618
EFFECT OF PROPHYLACTIC COLCHICINE THERAPY ON LEUKOCYTE FUNCTION IN PATIENTS WITH FAMILIAL MEDITERRANEAN FEVER CHARLES A. DINARELLO, MICHAEL J. CHUSID, ANTHONY S. FAUCI, JOHN I. GALLIN, DAVID C. DALE, and SHELDON M. WOLFF Patients with familial Mediterranean fever (FMF) who were part of a double-blind trial of daily colchicine as prophylaxis for their disease had leukocyte functions studied while receiving colchicine or placebo. Leukocytes taken from these patients while on prophylactic doses of colchicine produced normal quantities of leukocytic pyrogent ingested bacteria normally, and migrated normally in chemotactic chambers. In addition these patients had normal numbers of circulating T and B lymphocytes as well as normal blastogenic responses of their peripheral lymphocytes to mitogenic stimuli. The patients on colchicine, however, had significantly fewer neutrophils and monocytes accumulating at skin-window sites 24 hours after the initial abrasion. Because the early phase of the skinwindow response was normal in these patients, the decreased late response may be related to a failure to amplify the initial inflammatory reaction. The reduced capacity
to generate a normal inflammatory response may account for the failure of these patients to develop full attacks while taking colchicine.
From the Laboratory of Clinical Investigation, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland. Charles A. Dinarello, M.D.: Senior Investigator, Laboratory of Clinical Investigation (LCI), National Institute of Allergy and Infectious Diseases (NIAID), National Institutes of Health (NIH); Michael J . Chusid, M.D.: Clinical Associate, LCI, NIAID. NIH; Anthony S . Fauci, M.D.: Head, Clinical Physiology Section, LCI, NIAID, NIH; John I. Gallin, M.D.: Senior Investigator, LCI, NIAID, NIH; David C. Dale. M.D.: Senior Investigator, LCI. NIAID. N I H (presently Associate Professor of Medicine, University of Washington, School of Medicine, Seattle, Washington); Sheldon M. Wolff, M.D.: Clinical Director and Chief, LCI, NIAID, NIH. Address reprint requests to Charles A . Dinarello. M.D.. NIH, Building 10, Room IIBI5, Bethesda, Maryland 20014. Submitted for publication August I I , 1975; accepted October 30. 1975.
MATERIALS AND METHODS
Arthritis and Rheumatism, Vol. 19, No. 3 (May-June 1976)
Daily colchicine has recently been shown to be highly effective in preventing the attacks of familial Mediterranean fever (FMF) in patients with this disorder (l,2). Although there is evidence that colchicine interferes with cell migration and the metabolic processes of phagocytosis when the drug is added in vitro to human leukocytes (3-6), the mechanism of colchicine’s action in F M F remains unknown. During a doubleblind trial using colchicine (0.6-1.8 mg/day) in patients with FMF, the patients’ leukocytes were studied while they were taking either colchicine or placebo to determine any functional change induced by the drug in this disease.
Production of Leukocytic Pyrogen. Peripheral leukocytes separated by dextran sedimentation of whole blood were incubated with heat-killed Staphylococcus albus (30 bacteria: I leukocyte), Hank’s balanced salt solution (HBSS), and 10% normal human AB serum. After phagocytosis the leukocytes were resuspended in HBSS and incubated overnight at 37°C. Supernatants from these overnight incubations were injected intravenously into rabbits and rectal temperatures were recorded every 10 minutes (7). The peak rise in rectal temperature above baseline was recorded and used for statistical analyses. Phagocytic Uptake of “C-labeled Staphylococcus Aureus. “C-labeled heat-killed S aureus was incubated with
6 19
peripheral neutrophils from patients or controls in a ratio of 5-7 polymorphonuclear leukocytes (PMNs) for each bacterium. Phagocytosis occurred in the presence of human AB serum (10%) and HBSS on a rotating rocker at 37°C. Iced HBSS was added after 20 minutes to inhibit phagocytosis. Centrifugation of the incubation mixture at I 0 0 g and subsequent washings separated nonphagocytized bacteria from leukocytes and ingested bacteria. All experiments were done in triplicate. A previous study demonstrated that the amount of radioactivity present in the washed leukocyte pellet was indicative of the degree of phagocytosis as estimated by microscopic assay (8).
Lymphocyte Studies. Lymphocytes were separated from heparinized blood by the Hypaque-Ficoll method (9). The proportion of thymus-derived (T) lymphocytes was determined by the sheep red blood cell rosetting technique (10) whereas the proportion of bone-marrow-derived (B) lymphocytes was determined by the complement receptor (EAC) rosette method ( 1 1). The in vitro lymphocyte blastogenic responses to phytohemagglutinin (PHA), concanavalin A (Con A), and pokeweed mitogen (PWM) were studied using incorporation of tritiated thymidine (12). Neutrophil Migration in Vitro. Directed migration and random migration of neutrophils were studied by the T r labeled leukocyte double micropore filter technique (13). For this technique leukocytes are labeled with "Cr and placed in a modified Boyden chamber containing two 5-pm micropore filters (Millipore Corporation, Bedford, Massachusetts). After 3 hours of incubation at 37°C in 5% CO, and 100% humidity, the neutrophils traversed the upper and migrated into the lower micropore filter. After adjusting for the specific activity of the and for variability of the 61Crincorporation by the leukocytes, chemotaxis was expressed as corrected counts per minute in the lower filter (cor cpm in LF). For chemotactic studies endotoxin-activated whole serum from normal subjects was used as the chemotactic stimulus (14); for random migration studies buffer was used in the lower compartment of the chemotactic chamber (3). Quadruplicate chambers were used for each experimental condition and the means of different experiments were compared using the Student's r test. Cutaneous Inflammatory Responses. An area of skin on the volar surface of a forearm, approximately 3 x 3 mm, was abraded with a scalpel blade. A sterile, glass coverslip was placed on the abraded area and secured with tape. The coverslips were changed 1, 3, 5, 7, 9, 12, and 24 hours after abrasion, air-dried, and stained with Wright's stain. At the time of these tests the investigator was unaware of whether the patients were receiving placebo or colchicine. To measure the response, a 1x1-mm area of the most cellular field of each coverslip was photographed a t 50 X magnification; the number of cells in the field was counted by an examiner unaware of the circumstances of the test. Two hundred cell differential counts were also made for each slide. For 10 normal subjects and I 1 patients on and off colchicine, the geometric mean number of neutrophils and monocytes present at each time point was calculated. The geometric mean was used to decrease the influence of extreme 'observations because of the inherent variability of this test. The mean values were compared using the Student's t test.
RESULTS Leukocyte Pyrogen Production Eight patients taking two or three colchicine (0.6 mg) tablets per day for 2 weeks to 1 year were studied. Leukocytes obtained from these 8 patients while on colchicine produced normal amounts of pyrogen. In 34 rabbit assays the mean ( f S E M ) of the peak fever (rise above baseline temperature) in the colchicine group was 0.64"C f 0.05, and in 50 determinations in the control group the mean was 0.65"C f 0.08. Five patients were also studied while participating in a double-blind trial in which patients served as their own control (1). The mean peak fever from supernates of leukocytes from patients taking colchicine was 0.49"C & 0.06 and from patients taking placebo 0.65"C f 0.05. When a paired t test was used, this difference was not significant (P > 0.5).
Phagocytosis Six patients on colchicine therapy for periods of 2 weeks to over 1 year were studied. As shown in Table 1, phagocytic uptake of labeled S aureus by leukocytes from patients on colchicine Was not different from that of normal controls. The mean (* SEM) percent uptake of S aurew after 20 minutes was 54.3% f 4.0 in patients taking colchicine and 5 1.9% f 2.5 for concurrently studied controls. This difference was not significant.
Lymphocyte Studies The F M F patients receiving colchicine had total lymphocyte counts, proportions, and total numbers of circulating T and B lymphocytes that were no different from controls. The mean (f SEM) in vitro lymphocyte blastogenic responses expressed as difference between stimulated and unstimulated (control) cultures (A cpm) in the colchicine group were 202,442 f 34,747 cpm, 19 1,7 13 f 24,712 cpm, and 94,136 f 17,947 cpm for PHA, Con A, and PWM respectively. The values in the normal control group were 204,013 f 20,764 cpm, 150,127 f 20,118 cpm, and 78,477 f 11,971 cpm for PHA, Con A, and PWM respectively. There were no differences between the two groups with any of the mitogens studied (P > 0.2). In 2 patients who underwent lymphocyte studies both before and 1 to 3 months after initiation of colchicine therapy, there were no significant differences in any of the above lymphocyte studies.
DINARELLO ET AL
620
Neutrophil Migration in Vitro Leukocyte migrations in 9 patients taking colchicine were compared to those of 15 control subjects, including 4 patients taking placebo. As shown in Table 2, after 3 hours of incubation, no cellular abnormality of neutrophil chemotaxis or random migration was detected in cells obtained from F M F patients taking placebo (P > 0.05, t test). Similarly there was no detectable defect in the kinetics of in vitro random or directed neutrophil migration in the chemotactic chambers after 1 or 2 hours.
Cutaneous Inflammatory Responses in Vivo Skin-window responses were studied in 1 1 patients with F M F both on and off colchicine. The patients on colchicine had been on this therapy for at least 1 week. Compared to normal subjects, F M F patients off colchicine had normal skin-window responses (Figure 1). In FMF patients on colchicine the first 9 hours of the response were similar to those in normal controls; however after 9 hours the response was reduced (Figure I ) . The geometric mean number of neutrophils and monocytes at 24 hours was significantly reduced (P < 0.05) when compared to both normals and FMF patients off colchicine (Table 3).
DISCUSSION Attacks of FMF have clinical characteristics very similar to those of an acute inflammatory process, and this fact has been confirmed histologically with tissues removed at the time of the attack (16). Although there has been some speculation that F M F is an inborn error of metabolism (17), the inducing agent or mechanism of the acute attack remains unknown. Nevertheless daily colchicine is highly effective in preventing these attacks and hence may be important in understanding the pathogenesis of this disease. In a previous report the present authors noted that patients with F M F often had premonitory symptoms of an attack, but while the patients were taking daily colchicine these early symptoms rarely progressed into a full attack ( I ) . This finding suggested that the underlying excitatory process in F M F remained intact in these patients but that colchicine prophylaxis abated further development of the attack process. I n this report an attempt was made to correlate the therapeutic efficacy of colchicine in preventing these attacks with changes in host leukocyte function. There is ample evidence that colchicine rapidly binds to the pro-
Table 1. Phagocytosis by Granulocytes from Colchicine-treated Patients Phagocytosis (% uptake ofSaureus) Patient
Patient Uptake
Normal Uptake*
EG
51.8 68.6 58.5 57.8 39.3 49.6 54.3 f 4.0t
48. I 61.1 51.4 51.4 46.6 52.3 51.9 f 2.5t
sw RiS RoS
nw
GS Mean f S E
*Each patient was studied concurrently with a normal control subject. ?Significance level of difference between patients and normals, P > 0.20, paired I test.
tein subunits that form microtubules in cells (18,19). Thus leukocytes were obtained from patients taking therapeutic doses of the drug and then their cells were studied in vitro. Changes in the ability of leukocytes from patients on colchicine either to release leukocytic pyrogen or to ingest staphylococci could not be demonstrated. In addition there was no detectable change in numbers of circulating T and B lymphocytes. It is well known that in vitro colchicine interferes with cell division (20). In the present colchicine-treated patients, the blastogenic responses of lymphocytes to three commonly used mitogens were normal. This result may be related to the fact that the presence of the drug is required in vitro to interfere with the process of division in cell cultures. Despite the fact that the impairment of cell locomotion by colchicine has been repeatedly demonstrated (3,4), in the present study both random and directed migration of neutrophils obtained from patients on colchicine were normal. It should be emphasized that in the previous studies (3,4) this impairment occurred only when there were high concentrations of the drug in the incubating medium. The present study using the skin-window test was
Table 2. Efect of in biuo Colchicine on in Vitro Granulocyte Migration Granulocyte Source Normal ( I I)t FM F Placebo (4) Colchicine (9)
Chemotaxis*
Random Migration*
2288 f 281
223 f 26
2029 f 219 2406 f 28 I
209 f 28 296 f 44
*Mean f SE, cor cpm in L F after 3 hours of chamber incubation. ?Number of different subjects studied in parentheses.
COLCHICINE THERAPY IN FMF
62 1
able to demonstrate altered inflammatory responses in patients with F M F taking colchicine as compared to normal subjects or F M F patients taking a placebo. The abnormal responses were characterized by a reduction in the numbers of both neutrophils and monocytes accumulating at 9 to 24 hours after the initial stimulus. The authors believe that these observations are reliable because these tests were performed by an investigator who was blinded to the patients’ therapy and because cell responses were counted by a technician who was unaware of the circumstances of the tests. Both neutrophils and monocytes migrated in normal numbers during the early phase of the skin-window responses, both on and off colchicine. This finding suggests that the inflammatory stimulus was similar in both groups. It also suggests that reduced leukocyte adherence to the glass coverslips was not an artifactual reason for the results observed, although other investigators have shown that colchicine does reduce leukocyte adherence to glass and nylon (21,22). The abnormal late phase of the skin-window responses with colchicine therapy may be related to an impairment of the generation of certain factors that amplify the acute inflammatory response. Neutrophil immobilizing factor (23), a complement-activating substance (24,25), a neutrophil chemotactic factor (26,27), and a mononuclear cell chemotactic factor (28) have been demonstrated to exist preformed in neutrophils. These and other factors are probably released as a consequence of phagocytosis and degranulation and may play a role in the amplification and localization of the inflammatory process. Because colchicine decreases neutrophil degranulation (29), it may also modify release of cellular factors amplifying the acute inflammatory response. This failure to mobilize cellular factors associated with the amplification of the inflammatory response may account for both the diminished skinwindow response at 24 hours as well as the failure of progression of FM F attacks in patients taking colchicine.
-
NEUTROPHILS
Normal I101 FMF 1111
D----O
FMF on Colchicine 1111
MONOCYTES
4
8
12
16
Neutrophils
On Off On Off
Geometric Mean of Cells Counted
47.0 118.9 157.7 472.0,
24
Fig 1. The effect of daily colchicine on the skin-window response in patients with FMF. The numbers of individuals studied are in parentheses. Asterisked values are those that are significantly different (P < 0.05) from other measurements at the same time interval.
Table 3. Comparison of Skin- Window Responses at 24 Hours in FMF Patients on and offDaily Colchicine Colchicine
20
tiours
Mean Log,,,*
3.9 4.8 5.I 6.2
Standard Error
+0.4 f0.3 M onocytes f0.5 f0.3 *The logarithm (e) is used to reduce the influence of extreme observation of these data (15).
DINARELLO ET AL
622
REFERENCES 1. Dinarello CA, Wolff SM, Goldfinger SE, et al: Colchicine
therapy for familial Mediterranean fever. N Engl J Med 291~934-937, 1974 2. Zemer D, Revach M, Pras M, et al: A controlled trial of colchicine in preventing attacks of familial Mediterranean fever. N Engl J Med 291:932-934, 1974 3. Gallin JI, Rosenthal AS: The regulatory role of divalent cations in human granulocyte chemotaxis. J Cell Biol 62:594-609, 1974 4. Caner JEZ: Colchicine inhibition of chemotactic migration of human polymorphonuclear leukocytes. Arthritis Rheum 7:297-298, 1964 5. Goldfinger SE, Howell RR, Seegmiller JE: Suppression of metabolic accompaniments of phagocytosis by colchicine. Arthritis Rheum 8.1 112-1122, 1965 6. Malawista SE, Bodel P T The dissociation by colchicine of phagocytosis from increased oxygen consumption in human leukocytes. J Clin Invest 46:786-796, 1967 7 . Dinarello CA, Goldin NP, Wolff SM: Demonstration and characterization of two distinct human leukocytic pyrogens. J Exp Med 139:1369-1381, 1974 8. Root RK, Ellman L, Frank MM: Bactericidal and opsonic properties of CCdeficient guinea pig serum. J Immunol 109:477-486, 1972 9 . Boyum A: Isolation of mononuclear cells a n d granulocytes from human blood. Scand J Clin Lab Invest 21:77-88, 1969 (SUPPI97) 10. Jondal M, Holm G , Wigzell H: Surface markers on human T and B lymphocytes. I . A large population of lymphocytes forming nonimmune rosettes with sheep red blood cells. J Exp Med 136:207-215, 1972 I I . Bianco C, Patrick R, Nussenzweig V: A population of lymphocytes bearing a membrane receptor for antigenantibody-complement complexes. I . Separation and characterization. J Exp Med 132:702-720, 1970 12. Fauci AS, Dale DC: The effect of in viuo hydrocortisone on subpopulations of human lymphocytes. J Clin Invest 53~240-246, 1974 13. Gallin JI, Clark RA, Kimball HR: Granulocyte chemotaxis: an improved in vitro assay employing 61Cr-labeled granulocytes. J Immunol 110:233-240, 1973
14. Clark RA, Kimball HR: Defective granulocyte chemotaxis in the Chediak-Higashi syndrome. J Clin Invest 502645-2652, 197 1 15. Dale DC, Fauci AS, Wolff SM: Alternate-day prednisone. N Engl J Med 291:ll54-l158, 1974 16. Brick IB, Cajigas M: Benign paroxysmal peritonitis. N Engl J Med 244:786-790, 1951 17. Sohar E, Gafni J, Pras M , et al: Familial Mediterranean fever: a survey of 470 cases and review of the literature. Am J Med 43:227-253, 1967 18. Wilson L, Bamburg JR, Mizel SB, et al: Interaction of drugs with microtubule proteins. Fed Proc 33: 158-166, I974 19. Borisy G G , Taylor EW: The mechanism of action of colchicine. J Cell Biol 34525-548, 1967 20. Dustin P Jr: New aspects of the pharmacology of antimitotic agents. Pharmacol Rev 15:449-480, 1963 21. MacGregor RR: lnhibition of granulocyte adherence: potential mechanism of action of the anti-inflammatory drugs. Clin Res 22:423A, 1974 22. Penny R, Galton DAG, Scott JT, et al: Studies on neutrophi1 function. I. Physiological and pharmacologic aspects. Br J Haematol 12:623-632, 1966 23. Goetzl EJ, Austen KF: A neutrophil-immobilizing factor derived from human leukocytes. I. Generation and partial characterization. J Exp Med 136:1564-1580, 1972 24. Wright DG, Gallin J1: Generation of chemotactic activity from serum by a product released from human polymorphonuclear leukocytes during phagocytosis. Fed Proc 33:63 I , 1974 25. Goldstein IM, Weissmann G: Generation of C,-derived lysosomal enzyme-releasing activity (C5a) by lysates of leukocyte lysosomes. J Immunol 113:1583-1588, 1974 26. Wright DG, Gallin JI: Modulation of the inflammatory response by products released by human polymorphonuclear leukocytes during phagocytosis: generation and inactivation of the chemotactic factor C5a. Inflammation (in press) 27. Zigmond SH, Hirsch JG: Leukocyte locomotion and chemotaxis. J Exp Med 137:387-410, 1973 28. Ward PA: Chemotaxis of mononuclear cells. J Exp Med 128:1201-1221, 1968 29. Rajan KT: Lysosomes and gout. Nature 210959-960, 1966
EFFECT OF PROPHYLACTIC COLCHICINE THERAPY ON LEUKOCYTE FUNCTION IN PATIENTS WITH FAMILIAL MEDITERRANEAN FEVER CHARLES A. DINARELLO, MICHAEL J. CHUSID, ANTHONY S. FAUCI, JOHN I. GALLIN, DAVID C. DALE, and SHELDON M. WOLFF Patients with familial Mediterranean fever (FMF) who were part of a double-blind trial of daily colchicine as prophylaxis for their disease had leukocyte functions studied while receiving colchicine or placebo. Leukocytes taken from these patients while on prophylactic doses of colchicine produced normal quantities of leukocytic pyrogent ingested bacteria normally, and migrated normally in chemotactic chambers. In addition these patients had normal numbers of circulating T and B lymphocytes as well as normal blastogenic responses of their peripheral lymphocytes to mitogenic stimuli. The patients on colchicine, however, had significantly fewer neutrophils and monocytes accumulating at skin-window sites 24 hours after the initial abrasion. Because the early phase of the skinwindow response was normal in these patients, the decreased late response may be related to a failure to amplify the initial inflammatory reaction. The reduced capacity
to generate a normal inflammatory response may account for the failure of these patients to develop full attacks while taking colchicine.
From the Laboratory of Clinical Investigation, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland. Charles A. Dinarello, M.D.: Senior Investigator, Laboratory of Clinical Investigation (LCI), National Institute of Allergy and Infectious Diseases (NIAID), National Institutes of Health (NIH); Michael J . Chusid, M.D.: Clinical Associate, LCI, NIAID. NIH; Anthony S . Fauci, M.D.: Head, Clinical Physiology Section, LCI, NIAID, NIH; John I. Gallin, M.D.: Senior Investigator, LCI, NIAID, NIH; David C. Dale. M.D.: Senior Investigator, LCI. NIAID. N I H (presently Associate Professor of Medicine, University of Washington, School of Medicine, Seattle, Washington); Sheldon M. Wolff, M.D.: Clinical Director and Chief, LCI, NIAID, NIH. Address reprint requests to Charles A . Dinarello. M.D.. NIH, Building 10, Room IIBI5, Bethesda, Maryland 20014. Submitted for publication August I I , 1975; accepted October 30. 1975.
MATERIALS AND METHODS
Arthritis and Rheumatism, Vol. 19, No. 3 (May-June 1976)
Daily colchicine has recently been shown to be highly effective in preventing the attacks of familial Mediterranean fever (FMF) in patients with this disorder (l,2). Although there is evidence that colchicine interferes with cell migration and the metabolic processes of phagocytosis when the drug is added in vitro to human leukocytes (3-6), the mechanism of colchicine’s action in F M F remains unknown. During a doubleblind trial using colchicine (0.6-1.8 mg/day) in patients with FMF, the patients’ leukocytes were studied while they were taking either colchicine or placebo to determine any functional change induced by the drug in this disease.
Production of Leukocytic Pyrogen. Peripheral leukocytes separated by dextran sedimentation of whole blood were incubated with heat-killed Staphylococcus albus (30 bacteria: I leukocyte), Hank’s balanced salt solution (HBSS), and 10% normal human AB serum. After phagocytosis the leukocytes were resuspended in HBSS and incubated overnight at 37°C. Supernatants from these overnight incubations were injected intravenously into rabbits and rectal temperatures were recorded every 10 minutes (7). The peak rise in rectal temperature above baseline was recorded and used for statistical analyses. Phagocytic Uptake of “C-labeled Staphylococcus Aureus. “C-labeled heat-killed S aureus was incubated with
6 19
peripheral neutrophils from patients or controls in a ratio of 5-7 polymorphonuclear leukocytes (PMNs) for each bacterium. Phagocytosis occurred in the presence of human AB serum (10%) and HBSS on a rotating rocker at 37°C. Iced HBSS was added after 20 minutes to inhibit phagocytosis. Centrifugation of the incubation mixture at I 0 0 g and subsequent washings separated nonphagocytized bacteria from leukocytes and ingested bacteria. All experiments were done in triplicate. A previous study demonstrated that the amount of radioactivity present in the washed leukocyte pellet was indicative of the degree of phagocytosis as estimated by microscopic assay (8).
Lymphocyte Studies. Lymphocytes were separated from heparinized blood by the Hypaque-Ficoll method (9). The proportion of thymus-derived (T) lymphocytes was determined by the sheep red blood cell rosetting technique (10) whereas the proportion of bone-marrow-derived (B) lymphocytes was determined by the complement receptor (EAC) rosette method ( 1 1). The in vitro lymphocyte blastogenic responses to phytohemagglutinin (PHA), concanavalin A (Con A), and pokeweed mitogen (PWM) were studied using incorporation of tritiated thymidine (12). Neutrophil Migration in Vitro. Directed migration and random migration of neutrophils were studied by the T r labeled leukocyte double micropore filter technique (13). For this technique leukocytes are labeled with "Cr and placed in a modified Boyden chamber containing two 5-pm micropore filters (Millipore Corporation, Bedford, Massachusetts). After 3 hours of incubation at 37°C in 5% CO, and 100% humidity, the neutrophils traversed the upper and migrated into the lower micropore filter. After adjusting for the specific activity of the and for variability of the 61Crincorporation by the leukocytes, chemotaxis was expressed as corrected counts per minute in the lower filter (cor cpm in LF). For chemotactic studies endotoxin-activated whole serum from normal subjects was used as the chemotactic stimulus (14); for random migration studies buffer was used in the lower compartment of the chemotactic chamber (3). Quadruplicate chambers were used for each experimental condition and the means of different experiments were compared using the Student's r test. Cutaneous Inflammatory Responses. An area of skin on the volar surface of a forearm, approximately 3 x 3 mm, was abraded with a scalpel blade. A sterile, glass coverslip was placed on the abraded area and secured with tape. The coverslips were changed 1, 3, 5, 7, 9, 12, and 24 hours after abrasion, air-dried, and stained with Wright's stain. At the time of these tests the investigator was unaware of whether the patients were receiving placebo or colchicine. To measure the response, a 1x1-mm area of the most cellular field of each coverslip was photographed a t 50 X magnification; the number of cells in the field was counted by an examiner unaware of the circumstances of the test. Two hundred cell differential counts were also made for each slide. For 10 normal subjects and I 1 patients on and off colchicine, the geometric mean number of neutrophils and monocytes present at each time point was calculated. The geometric mean was used to decrease the influence of extreme 'observations because of the inherent variability of this test. The mean values were compared using the Student's t test.
RESULTS Leukocyte Pyrogen Production Eight patients taking two or three colchicine (0.6 mg) tablets per day for 2 weeks to 1 year were studied. Leukocytes obtained from these 8 patients while on colchicine produced normal amounts of pyrogen. In 34 rabbit assays the mean ( f S E M ) of the peak fever (rise above baseline temperature) in the colchicine group was 0.64"C f 0.05, and in 50 determinations in the control group the mean was 0.65"C f 0.08. Five patients were also studied while participating in a double-blind trial in which patients served as their own control (1). The mean peak fever from supernates of leukocytes from patients taking colchicine was 0.49"C & 0.06 and from patients taking placebo 0.65"C f 0.05. When a paired t test was used, this difference was not significant (P > 0.5).
Phagocytosis Six patients on colchicine therapy for periods of 2 weeks to over 1 year were studied. As shown in Table 1, phagocytic uptake of labeled S aureus by leukocytes from patients on colchicine Was not different from that of normal controls. The mean (* SEM) percent uptake of S aurew after 20 minutes was 54.3% f 4.0 in patients taking colchicine and 5 1.9% f 2.5 for concurrently studied controls. This difference was not significant.
Lymphocyte Studies The F M F patients receiving colchicine had total lymphocyte counts, proportions, and total numbers of circulating T and B lymphocytes that were no different from controls. The mean (f SEM) in vitro lymphocyte blastogenic responses expressed as difference between stimulated and unstimulated (control) cultures (A cpm) in the colchicine group were 202,442 f 34,747 cpm, 19 1,7 13 f 24,712 cpm, and 94,136 f 17,947 cpm for PHA, Con A, and PWM respectively. The values in the normal control group were 204,013 f 20,764 cpm, 150,127 f 20,118 cpm, and 78,477 f 11,971 cpm for PHA, Con A, and PWM respectively. There were no differences between the two groups with any of the mitogens studied (P > 0.2). In 2 patients who underwent lymphocyte studies both before and 1 to 3 months after initiation of colchicine therapy, there were no significant differences in any of the above lymphocyte studies.
DINARELLO ET AL
620
Neutrophil Migration in Vitro Leukocyte migrations in 9 patients taking colchicine were compared to those of 15 control subjects, including 4 patients taking placebo. As shown in Table 2, after 3 hours of incubation, no cellular abnormality of neutrophil chemotaxis or random migration was detected in cells obtained from F M F patients taking placebo (P > 0.05, t test). Similarly there was no detectable defect in the kinetics of in vitro random or directed neutrophil migration in the chemotactic chambers after 1 or 2 hours.
Cutaneous Inflammatory Responses in Vivo Skin-window responses were studied in 1 1 patients with F M F both on and off colchicine. The patients on colchicine had been on this therapy for at least 1 week. Compared to normal subjects, F M F patients off colchicine had normal skin-window responses (Figure 1). In FMF patients on colchicine the first 9 hours of the response were similar to those in normal controls; however after 9 hours the response was reduced (Figure I ) . The geometric mean number of neutrophils and monocytes at 24 hours was significantly reduced (P < 0.05) when compared to both normals and FMF patients off colchicine (Table 3).
DISCUSSION Attacks of FMF have clinical characteristics very similar to those of an acute inflammatory process, and this fact has been confirmed histologically with tissues removed at the time of the attack (16). Although there has been some speculation that F M F is an inborn error of metabolism (17), the inducing agent or mechanism of the acute attack remains unknown. Nevertheless daily colchicine is highly effective in preventing these attacks and hence may be important in understanding the pathogenesis of this disease. In a previous report the present authors noted that patients with F M F often had premonitory symptoms of an attack, but while the patients were taking daily colchicine these early symptoms rarely progressed into a full attack ( I ) . This finding suggested that the underlying excitatory process in F M F remained intact in these patients but that colchicine prophylaxis abated further development of the attack process. I n this report an attempt was made to correlate the therapeutic efficacy of colchicine in preventing these attacks with changes in host leukocyte function. There is ample evidence that colchicine rapidly binds to the pro-
Table 1. Phagocytosis by Granulocytes from Colchicine-treated Patients Phagocytosis (% uptake ofSaureus) Patient
Patient Uptake
Normal Uptake*
EG
51.8 68.6 58.5 57.8 39.3 49.6 54.3 f 4.0t
48. I 61.1 51.4 51.4 46.6 52.3 51.9 f 2.5t
sw RiS RoS
nw
GS Mean f S E
*Each patient was studied concurrently with a normal control subject. ?Significance level of difference between patients and normals, P > 0.20, paired I test.
tein subunits that form microtubules in cells (18,19). Thus leukocytes were obtained from patients taking therapeutic doses of the drug and then their cells were studied in vitro. Changes in the ability of leukocytes from patients on colchicine either to release leukocytic pyrogen or to ingest staphylococci could not be demonstrated. In addition there was no detectable change in numbers of circulating T and B lymphocytes. It is well known that in vitro colchicine interferes with cell division (20). In the present colchicine-treated patients, the blastogenic responses of lymphocytes to three commonly used mitogens were normal. This result may be related to the fact that the presence of the drug is required in vitro to interfere with the process of division in cell cultures. Despite the fact that the impairment of cell locomotion by colchicine has been repeatedly demonstrated (3,4), in the present study both random and directed migration of neutrophils obtained from patients on colchicine were normal. It should be emphasized that in the previous studies (3,4) this impairment occurred only when there were high concentrations of the drug in the incubating medium. The present study using the skin-window test was
Table 2. Efect of in biuo Colchicine on in Vitro Granulocyte Migration Granulocyte Source Normal ( I I)t FM F Placebo (4) Colchicine (9)
Chemotaxis*
Random Migration*
2288 f 281
223 f 26
2029 f 219 2406 f 28 I
209 f 28 296 f 44
*Mean f SE, cor cpm in L F after 3 hours of chamber incubation. ?Number of different subjects studied in parentheses.
COLCHICINE THERAPY IN FMF
62 1
able to demonstrate altered inflammatory responses in patients with F M F taking colchicine as compared to normal subjects or F M F patients taking a placebo. The abnormal responses were characterized by a reduction in the numbers of both neutrophils and monocytes accumulating at 9 to 24 hours after the initial stimulus. The authors believe that these observations are reliable because these tests were performed by an investigator who was blinded to the patients’ therapy and because cell responses were counted by a technician who was unaware of the circumstances of the tests. Both neutrophils and monocytes migrated in normal numbers during the early phase of the skin-window responses, both on and off colchicine. This finding suggests that the inflammatory stimulus was similar in both groups. It also suggests that reduced leukocyte adherence to the glass coverslips was not an artifactual reason for the results observed, although other investigators have shown that colchicine does reduce leukocyte adherence to glass and nylon (21,22). The abnormal late phase of the skin-window responses with colchicine therapy may be related to an impairment of the generation of certain factors that amplify the acute inflammatory response. Neutrophil immobilizing factor (23), a complement-activating substance (24,25), a neutrophil chemotactic factor (26,27), and a mononuclear cell chemotactic factor (28) have been demonstrated to exist preformed in neutrophils. These and other factors are probably released as a consequence of phagocytosis and degranulation and may play a role in the amplification and localization of the inflammatory process. Because colchicine decreases neutrophil degranulation (29), it may also modify release of cellular factors amplifying the acute inflammatory response. This failure to mobilize cellular factors associated with the amplification of the inflammatory response may account for both the diminished skinwindow response at 24 hours as well as the failure of progression of FM F attacks in patients taking colchicine.
-
NEUTROPHILS
Normal I101 FMF 1111
D----O
FMF on Colchicine 1111
MONOCYTES
4
8
12
16
Neutrophils
On Off On Off
Geometric Mean of Cells Counted
47.0 118.9 157.7 472.0,
24
Fig 1. The effect of daily colchicine on the skin-window response in patients with FMF. The numbers of individuals studied are in parentheses. Asterisked values are those that are significantly different (P < 0.05) from other measurements at the same time interval.
Table 3. Comparison of Skin- Window Responses at 24 Hours in FMF Patients on and offDaily Colchicine Colchicine
20
tiours
Mean Log,,,*
3.9 4.8 5.I 6.2
Standard Error
+0.4 f0.3 M onocytes f0.5 f0.3 *The logarithm (e) is used to reduce the influence of extreme observation of these data (15).
DINARELLO ET AL
622
REFERENCES 1. Dinarello CA, Wolff SM, Goldfinger SE, et al: Colchicine
therapy for familial Mediterranean fever. N Engl J Med 291~934-937, 1974 2. Zemer D, Revach M, Pras M, et al: A controlled trial of colchicine in preventing attacks of familial Mediterranean fever. N Engl J Med 291:932-934, 1974 3. Gallin JI, Rosenthal AS: The regulatory role of divalent cations in human granulocyte chemotaxis. J Cell Biol 62:594-609, 1974 4. Caner JEZ: Colchicine inhibition of chemotactic migration of human polymorphonuclear leukocytes. Arthritis Rheum 7:297-298, 1964 5. Goldfinger SE, Howell RR, Seegmiller JE: Suppression of metabolic accompaniments of phagocytosis by colchicine. Arthritis Rheum 8.1 112-1122, 1965 6. Malawista SE, Bodel P T The dissociation by colchicine of phagocytosis from increased oxygen consumption in human leukocytes. J Clin Invest 46:786-796, 1967 7 . Dinarello CA, Goldin NP, Wolff SM: Demonstration and characterization of two distinct human leukocytic pyrogens. J Exp Med 139:1369-1381, 1974 8. Root RK, Ellman L, Frank MM: Bactericidal and opsonic properties of CCdeficient guinea pig serum. J Immunol 109:477-486, 1972 9 . Boyum A: Isolation of mononuclear cells a n d granulocytes from human blood. Scand J Clin Lab Invest 21:77-88, 1969 (SUPPI97) 10. Jondal M, Holm G , Wigzell H: Surface markers on human T and B lymphocytes. I . A large population of lymphocytes forming nonimmune rosettes with sheep red blood cells. J Exp Med 136:207-215, 1972 I I . Bianco C, Patrick R, Nussenzweig V: A population of lymphocytes bearing a membrane receptor for antigenantibody-complement complexes. I . Separation and characterization. J Exp Med 132:702-720, 1970 12. Fauci AS, Dale DC: The effect of in viuo hydrocortisone on subpopulations of human lymphocytes. J Clin Invest 53~240-246, 1974 13. Gallin JI, Clark RA, Kimball HR: Granulocyte chemotaxis: an improved in vitro assay employing 61Cr-labeled granulocytes. J Immunol 110:233-240, 1973
14. Clark RA, Kimball HR: Defective granulocyte chemotaxis in the Chediak-Higashi syndrome. J Clin Invest 502645-2652, 197 1 15. Dale DC, Fauci AS, Wolff SM: Alternate-day prednisone. N Engl J Med 291:ll54-l158, 1974 16. Brick IB, Cajigas M: Benign paroxysmal peritonitis. N Engl J Med 244:786-790, 1951 17. Sohar E, Gafni J, Pras M , et al: Familial Mediterranean fever: a survey of 470 cases and review of the literature. Am J Med 43:227-253, 1967 18. Wilson L, Bamburg JR, Mizel SB, et al: Interaction of drugs with microtubule proteins. Fed Proc 33: 158-166, I974 19. Borisy G G , Taylor EW: The mechanism of action of colchicine. J Cell Biol 34525-548, 1967 20. Dustin P Jr: New aspects of the pharmacology of antimitotic agents. Pharmacol Rev 15:449-480, 1963 21. MacGregor RR: lnhibition of granulocyte adherence: potential mechanism of action of the anti-inflammatory drugs. Clin Res 22:423A, 1974 22. Penny R, Galton DAG, Scott JT, et al: Studies on neutrophi1 function. I. Physiological and pharmacologic aspects. Br J Haematol 12:623-632, 1966 23. Goetzl EJ, Austen KF: A neutrophil-immobilizing factor derived from human leukocytes. I. Generation and partial characterization. J Exp Med 136:1564-1580, 1972 24. Wright DG, Gallin J1: Generation of chemotactic activity from serum by a product released from human polymorphonuclear leukocytes during phagocytosis. Fed Proc 33:63 I , 1974 25. Goldstein IM, Weissmann G: Generation of C,-derived lysosomal enzyme-releasing activity (C5a) by lysates of leukocyte lysosomes. J Immunol 113:1583-1588, 1974 26. Wright DG, Gallin JI: Modulation of the inflammatory response by products released by human polymorphonuclear leukocytes during phagocytosis: generation and inactivation of the chemotactic factor C5a. Inflammation (in press) 27. Zigmond SH, Hirsch JG: Leukocyte locomotion and chemotaxis. J Exp Med 137:387-410, 1973 28. Ward PA: Chemotaxis of mononuclear cells. J Exp Med 128:1201-1221, 1968 29. Rajan KT: Lysosomes and gout. Nature 210959-960, 1966
