Overexpression of CXC Chemokines by an Adrenocortical Carcinoma: A Novel Clinical Syndrome
DAVID E. SCHTEINGART, THOMAS J. GIORDANO, RICARDO S. BENITEZ, MARIE D. BURDICK, MONICA N. STARKMAN, DOUGLAS A. ARENBERG, AND ROBERT M. STRIETER
Departments of Internal Medicine (D.E.S., R.S.B., D.A.A.), Pathology (T.J.G.), and Psychiatry (M.N.S.), University of Michigan Medical School, Ann Arbor, Michigan 48109; and Department of Medicine (M.D.B., R.M.S.), Division of Pulmonary and Critical Medicine, UCLA School of Medicine, Los Angeles, California 90024
A patient with adrenocortical carcinoma presented with fe- ver, leukocytosis, and increased acute phase reactants. The tumor was infiltrated with neutrophils. Immunohistochemi- cal staining of the tumor showed positive signal for epithelial neutrophil-activating protein-78, an angiogenic and chemo- tactic CXC chemokine. Conditioned medium from tumor-de- rived cells (RL-251) showed high concentration of IL-8, epi- thelial neutrophil-activating protein-78, Gro a, and Gro y, angiogenic CXC chemokines with a potential role in tumori- genesis. An adrenal cancer/severe combined immunodefi- ciency mouse chimera was developed. Mice grew tumors rapidly, and circulating levels of IL-8 and epithelial neutro- phil-activating protein-78 were detected. In contrast, animals transplanted with NCI-H295 cells, a nonchemokine-secreting cell line, grew tumors more slowly and did not have detectable chemokine levels. Similar to the patient, mice with RL-251
tumors developed marked leukocytosis and neutrophilia, and their tumors were infiltrated with neutrophils. Mice were pas- sively immunized with epithelial neutrophil-activating protein-78 antisera. A marked decrease in tumor growth was observed. Potential for chemokine production by other adre- nocortical tumors was investigated by RT-PCR in archival material. Six of seven adrenal carcinomas and one of three adenomas had cDNA for IL-8; six of seven carcinomas and the three adenomas had cDNA for epithelial neutrophil-activat- ing protein-78. We concluded that the clinical presentation of this case resulted from increased tumor production of che- motactic chemokines. Through their angiogenic and chemo- tactic properties these chemokines may play an important role in adrenal tumorigenesis. (J Clin Endocrinol Metab 86: 3968-3974, 2001)
A DRENOCORTICAL CARCINOMAS ARE rare but highly malignant neoplasms that occur with an inci- dence of one to two per million population per year (1). In approximately 50% of patients, the tumor presents with clin- ical and biochemical manifestations of excessive adrenal cor- tical hormone secretion. Cushing’s syndrome alone or in association with virilization occurs in 85% of functioning tumors. In the remaining 50%, tumors are silent in regards to the well-known metabolic effects of adrenal cortical ste- roids and may be discovered incidentally during abdominal imaging for unrelated abdominal complaints. When tumors are large, the presenting symptoms may be nonspecific such as abdominal pain and fever. We describe a novel clinical syndrome associated with primary adrenocortical carcinoma that appeared to be caused by tumor overproduction of CXC chemokines. These chemokines are cytokines that display four highly conserved cysteine amino acid residues, with the first two cysteines separated by a nonconserved amino acid residue. Chemokines with a three-amino acid motif, the glu- tamic acid-leucine-arginine motif that immediately precedes the first cysteine amino acid residue in the primary structure, have potent angiogenic activity essential for tumor growth. We present evidence that these chemokines contributed to the growth of this adrenocortical carcinoma.
Abbreviations: CT, Computerized tomography; ENA-78, epithelial neutrophil activating protein 78; FACS, fluorescence-activated cell sort- er; Gro a and Gro y, growth-related oncogenes; H&E, hematoxylin and eosin; NSCLC, non-small-cell lung cancer; PBGD, porphobilinogen deaminase; SCID, severe combined immunodeficiency.
Case description
A 75-yr-old man with a past medical history of arthritis and hypertension presented with recent onset of right flank and back pain and blood pressure of 219 /107 mm Hg. He had also noticed a recent onset of spells consisting of chills, warmth, sweating, and flushing. The patient had good ap- petite and had not experienced weight changes. He com- plained of bruising with minor trauma for 5 months before his evaluation. He had been initially evaluated at another hospital at which abdominal computerized tomography (CT) revealed a 4-by-5-cm right adrenal mass. A fine-needle CT guided aspiration biopsy of this mass was consistent with adrenocortical carcinoma. On physical examination the pa- tient was warm and flushed with a temperature of 39.5 C. He did not have a cushinoid appearance. His lungs showed occasional ronchi and his heart a grade ii/vi apical systolic murmur. The abdomen was nontender and no masses were palpable. Extremities revealed 4+ pretibial pitting edema. Initial laboratory studies revealed a normocytic, normochro- mic anemia (hematocrit 27.3%, hemoglobin 9.1 g/100, and marked leukocytosis [white count 27,300/mm3]) with neu- trophilia (83%). His erythrosedimentation rate was high (116 mm/hr) and ferritin levels were also high (962 µg/liter), consistent with elevation of acute phase reactants.
Because of the possibility of an acute infectious process, seven blood cultures were obtained during his febrile epi- sodes without evidence of bacterial growth. A trans-esoph- ageal echocardiogram was negative for vegetations or thrombi and a gallium scan was negative for a focal source
of infection. Repeat abdominal CT revealed a 5 × 6 × 7 heterogeneous mass in the right adrenal. Measurement of adrenal cortical hormones showed normal baseline cortisol and cortisol metabolite excretion measured as urinary free cortisol, 17-hydroxycorticosteroids and 17-ketosteroids. However, serum cortisol levels showed absence of circadian rhythm and corresponding ACTH levels were undetectable (Table 1). The suppressed ACTH levels occurred in spite of fever spikes at the time of sample collection. These studies were performed in the General Clinical Research Center and informed consent was obtained. The research protocol was reviewed and approved by the Institutional Review Board. The patient underwent a laparotomy with identification and resection of an well-encapsulated, necrotic adrenal mass. Six intraduodenal, umbilicated neoplastic lesions were ob- served. One of these lesions was resected. Culture of tissue samples obtained from the primary adrenal mass showed no bacterial growth. The pathological examination of the sur- gical specimens showed a poorly differentiated adrenocor- tical carcinoma and intraduodenal metastases. Figure 1 shows a high-power view of a histological section obtained from the primary tumor. The histological pattern was con- sistent with a high-grade adrenocortical carcinoma exhibit- ing nuclear pleomorphism and mitoses surrounded by neu- trophils. The metastatic lesion revealed a cellular pattern identical to the primary tumor. Postoperatively the patient became afebrile and the white count decreased by 40%. How- ever, he died several months later with tumor metastases.
Materials and Methods
Immunohistochemistry
Samples of tumor were fixed in 4% paraformaldehyde and sections prepared and stained for epithelial neutrophil activating protein 78 (ENA-78) using a modification of a previously described technique (2). Nonspecific binding sites were blocked with normal goat serum, washed, and overlaid with 1:1500 dilution of either control (rabbit) or rabbit antihuman ENA-78 serum. Slides were then rinsed and overlaid with secondary biotinylated goat antirabbit IgG (1:35) and incubated for 60 min. After washing twice with Tris-buffered saline, slides were over- laid with streptavidin, conjugated horseradish peroxidase (Vectastain ABC reagent) and incubated for 45 min. Diaminobenzidine was used for chromogenic localization of the chemokine antigen.
Establishment of RL-251 cell line
A cell suspension was prepared from a tumor specimen after treat- ment with collagenase and plated in fibronectin-coated plastic with RPMI 1640 medium containing 10% FBS and basic fibroblast growth factor. After several passes the cells were able to grow in uncoated plastic and in the absence of growth factors with a control population doubling time of 36 h. Colony formation experiments indicated a 35% plating
efficiency. These cells did not produce significant amounts of cortisol in vitro. The cells had an abnormal diploid karyotype with numerous deletions and translocations. NCI-H295 cells obtained from the Amer- ican Type Culture Collection (Manassas, VA) were used as control. These cells belong to a steroid-secreting, human adrenocortical carci- noma cell line established by Gazdar (3) with a 7-d doubling time. Cell growth for both cell lines was evaluated by measuring relative protein content using the sulforhodamine B assay (4) at baseline and at various times after incubation.
Measurement of CXC chemokines. IL-8, ENA-78, and growth-related on- cogenes (Gro & and Gro y) were measured in supernatant of tumor cell cultures using specific ELISAs (5, 6). Flat-bottom 96-well microtiter plates (NUNC IMMUNO-PLATE 196-F) were coated with a 50 pl/well of the polyclonal antichemokine antibodies for 24 h at 4 C and washed with PBS, 7.5, 0.005% TWEEN-20 (wash buffer). Microtiter plate non- specific-binding sites were blocked with 2% BSA in PBS and incubated for 60 min at 37 C. Plates were rinsed three times. A 50-pL sample was added followed by incubation for 1 h at 37 C. Biotinylated polyclonal rabbit anti-CXC chemokines, streptavidin-peroxidase conjugate, and chromogen substrate were sequentially added to the plates, each time followed by incubation and washings. After the addition of chromogen substrate, the plates were incubated at room temperature to the desired extinction and the reaction terminated with a 50-ml/well of 3 MH2SO4 solution. Plates were read at 490 nm in an automated microplate reader.
Establishment of an animal model. An adrenal cancer/severe combined immunodeficiency (SCID) mouse chimera was developed with each of the two cell lines, RL-251 and NCI-H295. Fox Chase SCID CB 17 mice, 4-6 wk of age, were injected sc in the flank with 1 × 106 cells suspended in 100 ul RPMI 1640 media. A 90% tumor take was observed. Animals were maintained under sterile conditions in laminar floor rooms in the Unit for Laboratory Animal Medicine, and studies were conducted in
| U-170HCS (mg/d) | U17KS (mg/d) | UFC (µg/d) | Cortisol/ACTH (µg/dl)/(pg/ml) | ||||
|---|---|---|---|---|---|---|---|
| 0800 h | 1200 h | 1600 h | MN | ||||
| Hormones | 7.0 | 7.8 | 29.2 | 24.4 | 26.0 | 21.7 | 22.5 |
| 0 | 0 | 0 | 0 | ||||
| Temp (C) | 38.7 | 38.0 | 38.6 | 36.3 | |||
ACTH and cortisol levels did not respond to fever. U17OHCS, Urinary 17-hydroxysteroids; normal: 2-10 mg/d; U17KS, urinary 17-ketosteroids; normal: 7-15 mg/d; UFC, urinary free cortisol; normal 20-90 µg/d; serum cortisol; normal 5-25 µg/dl (0800 h); 3-12 µg/dl (1600 h); ACTH, normal: 9-52 pg/ml.
accord with the principles and procedures outlined for the use of animals in research.
Passive immunization studies. To test the hypothesis that the CXC che- mokines promoted tumor growth in the RL-251/SCID mouse chimera, mice were passively immunized with anti ENA-78 antisera to block its angiogenic and chemotactic activity. Polyclonal rabbit antihuman ENA-78 sera were produced by immunization of rabbits with ENA-78 (Preprotech, Rocky Hill, NJ) in multiple intradermal sites with complete Freund’s adjuvant. Antibody specificity was confirmed by Western blot analysis against recombinant human ENA-78. The antibody was found to neutralize 30 ng of ENA-78 at a dilution of 1:1000. Twenty SCID mice were inoculated with RL-251 cells (1 × 106 cells/inoculate) sc in each flank. Ten animals were injected with anti-ENA-78 sera ip every 2 d for 6 wk beginning at the time of inoculation. The other 10 animals were injected with normal rabbit serum ip every 2 d and used as controls. At the completion of the study, animals were euthanized and tumors har- vested, weighed in a Mettler balance, and partitioned for endothelial cell and leukocyte markers.
Endothelial cell and leukocyte markers. Tumor tissue was digested with collagenase and DNAase and cell suspensions prepared. Specific anti- bodies against endothelial cell markers factor VIII and CD31 were added to wells of microtiter plates and cells delivered to each well and mixed. After incubation at 4 C for 20-40 min in the dark, they were washed and stained using fluorochromes and processed by flow cytometry. Results were expressed as percent of cells that stained positive for FVIII and CD31. Leukocyte infiltration was assessed by fluorescence-activated cell sorter (FACS) analysis of tumor digests.
Hormone assays. Serum cortisol levels were measured by an automated method, plasma ACTH by the Allegro Nichols assay, urinary 17-hy- droxycorticoids by the method of Silber & Porter (7) and 17-ketosteroids by the method of Drekter (8). Urinary free cortisol was measured by RIA using Diagnostic Products Corporation (Los Angeles, CA) kits.
RT-PCR for IL-8 and ENA-78. RT-PCR was performed by the method of Baumforth et al. (9). IL-8 primers used were AAGCTGGCCGTG- GCTCTCTTG and AGCCCTCTTCAAAAACTTCTC. The upstream primer was located at the 3’ end of exon 1 and the downstream primer was located at the 3’ of exon 3 and overlapped exon 4 by three base pairs (GenBank accession M28130). ENA-78 primers were GAACCCGCGAC- CGCTCGCAGC and CCTATGGCGAACACTTGCAGA. The upstream primer began at base 294 of ENA-78 messenger RNA and the down- stream primer began at base 428 (GenBank accession ×78686). Porpho- bilinogen deaminase (PBGD) primers for both IL-8/PBGD AND ENA- 78/PBGD were ACGATCCCGAGACTCTGCTTCG and GGTTACATT- AAAGGCTGTTGCTTG.
Results
In vitro studies
Immunoreactive ENA-78 was present in tumor sections stained by an immunoperoxidase method and suggested that the tumor was producing this chemokine (Fig. 2). We then measured CXC chemokine expression in vitro during incu- bation of RL and NCI-H295 tumor cells. Conditioned media was harvested from confluent cultures of both cell types at 1, 3, and 6 d. High levels of CXC chemokines (IL-8, ENA-78, Gro a and Gro y) were observed in cultures of the RL-251 cells (Fig. 3). Expression of CXC chemokines was undetect- able in culture supernatants of the NCI-H295 cells.
In vivo studies
An adrenal cancer/SCID mouse chimera was developed with each of these cell lines. Mice transplanted with RL-251 cells grew tumors within 2-3 wk. Animals were euthanized at the end of the study and peripheral blood obtained through cardiac puncture for analysis of IL-8 and ENA-78. Circulating chemokine levels were detected in every animal (Fig. 4). In contrast, blood obtained from animals trans- planted with NCI-H295 cells did not have detectable che- mokine levels. The mean peripheral white blood cell count of the RL mice (n = 5) was 56.8 x 103 with 90.4% neutrophils, compared with 2.0 × 103 with 70% neutrophils in the control mice without tumors (n = 3); P < 0.001. Tumors were dis- sected from the sc tissue, weighed, and examined histolog- ically. A high-power view of the hematoxylin and eosin (H&E) stained sections of the RL-251 CXC chemokine pro- ducing mouse tumor showed a carcinoma histologically sim- ilar to the original surgical specimen. Tumor cells were sur- rounded by an intense neutrophil infiltrate (Fig. 5). Tumor sections stained positive for ENA-78. In contrast, the tumor derived from NCI-H295 cells did not show a neutrophil infiltrate. Also and in contrast to animals with the NCI-H295 tumors, animals with the RL-251 tumors exhibited marked leukocyte infiltration of their livers. The growth rates of the two cell lines in vitro and tumors in vivo were different. Doubling time (mean ± SD) for the CXC chemokine-secreting
9
FIG. 3. CXC chemokines released in vitro by RL-251 cells. Conditioned cul- ture media was sampled at 1, 3, and 6 d of incubation and assayed for CXC che- mokines. Expression of CXC chemo- kines was undetectable in culture su- pernatants of the NCI-H295 cells.
4
3.5
IL-8
3
ENA-78
IL-8 ng/ml
2.5
2
1.5
1
0.5
0
1
2
3
sample
FIG. 4. CXC chemokine levels in animals transplanted with RL-251 cells. Each set represents one animal. Mice are not expected to have detectable human IL-8 or ENA-78
cells was shorter 44 ± 7.2 h than the nonchemokine-produc- ing cells, 109 ± 15.3; P < 0.001. Tumors derived from the CXC chemokine-producing cell line grew faster than the nonche- mokine-producing tumors. Mice inoculated with 10 × 106 RL-251 cells took 12 d to show tumors, compared with 80 d for a similar inoculate dose of the NCI-H295 cells.
IL-8
ENA-78
80
9000
70
8000
60
7000
60
6000
ng/ml
50
ng/ml
5000
40
4000
30
3000
20
2000
10
1000
0
0
-10
-1000
Baseline
Day 1
Day 3
Day 6
Baseline
Day 1
Day 3
Day 6
GRO &
GRO Y
140
0.25
120
0.2
100
ng/ml
80
ng/ml
0.15
60
0.1
40
20
0.05
0
0
-20
Baseline
Day 1
Day 3
Day 6
-0.05
Baseline
Day 1
Day 3
Day 6
4
3.5
3
ENA-78 ng/ml
2.5
2
1.5
1
0.5
0
4
Passive immunization with ENA-78 antisera. Mice bearing tu- mors from the chemokine-producing RL-251 cell line were subjected to passive immunization with a high titer neutral- izing antisera to the chemokine ENA-78, or control (preim- mune) serum. After 6 wk of tumor growth, the mean weight (in grams) of tumors from the anti-ENA-78 treated mice,
0.278 ± SE 0.093 (n = 9) was much lower than the weight of the tumors from the control mice, 0.854 ± SE 0.213 (n = 10); P < 0.03. Tumor take in the treated mice decreased by 20%. Endothelial cell markers decreased by 50% (Fig. 6). Leuko- cyte infiltration, as assessed by FACS analysis of tumor di- gests was also decreased (Fig. 7)
Potential for CXC chemokine production by other adrenocortical tumors. H&E stained slides of 16 archival adrenal carcinomas were examined for the presence of inflammatory cells. Five of the 16 tumors (30%) had inflammatory cells consisting mainly of lymphocytes, eosinophils, and plasma cells. One of the specimens had neutrophils within a scar. These tumors, however, were not simultaneously examined for the pres- ence of chemokines. The ability to produce IL-8 and ENA-78 was examined in tissue from another seven archival adrenal cortical carcinomas by RT-PCR. RL-251 and NIH-295 fresh cells were used as controls. Samples from three benign ad- renal adenomas were analyzed for comparison with the ma- lignant tumors. Results are summarized in Table 2. As ex- pected, RL-251 cells had cDNA for both IL-8 and ENA-78. H-295 cells that do not produce either chemokine in vitro or in vivo did not have cDNA for IL-8 but had cDNA for ENA- 78. Six of seven adrenal carcinomas and one of three ade- nomas had cDNA for IL-8. Six of seven carcinomas and the three adenomas had cDNA for ENA-78.
Discussion
We suggest that the patient’s clinical presentation resulted from increased tumor production of chemotactic chemokines that mobilized white cells and produced an inflammatory syndrome with fever and leukocytosis. This is supported by the presence of positive signal for ENA-78 obtained in tumor sections by immunohistochemistry. The patient also had bio- chemical evidence of ACTH-independent cortisol produc- tion by the tumor but urinary free cortisol levels were nor- mal, and he did not have clinical evidence of Cushing’s syndrome. This biochemical profile has been described as preclinical Cushing’s syndrome and found in association with incidentally discovered adrenocortical tumors (10). We investigated whether CXC chemokine production was an isolated occurrence in this patient or a feature of other ad- renocortical tumors by surveying archival adrenocortical tu-
20
FactorVIII
cell(X106)
15
FVIII+CD31
10
5
0
NRS
anti ENA-78
60
PM N’s (cells x105)
50
40
30
20
10
0
NRS
anti ENA-78
mors for presence of inflammatory cells and message for IL-8 and ENA-78. Inflammatory infiltrates were found in 30% of the tumors, and message for IL-8 and ENA-78 was detected in the majority of both malignant and benign adrenocortical tumors.
A unique cell line derived from our patient’s tumor was able to produce IL-8, ENA-78, Gro a, and Gro y. These four cytokines are members of the CXC chemokine family. In their monomeric form, the CXC chemokines range in mass from 7 to 10 kDa and are characteristically basic, heparin-binding proteins. These chemokines display four highly conserved cysteine amino acid residues, with the first two cysteines separated by a nonconserved amino acid residue (CXC) (11). The CXC chemokines are all clustered on human chromo- some 4 (q12-q21) and exhibit 20-50% homology on the amino acid level. The CXC chemokine family was originally rec- ognized for their leukocyte chemotactic activity. However, these cytokines can also behave as angiogenic or angiostatic factors depending upon the presence or absence, respec- tively, of a three amino acid motif, glu-leu-arg (ELR motif) that precedes the first conserved amino acid residue in the primary structure (12). The four CXC chemokines produced by this cell line contain the ELR motif and possess angiogenic activity.
The possibility that an adrenal neoplasm can overproduce chemokines is consistent with the observation that the nor- mal adrenal produces cytokines. Main sources are local im- mune cells (13) and adrenal cortical cells. When stimulated, these immune cells secrete a variety of cytokines, including IL-1, IL-6, and TNF-a and peptides such as VIF and TGF-B (14, 15) that can influence adrenal cortical function. Although there is extensive literature regarding cytokines in the ad- renal cortex, there are no reports examining the expression or secretion of the subclass of CXC chemokines. It is possible that these chemokines are locally produced and play an autocrine role in the growth and proliferation of adrenocor- tical cells. The interplay of growth factors, cytokines, and hormones in adrenal cortical growth and differentiation may also be important in adrenal tumorigenesis once the initial oncogenic event has occurred. The potential role of cytokines in adrenal tumors is also highlighted by a report of a cortisol- secreting adrenocortical adenoma infiltrated with macro- phages and leukocytes containing CD 45 and CD 8 markers (16). This adenoma had high expression of IL-1 receptors,
| Tumor | Type | IL-8 | ENA-78 |
|---|---|---|---|
| RL-251 | ACC | + | + |
| H-295 | ACC | - | + |
| 1 | ACA | - | + |
| 2 | ACA | + | + |
| 3 | ACA | - | + |
| 4 | ACC | + | + |
| 5 | ACC | + | + |
| 6 | ACC | + | + |
| 7 | ACC | + | + |
| 8 | ACC | + | + |
| 9 | ACC | + | - |
| 10 | ACC | - | + |
Chemokine-secreting RL-251 cells and nonchemokine-secreting H-295 cells were used as control. ACA, Adrenal cortical adenoma; ACC, adrenal cortical carcinoma.
and tumor cells in vitro showed a dose-dependent increase in cortisol secretion. This tumor, however, was not shown to produce either IL-1 or other cytokines. Although paraneo- plastic clinical syndromes associated with other tumors have been described as mediated by cytokines, this is the first report of a clinical syndrome associated with overproduction of CXC chemokines.
CXC chemokines have been shown to influence tumori- genesis in other types of neoplasms. IL-8 and ENA-78 have been found to promote tumor growth in non-small-cell lung cancer (NSCLC) (5, 17, 18). Using an NSCLC/SCID mouse chimera developed with two different human NSCLC cell lines, IL-8 production correlated directly with the rate of tumor growth (19). When IL-8 was depleted in vivo by pas- sive immunization with neutralizing antibodies, tumorigen- esis was markedly reduced through a reduction in tumor- associated vascular density and angiogenic activity (6). Similar findings have been shown for ENA-78. These find- ings indicate that IL-8- and ENA-78-mediated angiogenesis is critical to tumorigenesis of human NSCLC. Our studies with passive immunization of the RL-251/SCID mouse chi- mera suggest that this mechanism can now be extended to adrenal cortical carcinoma.
Neutrophil infiltration in association with tumor cells may contribute to the process of tumorigenesis. Interactions be- tween leukocytes and tumor cells can occur at different stages of tumor cell dissemination. For example, lympho- cytes and monocyte/macrophages can infiltrate the tumor mass and modulate tumor cell migration and proliferation by means of the release of cytokines, growth factors, and che- motactic peptides. In turn, tumor cells can modulate their own proliferation through the release of growth and trans- forming factors. (20-22).
There are indications that an inflammatory reaction sur- rounding any tumor may enhance tumor invasion. For ex- ample, infection may cause progression of some forms of adenocarcinoma of the ascending colon (23). Experimentally, tumor cells infected with microbial particles can grow faster, invade surrounding normal tissue, and metastasize more rapidly than noninfected ones (24). Murine fibrosarcoma cells, infected with Candida albicans, produced tumors that appeared sooner and grew more rapidly than tumors pro-
duced by uninfected control cells. In contrast to noninfected tumors, infected tumors were characterized by a marked polymorphonuclear leukocytic response. This leukocytic re- sponse was associated with widespread invasion of normal tissues surrounding the tumor that was not present in the tumors developed from noninfected cells. The infected an- imals developed micrometastases more rapidly throughout the lungs with small clusters of tumor cells visible in both intravascular and extravascular locations.
The presence of large numbers of leukocytes may facilitate tumor spread by degrading the normal tissue surrounding the tumor. A variety of substances originating in the neu- trophil granule are actively secreted by leukocytes in vitro upon stimulation. These substances could dissolve the tumor matrix, help mobilize tumor cells, and facilitate their migra- tion through tissues and endothelium for blood vessel in- vasion (25, 26). The combination of neovascularization and neutrophil-mediated disruption of tumor matrix together with other effects on capillary attachment and penetration may be important steps in tumor growth and metastasis. This process could be repeated in distant sites, further promoting the dissemination of tumor.
The frequency at which chemokine production is a feature of adrenocortical carcinoma remains to be confirmed. Our finding that 30% of archival adrenal carcinomas had inflam- matory cells and many of the adrenocortical tumors had message for IL-8 and ENA-78 by RT-PCR suggests that CXC chemokines may play an important role in adrenal tumori- genesis. The presence of cDNA for IL-8 and ENA-78 in ar- chival adrenocortical tumors does not prove active secretion of CXC chemokines by these tumors. It does, however, sup- port the possibility that these chemokines are secreted and enhance tumor growth by a systemic effect or modulate tumor growth by paracrine effects.
Acknowledgments
Received September 19, 2000. Accepted April 27, 2001.
Address all correspondence and requests for reprints to: David E. Schteingart. M.D., University of Michigan Medical Center, 5570 MSRB II, Box 0678, 1150 West Medical Center Drive, Ann Arbor, Michigan 48109-0001. E-mail: dschtein@umich.edu.
This work was supported in part by NIH Grants NCRR-M-01-RR- 000-42, CA- 87879, and HL-66027 and the Millie Schembechler Adrenal Cancer Research Fund of the University of Michigan Comprehensive Cancer Center.
References
1. Schteingart DE 1992 Treating adrenal cancer. Endocrinologist 2:149-157
2. Arenberg DA, Keane MP, DiGiovine B, et al. 1998 Epithelial-neutrophil activating peptide (ENA-78) is an important angiogenic factor in non-small cell lung cancer. J Clin Invest 102:465-472
3. Gazdar AF, Oie HK, Shakleton CH, et al. 1990 Establishment and charac- terization of a human adrenocortical carcinoma cell line that expresses multiple pathways of steroid biosynthesis. Cancer Res 50:5488-5496
4. Skehan P, Storeng R, Scudiero D, et al. 1990 New colorimetric cytotoxicity assay for anticancer drug screening. J Natl Cancer Inst 82:1107-1112
5. Koch AE, Kunkel SL, Harlow LA, et al. 1994 Epithelial neutrophil activating peptide-78: a novel chemotactic cytokine for neutrophils in arthritis. J Clin Invest 94:1012-1018
6. Arenberg DA, Kunkel SL, Polverini PJ, et al. 1996 Inhibition of interleukin-8 reduces tumorigenesis of human non-small cell lung cancer in SCID mice. J Clin Invest 97:2792-2802
7. Silber CC, Porter RH 1954 The determination of 17,21, dihydroxy 20-keto- steroids in urine and plasma. J Biol Chem 210:923-932
8. Chaney AL 1958 Standard 17 ketosteroids-urine methods of clinical chemistry New York: Academic Press, vol 2:79
9. Baumforth KR, Nelson PN, Digby JE, et al. 1999 Demystified. .. the poly- merase chain reaction. Mol Path 52:1-10
10. Reincke M, Niecke J, Krestin GP, et al. 1992 Preclinical Cushing’s syndrome in adrenal “incidentalomas”: comparison with adrenal Cushing’s syndrome. J Clin Endocrinol Metab 75:826-832
11. Miller MD, Krangel MS 1992 Biology and biochemistry of the chemokines: a family of chemotactic and inflammatory cytokines. Crit Rev Immunol 12:17-46
12. Strieter RM, Polverini PJ, Kunkel SL, et al. 1995 The functional role of the “ELR” motif in CXC chemokine-mediated angiogenesis. J Biol Chem 270: 27348-27357
13. Gonzalez-Hernandez JA, Bornstein SR, Ehrhart-Bornstein M 1994 Macro- phages within the human adrenal gland. Morphological data for a possible local immune-neuroendocrine interaction. Cell Tissue Res 278:201-205
14. Ottaway CA 1991 Vasoactive intestinal peptide and immune function. In: Ader R, Felten DL, Cullen N, eds. Psychoneuroimmunology. San Diego: Academy Press
15. Nathan CF 1987 Secondary products of macrophages. J Clin Invest 79:319-326
16. Willenberg HS, Stratakis CA, Marx C, et al. 1998 Aberrant interleukin-1 receptors in a cortisol-secreting adrenal adenoma causing Cushing’s syn- drome. N Engl J Med 339:27-31
17. Strieter RM, Polverini PJ, Arenberg DA, et al. 1995 Role of CXC chemokines as regulators of angiogenesis in lung cancer. J Leukoc Biol 57:752-762
18. Koch AE, Polverini PJ, Kunkel SL, et al. 1992 Interleukin-8 as a macrophage- derived mediator of angiogenesis. Science 258:1798-1801
19. Arenberg DA, Polverini PJ, Kunkel SL, et al. 1997 The role of CXC chemo- kines in the regulation of angiogenesis in non-small cell lung cancer. J Leukoc Biol 62:554-562
20. Goustin AS, Leof EB, Shipley GD, Moses HL 1986 Growth factors and cancer. Cancer Res 46:1015-1029
21. Heldin CH, Westermark B 1989 Growth factors as transforming proteins. Eur J Biochem 184:487-496
22. Lyons RM, Moses HL 1990 Transforming growth factors and the regulation of cell proliferation. Eur J Biochem 187:467-473
23. Parsonnet J 1995 Bacterial infection as a source of cancer. Environ Health Perspect Suppl 8:263-268
24. Ginsburg I, Fligiel SEG, Kunkel RG, et al. 1987 Phagocytosis of Candida albicans enhances malignant behavior of murine tumor cells. Science 238:1573- 1575
25. Snyderman R, Pike MC 1984 Transductional mechanisms of chemoattractant receptors on leukocytes. Contemp Top Immunobiol 14:1-28
26. Sklar LA, Jesaitis AJ, Painter RG 1984 The neutrophil N-formyl peptide receptor: dynamics of ligand-receptor interactions and their relationship to cellular responses. Contemp Topics Immunobiol 14:29-82