ARTICLE
INTRODUCTION
Intravesical BCG therapy has been successfully used
for the treatment of superficial bladder cancer [1, 2]. For BCG therapy
to be effective, the development of an immune response seems to be important,
although the exact mechanism remains to be established [3]. For the initiation
of the immune reaction, several steps during the first interaction of
BCG with the bladder wall have been suggested to be involved. These initial
reactions may include BCG adherence and internalization, cytokine production
and BCG antigen presentation by bladder (carcinoma) cells.
The immediate fate of BCG after intravesical instillation
is not known. Several lines of evidence suggest internalization of BCG
by urothelial cells in the initiation of the BCG-induced immune reaction
and/or anti-tumor effect. Attachment and internalization of BCG by bladder
tumor cells have been reported [4, 5]. Ingestion of mycobacteria both
in vitro and in vivo by other non-professional phagocytic
cells has also been reported [6]. The nature of the attachment as well
as the number of bacteria attached to the bladder wall and/or urothelial
cells are matters of controversy [7-11].
As a reaction to BCG therapy, urothelial cells are able
to express several molecules such as MHC class II and ICAM-1 antigens
on their cell wall [12, 13]. These molecules are fundamental to many immune
functions including antigen presentation and cell-mediated cytotoxicity.
BCG-antigen presentation to murine T-lymphocytes by urothelial cells has
been reported [13]. In addition, intravesical BCG instillation is associated
with an increased level of urinary cytokines [14-16]. The source of some
of these cytokines has not been conclusively determined, but epithelial
cells, including urothelial cells, do express the capacity to synthesize
cytokines [17, 18]. Recently, accumulating data have shown that BCG is
capable of upregulating cytokine synthesis such as IL-6 by tumor urothelial
cells grown in vitro [19, 20]. In patients, IL-6 levels in urine
following BCG instillation are reported to be related to the clinical
BCG anti-tumor response [20].
In this study the hypothesis that BCG internalization
depends on the degree of differentiation of TCC cells and is a prerequisite
for upregulation of IL-6 production by bladder tumor cells was tested.
METHODS
Cells and culture conditions
Stock cultures of a series of cell lines derived from
human bladder carcinomas consisting of the well differentiated cell lines,
RT-4 (grade 1), SBC-2 (grade 1), SBC-7 (grade 1) and the poorly differentiated
cell lines, T24 (grade 3), TCC-SUP (grade 3) and J82 (grade 3) [21-23],
were cultured as monolayers in standard Dulbecco's modified Eagle's medium
(DMEM) supplemented with 10% fetal calf serum, 1% L-glutamine, 100 U/ml
penicillin and 100 µg/ml streptomycin. Cells were grown at 37°
C in a humidified 5% CO2 atmosphere at pH 7.2 until subconfluency.
Cultures were routinely screened for mycoplasma.
For passage, cells were trypsinized with 0.05% trypsin
and 0.02% EDTA in phosphate buffered saline (PBS). All tissue culture
chemicals were obtained from Flow Laboratories (Irvine, Scotland). Tissue
culture plastics were from Costar (Cambridge, Mass., USA).
BCG preparation
Freeze-dried BCG, Connaught strain (3.2 x 108
CFU per vial), was reconstituted in 10 ml prewarmed DMEM. To eliminate
large clumps of bacteria, the suspension was centrifuged at 300 g for
3 min and the top 5 ml was used [5]. The concentration of CFU was set
by measuring the absorbance at a wavelength of 440 nm, compared to a standard
curve relating the CFU count to optical density.
Fluorescence activated cell sorter (FACS) analysis
of BCG internalization by bladder cancer cell lines
Attachment and internalization of BCG were quantitated
as described previously [11]. Briefly, cells (3 x 105/24-well)
were cultured overnight, and BCG (Connaught) labelled with fluorescein
isothiocyanate (FITC) was added. Cells were incubated with BCG-FITC for
4 hours, followed by washing, and harvesting with cell dissociation solution
(CDS) obtained from Sigma (St Louis, MO, USA). This procedure resulted
in cells containing both attached and internalized BCG-FITC. To exclusively
stain the fraction of extracellular or attached BCG, cells were subsequently
pre-incubated with polyclonal rabbit anti-BCG antibodies (alphaBCG) (1:100;
Dakopatts, Denmark) followed by incubation with phycoerythrin (PE) conjugated
goat anti-rabbit antibodies (1:25; Sigma, St Louis, MO, USA).
Cells (104) were measured with a fluorescence-activated
cell sorter (FACScan, Becton Dickinson Immunocytometry Systems, Mountain
View, CA, USA), to determine the fraction of cells "stained" with FITC
(Fl1+; intra- and extracellular BCG) and PE (Fl2+;
extracellular BCG). The percentage of Fl1+/Fl2
cells was taken as the percentage BCG-ingesting cells.
Inhibition studies of BCG
phagocytosis
Inhibition of internalization of BCG was studied using
alphaBCG (1:100-1:25) and rabbit polyclonal anti-FN antibodies (alphaFN)
(1:100-1:25; Dako Corp., Santa Barbara, CA), added 30 min prior to the
FACS internalization assay of BCG. To eliminate the effect of possible
aspecific binding of the Fc portion of alphaFN to BCG, inhibition experiments
were also performed with goat anti-human FN F(ab)2 fragments
(1:100-1:25; Organon Teknika, West Chester, PA). For inhibition studies
with Cytochalasin B (CytB;Sigma, St Louis, MO, USA), CytB at the indicated
concentrations was added to the cells 1 hour before and during the incubation
with BCG.
Determination and inhibition of constitutive and
BCG-induced upregulation of IL-6 production
Cells (3 x 105 cells) were continuously incubated
with BCG at the indicated concentrations. At the indicated time intervals,
medium was collected for cytokine determinations. IL-6 was quantified
with a human specific, oligoclonal ELISA obtained from Medgenix (Fleurus,
Belgium) [24]. Conditions for inhibition of IL-6 production by alphaBCG,
alphaFN and CytB were similar as those described for the inhibitory studies
of BCG internalization.
Statistical analysis
All experiments were performed at least twice. FACS analysis
measurements represent the mean of duplicate cultures, SD was less than
10%. All experiments concerning IL-6 measurements were performed in triplicate.
For comparison between groups the Mann-Whitney test or Student's t test
for independent samples were used (SPSS 6.1 for Microsoft Windows software).
RESULTS
Internalization of BCG
The kinetics of BCG internalization in T24 cells have
been reported in a previous study [11]. The percentage of cells internalizing
BCG appeared to be proportionally related to the time of incubation. As
presented in Figure 1,
the capacity for BCG internalization in a series of human bladder cancer
cell lines appeared to be cell-line dependent. The maximum number of cells
internalizing BCG during a period of 24 hours varied from 0-2% for the
well differentiated cell lines (RT-4, SBC-2 and SBC-7), to 34-68% for
the poorly differentiated cell lines (T-24, TCC-SUP and J-82). These data
suggest a relationship between the capacity to internalize BCG and the
differentiation grade of TCC cells. Only poorly differentiated tumor cell
lines internalized BCG.
BCG-induced and constitutive IL-6 synthesis
To study BCG internalization as a necessary initial step
of BCG-induced upregulation of IL-6, initially both the constitutive and
BCG-induced upregulation of IL-6 were determined in a series of well and
poorly differentiated cell lines. As illustrated in Figure
2 constitutive and BCG-mediated upregulation of IL-6 appeared
to be cell line-dependent. Determined after a 7 hours incubation period,
the well differentiated, non-phagocytic cell lines (SBC-2 and SBC-7) exhibited
low (0.4 to 0.5 ng/3 x 105 cells) constitutive IL-6 synthesis,
which appeared not to be upregulated in the presence of BCG (0.5 ng/3
x 105 cells). Of the poorly differentiated, BCG internalizing
cell lines, the constitutive IL-6 synthesis was found to be high for TCC-SUP
(1.1 ± 0.1 ng/3 x 105 cells) and J82 (1.7 ± 0.4 ng/3
x 105 cells) cells, but low (0.1 ± 0.0 ng/3 x 105
cells) for T24 cells. Stimulation with BCG did not significantly (p =
0.095) upregulate IL-6 synthesis in TCC-SUP cells. In contrast, both the
J82 and T24 exhibited a significant (p < 0.005) upregulation of IL-6
production when stimulated with BCG. This BCG-induced upregulation was
most pronounced in T24 cells, resulting in an IL-6 production of 2.6 ±
0.1 ng/ml compared to the constitutive synthesis of 0.1 ± 0.0 ng
per 3 x 105 cells over a period of 7 hours. This upregulation
of IL-6 production was BCG concentration-and time-dependent (Figure
3). So, the poorly differentiated TCC cell lines showed, in addition
to the capacity to internalize BCG, either a high constitutive (TCC-SUP
and J82) and/or a high BCG-inducible production of IL-6 (T24 and J82).
Relation between BCG phagocytosis and IL-6 production
The results outlined above suggest the possibility of
a causal relationship between BCG internalization and IL-6 upregulation,
which is especially illustrated in T24 cells. In a subsequent series of
experiments, the possibility of BCG internalization as a prerequisite
to BCG-induced IL-6 production was studied in detail. These experiments
were performed with T24 cells, since the characteristics of this cell
line, namely the low constitutive IL-6 synthesis associated with a high
degree of BCG-induced upregulation of IL-6 synthesis, seemed to be well
suited for this purpose.
Incubation with BCG, pretreated with alphaBCG, decreased
the percentage of T24 cells associated with BCG to 34% (1:100 alphaBCG)
and 12% (1: 25 alphaBCG) compared to 38% in the absence of alphaBCG (data
not shown). As illustrated in Figure
4, under similar conditions a significant (p = 0.028) inhibition
of BCG-induced upregulation of IL-6 synthesis occurred in the presence
of alphaBCG at a concentration of 1:25.
A condition previously found to reduce the internalization
of BCG by T24 cells specifically [11], namely incubation with CytB, was
subsequently tested for its ability to prevent IL-6 upregulation. Under
conditions where CytB reduced BCG internalization by T24 cells to 46%,
Cyt B inhibited BCG-induced IL-6 production, which was significantly stronger
than the inhibition of the constitutive IL-6 synthesis (Table
1).
Kuroda et al. [4] suggest that BCG internalization
is associated with fibronectin and can be inhibited by alphaFN. Using
identical antibodies the effect of alphaFN on BCG-induced upregulation
of IL-6 synthesis was studied. The results, presented in Figure
4, show that addition of alphaFN to T24 cells and subsequent incubation
with BCG did not inhibit BCG-induced IL-6 upregulation (2.3 ± 0.1
versus 2.2 ± 0.2 ng/ml per 3 x 105 cells in the
absence of alphaFN; p = 0.272). Since this observation contradicted our
expectations, the effect of alphaFN on BCG internalization was determined.
The results indicated that alphaFN, added at concentrations ranging from
1:100 to 1:25, did not inhibit BCG internalization (data not shown). So,
these results seemed to be in agreement with the absence of an effect
of alphaFN on the induction of IL-6 synthesis by BCG.
Overall, the data suggest that in T24 cells, internalization
of BCG is a prerequisite for BCG-induced IL-6 production.
DISCUSSION
Various cytokines can be detected in the urine of patients
treated with BCG intravesically [14, 15, 25]. Internalization of BCG by
urothelial cells in vivo has been reported [5]. This implies that
the present study concerning BCG internalization and cytokine production
of urothelial cells in vitro may be of clinical relevance.
In this study, the phagocytic capacity and high levels
of IL-6 production, either constitutive or BCG-induced or both, seemed
to be restricted to cell lines with a poor degree of differentiation (T24,
TCC-SUP and J82, grade 3). Tumor grade dependency has already been reported
with regard to adherence of BCG to bladder tumor cell lines [26], and
(BCG-induced) IL-6 production by bladder tumor cell lines [20]. In addition,
the present results suggested a relationship between BCG-internalization
and IL-6 production. Experiments with Cytochalasin B, a molecule which
specifically blocks phagocytosis [11], showed internalization of BCG to
be a prerequisite for the BCG-induced IL-6 production in T24 tumor cells.
Accepting the hypothesis that BCG internalization and (BCG-induced) cytokine
production of (tumor) urothelial cells may play a central role in the
mechanism of action of BCG, these findings are in accordance with clinical
observations that high grade tumors show a better response to BCG treatment
than lower grade tumors [27].
The observed capacity of high grade TCC tumor cells
to produce IL-6, either constitutively and/or during stimulation with
BCG, may reflect the clinical situation. Following instillations with
BCG, various cytokines can be detected in the urine of patients [14, 15,
25]. These observations suggest that, during intravesical BCG instillation,
urothelial (tumor) cells may be a significant source of urinary cytokines,
additional to the major cellular sources, such as macrophages and lymphocytes
infiltrating the bladder wall. It has been speculated that a BCG-induced
upregulation of cytokines in urothelial (tumor) cells of the bladder could
be of functional significance in vivo, initiating and/or modulating
the BCG-induced immune response [24, 25, 28]. It should be noted that
upregulation of the cytokine synthesis of urothelial tumor cells by microorganisms
seems to be a characteristic most extensively expressed in the presence
of BCG. It has been shown that the upregulation of IL-6 and TNF-alpha
synthesis in T24 cells by BCG is at least 10 times higher compared to
the upregulation by E. Coli or S. faecalis [19].
As indicated for T24 cells, upregulation of IL-6 production
appeared to be related to the dose of BCG. An optimal dose in vitro
was not established with the concentration range used. In patients however,
the dose may be limited by the adverse side effects of BCG [29].
In order to inhibit BCG internalization by T24 cells,
we used, in addition to Cytochalasin B, polyclonal alphaBCG and alphaFN
as described by Kuroda et al [4]. In accordance with these investigators,
we observed a decreased internalization due to alphaBCG, suggesting a
specific interaction between BCG and the T24 cellular membrane. This condition
also reduced the BCG-induced IL-6 upregulation by T24, which further consolidated
the findings with Cytochalasin B. In contrast to Kuroda et al.
we could not detect an inhibiting effect of alphaFN on BCG phagocytosis.
This discrepancy cannot be explained by the use of different sources of
antibodies. Furthermore, differences in BCG strains do not seem likely
since thusfar all strains have shown to be comparable with regard to clinical
efficacy against superficial bladder tumors. Absence of an inhibiting
effect of alphaFN for BCG phagocytosis by bladder carcinoma cell lines
has been reported by other investigators too [26, 30]. These observations
do not favour the previously suggested (opsonizing) role of FN in BCG
internalization by bladder tumor cells [4]. However, there may be differences
between the proposed role of FN in mediating BCG binding to the injured
bladder wall, initiation of an immune response and the associated antitumor
activity in a mouse bladder tumor model [10] and the possible role of
FN in the internalization of BCG in vitro systems.
In this study, alphaBCG inhibited in vitro internalization
of BCG in T24 cells, and IL-6 production. Anti-BCG antibodies (IgG and
IgA) have been shown to be present in serum and urine of patients following
BCG instillations [9] up to 12 months after BCG therapy, possibly interfering
with the effectiveness of subsequent BCG instillations. Whether the level
of anti-BCG antibodies in patients correlates with the anti-tumor effect
or not, and what the implications are for the efficacy of (maintenance)
instillations with BCG, remains to be established in further studies.
CONCLUSION
A high constitutive IL-6 production or IL-6 upregulation,
initiated by BCG internalization seems to be related to the degree of
differentiation of the bladder cancer cell lines. These observations suggest
that constitutive IL-6 production and/or BCG-mediated (after internalization)
upregulation of IL-6 production by poorly differentiated bladder carcinoma
cells may be part of the mode of action of intravesical BCG therapy. The
role of this phenomenon in vivo, however, remains to be established.
Acknowledgments
This study was supported by a grant from the Urological
Research Foundation Amsterdam (BUWO), The Netherlands and by Connaught
Laboratories Ltd., North York, Ontario, Canada.
REFERENCES
1. Herr H W, Pinsky C M, Whitmore W F Jr, Sogani P C, Oettgen H F, Melamed
M R. 1986. Long-term effect of intravesical bacillus Calmette-Guerin on
flat carcinoma in situ of the bladder. J. Urol. 135: 265.
2. Lamm D L, Blumenstein B A, Crawford E D, Montie J E, Scardoni P,
Grossman H B, Stanisic T H, Smith Jr J A, Sullivan J, Sarosdy M F, Crissman
J D, Coltman C A. 1991. A randomized trial of intravesical doxorubicin
and immunotherapy with bacille Calmette-Guerin for transitional-cell carcinoma
of the bladder. N. Engl. J. Med. 325: 1205.
3. Ratliff T L. 1989. Mechanism of action of intravesical BCG for bladder
cancer. (Review) Prog. Clin. Biol. Res. 310: 107.
4. Kuroda K, Brown E J, Telle W B, Russell D G, Ratliff T L. 1993. Characterization
of the internalization of bacillus Calmette-Guerin by human bladder tumor
cells. J. Clin. Invest. 91: 69.
5. Becich M J, Carroll S, Ratliff T L. 1991. Internalization of bacille
Calmette-Guérin by bladder tumor cells. J. Urol. 145: 1316.
6. Band A H, Chitamber S D, Bhattacharya A, Talwar G P. 1986. Mechanism
of phagocytosis of mycobacteria by Schwann cells and their comparison
with macrophages. Int. J. Lepr. other Mycobact. Dis. 54: 294.
7. Kavoussi L R, Brown E J, Ritchey J K, Ratliff T L. 1990. Fibronectin-mediated
Calmette-Guerin bacillus attachment to murine bladder mucosa. Requirement
for the expression of an anti-tumor response. J. Clin. Invest.
85: 62.
8. Teppema J S, de Boer E C, Steerenberg P A, van der Meijden A P. 1992.
Morphological aspects of the interaction of bacillus Calmette-Guerin (BCG)
with urothelial cells in vivo and in vitro: relevance for
antitumor activity? Urol. Res. 20: 219.
9. Van der Sloot E, Kuster S, Böhle A, Braun J, Wood W G. 1992.
Towards an understanding of the mode of action of bacillus Calmette-Guérin
therapy in bladder cancer treatment, especially with regard to the role
of fibronectin. Eur. J. Clin. Chem. Clin. Biochem. 30: 503.
10. Ratliff T L, Palmer J O, McGarr J A, Brown E J. 1987. Intravesical
BCG therapy for murine bladder tumors: Initiation of the response by fibronectin-mediated
attachment of BCG. Cancer Res. 47: 1762.
11. De Boer E C, Bevers R F, Kurth K H, Schamhart D H. 1996. Double
fluorescent flow cytometric quantification of bacterial internalization
and binding by epithelial cells. Cytometry 25: 381.
12. Jackson A M, Alexandroff A B, McIntyre M, Esuvara-
nathan K, James K, Chisholm G D. 1994. Induction of ICAM-1 expression
on bladder tumours by BCG immunotherapy. J. Clin. Pathol. 47: 309.
13. Lattime E C, Gomella L G, McCue P A. 1992. Murine bladder carcinoma
cells present antigen to BCG-specific CD4+ T cells. Cancer
Res. 52: 4286.
14. Böhle A, Nowc C H, Ulmer A J, Musehold J, Hofstetter A G, Flad
H D. 1992. Elevations of cytokines interleukin-1, interleukin-2 and tumor
necrosis factor in the urine of patients after intravesical bacillus Calmette-Guerin
immunotherapy. J. Urol. 144: 59.
15. De Boer E C, de Jong W H, Steerenberg P A, Aarden L A, Tetteroo
E, de Groot E R, van der Meijden P, Vegt P D, Debruyne F M, Ruitenberg
E J. 1992. Induction of urinary IL-1, IL-2, IL-6, and TNF during intravesical
immunotherapy with BCG in superficial bladder cancer. Cancer Immunol.
Immunother. 34: 306.
16. De Reijke T M, de Boer E C, Kurth K H, Schamhart D H. 1996. Urinary
cytokines during intravesical BCG therapy for superficial bladder cancer:
processing, stability and prognostic value. J. Urol. 155: 477.
17. Thompson A B, Robbins R A, Romberger D J, Sisson J H, Spurzem J
R, Teschler H, Rennard S I. 1995. Immunological functions of the pulmonary
epithelium. (Review) Eur. Resp. J. 8: 127.
18. Hedges S, Agace W, Svensson M, Sjögren A C, Ceska M, Svanborg
C. 1994. Uroepithelial cells are part of a mucosal cytokine network. Infect.
Immun. 62: 2315.
19. De Reijke T M, Vos P C, de Boer E C, Bevers R F, de Muinck Keizer
W H, Kurth K H, Schamhart D H. 1993. Cytokine production by the human
bladder carcinoma cell line T24 in the presence of bacillus Calmette-Guerin
(BCG). Urol. Res. 21: 349.
20. Esuvaranathan K, Alexandroff A B, McIntyre M, Jackson A M, Prescott
S, Chisholm G D, James K. 1995. Interleukin-6 production by bladder tumors
is upregulated by BCG immunotherapy. J. Urol. 154: 572.
21. Hansson Y, Vargas-Cortes M, Paulie S, Perlmann P. 1988. MHC nonrestricted
cytotoxic T cell clones with selective specificity from patients with
transitional cell carcinoma of the urinary bladder. Cancer Immunol.
Immunother. 27: 205.
22. Wang M H, Flad H D, Böhle A, Chen Y Q, Ulmer A J. 1991. Cellular
cytotoxicity of human natural killer cells and lymphokine-activated killer
cells against bladder carcinoma cell lines. Immunol. Lett. 27:
191.
23. Johansson S L, Unsgaard B, O'Toole C M. 1994. Cell lines from urinary
bladder tumors. In: Hay R J, Park J G, Gazdar A, Eds. Atlas of human
tumor cell lines, New York: Academic Press Inc., 341.
24. Schamhart D H, Kurth K H, de Reijke T M, Vleeming R. 1992. BCG treatment
and the importance of an inflammatory response. Urol. Res. 20:
199.
25. De Boer E C, Somogyi L, de Ruiter G J, de Reijke T M, Kurth K H,
Schamhart D H. 1997. Role of IL-8 in onset of the immune response in intravesical
BCG therapy for superficial bladder cancer. Urol. Res. 25: 31.
26. Schneider B, Thanhauser A, Jocham D, Loppnow H, Vollmer E, Galle
J, Flad H D, Ulmer A J, Böhle A. 1994. Specific binding of bacillus
Calmette-Guerin to urothelial tumor cells in vitro. World J.
Urol. 12: 337.
27. Melekos M D, Chionis H, Pantazakos A, Fokaefs E, Paranychianakis
G, Dauaher H. 1993. Intravesical BCG immunoprophylaxis of superficial
bladder cancer: results of a controlled prospective trial with modified
treatment schedule. J. Urol. 149: 744.
28. Jackson A M, James K. 1994. Understanding the most successful immunotherapy
for cancer. The immunologist 2/6: 208.
29. Kreider J W, Bartlett G L, Purnell D M. 1980. Immunotherapeutic
effectiveness of BCG inactivated by various modalities. Cancer
46: 480.
30. Vögeli T A, Geisler D, Schroten H, Ackermann R. 1993. Evidence
for fibronectin-independent binding of BCG Pasteur in cell culture systems
of urothelial cancer. J. Urol. 149: 337a (abstract).
|
Printable version
