Human Leukocytic Pyrogen Induces Release of Specific Granule Contents from Human Neutrophils

Human leukocytic pyrogen induces release of specific granule contents from human neutrophils.

M S Klempner, … , C A Dinarello, J I Gallin

J Clin Invest. 1978;61(5):1330-1336. https://doi.org/10.1172/JCI109050.

Research Article

The ability of highly purified human leukocytic pyrogen (LP) to induce neutrophil lysosomal protein release is described. Human peripheral blood neutrophils isolated by Ficoll-Hypaque and dextran sedimentation were exposed to purified human LP. The specific granule-associated proteins, lysozyme and lactoferrin were selectively released, whereas primary granule (beta-glucuronidase) and cytoplasmic (lactic dehydrogenase) enzyme markers were not. Optimum release was observed after 45 min in the presence of Ca++ and Mg++. Cytochalasin B (5 microgram/ml) had no effect on LP-induced lysosomal enzyme release. Since the pyrogenicity of LP is dependent on prostaglandin synthesis, the effect of two potent inhibitors of prostaglandin synthesis on lysozyme release was studied. Both indomethacin and naproxen failed to inhibit specific granule protein release. These observations suggest that the concommitance of fever, elevated serum or urine lysozyme and hypoferremia may, in part, be explained by the interaction of LP and peripheral blood neutrophils.

Find the latest version: https://jci.me/109050/pdf Human Leukocytic Pyrogen Induces Release of Specific Granule Contents from Human Neutrophils

MARK S. KLEMPNER, CHARLES A. DINARELLO, and JOHN I. GALLIN, Laboratory of Clinical Investigation, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland 20014

A B S T R A C T The ability of highly purified human markedly reduced in neutroph-ils obtained from pa- leukocytic pyrogen (LP) to induce neutrophil lysosomal tients with acute febrile infections (4). In addition, the protein release is described. Human peripheral blood constellation of fever, reduced serum iron, hypofer- neutrophils isolated by Ficoll-Hypaquie and dextran remic anemia, and elevated serumil or urinie lysozyme sedimentation were exposed to purified human LP. has beeni demonstrated in several inflammatory dis- The specific granule-associated proteins, lysozyme eases (5-7). These observations suggested to us that and lactoferrin were selectively released, whereas pri- LP, in addition to mediatinig fever, might directly inter- mary granule (,8-glucuronidase) and cytoplasmic (lactic act with the neutrophil and result in release of' neu- dehydrogenase) enzyme markers were not. Optimum trophil granule constituents. release was obsprved after 45 min in the presence of In this report, wve demiionistrate that smiiall amounts Ca++ and Mg++. Cytochalasin B (5 ,ug/ml) had no effect (:10 nM) of highly purified human LP caause selective on LP-induced lysosomal enzyme release. Since the release of specific grantule contents from neutrophils. pyrogenicity of LP is dependent on prostaglandin syn- Degranulation takes place in the absence of phagocyto- thesis, the effect of two potent inhibitors of prostaglan- sis, adherence to suirf;aces, or cytochalasin B. This is the din synthesis on lysozyme release was studied. Both first report of a physiologic substance whielh releases ly- indomethacin and naproxen failed to inhibit specific sosomal contents withouit other cellular maniipuilation granule protein release. These observations suggest and these observations mlay, in part, provi(le the basis that the concommitance of fever, elevated serum or for hypoferremia and elevated seruim or uirine lysozyme urine lysozyme and hypoferremia may, in part, be ex- in several febrile inflammatory dliseases. plained by the interaction of LP and peripheral blood neutrophils. METHODS INTRODUCTION Preparation of neutrophils. Heparinized venous blood was collected from healthy adult volunteers and centrifuged Fever, whether present in inflammatory or infectious on Ficoll-Hypa(Ique gradients followed by dextran sedimentation (8). Erythrocytes were lysed with hypotonic saline and diseases, is mediated by a small, molecular weight the resulting leukocyte suspension *was washed twice in (15,000 daltons) protein called leukocytic pyrogen Hanks' balanced salt solutioni (HBSS, National Instituites of (LP)l (1). In addition to its ability to elevate tempera- Health media unit) containing 1.0 mM CaCI2 and 0.5 mM ture, LP has been shown to acutely reduce serum iron, MgCl2. Leukocytes obtainie(d in this manniier conitaine(d 96- and this hypoferremia is on the 98% neutrophils. dependent presence Preparation of human LP. Human LP was preparecl from of neutrophils because it does not occur in neutropenic monontuclear cells as described previouisly (9). Crtude stuper- animals (2). Lactoferrin, an iron-binding protein found nates containing LP were concentrated in auitoclaved dialysis in neutrophil-specific granules is thought to be the nec- ttubing in front of a high-speed fan. Volumes were reducted essary granulocyte factor responsible for hypoferremia 1/20 and dialyzed against phosphate-buffere(d saline with 0.02% sodium azide. Concenitrated LP was l)laced over anl (3). Another specific granule constituent, lysozyme, is immunoadsorbant column conitaining rabbit anti-humlan LP antibody attached to CNBr-activated Sepharose. LP was eluted from this material in citric acid buffer, pH 3.2 (10), Received for publication 14 September 1977 and in revised neutralized with 0.1 M NaOH, anid stored in 0.02% Na azide. form 16 December 1977. LP eluted from the anti-LP immuntioadsorbanit was concen- ' Abbreviations used in this paper: HBSS, Hanks' balanced trated and chromatographed over Sephadex G-50 (fine) (Phar- salt solution; LP, leukocyte pyrogen; RPD, rabbit pyrogenic macia Fine Chemicals Inc., Piscataway, N. J.) (165 x 5.6 cm) dose. at 4°C in phosphate-buffered saline with 0.02% Na azide. The

1330 The Journal of Clinical Investigation Voluime 61 Aail 19h78 1330-13336 15,000 mol wt LP peak was isolated, dialyzed against H2O, assayed by measuring the consumption of j8-nicotinamide- and lyophilized. This material gave a single staining band adeninle dinucleotide (NADPH) during the coniversioni of on 7.5% polyacrylamide gels in 0.1% sodium dodecyl sulfate. pyruvate to lactate (16). Enzyme activity f'or lysosyme, 8- As described, further purification of human LP was carried glucuronidase and lactic dehydrogenase is expressed as the out using 1251-labeled LP (11, 12). A small (Itiantity of this percent of' total activity in a freeze-thawed lysate of 1.0 x 107 homogeneous radiolabeled LP (Fig. 1) (_ 1,000 epm) was neutrophils in 1.0 ml of' incubation media. Lactoferrin was added to unlabeled LP isolated from Sephadex G-50. This (luantitated by radial immunodiffusion in agarose gels (17). material was chromatographed over G-15 followed by DEAE For these studies, neutrophils were suspended in HBSS at ion-exchange and, in each case, the radioactivity peak was 5 x 107 cells/ml and incubated with LP (10 RPD/ml) as pre- used to indicate purified unlabeled LP. Since biologic activity viously described. After incubation, supernates and freeze- of LP decreased significantly during the ion-exchange step, thawed cell lysates were extracted in 1 M NaCl at 4°C over- 0.1% albumin was added to each collection tube to prevent night (18). The specimens were then centrifuged for 30 min at nonspecific adsorbtion. Recovery of biologic activity was be- 20,000 g 4°C to remove insoluble material. Radial immuno- tween 1 and 2% of the starting concentrated crude LP. diffusion plates were prepared using commercially obtained Rabbit pyrogen assay. Details concerning housing, train- rabbit antisera to human lactof'errin (Behring Diagnostics, ing, and temperature recordings have been reported else- Somerville, N. J.). Purified human colostral lactof'errin where (13). To determine the number of' pyrogenic doses in (Calbiochem, San Diego, Calif:) and neutrophil extracts at preparations of human LP, a two-point dose-responise was various concentrations were used as the immunodiffusion employed using six rabbits as described elsewhere (10, 13). standards. Using these methods, the minimum concentration For the present study, peak fever occurring in rabbits between of detectable lactoferrin was 20,g/ml. Results are expressed as 0.6 and 1.0°C was considered a rabbit pyrogenic dose (RPD) either micrograms equivalent of human colostral lactof'errin or and in a previouis report has been estimated to contain 50 ng the percent of' total lactof'errin released. or less of the purified LP protein (12). Based on the 15,000- Prostaglandin assay. PGF2o was determined by radioim- dalton mol wt of' human LP, 1 RPD/ml is s30 nM. munoassay (kindly performed by Dr. Michael Kaliner) (19). Incubation conditions. Neutrophils were adjusted to 1.0 100-,ul experimental samples were incubated with 6,000 epm x 107 or 5 x 107 cells/ml in HBSS. For some experiments, [3H]PGF2, and 50 ,ul of rabbit anti-prostaglandin in a final the concentrations of CaCl2 and MgCl2 were varied to examine volume of 450 ,ul trizma (0.012%), NaCI (0.083%) and gelatin the divalent cation effect on lysosomal protein release. Neu- (0.1%), pH 7.4 at 4°C for 12-16 h. The bound [3H]PGF2,, was trophil suspensions were then incubate'd for 45 min at 37°C separated from unbound [3H]prostaglandin by adding 1.0 ml in a shaking water bath in media alone or in the presence of iced charcoal (0.25%) dextran (0.025%) in Tris-NaCl buffer of LP. LP was dialyzed against HBSS or H2O to remove sodium and incubating at 4°C for 20 min. After centrifugation (200 g azide. For inhibitor or enhancement experiments, cells were for 10 min), the supernate was placed into scintillation vials preincubated with the pharmacologic agent or media alone with 10 ml Aquasol (American Cyanamid Co. Pearl River, followed by the usual 45-min incubation period. After incuba- N. Y.) and the radioactivity was determined in an LS-350 tion, cells were centrifuged at 400 g for 10 min at 4°C and (Beckman Instruments, Inc., Fullerton, Calif.). The results the supernates were collected. are expressed as micrograms per 107 neutrophils. Enzyme determinations. Lysozyme was determined by Statistics. Student's t test was used to compare the means measuring the rate of lysis of Micrococcus lysodeikticus and standard errors for significance except where otherwise (Worthington Biochemical Corp., Freehold, N. J.) at pH 6.2 noted. according to a turbidometric method (14). 6-Glucuronidase was assayed by measuring the release of phenolphthalein from its f-glucuronate (Sigma Chemical Co., St. Louis, Mo.) after RESULTS 6 h of incubation at pH 4.5 (15). Lactic dehydrogenase was Enzyme release. Semipurified preparations of human LP which had been eluted from anti-LP immuno- 7 adsorbants followed by gel filtration on Sephadex G-50 contained no detectable lysozyme, f3-glucuronidase, 6 8 lactic dehydrogenase, or lactoferrin. Over a wide concentration range, LP caused selective extracellular re- 5 7 lease of lysozyme with little or no release of,8-glucuron- 4 ~~~~~~~~~~~~~~~6idase (Fig. 2). Furthermore, there was consistently <4% of total lactic dehydrogenase activity in the incu- ~335 bation media, indicating that lysozyme release was not a result of cell death. Significant release of lysozyme 2 4 was observed at LP concentrations of 1 RPD/ml. Maxi- mal enzyme release occurred by 20 RPD/ml and higher 1 ~~~~~~~~~~~~~~~3LP concentrations did not result in additional lysozyme release. By light microscopy, neutrophils exposed to 10 FRACTIONS RPD/ml, which had released _ 20% of total lysozyme, did not appear vacuolated and there was no gross FIGURE 1 Isoelectric focusing of '25I-labeled human LP. 1% change in cytoplasmic granularity. Heating these prep- ampholine (3-10) and 4.5% acrylamide. Gel lengths were 2.5 x 100 mm. Constant voltage, 200 V, was applied for 18 h at arations of LP at 80°C for 15 min or precipitation in 4°C. Gels were cut every 2 mm and the radioactivity and pH the presence of NaCl during lyophilization, procedures were measured on each fraction. which destroy the pyrogenicity of human LP, destroyed

Pyrogen-Induced Neutrophil Degranulation 1331 30

LLIw (nwi w a: z

0 LLJ z w 0 w

0 0 0.5 1.0 5.0 10.0 20.0 30.0 HEATED LP NaCI 10 RPD/ml PPT 10 RPD/ml

LP CONCENTRATION (RPD/ml)

FIGURE 2 Dose-response curve of LP-induccd enzynei rleas. ReHsults arc the mcean -SEN of three separate experimenits.

its ability to release lvsozyme. Fuirthermiiore, large pv ro- between 5 ani( 1.5 min and <5% additional enzyme genic doses of endotoxin (300 gg/ml) (E.scherichia coli Tas releasecl over the next 30 mini of incuatioll. endotoxin, Difco Laboratories, Detroit, Mich.) ha(l I10 Lactoferrin release is shown in Table I. A freeze- effect on lysosomal release utider these conditions. thwved lvsate of .5 x tO( p)olymorplonticlear leukocvtes. Therefore, the ability of these semiptirified prepara- extracted as dtescrihed, conitainied 276 _ 14 ,tg of lacto- tions of human LP to release lvsozvmI1re wvas not dute to fe rrini. WNheni neutrol)hils were incubated with LI) at 10 contamination hv endotoxin. In addition, preparations of LP were negative in the Lirnulwis amehocvte lvsate test (20). 30r- To ascertain that the LP molecule was responisible for releasing lysozyme in the above experimiients, fuir- ther purification methods were employed. 500 RPD were purified using immuinoadsorbance, gel filtration over G-50, G-15, and DEAE ion-exchange as (lescribedl 0 in Methods. Humilan serulm alblumini (0.1'7%) was acddedl C,) to the collection tules f'rom01 the final ion-exchange LJ 20k column to protect the molecule againist nonispecific LLJ losses. Despite stuch measures, only five RPI) of puri- ULJ fied LP were recovered. \Vhen this preparationi was divided into two samples and incuhated with neuitro- 0 phils, the quantity of lysosyme releatse was 16.54()0.2%c -J -J which is comparable to the amiiotunit of lysozyvmle released Iv 2.5 RPD/ml (Fig. 2). Htumiiani seruim albumin, 0 10 adjusted to the samiie proteini coniicentrationi (at 280 nm-1) served as a control and did not release lvsozvmle. Be- cauise of the low recovery (1%) of pvrogeniic (loses eni- couintered (luring purificationi of LP, subsequent experiments were carriedl out uisinlg semilipturifiecl )rel)arations. The kinetics of LP-indcted lvsozvlme release in tw(o 0 10 20 30 40 50 separate experiments are shownl in Fig. 3. By 1 min, TIME (min) 9.8+ 1.0% of total lysozymile was releasedl. Neutrophils incubated in buffer released <3%' of lysozvme at this Fic;cuRn; 3 Kinietics of LP-induced Ivysozvnc release. Restilts time (not shoown). A 50%' maximal response was reaclhecl are the mean + SEMt of two separate expesrients.

1332 M\. S. Klempner. C. A. Dinarcllo, au(II 1. 1. G"rallicc TABLE I sponse was observed. Nevertheless, enzyme release LP-Intwiuced Netitroph il Lactoferri n Release wvith either divalent cation alone was not simply addi- tive and the two cations together act synergistically Preparationi Lactoferrin Total released (four sample paired t test P < 0.05). 8-Glucuronidase Uglml % release remained <8% in all these experiments (data Whole cell l\sate* not shown). Exp 1 263 Failure to effect LP-induced enzyme release by indo- Exp 2 290 methacitn, naproxent, or cytochalasitn B. LP is thought NIedia from neuitrophiils to produce fever by inducing local prostaglandin incuibated in: syn- Btuffer thesis in the hypothalmus (23). WAe therefore investi- Exp 1 <20 0 gated the effect of two potent prostaglandin inhibitors Exp) 2 <20 0 on enzyme release induced by LP. Neutrophils were LP preincubated for 15 min with various concentrations Exp 1 62 23.5 of indomethacin (1-100 ,uM) or naproxen at 100 juM. Exp 2 105 36.2 After preincubation, LP was added at 5 or 20 RPD/ml and the incubation continued for an additional 45 min. * Neitrophils were at .5 x 107 cells/mi. As shown in Fig. 4, there was no significant effect on enzyme release by either agent. Even on the steep RPD/ml, 29.8+6.3% of the total neutrophil lactoferrin portion of the dose-response curve (5 RPD/ml) no in- contenit wvas release(I into the media. Nledia froimi cells hibitorv effect could be demonstrated. To insure that incubated in buffer alone cointainied no detectable lac- these agents inhibited prostaglandin synthesis, netutro- toferrini. LP-indluced lactoferrini release wvas unaffected phils (1 x 107/ml in HBSS) were preincubated with in- by the presenice of normal AB serumll in the incubation domethacin (1 or 100 ,uM) or buffer at 37°C for 15 min. mixture. Arachadonic acid (10 ,ug/ml, Sigma Chemical Co.) was Divalenit catiotn eflect otn LP-iniduced enzynme release. then added and the incubation continued for 15 min. Several investigators have demiionistrated the divalent Supernates were collected and assayed for prostaglan- cation re(luiremiieint for optimal lysosomal enzyme release induced by phagocytic and pharmocologic stim- uli (21, 22). Therefore, we studied the ability of LP to induce enzyme release in the absence of Ca, Mg, or both. Results are shown in Table II. LP, in the absence of either Ca or Mg, caused significant lysozyme release T when compared to the buffer. Maximum release was observed when Ca and Mg were present. W7ith either cnC,J Ca or Mg alonie, an intermediate yet significant re- w w-j TABLE II N Effect of Calcitm and Magniesium on LP-Induced 0 Lilsozi nme Release

(Cation 0

LP Ca' Nig+ Lvosozme P* z u, ,W w 5. RPDItIl 1. 11 ('/% release Q. 0 ) 0) 0 ) 5.5±0.8 (5)4 + 0 0 11.8±0.8 (5) <0.001 + 1.50) 14.6±1.0 (2) <0.01 + 2.5 0 14.6±0.3 (1) <0.02 + 0 1.( 13.8±0.5 (2) <0.01 + 0 2.5 14.7±)0.1 (1) <0.02 + 1.5 1.0 21.9± 1.2 (5) <0.001 MOLJiI -_ _J L _ LP INDOMETHACIN NAPROXEN I,LM IO,fM IOO,sM IOOiM * Significance level of differenee compared to percent enzyme release in the absence of' LP, Ca', and Mg++. LP 5 RPD/ml 4 Mean±SEN1 perceent lysozyme release (NIethods). Number FIGURE 4 Effect of' indomethacin and naproxen on LP- of different experiments in parenitheses. For one experiment, induced lysozyme release. Results are the mean±SEM of mean ±SEM of' triplicate samples are shown. three separate experiments.

Pyrogei-Itnduced Neutrophil Degrantulation 1333 din F2,. Neutrophils preincubated in buffer and fol- zyme and lactoferrin in the absence of'/3-glucuronidase lowed by arachadonic acid, produced 193.9+24.1 ug provides strong evidence that the enzyme-mobilizing PGF2,,Q/107 neutrophils. Cells exposed to indomethacin effect of' LP is primarily on specific granules (33). Fur- at 100 and I ,uM before addition of arachadonic acid thermore, assaying for 8-glucuronidase during our ki- produced 20.1±2.8 and 54.2±13.0()tg/107 neutrophils, netic studies failed to show significant release of' this respectively. einzyme even at early time points (not shown). Because Cytochalasini B has been shown by several inlvesti- recovery of LP from the purification procedure is low gators to f;acilitate neutrophil enzyme release when (1%), it was not possible to assess the effect of higher cells are exposed to a variety of soluble and particulate concentrations on enzyme release. As is the case with stimulanits (24). To determiinie if cytochalasin B caused a ionophore A23187, it is possible that the selectivity for similar enhancement of enzyme release when LP was release of' specific granule contents might be abolished the stimulus, neutrophils vere preincubated vith cyto- at greater LP concentrations (32). The lack of an in- chalasin B (5 ,g/ml) for 10 min. After preincubation hibitory effect of' indomethacin and naproxen suggests LP (5 RPD/ml) was added and the incubation contin- that the LP-induced release is independent of de novlo ued for 45 mill In three separate experiments, cells prostaglandini synithesis. in buffer or cytochalasini B without LP released 8.2 Our in vitro observationi that LP is a sufficient stimu- ±1.4 and 7.3±0.6% total lysozymlle, respectively (P lus to induce neutrophil specific granule protein re- > 0.05). Similarly, in the presenice of' LP there was no lease when neuitrophils are in suspension may have enihancement of lysozyme release witlh cytochalasin B direct cliniical relevance. Acute hypoferremia has been (20.1±1.2% for LP in buffer vs. 21.2±+1.4% for LP after slhown to oceur in both man and animals after the cytochalasin B, P > 0.05). ,-Gllucuronidase release was injection of bacterial endotoxin and to accompany a vari- also unlaffected I)v cytochalasinl B (not showni). ety of inflammatory diseases (34, 35, 6). Even apyro- genic doses of endotoxin are able to produce hypofer- DISCUSSION remila in man, and the lowering of serum iron corresponds to accumulation of iron in reticuloendothelial The results of this study demonstrate that a biologically cells (6). Similarly, intravenous injection of LP also produced human protein, LP, is capable of induciing results in acute hypoferremia (36). Moreover, LP- lysosomal protein release in the absence of phagocyto- induced hypoferremia is dependent on neutrophils sis, adherence, or cytochalasin B treatment. Other na- since it does not occur in neutropenic animals (2) and naturally occurring serum substances which induce the factor in neutrophils responsible for lowering of degranulation, such as chemotactic f;ators, all re(quire serumii iron has been shown to be lactoferrin (3). This perturbation of' the neutrophil membrane with cyto- glycoprotein, which exists in the neutrophil in the iron- chalasin B or adherence to a filter to result in secretion free state (apolactoferrin), avidly binds serum iron, (25, 26). Thus, LP represents a naturally occurring lowers serum iron in a dose-dependent fashion, and af- small molecular weight protein which can initiate ter the injection of 59Fe has been shown to accumulate release of specific granule contenits fromii cells in sus- in the satuirated state (59Fe-lactoferrin) in the reticulo- pension without other perturbation of the cell endothelial systeimi (3). membrane. Our observations provide a possible explanation for The selectivity for the release of' specific granule the intermiiediate step of' apolactoferrin release from contents is in keeping with several previous observa- neutroplhils in inflamnmatory states. In addition, since tions that these proteins are more accessible for extra- apolactoferrin is capable of removing Fe from trans- cellular release. Studies by Wright and Malawista (27) f'errin-Fe2 unider conditionis found in inflammatory sites have shown that under coniditions of phagocytosis, lyso- (e.g. acidic pH) and Fe-lactoferrin is preferentially zyme release was easily demonstrated, whereas certain cleared from the circulation (3), the equilibrium reaction: acid hydrolases associated with peroxidase-positive granules were only minimally released. Lefell and Transferrin-Fe., + apolactoferrin Spitznagel (28) have made similar observations when i apotransferrini + lactoferrin-Fe2, the degranulating stimulus was immune complex. More recently, selective mobilization of' specific gran- is constantly shifted to the right. The clinical observa- ule associated enzymes has been shown with the phar- tion that the hypof'erremic hypochromic anemia macologic agents coneanavalin A, phorbol myristate accompanying inflammation is generally unresponsive acetate, and ionophore A23187 (29-32, 22). In this re- to Fe therapy is not surprising since even small gard, it is of particular interest that LP follows a similar amounts of released apolactoferrin would favor pattern of selective release of specific granule contents accumulation of' Fe-lactoferrin in the reticuloendothe- including a re(quirement for calcium and magnesium lial systemii and thereby reduce the amount of for optimal release. The extracellular release of lyso- iron available for hematopoiesis. Whether other sub-

1334 M. S. Klempner, C. A. Diniarello, ald J. I. Gallint stances accompanying acute inflammation are also able tents which may be important in a number of inflamma- to induce neutrophil apolactoferrin release remains to tory diseases provides new areas for assessing the clin- be determined. ical imlportance of both LP and neutrophil secretory The kinetic factors responsible for elevated serum products in the pathophysiology of inflammation. lysozyme in inflammatory and myeloproliferative disorders have been extensively investigated. It is well REFERENCES established that renal clearance accounts for the major- ity of lysozyme elimination. In anephric patients, for 1. Dinarello, C. A., and S. XI. WZolff. 1976. Exogenous and endogenous pyrogens. In Braiin Dysfunction in Infantile example, lysozyme clearance is only 15% of the rate Febrile Convulsions. M. A. Brazier and F. Coceni, editors. found in normals (37). In regard to production and re- Raven Press, New York. 117-128. lease, txvo major f:actors have been defined that relate 2. Kampschmidt, R. F., and H. F. Upchurch. 1969. Lowering to serum lysozyme: (a) the syinthesis and release of of plasma iroin concenitration in the rat with leukocyte of the extracts. A m. J. Ph,ysiol. 217: 1287-1291. lysozyme from leuikocytes and (b) the total size 3. Vlani Snick, J. L., P. L. Masson, and J. F. Heremans. 1974. granulocyte and monocyte pools. In studies by Bodel The involvement of lactoferrini in the hyposideremia of et al. (38), the lysozyme secretory rate by monocytes acute inflaimmniationi.J. Exp). Aled. 140: 1068-1084. from patients with sareordosis has been shown to be 4. Haniseni, N. E., and V. Andersen. 1973. Lysozyme activity significantly greater than controls and to directly cor- in hutimnan neutrophilic granuilocytes. Br. J. Haetmiatol. 24: 613-623. relate with serumii lysozyme levels. Furthermore, these 5. Perillie, P. E., K. Khani, and S. C. Finch. 1973. Serumn moniocytes produce excessive LP in response to endo- lysozyme in pulmoniary tulbercuLlosis. Am. J. Med. Sci. 265: toxin. Hansen and Anderseni (4) have demonstrated that 297-302. neutrophils isolated fromii patienits with acute bacterial 6. Cartwright, G. E. 1966. The anemiiia of chronic disorders. conitrol Semini. Hemnlatol. 3: 351-375. infections contained 50% less lysozyme than 7. Paseule, R. S., J. B. L. Gee, anid S. C. Finch. 1973. Usefull- neutrophils. To account for this finding, these authors ness of sertum lysozyme measuremenit in diagnosis and suggested that bacterial infection suppressed neutro- evaluation of sarcoidosis. N. Engl. J. Med. 287: 1074- phil lysozyme synthesis. However, in preliminary stud- 1076. ies, we have observed that after a single intraveniouis 8. Bdvyuni, A. 1968. Isolationi of mononuclear cells and gran- ulocvtes f'rom humani blood. Isolation of' monionuclear injectioni of bacterial enidotoxin into two normal vol un- cells bv one centrif'ugationi and of granulocytes by com- teers, mature neutrophils isolated from peripheral bininig (entrifugation and sedimentation at 1 g. Scand. blood contained significantly less total lysozyme at 4 h J. Clini. Lab. Intvest. Stl)1)1. 97: 77-89. than at 0 or 24 h (27.6+(0.2, 23.6+0.4, and 29.2+0.5 9. Dinarello, C. A., N. P. Goldin, and S. NI. W7olff. 1974. 4, and 24 h, Demonistrationi and characterization of two distinct hunia gg/5 x 106 neutrophils at 0, respectively, leuikocytic pyrogens. j. Ex1). Med. 139: 1369-1381. P < 0.02 comparing the 4-h lysozyme content to 0 and 10. Dinairello, C. A., L. Renfer, and S. M. WN'olff. 1977. Produc- 24 h). Thus, our in vitro and in vivo data suggest that tion of antibody againist humani leuikocytic pyrogen. J. the reduced neutrophil lysozymie content may reflect, Clitc.It vest. 60: 465-472. at least in part, netutrophil secretion in response to py- 11. Dinairello, C. A., and S. NM. WVolff. 1977. Partial ptirification of' human leuikocytic pyrogen. Inflammnation. 2: 179-189. rogen. 12. Dinarello, C. A., L. Renfer, and S. M. W\olf'. 1977. Hti- In myeloproliferative disorders, the size of the gran- miian letikocvtic pyrogen: p)urification anicl development of uilocyte andI monocvte pools has been repeatedly a ra(lioinimunoassav. Proc. Natl. Acad. Sci. U. S. A. 74: shown to correlate with serumil lysozyme (39). In both 4624-4627. in 13. WN'olf'f, S. I., J. H. Mlulholland, and S. B. \Ward. 1965. man and animals, the nadir of grainulocytes drug- Quantitative aspects of'the pyrogenic response of' rabbits induced leukopeniia corresponds to the nadir of serumn to endotoxini.J. Lab. Clitn. Aled. 65: 268-276. lysozyme. Similarly, the highest serum lysozyme levels 14. Litwack, G. 1955. Photometric determination of lysozyme are observed during marrow hyperplasia with in- activity. Proc. Soc. Exp). Biol. Med. 89: 401-403. creased aind moniocyte pools. It is interest- 15. Talalay, P., WV. H. Fishman, and C. Htiggins. 1946. Chro- granuilocyte magenic substrates. I1. Pheniolphithaleini glutcuronic acidl ing to note that LP has kAn effect on granulocyte turn- as substrate for the assay of glucuronidase activity.J. Biol. over. Kampschimiidt and Upchutrch (40) have shown that C/lienz. 166: 757-772. repeated inijectionis of LP results in progressive in- 16. Bergmever, HI. U., e(litor. 1963. In NMethods in enzvymatic crease in peripheral blood leukocytes. Moreover, in analysis. Academic Press Ine., New% York. 737-739. bone marrow culture, LP dramatically stimulated 17. Mancini, G., A. 0. Carb)onara, and J. F. Heremans. 1965. Iimmuiiiinoche mical (Itianititation of antigens by single radial colony form-lation. Since LP hals multiple effects on fac- inmmuniiiiodiffutsion. In inu nocheim istri . 2: 235-254. tors known to effect serumil lysozyme, our observations 18. Nlasson, P. L., J. F. Heremans, and E. Schonne. 1969. of LP-induced lysozyme release may relate to only a Lactoferrin, anl ironi binding protein in netitrophilic leti- portion of elevated sertumii lysozymne in inflamimatory kocytes. j. Exp). Ml(d. 130: 643-658. 19. Jaffe, B. NI., J. XW. Smlith, XX. T. Newtoni, and C. XW. Parker. and myeloproliferative disorders. 1971. Radioimmutnoassav for prostaglandinis. Science The demonstration that a biologic substance induces (Wl(aslh. D. C.). 171: 494-496. selective release of neutrophil specific granule con- 20. Elimi, R. J., and( S. MI. XXolff. 1973. Non-specificity of the

Pilrogen-I nd(uiccdl Neutrop)h il Degra- n nlationt 133.5 liniulus amebolvte lvsate test: positive reactions with 30. Goldstein, I. M., S. T. Hoffstein, and G. Weissmann. 1975. polvnuceleotides and proteins. J. Infect. Dis. 128: 349- Mechanisms of lysosomal enzyme release from humani 352. polymorphontclear leukocytes: effects of phorbol my- 21. Goldstein, I. M., G. Weissmann, P. B. Dunham, and R. ristate acetate. J. Cell Biol. 66: 647-652. Soberman. 1975. The role of calcium in the secretion of 31. White, J. A., and R. D. Estensen. 1974. Selective labiliza- enzymes by huiman polymorphonuclear leukocytes. In tion of'specific graniules in polymorphonuclear leukocytes Calciulm Transport in Contraction and Secretion. E. Cara- hy phorbol myristate acetate. Am. J. Pathol. 75: 45-54. Itili, editor. North-Holland(l Publishing Co., Amsterdamll. 32. wright, D. G., D. A. Bralove, and J. I. Gallin. 1977. The 185- 193. dif'ferential mobilization of' human neuitrophil granules. 22. Goldstein, I. M., S. T. Hoffstein, and G. Weissmann. 1975. Ef'fects of phorbal myristate acetate and ionophore Influence of divalent cations tupon complement-mediated A23187. Am. J. Pathlol. 87: 273-284. enzyme release from huiman polymorphonuiclear leuko- 33. Spitznagel, J. K., F. G. Dalldorf, M. S. Lef'ell, J. D. Folds, cytes.J. Immnunol. 115: 665-670. 1. R. H. Walsh, M. H. Cooney, and L. E. Martin. 1974. 23. Milton, A. S., and S. Wenidlandt. 1970. A possible role Character of'azurophil and specific granules purified f'rom for prostaglandin E, as a modulator for temperature regu- human polymorphonuclear leukocytes. Lab. Invest. 30: lation in the central nervous systemii of the cat. J. Physiol. 774-785. (Lout(d.). 217: 33-34P. 34. Ban, A., L. Kosear, J. Kacsko, P. Demeny, J. Csongar, and(l 24. Zurier, R. B., S. Hoffstein, and G. Weissmann. 1973. Cyto- B. Siro. 1962. Die wirkung eines bakeriellen pyrogens- chalasin B: Effect on lysosomal enzyme release from pyrexal- auf'das serumiiieisen. Dtsch. Arch. Klin. Med. 208: humani leuikocytes. Proc. Natl. Acad. Sci. U. S. A. 70: 844- 45-52. 848. 35. Kampschmidt, R. F., and H. F. Upchuirch. 1962. Effects 25. Goldsteini, I. M., S. Hoff;stein, J. 1. Gallin, and G. Weiss- of hacterial en(dotoxini on plasma iron. Proc. Soc. Exp). mann. 1973. Mechanisms of lysosomal enzyme release Biol. Med. 110: 191-193. from lhumilani leukocytes: Microtubule assembly and membrane fusion induced bv a component of complement. 36. Merriman, C. R., L. A. Pulliam, and R. F. Kampschmidt. Proc. Natl. Acad. Sci. U. S. A. 70: 2916-2920. 1977. Comparisoni of' leukocytic pyrogen and leukocytic 26. Becker, E. L., H. J. Showell, P. M. Henson, and L. S. endogeneous mediator. Proc. Soc. Ex). Biol. Med. 154: Hsti. 1974. The ability of chemotactic factors to induce 224-227. lysosomal enzyme release. I. The characteristics of the 37. Hensen, N. B., H. Karle, and V. Andersen. 1974. Produc- release, the importance of surfaces and the relation of tion and elimination of' plasma lysozyme. In Lysozyme. enizymiie release to chemotactic responsiveness. J. Im- E. F. Osserman, R. E. Canfield, and S. Beychok, editors. munol. 12: 2047-2054. Academic Press Inc., New York. 307-319. 27. Wright, D. G., and S. E. Malawista. 1972. The mobiliza- 38. Bodlel, P. T., P. T. Major, and J. B. Gee. 1976. Sarcoidosis: tion ancl extracellular release of granular enzymes from enhanced secretion of'endogenouis pyrogen and lysozyme human leukocytes during phagocytosis. J. Cell Biol. 53: by peripheral blood monocytes. Clin. Res. 24: 587A. 788-797. (Abstr.) 28. Lefell, M. S., and J. K. Spitznagel. 1974. Intracellular and 39. Finch, S. C., 0. Castro, M. E. Lippman, J. A. Donadio, extracellular degranulation of human polymorphonuclear and P. E. Perillie. 1974. Lysozyme in leukopathic states. azurophil and specific graniules induced by immune com- In Lysozyme. E. F. Osserman, R. E. Canfield, and S. plexes. Itnfect. Immun. 10: 1241-1249. Beychok, editors. Academic Press, Inc., New York. 335- 29. Hoffstein, S., R. Soberman, I. Goldstein, and G. Weiss- 345. mann. 1976. Concanavalin A induces microtubule as- 40. Kampschmidt, R. F., and H. F. Upchurch. 1977. Possible sembly and specific granule discharge in human poly- involvement of leukocytic endogenous mediator in granu- morphonuclear leukocytes.J. Cell Biol. 68: 781-787. lopoesis. Proc. Soc. Exp. Biol. Med. 155: 89-93.

1336 M. S. Klempner, C. A. Dinarello, andJ. 1. Gallin