Skip Navigation

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (6)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Haqqani, A. S.
Right arrow Articles by Birnboim, H.C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Haqqani, A. S.
Right arrow Articles by Birnboim, H.C.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Carcinogenesis, Vol. 22, No. 2, 243-250, February 2001
© 2001 Oxford University Press

CANCER BIOLOGY

Constitutive expression of interleukin-8 by Mutatect cells markedly affects their tumor biology

Arsalan S. Haqqani, Jagdeep K. Sandhu1, and H.Chaim Birnboim2,

Department of Biochemistry, Microbiology and Immunology, University of Ottawa and the Ottawa Regional Cancer Centre, Ottawa, Ontario K1H 1C4, Canada


   Abstract
Top
 Abstract
Introduction
 Materials and methods
 Results
 Discussion
 References
 
Interleukin-8 (IL-8) is a chemokine for neutrophils and an angiogenic factor. Human tumors that express IL-8 may exhibit intense neutrophil infiltration and increased vascularization. Mutatect cells are a murine fibrosarcoma that can be grown as subcutaneous tumors in syngeneic C57BL/6 mice. Since neutrophils are a source of cytotoxic and genotoxic species, we constructed Mutatect cell lines that constitutively express human IL-8 to explore the involvement of neutrophils in tumor biology and genetic instability. An IL-8/neo expression plasmid was stably transfected into Mutatect MC17-51 cells and clone MIL-4 was isolated. Tumors initiated with 5x105 MIL-4 cells grew very slowly compared with tumors from pure MC17-51 cells or from 0.5 to 4x105 MIL-4 cells mixed with 5x105 MC17-51 cells. Over 95% of cells recovered from slow-growing pure MIL-4 tumors lost the transgene as measured by loss of (i) resistance to G418, (ii) expression of IL-8 protein and (iii) IL-8-specific DNA sequences. When tumors from mixed cell types were examined, loss of the transgene did not occur; rather, IL-8 producing cells appeared to have some growth advantage. The neutrophil content of tumors (as measured by myeloperoxidase) was directly proportional to the level of IL-8 expressed at the time tumors were excised. As reported earlier, the frequency of mutations at the hypoxanthine phosphoribosyltransferase locus was also directly proportional to neutrophil content. To explain some of these biological findings, we postulate that early in development of pure MIL-4 tumors, genotoxic/cytotoxic neutrophils are attracted by IL-8, which in turn leads to loss of the transgene and to localized cytotoxicity of IL-8 producing cells. In mixed tumors, where the initial IL-8 concentration may be lower, tumors might become established more readily because fewer neutrophils may be attracted. This relatively simple experimental paradigm has revealed some of the complex biological changes that can occur as a result of IL-8 in tumors.

Abbreviations: Br-CL, bromide-dependent chemiluminescence; BSA, bovine serum albumen; CMV, cytomegalovirus; DMEM, Dulbecco's modified Eagle's medium; FCS, fetal calf serum; Hprt, hypoxanthine phosphoribosyltransferase; IL-8, interleukin-8; MPO, myeloperoxidase; neo, neomycin-resistance gene; PBS, phosphate buffered saline; PCR, polymerase chain reaction; rhIL-8, recombinant human IL-8; 6-TG, 6-thioguanine.


   Introduction
Top
 Abstract
 Introduction
Materials and methods
 Results
 Discussion
 References
 
Interleukin-8 (IL-8) is a C-X-C chemokine for neutrophils and an angiogenic factor (1–7). It is also a chemoattractant for neutrophils from mouse and other species (8,9). The presence of this cytokine has been reported in human melanoma (10,11), non-small-cell lung tumors (12), bronchioloalveolar carcinoma (13), head and neck squamous carcinoma (14), gastric carcinomas (15), colon carcinoma (16), mesothelioma (17) and various glioblastoma (18,19). IL-8 expression has been associated with a high level of neutrophil infiltration and increased vascularization, and with poor prognosis in some cases (13–15). There have been a limited number of studies in which the effect of IL-8 expression in experimental tumor models has been explored. IL-8 expression by CHO cells in a nude mouse tumor model decreased subcutaneous tumorigenicity (20). The opposite effect of IL-8 expression was seen in human gastric carcinoma TMK-1 cells grown at an orthotopic site in nude mice. IL-8 expressing TMK-1 cells produced rapidly growing and highly vascular tumors in the gastric wall (21). Thus, expression of IL-8 is found widely in tumors and may have complex effects.

We have established the Mutatect model as an experimental paradigm to study the biological role of neutrophils and reactive nitrogen and oxygen species on genetic instability and tumorigenicity of a fibrosarcoma in syngeneic C57BL/6 mice (22). Mutatect tumors are infiltrated with host cells, predominantly neutrophils, and direct injection of IL-8 into tumors increased the number of infiltrating neutrophils (23). To explore the effect of IL-8 in Mutatect tumors in more detail, we have constructed Mutatect cell lines that constitutively express IL-8 and describe the growth properties of tumors formed with these cells as well as the stability of the transfected IL-8 gene.


   Materials and methods
Top
 Abstract
 Introduction
 Materials and methods
Results
 Discussion
 References
 
Mutatect cell lines
Derivation of the Mutatect model has been described previously (22). Conditions for cell culture, detection of hypoxanthine phosphoribosyltransferase (Hprt–) mutants and removal of pre-existing mutant cells, prior to formation of subcutaneous tumors, have been reported elsewhere (23,24). The parental Mutatect cell line used for transfection experiments was MC-TGS17-51 (24) (referred to as MC17-51 in this report). Unless otherwise stated, all cell lines were maintained in Dulbecco's modified Eagle's medium (DMEM) plus 10% fetal calf serum (FCS) (Gibco BRL, Burlington, Canada) in a humidified atmosphere of 5% CO2/95% O2 at 37°C.

Mutatect tumor formation and size measurements
Subcutaneous tumors were formed by injecting 5x105 MC17-51, MIL-1 or MIL-4 cells in 0.1 ml of phosphate buffered saline (PBS) into 8–10 week-old C57BL/6 female mice. Mixed tumors were formed by inoculating mixtures of MC17-51 and MIL-4 cells as indicated in the table and figure legends. For in vitro controls, the same cell mixtures were grown in culture for the entire period of in vivo tumor growth, with subculturing twice weekly. Once the tumors reached a measurable size (~1 week after injection), dimensions were estimated every 2–3 days. Tumor volumes were calculated from the equation of a hemi-ellipsoid:


where V is the volume of the tumor in mm3. When the largest diameter of control MC17-51 tumors reached 1 cm (i.e. ~21 days), all tumors were harvested. Each tumor (in 5 ml DMEM plus 10% FCS) was lightly minced and mechanically dispersed by passage through a syringe, then allowed to settle for 5 min. The supernatant (cellular fraction) was used for measurement of myeloperoxidase (MPO) and for ex vivo growth of tumor cells for 2–3 days. These growing cells were analyzed for G418 resistance, IL-8 protein expression, IL-8 gene loss and the frequency of Hprt– mutations.

Construction of an IL-8/neo plasmid
I.M.A.G.E. Consortium LLNL cDNA clone #328322 (Research Genetics Inc., Huntsville, AL) contains human IL-8 sequences (25). This plasmid was first digested to completion with NheI, then partially digested with EcoRI and finally blunt-ended with Mung bean nuclease to obtain an 833 bp fragment containing the entire coding region of IL-8. Mammalian expression vector pcDNA3 (Invitrogen, Carlsbad, CA) was digested with BamHI and blunt-ended with T4 DNA polymerase. The fragment and the vector were ligated to generate a construct containing the IL-8 coding region under a cytomegalovirus (CMV) promoter. The vector also contained an ampicillin gene for bacterial selection and a neomycin (neo) gene for mammalian selection. The construct was transformed into competent SURE Escherichia coli (Stratagene Cloning Systems, La Jolla, CA) and ampicillin-resistant clones were selected. pcDNA3–IL8 (clone L25-13) was isolated. It was confirmed that it contained the IL-8 coding region in a sense orientation by restriction nuclease analysis and the correct nucleotide sequence by sequencing from the T7 promoter of pcDNA3. pcDNA3–IL8 is also referred to as the IL-8/neo trangene in this report.

Transfection of the IL-8/neo construct into Mutatect cells
MC17-51 cells were transfected with either pcDNA3-alone or pcDNA3–IL8 using liposomes. In transient transfection experiments, the culture medium was analyzed after 48 h for IL-8. To produce stable transfectants, the pcDNA3–IL8 construct was linearized with ScaI before transfection into MC17-51 cells. After 48 h in culture, 500 µg/ml G418 was added; G418-resistant (MIL) clones were selected and the culture media screened by ELISA for IL-8 production.

Competitive ELISA for detection of IL-8 in culture medium
A competitive ELISA was developed to permit sensitive detection of IL-8 in the medium of Mutatect cells. Neutralite avidin (Molecular Probes, Leiden, The Netherlands) was bound (50 µl/well of 20 µg/ml in PBS at room temperature for 18 h) to a 96-well polystyrene microtitre plate, then blocked with 1% bovine serum albumen (BSA) for 2 h. Recombinant human IL-8 (rhIL-8, from R&D Systems, Minneapolis, MN, supplied in solution as 300 µg/ml in 30 mg/ml BSA) was biotinylated as follows. A mixture was prepared containing 0.1 M sodium borate (pH 8.8), 50 µM sulfosuccinimidyl biotin, 0.6 µg rhIL-8 and 60 µg BSA in 200 µl and allowed to react overnight at room temperature. Unreacted functional groups of sulfosuccinimidyl biotin were blocked with 5 mM ammonium chloride at room temperature for 30 min. Free biotin was removed by extensive dialysis against PBS. As a control, biotinylated BSA was prepared in a similar manner using a mixture lacking rhIL-8. Approximately 1 ng of dialyzed biotinylated rhIL-8 in 100 µl of ELISA buffer (0.02 M Tris–HCl, pH 7.5, 0.15 M NaCl, 0.05% v/v Tween-20) was added per well of the avidin-coated, BSA-blocked 96-well plate and incubated at 4°C overnight. The wells were washed three times with ELISA buffer and stored at 4°C for up to 1 month. Using anti-IL-8 antibody as described below, bound biotinylated-rhIL-8 could be detected while avidin-coated, BSA-blocked wells containing bound biotinylated-BSA, free rhIL-8 or no additive produced no signal.

Analysis of IL-8 in culture medium was carried out as follows. Cells were grown in DMEM plus 10% FCS at a density of 5x104/cm2. Eighteen hours before analysis, the medium was removed and replaced with DMEM plus 0.1% FCS. (Reducing the amount of FCS was found to lower the background and increase the sensitivity of the ELISA; data not shown.) The medium (0.5 ml) was mixed with rabbit anti-human IL-8 polyclonal antibody (0.5 µl of 1 mg/ml, Endogen, Woburn, MA) and incubated overnight at 4°C. Wells were washed before use with ELISA buffer and 100 µl of culture medium preincubated with anti-IL8 antibody was added and then incubated for 1 h at room temperature. Unbound antibody was removed by three washes with ELISA buffer. Secondary antibody, 1:1000 diluted phosphatase-conjugated goat anti-rabbit IgG (Kirkegaard & Perry, Gaithersburg, MD) (100 µl/well) was added and incubated for 1 h at 37°C. Wells were washed three times with ELISA buffer, 100 µl/well of substrate (3 mM p-nitrophenyl phosphate, 50 mM Na2CO3, 0.05 mM MgCl2) was added and color was allowed to develop overnight at 4°C. Absorbance was measured at 405 nm in a Benchmark Microplate Reader (Bio-Rad, Mississauga, Canada). For correction of non-specific binding, the absorbance of wells without biotinylated-rhIL8 was subtracted from the absorbance of corresponding biotinylated-rhIL8-coated wells.

A standard curve was constructed using 0–2 ng/ml of rhIL-8 added to DMEM plus 0.1% FCS. IL-8 produced by Mutatect cell lines was quantified with reference to this standard curve. The assay was nearly linear in the range 0.02–1.50 ng of IL-8/ml (data not shown). Each assay was performed in triplicate.

Biological activity of IL-8 produced by Mutatect cells
An in vitro chemotaxis assay for neutrophils was used to verify that human IL-8 produced by Mutatect cells had biological activity. Neutrophil-rich granulocytes were purified from the peripheral blood of normal healthy individuals using Ficoll-Hypaque (Pharmacia Biotech, Sweden) density gradient centrifugation followed by hypotonic lysis to remove erythrocytes (26). Culture medium (0.5 ml) from transiently transfected Mutatect cells was added to a 24-well plate. Nunc 10 mm Tissue Culture Inserts (Nalge Nunc International, Naperville, IL) with a 8 µm polycarbonate membrane were placed into each well as the upper compartment of a Boyden chamber. A neutrophil-rich granulocyte suspension (0.4 ml containing 4x105 cells in PBS plus 0.9 mM CaCl2, 0.5 mM MgCl2 and 0.1% BSA) was added into each insert. The plate was incubated at 37°C for 60 min to permit chemotaxis of neutrophils through pores in the membrane. The membrane was removed, fixed in methanol, stained with hematoxylin, transferred to a glass cover slip, air-dried and mounted on a glass slide with a drop of immersion oil such that the original lower surface of the membrane was uppermost. Neutrophils on this surface represent cells that have migrated through the pores of the membrane. The average number of neutrophils per x400 magnification field was calculated from a count of eight fields. Experiments were carried out in duplicate.

Analysis of G418R and Hprt– mutant cells
The percentage G418R and the frequency of Hprt– mutant cells was measured in Mutatect cells growing ex vivo and in vitro. For estimating the percentage of cells expressing a functional neo gene, 200 viable cells were plated in triplicate on 6-cm dishes in the absence or presence of 500 µg/ml G418. After 8 days of incubation, the number of G418-resistant colonies was counted and the percentage of G418R cells as a fraction of the total cells plated was calculated. For estimation of the Hprt– mutation frequency, 1x105 viable cells from each tumor were plated in triplicate in 10-cm dishes in medium containing 50 µM 6-thioguanine (6-TG). After 14 days of incubation, the number of 6-TG-resistant colonies (representing Hprt– mutants) was counted and corrected for plating efficiency. Mutation frequency is the number of Hprt– mutants per 1x105 viable cells (24).

Detection of IL-8 gene loss by PCR
Mutatect cells recovered from tumors (i.e. growing ex vivo) and in vitro controls were analyzed for loss of a fragment of the transfected gene. Genomic DNA was extracted and a 214 bp fragment from the integrated IL-8 transgene was amplified by polymerase chain reaction (PCR) analysis (forward primer, GAGGCCTATATAAGCAGAGC; reverse primer, AGAGCTGCAGAAATCAGGAA). The 214 bp PCR product corresponded to 110 bp of the pcDNA3 CMV promoter and 104 bp of the 5' end of IL-8 cDNA (i.e. the entire 5' untranslated region and the first 56 nt of the coding region). Failure to detect the specific 214 bp PCR, while detecting a non-specific 670 bp fragment product on an ethidium bromide-stained agarose gel, was taken as evidence of IL-8 gene loss.

Tumor myeloperoxidase
One-tenth of each tumor cell suspension was centrifuged and resuspended in 0.2 ml PBS. Cells were lysed by sonication; MPO was solublized by addition of cetyltrimethylammonium bromide to 0.2%, followed by sonication. Protein was quantified using fluorescamine (27), an amine-reactive fluorophore, as described elsewhere (28). MPO activity in the sonicates was measured by a specific assay based on bromide-dependent chemiluminescence (Br-CL) of luminol at pH 5 (29). One unit of MPO activity is defined as the corrected Br-CL 3-min reading (29). Each sample was assayed in triplicate.

Reagents
Sulfosuccinimidyl biotin, p-nitrophenyl phosphate, fluorescamine, hypoxanthine, aminopterin, thymidine, 6-TG, BSA and cetyltrimethylammonium bromide were from Sigma-Aldrich (Oakville, Canada). G418 was from Gibco BRL.

Statistical analysis
For all statistical analyses, non-parametric tests were used. The Mann–Whitney U test was used to compare two unpaired groups. For three or more unpaired groups, a Kruskal–Wallis test was used. Spearman rank correlation was used to determine a correlation between two measured variables. Curve fitting and statistical tests were performed using GraphPad Prism version 3.00 (GraphPad Software, San Diego, CA). A value of P < 0.05 was considered to be statistically significant. In the figures, statistical significance is indicated as * (for P < 0.05), ** (for P < 0.01) or *** (for P < 0.0001).


   Results
Top
 Abstract
 Introduction
 Materials and methods
 Results
Discussion
 References
 
Construction of an IL-8-expressing cell line
A human IL-8 cDNA clone from I.M.A.G.E. Consortium was used as the starting point to prepare an IL-8-expressing construct. The coding region was subcloned into a mammalian expression vector, pcDNA3, downstream of a CMV promoter. To verify its ability to express biologically active IL-8, the pcDNA3–IL8 construct was transiently transfected into Mutatect MC17-51 cells; as a negative control, cells were transfected with vector alone. IL-8 levels in the culture medium from transiently transfected cells were measured by ELISA. Medium from pcDNA3–IL8 transfected cells was estimated to contain ~1.25 ng IL-8/ml; control medium (cell-free medium) and medium from vector-transfected cells contained <0.1 ng/ml (Figure 1Go). To test the biological activity of the expressed IL-8, an in vitro chemotaxis assay was carried out using a Boyden chamber. Culture medium from pcDNA3–IL-8 transfected cells attracted significantly more human neutrophils than control medium (P < 0.0001) (Figure 1Go). These experiments demonstrate the ability of the pcDNA3–IL8 construct to express biologically active IL-8.



View larger version (42K):
[in this window]
[in a new window]
  		Fig. 1. Mutatect MC17-51 cells, transiently transfected with a pcDNA3–IL8 construct, express IL-8 that is biologically active. Culture media from control (no cells), pcDNA3 or pcDNA3–IL8 transfected cells were examined. Hatched bars (left axis), IL-8 protein detected using a competitive ELISA. Solid bars (right axis), biological activity of IL-8 using an in vitro chemotaxis assay for human neutrophils. Other details are provided in Materials and methods. Each bar represents mean ± SEM of at least triplicate values. Differences between pcDNA3 and pcDNA3–IL-8 culture supernatants were highly statistically significant (P < 0.0001, indicated by ***). When normalized, 1 ng/ml IL-8 in the ELISA is equivalent to 7.5 ng IL-8/1x106 cells/24 h.

 
pcDNA3–IL8 (which also contains a neo gene) was transfected into Mutatect MC17-51 cells to construct stable cell lines that secrete IL-8. G418-resistant clones were selected and screened for IL-8 expression (Table IGo). Two clones were selected for further experiments: MIL-4, which produced 2.4 ng/106 cells/24 h and MIL-1, which produced < 0.05 ng/106 cells/24 h.


View this table:
[in this window]
[in a new window]
  		Table I. IL-8 produced by G418-resistant MIL clonesa
 
Tumorigenicity of IL-8 secreting MIL-4 cells
MIL-4, MIL-1 and MC17-51 cells did not differ significantly in their in vitro growth rates; doubling times were 14–16 h. To determine their tumorigenicity and in vivo growth rates, 5x105 cells of each type were injected subcutaneously into groups of C57BL/6 mice and tumor size was measured at day 21. Mice injected with non-IL-8-secreting cells (MC17-51 or MIL-1) produced tumors that reached ~10 mm in diameter at day 21 (Figure 2AGo). In contrast, IL-8 secreting MIL-4 cells produced very slow growing tumors, whose average size was <3 mm diameter at day 21 (P < 0.01).



View larger version (18K):
[in this window]
[in a new window]
  		Fig. 2. Comparison of the tumorigenicity of Mutatect MIL-1, MIL-4, MC17-51 and mixtures of MIL-4 and MC17-51 cells in syngeneic C57BL/6 mice. (A) Tumor size (largest diameter) measured 21 days after subcutaneous injection of 5x105 MC17-51, MIL-1 or MIL-4 cells. (B) Volume of tumors at the indicated time following injection of mixtures of 0–4x105 MIL-4 cells and 0–5x105 MC17-51 cells; the values 0:5, 0.5:5, 1:5, 2:5, 4:5 and 5:0 represent the number of MIL-4 and MC17-51 cells, respectively, in the mixtures. Tumor volume was calculated as described in Materials and methods. Each point represents the mean volume of five tumors in each group. Tumor volume at day 24 for group 5:0 was significantly smaller than other groups (P = 0.006, Kruskal–Wallis test). For clarity, standard error bars are only shown for volumes at day 24 and the 2:5 curve is omitted. Lines represent curves fitted according to an exponential growth model.

 
IL-8 is a chemokine for neutrophils and slow growth of MIL-4 tumors could be due to the cytotoxicity of a high number of neutrophils expected to be attracted to the site of tumor injection (20). In an attempt to manipulate the local concentration of IL-8, we mixed IL-8-producing MIL-4 cells with non-producing MC17-51 cells prior to injection. In an initial experiment, the total number of cells was kept constant at 5x105 per inoculation, with varying ratios of the two cell types (Table IIGo). With increasing numbers of MIL-4 cells and decreasing numbers of MC17-51 cells, the tumor volume was progressively smaller. In a larger series involving 30 animals, cells were mixed in the following ratios and injected into six groups of mice (five animals per group): 0:5, 0.5:5, 1:5, 2:5, 4:5 and 5:0 (x105 MIL-4:MC17-51). The same cell mixtures were also grown in vitro for an equivalent period of time. Tumor volumes were monitored until excision at day 24 (Figure 2BGo). In agreement with the results shown in Figure 2AGo and Table IIGo, injection of pure MIL-4 cells (i.e. 5:0) produced smaller tumors than any of the MIL-4/MC17-51 mixtures (P = 0.006), with growth kinetics indicating a lag phase that was at least 1 week longer than pure MC17-51 tumors. Tumors initiated with the 0.5:5 mixture tended to grow more slowly than 0:5 (pure MC17-51) tumors, although the difference was not statistically significant. Other combinations (1:5, 2:5 and 4:5) formed tumors whose growth was intermediate. (The growth curve of 2:5 tumors has been omitted for clarity.) Thus, pure MIL-4 tumors grew very slowly and the growth rate of MC17-51 tumors was modestly decreased by addition of 0.5–4x105 MIL-4 cells.


View this table:
[in this window]
[in a new window]
  		Table II. Number of Mutatect MC17-51 and MIL-4 cells used to form mixed tumors and the size of each tumor at day 20a
 
Fate of MIL-4 cells in mixed tumors
MIL-4 cells contain a neo gene that allows them to be distinguished from MC17-51 cells on the basis of resistance to G418. Mixtures of MIL-4 and MC17-51 cells were cultured and the percentage of G418R cells was measured after 30 days. The percentage of MIL-4 cells in the population was essentially identical to the initial percentage of MIL-4 cells used in the culture (Figure 3AGo). This provides further evidence that MIL-4 and MC17-51 cells have very similar growth rates in vitro. Tumors formed from the same mixtures of MIL-4 and MC17-51 cells (as in Figure 2BGo) were next examined to determine whether either cell type had a growth advantage in vivo. These tumors were excised at day 24 and cultured ex vivo in the absence or presence of G418. It was anticipated that the percentage of G418R cells in mixed tumors would decrease, since tumors from 100% MIL-4 cells grew very slowly (see Figure 2Go). Contrary to expectation, the percentage of G418R cells recovered from mixed tumors was as high or higher than corresponding in vitro cultures (Figure 3AGo), suggesting a growth advantage to the IL-8-expressing MIL-4 cells in mixed tumors. We then examined the level of ex vivo IL-8 expression in cells from the same set of tumors. IL-8 was measured in the culture media using ELISA. Cells from mixed MIL-4/MC17-51 tumors produced a significant amount of IL-8, comparable with or even higher than corresponding in vitro mixed cultures (Figure 3BGo), again suggesting that MIL-4 cells had a growth advantage in mixed tumors.



View larger version (25K):
[in this window]
[in a new window]
  		Fig. 3. Expression of neo and IL-8 by cells recovered from Mutatect tumors. (A–B) Percentage of neo-expressing (G418-resistant) cells (A) or detection of IL-8 levels in cells (B) recovered from tumors (solid bars) and cells grown in vitro for the equivalent period of time (hatched bars). The values 0:5, 0.5:5, 1:5, 2:5 and 4:5 refer to the ratio of MIL-4 and MC17-51 cells used in the initial injection as described in the legend to Figure 2Go. Each bar represents mean ± SEM of five tumors or three cultures. Open bars represent the ratio of G418-resistant cells used or the IL-8 produced in the initial injection. (C) Percentage of G418-resistant MIL-1 and MIL-4 cells grown in vitro (closed symbols) or recovered from tumors (open symbols). Two independent experiments with MIL-4 cells are shown. Cells recovered from MIL-1 tumors are not 100% G418R because they contain 10–20% host cells (24). (D) Detection of IL-8 levels in cells grown in vitro (closed symbols) or recovered from tumors (open symbols). In all cases (A–D), tumors were harvested after 24 days. Other details are provided in Material and methods.

 
G418 resistance and IL-8 expression in pure MIL-4 tumors
Pure MIL-4 cells formed very slow growing tumors (Figure 2Go) but by day 24 at least 4x106 viable cells could be recovered from each tumor. When examined for G418 resistance, it was found that the majority (>90%) of cells cultured ex vivo were sensitive to the drug while cells not passaged through the animal remained resistant (Figure 3CGo). Tumors were also initiated with a G418R, non-IL-8-producing line (MIL-1) to determine whether the absence of IL-8 expression could influence loss of neo expression. As shown in Figure 3CGo, cells cultured ex vivo from MIL-1 tumors did not lose G418 resistance. This suggested that expression of IL-8 might be needed for high frequency in vivo loss of neo gene expression. Cells recovered from MIL-4 tumors were examined for their capacity to produce IL-8 in culture. It was found that, as with neo gene function, most cells had lost their capacity to produce IL-8 (Figure 3DGo). Taken together, these observations raised the possibility that the IL-8/neo transgene itself was very unstable in vivo.

IL-8/neo transgene in MIL-4 and mixed MIL-4/MC17-51 tumors
To examine the stability of the IL-8/neo transgene in tumors, genomic DNA from the mixed MIL-4/MC17-51 and pure MIL-4 tumors was analyzed by PCR for the presence of a 214 bp fragment corresponding to the 5'-end of the IL-8 cDNA. Little or none of this PCR fragment could be detected in DNA from six separate MIL-4 tumors (Figure 4Go). In contrast, DNA from all mixed MIL-4/MC17-51 tumors contained these IL-8 specific sequences; the PCR fragment was detectable in all 13 of the mixed tumors analyzed (data not shown). These results strongly suggest that loss of neo and IL-8 function in MIL-4 tumors was due to physical loss of the transgene and, conversely, conservation of neo and IL-8 function in mixed tumors was associated with retention of the transgene.



View larger version (45K):
[in this window]
[in a new window]
  		Fig. 4. Loss of the IL-8/neo transgene from MIL-4 tumors. Genomic DNA was extracted from 24-day MIL-4 tumors and from cultured MC17-51 or MIL-4 cells. A 214-bp fragment corresponding to the 5'-end of the IL-8 cDNA in the pcDNA3–IL8 transgene was amplified by PCR. The band corresponding to a non-specific 670 bp fragment was used as an internal control. Other details are presented in Materials and methods.

 
Neutrophil myeloperoxidase in MIL-4 and mixed MIL-4/MC17-51 tumors
IL-8 is a chemoattractant for human neutrophils (Figure 1Go) and was expected to increase the neutrophil content of Mutatect tumors containing the IL-8/neo transgene. To quantify the infiltration of neutrophils into MIL-4 and mixed tumors, the neutrophil-specific marker, MPO, was measured in tumor extracts (29). The level of MPO in pure MIL-4, pure MC17-51 and in the mixed MIL-4:MC17-51 ratio of 0.5:5 tumors were all very similar (Figure 5AGo). In 1:5, 2:5 and 4:5 mixed tumors, the MPO levels were statistically significantly higher. This pattern parallels the level of IL-8 measured in the same tumors (Figure 3BGo). Ex vivo cultures expressing >0.5 ng IL-8/106 cells/24 h consistently had higher levels of MPO than cultures expressing <0.25 ng IL-8/106 cells/24 h (P < 0.0001) (Figure 5BGo). These experiments confirm that IL-8 is functional as a chemoattractant for mouse neutrophils in Mutatect tumors.



View larger version (23K):
[in this window]
[in a new window]
  		Fig. 5. MPO (U/µg protein) in mixed MIL-4/MC17-51 tumors and its correlation with IL-8 levels. (A) MPO (a neutrophil-specific marker) was measured in tumor extracts as described in Materials and methods. Each point represents a single tumor and the bar represents the median. (B) Correlation between the IL-8 expression in cultured cells recovered from tumors and MPO in the tumor extracts. The values 0:5, 0.5:5, 1:5, 2:5, 4:5 and 5:0 refer to the ratio of MIL-4 and MC17-51 cells used in the initial injection as described in Figure 2Go. The symbols in (B) correspond to the same mixed tumors as in (A).

 
Neutrophil myeloperoxidase and the level of Hprt mutations
Mutatect cells were developed to permit detection of Hprt mutations that arise in vivo due to factors in the tumor microenvironment (22). The frequency of Hprt mutants was recently shown to correlate strongly with the number of intratumoral neutrophils (23). The data shown in Figure 6Go are consistent with our earlier observations. When results from all tumors were pooled, there was a clear correlation between the neutrophil content (MPO) and the Hprt mutant frequency.



View larger version (17K):
[in this window]
[in a new window]
  		Fig. 6. Correlation between the level of MPO and the in vivo frequency of Hprt mutations (MF, number of Hprt mutants per 105 clonable cells) in mixed MIL-4/MC17-51 tumors. MPO was measured in tumor extracts and Hprt mutant cells were detected as described in Materials and methods. The values 0:5, 0.5:5, 1:5, 2:5, 4:5 and 5:0 refer to the ratio of MIL-4 and MC17-51 cells used in the initial injection as described in Figure 2Go. The symbols correspond to the same mixed tumors as in Figure 5AGo. Spearman rank correlation r = 0.4, P = 0.05, n = 26 (r = 0.52, P = 0.007 excluding the outlier). When the 5:0 data are excluded, correlation of r = 0.7, P = 0.0006, n = 20 is obtained.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
References
 
IL-8 is a potent neutrophil chemokine and angiogenic factor, and expressed by many human tumors. Expression has been associated with a high level of neutrophil infiltration, increased vascularization and poor prognosis (13–15). We have previously shown that direct injection of IL-8 into established Mutatect tumors can increase the number of infiltrating neutrophils, and that this is associated with an elevated mutation frequency in the tumor cells (23). To more closely model the situation that may exist in a human tumor, we constructed a Mutatect cell line (MIL-4) that constitutively expresses IL-8. The presence of IL-8 expressing cells in tumors had a marked effect on their growth. Pure MIL-4 tumors grew very slowly compared with pure non-IL-8 secreting MC17-15 tumors or mixed MIL-4/MC17-51 tumors. Perhaps because it is a multifunctional cytokine with chemoattractant and angiogenic properties, IL-8 can apparently be both growth-promoting and growth-inhibiting. In CHO tumors in nude mice (20) and pure MIL-4 tumors, production of IL-8 reduced tumorigenicity whereas in TMK-1 tumors in nude mice IL-8 production enhanced tumorigenicity (21). To explain the differences in the growth properties of these tumors and the tumors initiated with pure MIL-4 cells and MIL-4/MC17-51 cells, we assume that differences in the local concentration of IL-8 are responsible. In pure MIL-4 tumors, high infiltrating cytotoxic neutrophils early in tumor development may prevent tumors becoming established. In mixed tumors, low IL-8 concentration may result in fewer infiltrating cytotoxic neutrophils early in tumor development, permitting tumors to become established more readily. Eventually, as mixed tumors grow in size, the number of infiltrating neutrophils relative to the tumor mass may be insufficient to alter the course of tumorigenesis. We observed enrichment of IL-8-secreting MIL-4 cells over MC17-51 cells in mixed tumors, which might be due to a growth advantage conferred by IL-8-mediated angiogenesis (21) in MIL-4-rich segments of the tumor. The ultimate effect of IL-8 on the biology of a tumor may depend on its local concentration and on the presence or absence of other chemoattractant or angiogenic factors in the tumor.

The presence of phenotypic and genotypic markers in MIL-4 cells allowed their fate to be studied in pure and mixed tumors. Tumors initiated with pure MIL-4 cells not only grew very slowly, with a long lag phase, but the small tumors that eventually did grow lost their IL-8/neo transgene with a remarkably high frequency. Despite the fact that the transgene is very stable in cultured MIL-4 cells, only <10% of cells recovered from tumors were resistant to G418; their IL-8 level was also <10% of controls, and most cells appeared to have lost the transgene (as measured by PCR). In striking contrast, the fate of the transgene in tumors initiated with mixtures of MC17-51 cells and MIL-4 cells was different. Not only were G418-resistant cells detected in mixed tumors, in most cases their proportion actually exceeded that in the initial injection mixture. Similarly, IL-8 expression in cells cultured ex vivo was as high or higher than in the in vitro control mixtures. Consistent with these findings, the IL-8/neo transgene was retained. In the case of the 0.5:5 mixture, the proportion of G418-resistant cells and the level of IL-8 was similar or slightly lower than the proportion in the initial injection mixture; this mixture also produced slow growing tumors. PCR analysis of DNA from MIL-4 cells in pure and mixed tumors indicates that the loss of G418 resistance and IL-8 expression is best explained by loss of the transgene.

To explain the high frequency loss of the IL-8/neo transgene in pure MIL-4 tumors, we postulate that neutrophils, attracted by IL-8 early in the development of tumors, are responsible. Reactive nitrogen and oxygen species, products of mouse neutrophils (23), can be both genotoxic and cytotoxic (30). Since initially all MIL-4 cells were G418 resistant and IL-8 producing, and remained so for at least 5 weeks in vitro (Figure 3Go), it is probable that genotoxicity mediated in vivo by a neutrophil-derived reactive species was responsible for the instability of the transgene. This notion is supported by recent evidence from our laboratory that an antioxidant, vitamin E, can markedly reduce the mutagenicity (based upon Hprt mutant frequency) in Mutatect tumors (31). Neutrophils probably also mediated the cytotoxicity responsible for the observed lag in tumor growth. To explain selective killing of IL-8-producing MIL-4 cells compared with non-producing MIL-4 cells, we postulate that a higher concentration of IL-8 surrounding the former delays apoptosis of neutrophils, locally prolonging their cytotoxic capability. A high concentration of IL-8 has been shown to delay spontaneous apoptosis in human neutrophils (32). The combination of genotoxicity leading to loss of the transgene and selective cytotoxicity of IL-8 producing cells may explain why MIL-4 cells lacking the transgene preferentially escape killing in pure MIL-4 tumors.

In summary, IL-8 is a multifunctional cytokine that can have complex and diverse effects on tumor biology. It can attract neutrophils and stimulate neovascularization. Its net effect on the biology of a tumor may depend on the size of the tumor and amount of IL-8 produced. Under some conditions, IL-8 may attract cytotoxic neutrophils that can delay tumor growth. Under other conditions, angiogenesis may predominate and IL-8 enhance tumor growth. Neutrophils can also be genotoxic, increasing the Hprt mutant frequency in MIL-4 tumors (Figure 6Go). Thus, the potential effects of IL-8 in tumors are complex and may include cytotoxicity and mutagenicity mediated by neutrophils.


   Notes
 
1 Present address: Institute for Biological Sciences, National Research Council of Canada, Ottawa, Ontario K1A 0R6, Canada Back

2 To whom correspondence should be addressedEmail: birnboim{at}uottawa.ca Back


   Acknowledgments
 
We thank Donna Grant and Denise Proulx for their excellent technical assistance. This work was supported by Medical Research Council of Canada, grant no. MT-8728 to H.C.B. A.S.H. is a recipient of a doctoral research award from Medical Research Council of Canada. H.C.B. is a Senior Career Scientist of Cancer Care Ontario.


   References
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 

  1. Schröder,J.M., Mrowietz,U., Morita,E. and Christophers,E. (1987) Purification and partial biochemical characterization of a human monocyte-derived, neutrophil-activating peptide that lacks interleukin 1 activity. J. Immunol., 139, 3474–3483.[Abstract]
  2. Furuta,R., Yamagishi,J., Kotani,H., Sakamoto,F., Fukui,T., Matsui,Y., Sohmura,Y., Yamada,M., Yoshimura,T. and Larsen,C.G. (1989) Production and characterization of recombinant human neutrophil chemotactic factor. J. Biochem. (Tokyo), 106, 436–441.[Abstract/Free Full Text]
  3. Oppenheim,J.J., Zachariae,C.O., Mukaida,N. and Matsushima,K. (1991) Properties of the novel proinflammatory supergene intercrine cytokine family. Annu. Rev. Immunol., 9, 617–648.[ISI][Medline]
  4. Koch,A.E., Polverini,P.J., Kunkel,S.L., Harlow,L.A., DiPietro,L.A., Elner,V.M., Elner,S.G. and Strieter,R.M. (1992) Interleukin-8 as a macrophage-derived mediator of angiogenesis. Science, 258, 1798–1801.[Abstract/Free Full Text]
  5. Roberts,P.J., Pizzey,A.R., Khwaja,A., Carver,J.E., Mire-Sluis,A.R. and Linch,D.C. (1993) The effects of interleukin-8 on neutrophil fMetLeuPhe receptors, CD11b expression and metabolic activity, in comparison and combination with other cytokines. Br. J. Haematol., 84, 586–594.[ISI][Medline]
  6. Yuo,A., Kitagawa,S., Kasahara,T., Matsushima,K., Saito,M. and Takaku,F. (1991) Stimulation and priming of human neutrophils by interleukin-8: cooperation with tumor necrosis factor and colony-stimulating factors. Blood, 78, 2708–2714.[Abstract/Free Full Text]
  7. Elbim,C., Bailly,S., Chollet-Martin,S., Hakim,J. and Gougerot-Pocidalo,M.A. (1994) Differential priming effects of proinflammatory cytokines on human neutrophil oxidative burst in response to bacterial N-formyl peptides. Infect. Immun., 62, 2195–2201.[Abstract/Free Full Text]
  8. Rot,A. (1991) Chemotactic potency of recombinant human neutrophil attractant/activation protein-1 (interleukin-8) for polymorphonuclear leukocytes of different species. Cytokine, 3, 21–27.[ISI][Medline]
  9. Sugawara,T., Miyamoto,M., Takayama,S. and Kato,M. (1995) Separation of neutrophils from blood in human and laboratory animals and comparison of the chemotaxis. J. Pharmacol. Toxicol. Methods, 33, 91–100.[ISI][Medline]
  10. Singh,R.K., Varney,M.L., Bucana,C.D. and Johansson,S.L. (1999) Expression of interleukin-8 in primary and metastatic malignant melanoma of the skin. Melanoma Res., 9, 383–387.[ISI][Medline]
  11. Singh,R.K., Gutman,M., Radinsky,R., Bucana,C.D. and Fidler,I.J. (1994) Expression of interleukin 8 correlates with the metastatic potential of human melanoma cells in nude mice. Cancer Res., 54, 3242–3247.[Abstract/Free Full Text]
  12. Colasante,A., Mascetra,N., Brunetti,M., Lattanzio,G., Diodoro,M., Caltagirone,S., Musiani,P. and Aiello,F.B. (1997) Transforming growth factor beta 1, interleukin-8 and interleukin-1, in non-small-cell lung tumors. Am. J. Respir. Crit. Care Med., 156, 968–973.[Abstract/Free Full Text]
  13. Bellocq,A., Antoine,M., Flahault,A., Philippe,C., Crestani,B., Bernaudin,J.F., Mayaud,C., Milleron,B., Baud,L. and Cadranel,J. (1998) Neutrophil alveolitis in bronchioloalveolar carcinoma: induction by tumor-derived interleukin-8 and relation to clinical outcome. Am. J. Pathol., 152, 83–92.[Abstract]
  14. Eisma,R.J., Spiro,J.D. and Kreutzer,D.L. (1999) Role of angiogenic factors: coexpression of interleukin-8 and vascular endothelial growth factor in patients with head and neck squamous carcinoma. Laryngoscope, 109, 687–693.[ISI][Medline]
  15. Kitadai,Y., Haruma,K., Sumii,K. et al. (1998) Expression of interleukin-8 correlates with vascularity in human gastric carcinomas. Am. J. Pathol., 152, 93–100.[Abstract]
  16. Sakamoto,K., Masuda,T., Mita,S., Ishiko,T., Nakashima,Y., Arakawa,H., Egami,H., Harada,S., Matsushima,K. and Ogawa,M. (1992) Interleukin-8 is constitutively and commonly produced by various human carcinoma cell lines. Int. J. Clin. Lab. Res., 22, 216–219.[ISI][Medline]
  17. Galffy,G., Mohammed,K.A., Dowling,P.A., Nasreen,N., Ward,M.J. and Antony,V.B. (1999) Interleukin 8: an autocrine growth factor for malignant mesothelioma. Cancer Res., 59, 367–371.[Abstract/Free Full Text]
  18. Van Meir,E., Ceska,M., Effenberger,F., Walz,A., Grouzmann,E., Desbaillets,I., Frei,K., Fontana,A. and De Tribolet,N. (1992) Interleukin-8 is produced in neoplastic and infectious diseases of the human central nervous system. Cancer Res., 52, 4297–4305.[Abstract/Free Full Text]
  19. Desbaillets,I., Diserens,A.C., Tribolet,N., Hamou,M.F. and Van Meir,E.G. (1997) Upregulation of interleukin 8 by oxygen-deprived cells in glioblastoma suggests a role in leukocyte activation, chemotaxis and angiogenesis. J. Exp. Med., 186, 1201–1212.[Abstract/Free Full Text]
  20. Hirose,K., Hakozaki,M., Nyunoya,Y., Kobayashi,Y., Matsushita,K.X.T.T., Mikata,A., Mukaida,N. and Matsushima,K. (1995) Chemokine gene transfection into tumour cells reduced tumorigenicity in nude mice in association with neutrophilic infiltration. Br. J. Cancer, 72, 708–714.[ISI][Medline]
  21. Kitadai,Y., Takahashi,Y., Haruma,K. et al. (1999) Transfection of interleukin-8 increases angiogenesis and tumorigenesis of human gastric carcinoma cells in nude mice. Br. J. Cancer, 81, 647–653.[ISI][Medline]
  22. Birnboim,H.C., Wilkinson,D., Sandhu,J.K., McLean,J.R. and Ross,W. (1999) Mutatect: a mouse tumour model for detecting radiation-induced mutations in vivo. Mutat. Res., 430, 275–280.[ISI][Medline]
  23. Sandhu,J.K., Privora,H.F., Wenckebach,G. and Birnboim,H.C. (2000) Neutrophils, nitric oxide synthase and mutations in the Mutatect murine tumor model. Am. J. Pathol., 156, 509–518.[Abstract/Free Full Text]
  24. Wilkinson,D., Sandhu,J.K., Breneman,J.W., Tucker,J.D. and Birnboim,H.C. (1995) Hprt mutants in a transplantable murine tumour arise more frequently in vivo than in vitro. Br. J. Cancer, 72, 1234–1240.[ISI][Medline]
  25. Lennon,G., Auffray,C., Polymeropoulos,M. and Soares,M.B. (1996) The I.M.A.G.E. Consortium: an integrated molecular analysis of genomes and their expression. Genomics, 33, 151–152.[ISI][Medline]
  26. Cowling,R.T. and Chaim Birnboim,H. (1994) Incorporation of [32P]orthophosphate into inorganic polyphosphates by human granulocytes and other human cell types. J. Biol. Chem., 269, 9480–9485.[Abstract/Free Full Text]
  27. Udenfriend,S., Stein,S., Bohlen,P., Dairman,W., Leimgruber,W. and Weigele,M. (1972) Fluorescamine: a reagent for assay of amino acids, peptides, proteins, and primary amines in the picomole range. Science, 178, 871–872.[Abstract/Free Full Text]
  28. De Vouge,M.W., Yamazaki,A., Bennett,S.A.L., Chen,J.-H., Shwed,P.S., Couture,C. and Birnboim,H.C. (1994) Immunoselection of GRP94/endoplasmin from a KNRK cell-specific lambdagt11 library using antibodies directed against a putative heparanase amino-terminal peptide. Int. J. Cancer, 56, 286–294.[ISI][Medline]
  29. Haqqani,A.S., Sandhu,J.K. and Birnboim,H.C. (1999) A myeloperoxidase-specific assay based upon bromide-dependent chemiluminescence of luminol. Anal. Biochem., 273, 126–132.[ISI][Medline]
  30. Sandhu,J.K. and Birnboim,H.C. (1997) Mutagenicity and cytotoxicity of reactive oxygen and nitrogen species in the MN-11 murine tumor cell line. Mutat. Res., 379, 241–252.[ISI][Medline]
  31. Sandhu,J.K., Haqqani,A.S. and Birnboim,H.C. (2000) Effect of dietary vitamin E on spontaneous or nitric oxide donor-induced mutations in a mouse tumor model. J. Natl Cancer Inst., 92, 1429–1433.[Abstract/Free Full Text]
  32. Kettritz,R., Gaido,M.L., Haller,H., Luft,F.C., Jennette,C.J. and Falk,R.J. (1998) Interleukin-8 delays spontaneous and tumor necrosis factor-{alpha}-mediated apoptosis of human neutrophils. Kidney Int., 53, 84–91.[ISI][Medline]
Received July 19, 2000; revised October 30, 2000; accepted November 9, 2000.


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 J. Immunol.Home page
J.-E. Murphy, R. E. Morales, J. Scott, and T. S. Kupper
IL-1{alpha}, Innate Immunity, and Skin Carcinogenesis: The Effect of Constitutive Expression of IL-1{alpha} in Epidermis on Chemical Carcinogenesis
J. Immunol., June 1, 2003; 170(11): 5697 - 5703.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (6)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Haqqani, A. S.
Right arrow Articles by Birnboim, H.C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Haqqani, A. S.
Right arrow Articles by Birnboim, H.C.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?