- Email: [email protected]
macrophages may contribute to tissue inflammation in AIDS patients
142
/ SECOND
INTERNATIONAL
WORKSHOP
ON
CYTOKINES
415
418
A BIOASSAY FOR HUMAN INTERLEUKIN-4. J. P. Siesel and H. S. Mostowski. CBER, FDA, Bethesda, MD 20892 Most biological assays for human interleukin-4 (IL-4) are based upon its costimulatory effects on tonsillar B cell proliferation. These assays have several drawbacks including: limited availability of tonsils, difficulty of cell preparation, and variability amongst cell preparations. We have developed a bioassay for human IL-4 based upon its ability to upregulate CD23 (low affinity IgE receptor) expression. Ramos, an EBV-transformed B cell line, was subcloned yielding a clone, Ramos-2G6, which is several fold more sensitive to this effect of IL-4. In microtiter plates, Ramos-2G6 cells were cultured for 48 hours in the presence of dilutions of recombinant human IL-4 (Immunex) or samples, stained with murine anti-human CD23 and goat anti-mouse IgG-FITC. IL-4 induced a 3- to E-fold increase (35 to 60 channel shift, 255 channels = 4 logs) in fluorescence intensity as measured by flow cytometry. Significant effects were observed at an IL-4 concentration of 12.5 pgiml and increased with concentrations up to 800 pg/ml. Fluorescence intensity of replicates rarely varied by greater than 2 channels; thus, by comparison with a standard curve, the IL-4 content of unkown specimens was determined with coefficients of variation of approximately 0.1 for samples tested at 100 to 200 pg/ml and approximately 0.3 for samples tested elsewhere in the 12.5 to 800 pg/ml range. Interassay variation and the effects of other cytokines are under investigation.
SECONDARY CYTOKINE PRODUCTION AND CELLULAR INFILTRATE FOLLOW INTERLEUKIN-2 INJECTION INTO CEREBRAL SPINAL FLUID (CSF) M.R. Steuer’@, W.G. Loudon, J.B. Blacklock*, R.P. Moser, J. Carinhas, and E.A. Grimm. *Baylor College of Medicine, and U.T. M.D. Anderson Cancer Center, Houston, TX 77030 The half-life and effect of IL-2 injected into the CSF of patients with metastatic brain tumors was addressed in 8 patients. Leptomeningeal involvement was a prerequisite for inclusion into the protocol, which consisted of bi-weekly, recombiit IL-2 injection into the ventricles at 0.52 x lOE6winjection. All patients exhibited a significiant decrease or total elimination of tumor in the CSF, concomitant with a leukocyte influx. The infiltrate consisted mainly of PMNs for the first 6 hrs, followed by a predominately lymphocytic infiltrate. These results, combined with a delayed (3-4 hrs) toxicity of fever, headache, nausea and vomiting, obtundation, and elevated intmcmnial pressure, prompted us to explore the secondary cytokine content in the CSF. Ventricular CSF was sampled serially to quantify IL-2, TNFa, and IFNSby ELISA and RIA. An IL-2 half-life of S-10 hrs was determined in the CSF of 6/6 patients. TNF was detectable at 2 hrs post IL-2, peaked at 4 hrs, and returned to baseline at 12 hrs. Peak TNF levels (3/4) ranged from 50 to 100 pg/ml; however, l/4 had a level of 1500 pdml. IFNllevels peaked later, between 6-12 hrs, approaching 30 NIH units/ml. These data conclusively demonstrate a secondary cytokine response consisting of TNF and IFN-~occurring in response to IL-2. The source of these cytokines is unknown; however, it appears that peak levels precede and may therefore affect leukocyte influx, suggesting a CNS origin.
416
419
FIBRONECTIN AND THE SPONTANEOUS GROWTH OF SCLERODEEMA LUNG FIBROBIASTS CULTURED FROM BRONCHOALVEOLAR LAVAGE FLUID. R.M. .Silver E.A Smith and E.G. _LeRou. Med Univ of South Carolina, Charleston, SC 294'25. Scleroderma frequently affects the lungs and often results in interstitial pulmonary fibrosis. We sought to determine whether fibroblssts could be cultured from fluid obtained from the the lungs by non-invasive technique of bronchoslveolar lavsge (BAL) and how the capacity of fibroblasts to grow from such cultures related to features of lung inflammation ("alveolitis"). Six normal volunteers and 11 scleroderms patients underwent BAL. Adherent alveolar cells (mostly slveolsr mscrophsges, AM) were tested for fibronectin (FN) release at 24 hr by ELISA and were maintained in culture with DMEM + 10% FBS. Fibroblsst cultures were established from 5/11 (45%) scleroderms patients and O/6 normals, The outgrowth of scleroderms lung fibroblssts was associated with significantly greater slveolitis and FN release by AM (p=.OOl vs normal controls and p-.05 vs scleroderms patients without fibroblsst outgrowth). Preliminary in situ hybridization studies show increased mRNA FN transcripts in AM of scleroderms patients with slveolitis. FN and cytokines produced by scleroderms AM may stimulate and/or recruit lung fibroblssts with a high proliferative capacity.
QUANTITATION OF HUMN INTERLEUKIN-6 IN PERIPHERALBLOODWiONUCLEAR CELL CULTURESUPERNATANTS BY ELISA. J. Warren. T. Hickewicz. and .l. Willibmp Genzyme Corp., Boston, PA 02111 Measurement of human IL-6 by conventional bioassay is cmberscme, and may be subject to artifact due to overlapping biowtivities expressed by other cytokines (i.e. IL-l, IL-4, TNF). Therefore, L solid phase sandwich ELISA was constructed to selectively quantitate human IL-6. The assay provides a useful detection range of -0.4-25 ng human ILdhl. with a minimum detection limit that is similar to that displayed by the WHO international standard IL-6 bioassay (1 unit I -200 pg IL-6/ml). The assay is specific for human IL-6 in that 110 rig/ml of TNF-alpha. IFN-garma, IL-l, 11-2, IL-3, IL-4, G-CSF, W-CSF or GH-CSF are not detected. The ELISA provides results in 24 hours. compared to 3 days for bioassay. The method allows accurate detection of reccmbinant or natural IL-6 in culture media (RPHI-1640 supplemented with 10% FCS) or human serum. as "spiked" samples exhibited < 15% variation from expected valuer. Host importmtly, IL-6 levels produced in by peripheral blood mononuclear cell (PBMC) cultures were easily detected by this assay. IL-6 production by PBMCwas dependent upon stimulus applied and culture duration. PBMCstimulated for 24 hours with 10 or 30 ug/nl Pt!A-p produced approximately 5 or 17 ng IL6/ ml respectively. Addition of 10 rig/ml phorbol myristate acetate to the PHA-p stimulus resulted in 15 fold increase in IL-6 production at 24 hours. Continuation of cultures to 72 hours resulted in 2-3 fold higher IL-6 levels. Parallel resting PBMCcultures did not release detectable levels of IL-6 (< 200 pglml). This ELISA allows for specific quantitation of imwnoreactive IL-6 present in biological samples. thus it can serve as a valuable adjunct to conventional IL-6 bioassays.
417
420
CYTOMEGALOVIRUS (CMV) INDUCTION OF TUMOR NECROSIS FACTOR (TNF) SECRETION BY MONOCYTES/MACROPHAGES MAY CONTRIBUTE TO TISSUE INFLAMMATION IN AIDS PATIENTS. mSmith. CL Lamerson. and SM Wshl. NIH, Bethesda, MD. CMV, an important pathogen in immunocompromised hosts, We investigated whether causes localized tissue pathology. CMV might mediate this pathology by inducing monocyte and tissue mscrophage expression of TNF. CMV was isolated from an AIDS patient with CMV colitis and inoculated into cultures of purified human monocytes. Infection was confirmed by the presence of the 68 kD immediate early antigen. CMV-infected, but not mock-infected, monocytes expressed TNF mRNA by Northern blot analysis. In parallel experiments, CMV-infected monocytes produced several-fold greater amounts of TNF by bioassay than mock-infected monocytes in response to a second stimulus. By in situ hybridization, TNF-specific RNA was shown to be expressed in mucossl macrophsges in colon specimens of AIDS patients with CMV colitis, but not in specimens from healthy subjects or asymptomatic AIDS patients. The mucosal mononuclesr cells that expressed TNF-a contained CMV inclusions. Thus, TNF is expressed in increased amounts in Cm-infected monocytes in vitro and in CMV-positive mucossl cells during CMV enteritis. TNF may contribute to the inflammatory changes that occur in CMV-induced organ disease in AIDS patients and, since TNF upregulates HIV expression in monocytes. CMVinduced TNF secretion may provide a mechanism by which CMV serves as a cofactor in AIDS patients.
TOTAL HUMAN PERITONEAL FLUID (PF) M-CSF LEVELS ARE HIGHER THAN PLASMA LEVELS AND RELATE TO TOTAL PERITONEAL MACROPHAGE CONTENT. J.B. Weinberg, S.
Ramakrishnan, F.J., A.F. Haney, VA and Duke, Durham, NC 27705. Women infertile from various causes have increased numbers of peritoneal macrophages (mat). This increase results from the influx of monocytes into the peritoneum, and/or from local pr’ation of peritoneal mat. The purpose of this study was to determine levels of plasma (Pl) and PF M-CSF. Pl and PF were collected from women undergoing laparoscopy or laparotomy for the evaluation of gynecologic conditions. PF cells were >90% mat. M-CSF was measured using a rabbit anti-human recombinant M-CSF antibody in a radio-immunoassay. Matched sets of Pl and PF from seventy nine (79) women were tested. In 78/79 sets, the PF levels were higher than the corresponding Pl levels, with meat&EM of 0.95M.06 (Pl) and 2.49zhO.13 (PF). Pl and PF M-CSF concentrations were essentially the same in the different groups (normal, uterine fibroids, inactive pelvic inflammatory disease, adhesions, and endometriosis). There was no correlation between M-CSF cone and PF volume, PF cell cone or absolute number, or tubal patency status (open or closed). However, the absolute amount of PF M-CSF (cone X volume) correlated closely with the absolute number of PF cells [total M-CSF range (ng)/total mat (milliotiSEM): 0 to 5/7.8&4.0; 5 to 15/10.6?2.0; 15 to 25115.3f2.2; 25 to 35/23.3f8.7; and >35/72.4&13.2]. M-CSF is known to be synthesized by mat, tibroblasts, endothelial cells, and some tumor cells. This growth factor can stimulate mat growth, and also enhance differentiation of monocytes and mat. Our results suggest that PF M-CSF may be elaborated by the PF mat. This M-CSF could then function as a mitogen for PF mat, and/or mediate the differentiation of newly arrived blood monocytes into the phenotypically different tissue macrophages.
/ SECOND
INTERNATIONAL
WORKSHOP
ON
CYTOKINES
415
418
A BIOASSAY FOR HUMAN INTERLEUKIN-4. J. P. Siesel and H. S. Mostowski. CBER, FDA, Bethesda, MD 20892 Most biological assays for human interleukin-4 (IL-4) are based upon its costimulatory effects on tonsillar B cell proliferation. These assays have several drawbacks including: limited availability of tonsils, difficulty of cell preparation, and variability amongst cell preparations. We have developed a bioassay for human IL-4 based upon its ability to upregulate CD23 (low affinity IgE receptor) expression. Ramos, an EBV-transformed B cell line, was subcloned yielding a clone, Ramos-2G6, which is several fold more sensitive to this effect of IL-4. In microtiter plates, Ramos-2G6 cells were cultured for 48 hours in the presence of dilutions of recombinant human IL-4 (Immunex) or samples, stained with murine anti-human CD23 and goat anti-mouse IgG-FITC. IL-4 induced a 3- to E-fold increase (35 to 60 channel shift, 255 channels = 4 logs) in fluorescence intensity as measured by flow cytometry. Significant effects were observed at an IL-4 concentration of 12.5 pgiml and increased with concentrations up to 800 pg/ml. Fluorescence intensity of replicates rarely varied by greater than 2 channels; thus, by comparison with a standard curve, the IL-4 content of unkown specimens was determined with coefficients of variation of approximately 0.1 for samples tested at 100 to 200 pg/ml and approximately 0.3 for samples tested elsewhere in the 12.5 to 800 pg/ml range. Interassay variation and the effects of other cytokines are under investigation.
SECONDARY CYTOKINE PRODUCTION AND CELLULAR INFILTRATE FOLLOW INTERLEUKIN-2 INJECTION INTO CEREBRAL SPINAL FLUID (CSF) M.R. Steuer’@, W.G. Loudon, J.B. Blacklock*, R.P. Moser, J. Carinhas, and E.A. Grimm. *Baylor College of Medicine, and U.T. M.D. Anderson Cancer Center, Houston, TX 77030 The half-life and effect of IL-2 injected into the CSF of patients with metastatic brain tumors was addressed in 8 patients. Leptomeningeal involvement was a prerequisite for inclusion into the protocol, which consisted of bi-weekly, recombiit IL-2 injection into the ventricles at 0.52 x lOE6winjection. All patients exhibited a significiant decrease or total elimination of tumor in the CSF, concomitant with a leukocyte influx. The infiltrate consisted mainly of PMNs for the first 6 hrs, followed by a predominately lymphocytic infiltrate. These results, combined with a delayed (3-4 hrs) toxicity of fever, headache, nausea and vomiting, obtundation, and elevated intmcmnial pressure, prompted us to explore the secondary cytokine content in the CSF. Ventricular CSF was sampled serially to quantify IL-2, TNFa, and IFNSby ELISA and RIA. An IL-2 half-life of S-10 hrs was determined in the CSF of 6/6 patients. TNF was detectable at 2 hrs post IL-2, peaked at 4 hrs, and returned to baseline at 12 hrs. Peak TNF levels (3/4) ranged from 50 to 100 pg/ml; however, l/4 had a level of 1500 pdml. IFNllevels peaked later, between 6-12 hrs, approaching 30 NIH units/ml. These data conclusively demonstrate a secondary cytokine response consisting of TNF and IFN-~occurring in response to IL-2. The source of these cytokines is unknown; however, it appears that peak levels precede and may therefore affect leukocyte influx, suggesting a CNS origin.
416
419
FIBRONECTIN AND THE SPONTANEOUS GROWTH OF SCLERODEEMA LUNG FIBROBIASTS CULTURED FROM BRONCHOALVEOLAR LAVAGE FLUID. R.M. .Silver E.A Smith and E.G. _LeRou. Med Univ of South Carolina, Charleston, SC 294'25. Scleroderma frequently affects the lungs and often results in interstitial pulmonary fibrosis. We sought to determine whether fibroblssts could be cultured from fluid obtained from the the lungs by non-invasive technique of bronchoslveolar lavsge (BAL) and how the capacity of fibroblasts to grow from such cultures related to features of lung inflammation ("alveolitis"). Six normal volunteers and 11 scleroderms patients underwent BAL. Adherent alveolar cells (mostly slveolsr mscrophsges, AM) were tested for fibronectin (FN) release at 24 hr by ELISA and were maintained in culture with DMEM + 10% FBS. Fibroblsst cultures were established from 5/11 (45%) scleroderms patients and O/6 normals, The outgrowth of scleroderms lung fibroblssts was associated with significantly greater slveolitis and FN release by AM (p=.OOl vs normal controls and p-.05 vs scleroderms patients without fibroblsst outgrowth). Preliminary in situ hybridization studies show increased mRNA FN transcripts in AM of scleroderms patients with slveolitis. FN and cytokines produced by scleroderms AM may stimulate and/or recruit lung fibroblssts with a high proliferative capacity.
QUANTITATION OF HUMN INTERLEUKIN-6 IN PERIPHERALBLOODWiONUCLEAR CELL CULTURESUPERNATANTS BY ELISA. J. Warren. T. Hickewicz. and .l. Willibmp Genzyme Corp., Boston, PA 02111 Measurement of human IL-6 by conventional bioassay is cmberscme, and may be subject to artifact due to overlapping biowtivities expressed by other cytokines (i.e. IL-l, IL-4, TNF). Therefore, L solid phase sandwich ELISA was constructed to selectively quantitate human IL-6. The assay provides a useful detection range of -0.4-25 ng human ILdhl. with a minimum detection limit that is similar to that displayed by the WHO international standard IL-6 bioassay (1 unit I -200 pg IL-6/ml). The assay is specific for human IL-6 in that 110 rig/ml of TNF-alpha. IFN-garma, IL-l, 11-2, IL-3, IL-4, G-CSF, W-CSF or GH-CSF are not detected. The ELISA provides results in 24 hours. compared to 3 days for bioassay. The method allows accurate detection of reccmbinant or natural IL-6 in culture media (RPHI-1640 supplemented with 10% FCS) or human serum. as "spiked" samples exhibited < 15% variation from expected valuer. Host importmtly, IL-6 levels produced in by peripheral blood mononuclear cell (PBMC) cultures were easily detected by this assay. IL-6 production by PBMCwas dependent upon stimulus applied and culture duration. PBMCstimulated for 24 hours with 10 or 30 ug/nl Pt!A-p produced approximately 5 or 17 ng IL6/ ml respectively. Addition of 10 rig/ml phorbol myristate acetate to the PHA-p stimulus resulted in 15 fold increase in IL-6 production at 24 hours. Continuation of cultures to 72 hours resulted in 2-3 fold higher IL-6 levels. Parallel resting PBMCcultures did not release detectable levels of IL-6 (< 200 pglml). This ELISA allows for specific quantitation of imwnoreactive IL-6 present in biological samples. thus it can serve as a valuable adjunct to conventional IL-6 bioassays.
417
420
CYTOMEGALOVIRUS (CMV) INDUCTION OF TUMOR NECROSIS FACTOR (TNF) SECRETION BY MONOCYTES/MACROPHAGES MAY CONTRIBUTE TO TISSUE INFLAMMATION IN AIDS PATIENTS. mSmith. CL Lamerson. and SM Wshl. NIH, Bethesda, MD. CMV, an important pathogen in immunocompromised hosts, We investigated whether causes localized tissue pathology. CMV might mediate this pathology by inducing monocyte and tissue mscrophage expression of TNF. CMV was isolated from an AIDS patient with CMV colitis and inoculated into cultures of purified human monocytes. Infection was confirmed by the presence of the 68 kD immediate early antigen. CMV-infected, but not mock-infected, monocytes expressed TNF mRNA by Northern blot analysis. In parallel experiments, CMV-infected monocytes produced several-fold greater amounts of TNF by bioassay than mock-infected monocytes in response to a second stimulus. By in situ hybridization, TNF-specific RNA was shown to be expressed in mucossl macrophsges in colon specimens of AIDS patients with CMV colitis, but not in specimens from healthy subjects or asymptomatic AIDS patients. The mucosal mononuclesr cells that expressed TNF-a contained CMV inclusions. Thus, TNF is expressed in increased amounts in Cm-infected monocytes in vitro and in CMV-positive mucossl cells during CMV enteritis. TNF may contribute to the inflammatory changes that occur in CMV-induced organ disease in AIDS patients and, since TNF upregulates HIV expression in monocytes. CMVinduced TNF secretion may provide a mechanism by which CMV serves as a cofactor in AIDS patients.
TOTAL HUMAN PERITONEAL FLUID (PF) M-CSF LEVELS ARE HIGHER THAN PLASMA LEVELS AND RELATE TO TOTAL PERITONEAL MACROPHAGE CONTENT. J.B. Weinberg, S.
Ramakrishnan, F.J., A.F. Haney, VA and Duke, Durham, NC 27705. Women infertile from various causes have increased numbers of peritoneal macrophages (mat). This increase results from the influx of monocytes into the peritoneum, and/or from local pr’ation of peritoneal mat. The purpose of this study was to determine levels of plasma (Pl) and PF M-CSF. Pl and PF were collected from women undergoing laparoscopy or laparotomy for the evaluation of gynecologic conditions. PF cells were >90% mat. M-CSF was measured using a rabbit anti-human recombinant M-CSF antibody in a radio-immunoassay. Matched sets of Pl and PF from seventy nine (79) women were tested. In 78/79 sets, the PF levels were higher than the corresponding Pl levels, with meat&EM of 0.95M.06 (Pl) and 2.49zhO.13 (PF). Pl and PF M-CSF concentrations were essentially the same in the different groups (normal, uterine fibroids, inactive pelvic inflammatory disease, adhesions, and endometriosis). There was no correlation between M-CSF cone and PF volume, PF cell cone or absolute number, or tubal patency status (open or closed). However, the absolute amount of PF M-CSF (cone X volume) correlated closely with the absolute number of PF cells [total M-CSF range (ng)/total mat (milliotiSEM): 0 to 5/7.8&4.0; 5 to 15/10.6?2.0; 15 to 25115.3f2.2; 25 to 35/23.3f8.7; and >35/72.4&13.2]. M-CSF is known to be synthesized by mat, tibroblasts, endothelial cells, and some tumor cells. This growth factor can stimulate mat growth, and also enhance differentiation of monocytes and mat. Our results suggest that PF M-CSF may be elaborated by the PF mat. This M-CSF could then function as a mitogen for PF mat, and/or mediate the differentiation of newly arrived blood monocytes into the phenotypically different tissue macrophages.