line) and uninfected (faint line) cells were reacted with irrel-
evant γ1 control antibody (A), and antimucin antibodies BC-2 (B) and 4H5 (C), followed by
immunostaining with FITC labeled secondary antibody (Reprinted with permission from ref. 8).
484 Ciborowski and Finn
ration and shortening the lifetime of these resins. Recombinant MUC1 binds, although
with low yield, to CM-Sephadex C-50 at pH 6.5 and can be eluted with a linear gradi-
ent of 0–1.0 M NaCl.
Recently, we developed a double-step affinity chromatography procedure to purify
recombinant MUC1.
1. Anti-MUC1 antibody (3C6 MAb) on ProteinA-Sepharose (Pharmacia) column and
washed with 10 bed volumes of PBS.
2. Insect cells expressing MUC1 are lysed in 50 mM Tris-HCl buffer, pH 6.5 containing 1%
NP-40.
3. The lysate is precleared by two consecutive centrifugations of 15 min at 3000 rpm (Sorvall,
Newtown, CT) and 1 h ultracentrifugation at 100,000g (Beckman, Palo Alto, CA).
4. Clear cell lysate is diluted with PBS at 1:1 ratio and loaded onto a Protein A–Sepharose-
3C6 MAb affinity column. MUC1 binds to 3C6 antibody, creating MUC1-3C6 complexes
that are eluted from the column with 50 mM glycine buffer, pH 2.8.
5. Mucin is released from the antibody by dissociation of the antibody chains with 1 mM
dithiothreitol, followed by alkylation with 1 mM iodoacetamide.
6. After dialysis, heavy and light chains of 3C6 MAb are removed by affinity chromatogra-
phy using antimouse IgG antibody covalently bound to CNBr-Sepharose (Sigma).
This method requires a ready supply of monoclonal anti-MUC1 antibody. Further sep-
aration of different MUC1 forms can be accomplished using a molecular sieving step.
4. Notes
1. pBlueBacIII is one of the early baculoviral vectors developed. It is large (10.2 kbp) but it
can still handle very well MUC1 cDNA inserts as large as 3.1 kbp. We successfully
inserted the HindIII cassettes into the polylinker, but were not successful in ligating the
BamHI cassettes cut out of the pcDNA3 vector. Recently, pBlueBacIII transfer vector has
been replaced with pBlueBac4.5 vector and linearized Bac-N-Blue™ AcMNPV DNA,
which are now available from Invitrogen as an easy-to-use transfection kit. pBlueBac4.5
vector is 4.9 kbp and offers more sites in the polylinker, including HindIII and BamHI.
We expect that it will also handle MUC1 cDNA properly.
2. By “standard methodology” we imply methodology described in several commonly used
laboratory manuals such as Current Protocols in Molecular Biology (8) or similar.
3. Neither pIE1-4 nor pIE1-3 have HindIII restriction site in the polylinker. It makes it more
complicated to subclone a HindIII cassette, such as 22TR- or 2TRMUC1 cDNAs.
4. We usually grow Sf-9 cells in 5% FBS and we have not noticed any difference in the
expression of MUC1 when compared with cells cultured in 10% FBS. In some instances,
e.g., in cultures started from thawed cells, we use 10% FBS to support better their recov-
ery from the shock of freezing and thawing.
5. We recommend the use of CsCl purified plasmid DNA in this procedure. Moreover, vari-
ous ratios between viral and vector DNAs help to obtain a good yield of recombinants in
at least one of five transfections. We also recommend the use of more liposomes than the
manufacturer suggests in order to increase further the yield of transfection.
6. Although we successfully expressed MUC1 mucin in the pIE-4/pIE-neo system, it was
difficult to maintain long-term expression. The best level of expression was approx 30%
of MUC1-positive cells in the Neo-resistant cell population. Expansion of such culture
resulted in a progressive decrease in the percentage of MUC1/neo-positive cells. The
Không có nhận xét nào:
Đăng nhận xét