Cyclic nucleotide changes in human neutrophils: induced by chemoattractants and chemotactic modulators

Two classes of agents have effects on the chemotactic response of human neutrophils. Chemoattractants, such as E. coli bacterial factor (BF) and the recently discovered N-formylmethionyl peptides, initiate directed cellular movement, while chemotactic modulators enhance or depress cell movement initiated by chemoattractans. We have previously shown that cyclic adenosine 3',5'-monophosphate (cAMP) depresses neutrophil migration toward E. coli BF, while cyclic guanosine 3',5'-monophosphate (cGMP) enhances the same process. In the present study, we examined the effects of chemoattractants and modulators on neutrophil levels of cGMP and cAMP. A simplified radioimmunoassay procedure was developed which employed a rapid heat inactivation of cellular samples rather than the traditional acid inactivation. Three chemoattractants and six chemotactic modulators were tested in concentrations which initiate or modulate cell movement. The chemoattractants, BF, N-formylmethionylalanine, and trypsinized human complement, stimulated the accumulation of neutrophil cGMP (40-70% average increase) but had little effect on cAMP concentrations. The enhancing chemotactic modulators, phenylephrine, prostaglandin F1-alpha, carbachol, and phorbol myristate acetate, also caused significant (p ? 0.05) increases (19-34%) in neutrophil cGMP levels without affecting cAMP concentrations. In contrast, two agents which inhibit the chemotactic response; isoproterenol and prostaglandin E1, significantly elevated cAMP concentrations in these cells. These results suggest that the ability to elevate cGMP is inherent in the action of both chemoattractants and enhancing chemotactic modulators. Increases in cAMP appear to be correlated with inhibition of chemotaxis.

CYCLIC NUCLEOTIDE CHANGES IN HUMAN NEUTROPHILS INDUCED BY CHEMOATTRACTANTS AND CHEMOTACTIC MODULATORS by Gary E. Hatch A dissertation sublnitted to the faculty of the Uni vers ity of Utah in part ial fulfillment of the requirements for the degree of Doctor of Philos ophy Department of Pharmacology University of Utah March 1977 THE UNIVERSITY OF UTAH GRADUATE SCHOOL SUPERVISORY COMMITTEE APPROVAL of a dissertation submitted by GARY E. HATCH I have read this dissertation and have found it to be of satisfactory quality for a doctoral degree. J~Y\ ()..S) Iq77 Date William K. Nichols, Ph. D. Chairman, Supervisory Committee I have read this dissertation and have found it to be of satisfactory quality for a doctoral degree. ~S- 31&., fCZ?2 ;I /-.1 7 /(, j?-Lf Date I Harry R. Hill, M. D. Member, Supervisory Committee I have read this dissertation and have found it to be of satisfactory quality for a doctoral degree. tJl J~. ~ 1977 Dale Stewart C. Harvey, Member, Supervisory Committee I have read this dissertation and have it to be of ~~at.i.6.factory quality for a doctoral degree. ~:.:~. ' 19 71' CZ.~-,J~~~~~~~~~- ~ Gerald Cf. Krue ,M. D. Member, S}r'pervisory Committee I have read this dissertation and have found it to be of satisfactory quality for a doctoral degree. Z-J; 1977 Dale Lester M. Partlow, Ph. D. Member, Supervisory Committee THE UNIVERSITY OF UTAH GRADUATE SCHOOL FINAL" READING APPROVAL To the Graduate Council of The University of Utah: I have read the-dissertation of Gary E. Hatch in its final form and have found that (I) its format, citations, and bibliographic style are consistent and acceptable; (2) its illustrative materials including figures, tables, and charts are in place; and (3) the final manuscript is satisfactory to the Supervisory Committee and is ready for submission to the Graduate School. JV\v--. ~S-)} q 7 J Date William K. Nichols, Ph. D. Member, Supervisory Committee Approved for the Major Department Approved for the Graduate Council Sterling McMurrin De n of The Graduate School ABSTRACT Two classes of agents have effects on the chemotactic response of human neutrophils. Chemoattractants, such as E. coli bacterial factor (BF) and the recently discovered N-formylmethionyl peptides, initiate directed cellular movement, while chemotactic modulators enhance or depress cell movement initiated by chemoattractants. We have previously shown that cyclic adenosine 3',5' -monophosphate (cAMP) depresses neutrophil migra­tion toward E. coli BF, while cyclic guanosine 3', 5'-monophosphate (cGMP) enhances the sam.e process. In the present study, we examined the effects of chemoattractants and modulators on neutrophil levels of cGMP and cAMP. A simplified radioimmunoassay procedure was developed which employed a rapid heat inactivation of cellular samples rather than the traditional acid inactivation. Three chemoattractants and six chemotactic modulators were tested in concentrations which initiate or modulate cell movement. The chemo­attractants, BF, N-formylmethionylalanine, and trypsinized human comple­ment, stimulated the accumulation of neutrophil cGMP (40-70% average increase) but had little effect on cAMP concentrations. TIle enhancing chemotactic modulators, phenylephrine, prostaglandin F2a, carbachol, and phorbol myristate acetate, also caused significant (p~O. 05) increases (19-34%) in neutrophil cGMP levels without affecting cAMP concentrations. In contrast, two agents which inhibit the chemotactic response; isoproterenol and prostaglandin E l' significantly elevated cAMP concentrations in these cells. These results suggest that the ability to elevate cGIv1P is inherent in the action of both chemoattractants and enhancing chemotactic modulators. Increases in cA1\1P appear to be correlated with inhibition of chemotaxis. v ACKNOWLEDGEMENTS The help of many people have m.ade my graduate experience at the University of Utah both rewarding and enjoyable. Dr. Dixon M. Woodbury and all of the faculty of the Department of Pharmacology are to be thanked for making this experience possible. Dr. Lester M. Partlow gave me many valuable experiences during my first research rotation in his laboratory. The help of my supervisory committee is gratefully acknowledged: Dr. Gerald Krueger for a valuable journal club experience, and Dr. Stewart C. Harvey for his help in reading the manuscripts. Dr. Charles J. Nabors should als 0 be thanked for generously providing the use of his calculator and filtering apparatus. I am particularly indebted to Drs. Harry R. Hill and William Ko Nichols for their friendship and concern. Dr. Hill gave many hours of individual help, encouragement, and expertise. His laboratory facilities and supplies made this project possible. Dr. Nichols was exceedingly helpful in teaching me the methods and introducing me to the people· working with cyclic nucleotides, as well as serving as chairnlan of my thesis committee. My experiences in the laboratory were also made more enjoyable because of the association and help of Kim Rawlinson, Nancy Hogan and Mary P artas. Finally, the loving support of my wife Gayle, and our children Sarah, Ryan, and Kevin have made this experience worth any effort that it might have required. The encouragement of my parents, Ephraim and Verena Hatch, and my parents-in-law, Glenn and Emmy Collette, has also been appreciated. This work was supported by grants from the National Institutes of Health (GM00153, AM18354, AL 13150, and AM16330), and from the Howard Hughes Medical Institute (Dr. Hill). vii TABLE OF CONTENTS ABSTRACT ACKNOWLEDGEMEN~S LIST OF ILLUSTRATIONS PART ONE: A Simplified Procedure for the Radioimmunoassay of Cyclic Nucleotides and its Application to Human Leukocytes Summary Introduction Materials and Methods Results Discussion References PA RT TWO: TIle Effect of Initiators and Modulators of Cell Movement on the Concentration of Cyclic Nucleotides in Human Neutrophil Suspensions Summary Introduction Materials and Methods Results Discussion References Page v vii x 2 3 4 7 15 17 21 22 24 27 41 45 LIST OF ILLUSTRATIONS PART ONE Tables 1. 2. 3. 4. S. Loss of sample volume due to evaporation during heat treatment. The measurement of cyclic nucleotide standards added to heat-killed cellular samples and the effect of phosphodiesterase treatment. Heat treatment of guanosine Sf-triphosphate (GTP) and its effect on cyclic GMP concentrations. Recovery of cyclic nucleotides after heat inactivation, sonication, and micropore filtration. The effect of heat-treated PMNs, erythrocytes, and serum on the no-antibody control of a cyclic GMP radioimmunoassay. PART TWO Tables 1. 2. 3. 4. TI1e effect of chenl0attractants all cyclic nucleotide content in preincubated neutrophils. Changes in neutrophil cyclic GMP concentrations produced by chemoattractants and modulators when cells were not pre incubated. The effect of chemotactic modulators on cyclic nucleotide content in preincubated neutrophils. Cyclic GMP effects of combined chemoattractants and chemotactic modulators. Page 8 9 10 12 13 28 32 36 40 Figures L 2. 3. 4. Accumulation of cGMP in human neutrophils in response to E. coli chemotactic factor (BF). Time courses of cGMP a ccumulation in neutrophils caused by N-formylmethionylalanine (fMet-Ala) in pre incubated and non -preincubated cells. Tirne courses of cGMP accumulation in neutrophils produced by carbachol and phorbol myrista te acetate. Time course of neutrophil cAMP accumulation produced by isoproterenol. x Page 29 33 37 39 PART ONE A SIMPLIFIED PROCEDURE FOR RADIOIMMUNOASSAY OF CYCLIC NUCLEOTIDES AND ITS APPLICATION TO HUMAN LEUKOCYTES SUMMARY Rapid heat-treatment of leukocyte suspensions was found to be an effective alternative to acid treatment in the preparation of these cells for radioimmunoassay of cyclic guanosine 3',5' -monophosphate (cGMP)land cyclic adenosine 3',5' -monophosphate (cAMP). A two-second heating to boiling temperatures, followed by sonication and micropore filtration, was employed. This procedure adequately inactivated or removed enzymes and binding proteins that can alter cyclic nucleotide concentrations or otherwise interfere with the radioimmunoassay. Moreover, heating in this manner did not appear to affect the stability of cyclic nucleotides or cause significant formation of cGMP from endogenous GTP. Recovery of cyclic nucleotides after heating and filtration was high (89-96%), making possible the meas-urement of both cGMP and cAMP in small cellular samples. Variation in cyclic nucleotide recovery was small (SD=1. 4-4. 0 %); therefore, individual recovery determinations were unnecessary" lAbbreviations used in this paper: cGMP, guanosine 3',5'-monophos­phate; cAMP1 adenosine 3',5'-n10nophosphate; GTP, guanosine 5' -triphos­phate.; PMNs, human polymorphonuclear leukocytes; PD, phosphodiesterase. INTRODUCTION Studies of cellular cyclic nucleotides have traditionally been difficult because of the laborious procedures required for preparing radioimmuno­assay samples. These procedures require the denaturation of macro­molecules involved in the synthesis, degradation, and binding of the cyclic nucleotides. Protein precipitants, such as trichloroacetic, perchloric, or hydrochloric acids,are en1ployed for this purpose. These precipitants must later be removed from samples,because they interfere with the radio­immunoassay. Ether extraction of trichloroacetic acid (1), precipitation of perchloric acid with KOH (2) or its elimination by use of alumina columns (3), or evaporation of ethanolic HCl (4), are procedures commonly employed to remove these agents. In addition to being time-consuming, these extraction steps sometimes cause unpredictable losses of cyclic nucleotide and necessitate a separate determination of cyclic nucleotide recovery for each sample (2,5). The present studies were designed to determine if a rapid heat-inactivation could be employed to simplify the preparation of immunoassay samples. Previous studies have shown that rat brain cyclic nucleotide concentrations measured after two-second focused Inicro­wave heating were comparable to those obtained by rapid freezing followed by perchloric acid treatment (6). We report that a two-second heat treat­ment of purified human blood neutrophil suspensions simplifies the proce­dure for preparing these cells for radioimmunoassay. MATERIALS AND METHODS Iodinated cyclic nucleotide derivitives (1 125 -succinyl tyrosine methyl ester), cyclic nucleotide standards, and antibodies to cGMP and cAMP were obtained from Collaborative Research, Waltham, Mass. Antibodies used had more than a 5000 fold greater affinity for the cyclic nucleotides measured than for closely related cyclic and non -cyclic nucleotides (manufacturers data included with shipments). Ficoll-hypaque (Ficoll­Paque) was obtained from Pharmacia, Piscataway, N. J. ; micropore filters (No. HAWP02500) from Millipore Corp., Bedford, Mass.; Medium 199 with Hanks salts from Microbiological Associates Inc., Bethesda,Md.; GTP, sodium azide, and Hepes buffer from Sigma, St. Louis, Mo; and beef heart phosphodiesterase from Boehringer Mannbeim, San Francisco, Ca. Tissue preparation. Leukocyte-rich plasma was obtained by sedimen­tation (1 hour, 37oC) of heparinized venous blood (10 V/ml) from normal adult donors. Suspensions of polymorphonuclear leukocytes (PMNs) were separated from leukocyte-rich plasma by density gradient centrifugation (1/2 hour, 400 g) on ficoll-hypaque (7). Cells were suspended at a final concentration of 5 X 106 PMNs per ml in Medium 199 to which Hepes buffer had been added to a concentration of 15 mM. The final pH of this solution was 7. 4. Cell suspension was added in O. 5 ml aliquots to 15 X 100 mm glass culture tubes. Heat treatment was performed by holding the tubes containing cells in a large Bunsen burner flalne until boiling began (2 sec. ) Samples heated in this manner were stored at -200 C until further use. Heated cellular suspensions were sonicated (cell disruptor model W140, Heat Systems -Ultras onics Inc. Plainview, N. Y. ) and then filtered (0. 45 ~ pore size) to remove cellular debris. Filtrate cyclic nucleotides were acetylated as previously described (8) by adding 30 ~l of a freshly prepared solution of triethylamine and acetic anhydride (2:1 vol/vol) per nll of sample. Radioimmunoassay_ Cyclic nucleotide standards used in the assay 5 were prepared in Medium 199-Hepes. Sodium azide (final conc. of 0.05%) added to the media was found to be effective in inhibiting microbial growth. This agent did not interfere with acetylation, antibody binding, or subsequent phosphodiesterase treatments of cyclic nucleotide samples and was therefore added to cyclic nucleotide standards stored at 40 C for up to 1 month. hnmunoassays were performed by adding 100 1-11 of the cyclic nucleotide standard or of the filtered, heat-inactivated cell suspension to each assay tube. Antibody and iodinated cyclic nucleotides were prepared in 50 mM sodium acetate buffer, pH 4. 75, containing 20 mM calcium chloride. The low cross-reactivity of the cGMP antibody with cAMP and with GTP reported by the manufacturer was confirmed by our experiments. Imnlunoassay tubes containing a total volume of 300 1-11 were incubated overnight at 40 C. Separation of antibody-bound cyclic nucleotide was performed by micropore filtration (Millipore application manual AM304). The radioactivity on each filter disc was determined from the time required to generate 4000 counts, in order to eliminate the necessity of adjusting the concentration of iodinated cyclic nucleotide during radioactive decay of these compounds. Time-values obtained were analyzed using a Hewlett Packard Model 10 calculator (Loveland, Co., program No. 9810A). Phosphodiesterase-treatment was performed by adding beef heart phosphodiesterase (PD; final concentration of 0.3 mg/ml) to samples containing cyclic nucleotides and incubating at room temperature for 2 hours, after which the enzyme was inactivated by rapid heat-treatment, then fUte re d. 6 RESULTS Removal of enzymes and binding molecules. Heating of cell suspen­sions until boiling occurred required about 2 seconds and resulted in a volume loss of about 2% of the sample (Table 1). Table 2 shows the effect of adding known amounts of cyclic nucleotide to Medium 199 -Hepes alone and to Medium 199-Hepes containing heat-killed neutrophils. Concen­trations of cyclic nucleotides determined after addition of standards were comparable with the expected values. The fact that heat- killed cells did not affect the radioimmunoassay of added cyclic nucleotides indicateS that phosphodiesterases, cyclases, and binding proteins were inactivated or removed by heating and filtration. Cyclic nucleotides were undetectable in PD-treated samples, indicating that "blank" materials which mimic the cyclic nucleotides in the immunoassay are not present in heated PMNs as previously reported in other tissue extracts (1). Formation of cGMP from GTP was shown by Kimura and Murad (9) to result from heating 1 roM GTP in the presence of divalent cations. We have sought to assess the contribution of this reaction to cGMP concentrations in PMNs as deterrnined by the heat treatnlent described here. Table 3 shows the effect of two-second heating on cGMP formation from various concentrations of added GTP. No more than 0.0022% of the added GTP was converted to cGMP under any of the conditions examine d. This percentage Table 1. Loss of sample volume due to evaporation during heat treatment. Percent loss Time to boiling (sec) of volume 1 1.8 2.16 2 2.2 2.68 3 1.8 1. 78 4 2.0 2.88 '5 1.8 1. 70 6 2.2 2.75 7 2.2 2. 12 8 1.8 2.34 Mean + SD 1. 98 + O. 19 2.30 + 0.44 Culture tubes (15 X 100 mm) containing O. 5 ml each of cellular suspension in Medium 199-Hepes were weighed, heat-treated, and allowed to stand at room temperature until condensed water which collected on 8 the inside walls of the tubes during heating had evaporated (1 hour). Tubes were then weighed again. Percent weight loss is calculated to represent the percent of O. 5 ml of solution that was lost due to the heat treatment. The percent volume loss due to standing one hour at room temperature without prior heat treatment was O. 9%. 9 Table 2. The measurement of cyclic nucleotide standards added to heat-killed cellular samples and the effect of phosphodiesterase treatment. Test sample fmoles cG:tvfP fmoles cAMP l. 100 fmoles cyclic nucleotide 102 + 3 106 +5 2. Heat-killed PMNs 36 +2 178 +5 :E = 138 + 3 L =284 +7 3. 100 fmoles cyclic nucleotide + heat-killed PMNs 138 +4 280 + 11 4. Heat-killed PMNs + PD <5 <5 Equal amounts of cGMP and cA"MP standards were added to cell-free Inedia and to heat- killed PMNs. Sample 1 shows the cGMP and cA1V1P levels, as determined by radioimmunoassay, of Medium 199-Hepes plus 100 fmoles of these agents. Sample 2 depicts the levels of these cyclic nucleotides in heat- killed PMNs. Sample 3 shows the levels in the same PMN prepar­ations to which 100 fmoles of cyclic nucleotide were added. Sample 4 shows the effect of phosphodiesterase (PD) treatment on cyclic nucleotide concentrations. Values represent mean ± SE of 3-5 incubations with 3 assay tubes per incubation. Table 3. Heat treatment of guanosine Sf-triphosphate (GTP) and its effect on cGMP concentrations. Experiment 1. Medium 199 + O. 01 mM GTP Heat-killed PMNs + O. 01 mM GTP 2. Medium 199 + 0.1 mM GTP Heat-killed PMNs + 0.1 mM GTP 3. Medium 199 + 1. 0 mM GTP Heat- killed PMNs + 1. 0 mM GTP 4. Medium 199 + 10. a mM GTP Heat-killed PMNs + 10.0 mM GTP Percent conversion of GTP to cGMP 0.0019 0.0011 0.0009 0.0022 0.00023 0.00019 0.00001 0.00001 10 Guanosine 5' -triphosphate was added to Medium 199-Hepes and to heat­killed PMN suspensions in the same medium. Part of each sample was then subjected to a two-second heat treatment. Differences in cG:NIP concentra­tions between heated and unheated samples, as determined by radioimmuno­assay, were used to calculate the precent of added GTP that was converted to cGMP by the heat treatment. Radioligand displacement in the imlnuno­assay due to 10 mM GTP was equivalent to the displacement seen with 1 nM cGMP. Values represent the mean ± SE for 3 incubations with 2 assay tubes per incubation. 11 was not increased at lower GTP concentrations and was decreased at higher GTP levels. Recovery of cyclic nucleotides after filtration. The removal of heat­coagulated cellular debris was found to be necessary after heating, because its presence interfered with later separation of free and antibody-bound cyclic nucleotides. Micropore filtration of each cellular sample was found to be the most rapid and effective method of removing heat-coagulated debris. Table 4 shows the recovery of cyclic nucleotides after micropore filtration. Both neutrophil and mononuclear cell suspensions yielded high recovery values for both cyclic nucleotides (89-96%). Plasma samples, \vhich contained larger amounts of protein, showed lower recoveries (60- 70 %). The variability in recoveries, however, was always low (SE 1. 4 to 4.0%). Thus, individual recovery determinations for each sample are unnecessary when heating and micropore filtration are employed. Effects of high protein concentrations. Under most conditions, heat treatment followed by sonication and micropore filtration provided a convenient method of rapidly preparing radioimmunoassay samples. Occasionally, however, large amounts of protein are required in cellular incubations, the presence of which slows down the filtration step and inter­feres with later separation of bound and free radiolabeled cyclic nucleotide. Table 5 shows the effect of protein in cellular samples on the no-antibody control of a cGW radioimmunoassay. Similar results were seen in cAMP assays. Proteins normally present in heated suspensions of PMNs only 12 Table 4. Recovery of cyclic nucleotides after heat inactivation, sonication, and micropore filtration. Percent Recovery ± SD Exp. Heat-treated sample Cyclic GMP Cyclic AMP l. Polymorphonuclear leukocytes 92.5 ± 2. 9 ( 8) not done 2. Polymorphonuclear leukocytes 95.0 ± 2. 3 (10) 94 .. 3±2.0(10) 3. Mononuclear leukocytes 96.0 ± 2.3 ( 9) 93.3 ± 1. 7 (10) 4. Mononuclear leukocytes 94. 1 ± 2. 0 ( 7) 89. 1 ± 1. 4 ( 8) 5. Blood plasma 76. 8 ± 4. 0 (10) 60. 3 ± 2. 6 (10) Radioactive (1125) cyclic nucleotides were added to 0.6 ml of heat­inactivated cell suspensions or plasma samples (30 cpm / III final concen­tration). Samples were then sonicated and filtered.. Counts obtained from 100 III of the cell suspension before filtering were compared with the counts obtained in 100 1-11 of solution after filtration to obtain the percent recovery value. NeutrophiJ and mononuclear cell samples were suspensions containing 5 X 10 cells / ml in Medium 199-Hepes. Plasma samples were 1: 10 dilutions of fresh plasma in phosphate buffered saline. The number of separate samples is in parentheses. 13 Table 5. TI1e effect of heat-treated PMNs, erythrocytes, and serum on the no-antibody control of a cyclic GMP radioimmunoassay. Agents added No- antibody control (Percent of total counts added*) Medium 199-Hepes 6 Medium 199-Hepes + PMNs (5 X 10 / ml) Medium 199-Hepes + PMNs + 1. 0 % serum Medium 199-Hepes + PMNs + 2. 1 % serum Medium 199-Hepes + PMNs + 3. 8 % serum Medium 199-Hepes + PMNs + 5. 7 % serum Medium 199-Hepes + PMNs + 9.0 % serum Medium 199 + Erythrocytes (1. 5 X 109 / mO 3.0 % 3.2 % 3.6% 3.8% 4.4% 7.0 % 8.4 % 5.7% A typical radioimmunoassay was performed as described in Methods using micropore filtration for the final separation of bound and free radio­label. Values given in this table represent the magnitude of the control to which no antibody was added expressed as a percentage of the total counts per minute of iodinated cGMP added. Heated neutrophils slightly increased the blank and erythrocytes significantly increased the blank. Each value is calculated from the mean of duplicate assay tubes. * Counts added were 6800 cpm per assay tube. slightly increased the no-antibody control over that seen with Medium 199- Hepes alone. However, the presence of serum (9 % vol/vol) increased 14 this value almost 3 fold (3.2 % increased to 8.4%). Concentrations of heated serum higher than 10 % are very dlfficult to filter. Similar increases in the no-antibody control of high protein samples were seen when ammonium sulphate precipitation was employed in the final separation of bound and free radiolabel. TI1e increased no-antibody control is therefore due to proteins or other macromolecules that are not removed by the heat-treatment and initial micropore filtration. It can be concluded that the heat- kill procedure is most useful when cell suspensions contain relatively little protein. DISCUSSION We have shown (Table 1) that heat treatment of neutrophil suspensions can be performed rapidly (2 seconds) with little loss of sample volume due to evaporation. This procedure appears to inactivate inhibitors of cAMP-and cGMP-binding that have been reported to be present in neutrophils (10, 11, 12). Phosphodiesterase treatment of cellular samples (Table 2) indicates that the nucleotides measured are cGJv1P and cAMP. Experiments in which known an10unts of the cyclic nuclectides were added to heat- killed cells indicate that the cyclic nucleotides are not degraded by the heat-treatInent or by remaining cellular phosphodiesterases. The effectiveness of heat-treatment jn inactivating cellular enzyn1es was also indicated by Guidotti et al.(7) who showed that rat brain adenylate cyclase and phosphodiesterase were inactivated by a two-second exposure to microwave heating. TI1e possible formation of cGMP from GTP during the two-second heat;:reatment was also investigated. We have found (Table 3) that no n10re thCUl 0.002 % of the GTP added to Inedium 199 or to heat-killed PMNs is converted to cGMP during the two-second heat treatment. Assuming intracellular GTP concentrations are about 0.3 to 0.5 mM (13), a 0.002 % conversion of endogenous GTP to cGMP would cause an increase in cellular cGMP of about 6 -8 13 1 fmole / 10 cells (or 10 moles / 10 cells assuming 10 13 wet cells per liter). Kimura and Murad (9) indicate that 0.05 % of the 16 added GTP can be converted to cGMP after 3-5 minutes of heating at 1000 C, however, their data also show that less prolonged heating times result in less formation of cGMP. We conclude that at the probable intracellular GTP concentrations (0.5 mM or lower) and using the two-second heat treat-ment the contribution of this reaction to cGMP measurements would be negligible. Similar conversion of ATP to cAMP can apparently occur under o conditions of prolonged heating (30 min, 100 C) in 0.4 N barium hydroxide (14, 15), however, this effect i.s not observed in physiologic buffers during shorter heating times (9). In comparison with the ethanolic HCl-evaporation technique (4), or trichloroacetic acid-ether extraction (1), the heat-filtration procedure described here appears to provide equivalent or increased percent recovery of both cyclic nucleotides.. A disadvantage of the heat- kill method is that large amounts of protein (greater than 5 % plasma) can reduce cyclic nucleotide recovery, make micropore filtration difficult, and interfere with the final separation of bound and free radiolabel. Moderate amounts of protein (up to 5 % plasma) appear to present few problems if they are present in all incubations or if adequate controls are performed. The advantage of the heat treatment over the traditional proce dures is that the tin1e -consuming extraction of protein preCipitants is eliminated. REFERENCES 1. Steiner, A. L., A. S. Pagliara, L. R. Chase, and D. M. Kipnis. 1972. Radioimmunoassay for cyclic nucleotides: II. Adenosine 3',5'­monophosphate and guanosine 3',5' -monophosphate in mammalian tissues and body fluids. J. BioI. Chern. 247: 1114-1120. 2. Sandler, J. A., R. I. Clyman, V. C. Manganiello, and M. Vaughan 1975. The effect of serotonin (5-hydroxytryptamine) and derivitives on guanosine 3',5'-monophosphate in human monocytes. J. Clin. Invest. 55:431-435. 3. Haddox, M. K., L. T. Furcht, S. R. Gentry, J. H" Stephenson, and N. D. Goldberg. 1976. Periodate-induced increase in cyclic GMP in mouse and guinea pig splenic cells in association with mitogenesis. Nature 262: 146-148. 4. Kebabian, J. W., A. L. Steiner, and P. Greengard. 1975. Muscarinic cholinergic regulation of cyclic guanosine 3',5' -monophosphate in autonomic ganglia: possible role in synaptic transmission. J. Pharm. Exp. Ther. 193:474-488. 5. Goldberg, N. D., and A. G. O'Toole. 1972. Analysis of cyclic 3',5'­n1. onophosphate. Methods Biochem. Anal. 20:1-39. 6. Guidotti, A., D. L. Cheney, M. Trabucchi, C. Wang, and R. A. Hawkins. 1974. Focussed n1.icrowave radiation: a technique to minimize post mortem changes of cyclic nucleotides, DOPA, and choline and to preserve brain morphology. 7. Boyum, A. 1967. Isolation of mononuclear cells and granulocytes from human blood. Scand. J. ClinG Lab. Invest. 2]; 77-89. 8. Harper, J. F. and G. Brooker. 1975. Femtomole sensitive radio­immunoassay for cyclic AMP and cyclic GMP after 2' 0 acetylation j by acetic anhydride in aqueous solution. J. Cyclic Nucleotide Res. 1:207-218. 9. Kimura, H., and F. Murad. 1974. Nonenzymatic formation of guanosine 3!, 5 '-monophosphate from guanosine triphosphate. J. BioI. Chern. 249: 329-331. 10. Zurier, R. B., G. Weissmann, S. Hoffstein, S. Kammerman, and H. H. Tai. 1974. Mechanisms of lysosomal enzyme release from human leukocytes. II. Effects of cyclic AMP and cyclic GMP, auton·omic agonists, and agents which affect microtubule function. J. Clin Invest. 53:297-309. 18 11. Tsung, P-K., and G. Weissmann. 1973. Inhibitor of cyclic AMP binding and protein kinase activity in leukocyte lysosomes. Biochem. Biophys. Res. Commun. 51: 836-842. 12. Goldstein, I., S. Hoffstein, J. Gallin, and G. Weissmann. 1973. Mechanisms of lysosomal enzyme release from human leukocytes; Microtubule assembly and melnbrane fusion induced by a component of complement. Proc. Natl. Acad. Sci. U. S. A. 71:2916-2920. 13. Goldberg, N. D., R. F. O'Dea, and M. K. Haddox.· 1973. In Adv. Cyclic Nucleotide Res. Vol. 3. Edited by P. Greengard and G. A. Robison. Raven Press, New York. P. 155. 19 14. Cook, W. H., D. Lipkin, and R. Markham. 1957. The formation of a cyclic dianhydrodiadenylic acid by the alkaline degradation of adenosine 5' -triphosphoric acid. J. Am. Chern. Soc. 79: 3607. 15. Sutherland, E. W., and T. W. Rall. 1957. The properties of an adenine ribonucleotide produced with cellular particles, ATP, Mg ++, and epinephrine or glucagon. J. Am. Chern. Soc. 79:3608. PART TWO THE EFFECT OF INITIATORS AND fv10DULATORS OF CELL MOVEMENT ON THE CONCENTRATION OF CYCLIC NUCLEOTIDES IN HUMAN NEUTROPHIL SUSPENSIONS SUMMARY Two classes of agents have effects on the chemotactic response of hun1an neutrophils. Chemoattractants such as activated complement components, E. coli bacterial chemotactic factor (BF) and the recently discovered N-formyllnethionyl peptides, initiate directed movement, while chenlotactic nlodulators enhance or depress cell movement initiated by chemoattractants. We and others have previously shown that cyclic adenosine 3', S'-monophosphate (cAMP) depresses neutrophil migration toward BF and cyclic guanosine 3' t 5' -monophosphate (cGMP) enhances the same process. In the present study, we examined the effects of several chemoattractants and chemotactic modulators on neutrophil levels of cG:rv1P and cAMP. The chemoattractants; trypsinized human complement, BF, and N-formylmethionylalanine, stimulated the accumulation of neutrophil cGMP (40-70 % average increase) but had little effect on cAMP concentrations. The enhancing chemotactic modulators, phenylephrine, prostaglandin F 2a' carbachol, and phorbol myristate acetate, also produced increases in cGMP levels without affecting cAMP concentrations in neutrophils. In contrast, two agents which inhibit the chemotactic response, isoproterenol and prostaglandin El' elevated cAMP concentrations in these cells. The ability to elevate cGMP appears, therefore, to be inherent in the action of both chemoattractants and enhancing chemotactic modulators. hlcreases in cAMP appear to be correlated with inhibition of chemotaxis. INTRODUCTION Chemotaxis, the unidirectional movement of cells toward a chemical stimulus, is a basic and extremely important function of some mammalian cells. Two classes of agents have pronounced effects on this process. Chemoattractants initiate directed movement when presented to the cell in a concentration gradient, and chemotactic modulators, which are not chemotactically active by themselves, can enhance or depress cell move" ment initiated by chemoattractants. In previous studies, we have demon­strated that cyclic guanOSine 3', S'-monophosphate (cGMP)I, as well as agents believed to increase its cellular concentrations, function as chemo-tactic modulators, enhancing the chemotactic movement of human poly-morphonuclear leukocytes (PMNs) (1,2). Cyclic adenosine 3',5' -monophos-phate (cAMP) and agents believed to increase its cellular levels are, on the other hand, inhibitory to the same process (1,2,3). While it is assumed that the cyclic nucleotides are important in the action of both chemoattract-ants and chemotactic modulators, few direct measurements have been made of cyclic nucleotides in PMNs in response to these agents. Zymosan-treated serum, which possesses chemotactic activity, has been reported * Abbreviations used in the paper: PMN, human polymorphonuclear leukocyte; cAtvrr', cyclic adenosine 3t,St-monophosphate; cGMP, cyclic guanosine 3',5' -monophosphate; BF, bacterial chemotactic factor derived from E. coli; f~1et-Ala, N-formylmethionylalanine. to elevate cGMP levels in hUlnan neutrophils (4). The effects of other chemoattractants on cGMP levels have not been investigated. The mode of action of the chemotactic modulators also remains uncertain (I, 2,4,5). The studies reported here were designed to determine 1) if other chemo­tactically active agents might also affect cGMP levels, and 2) whether chen1.otactic modulators exert their effects through alterations in cellular cyclic nucleotides. 23 MATERIALS AND METHODS Reagents. Phenylephrine, formylmethionylalanine, isoproterenol, Hepes buffer, trypsin, soybean trypsin inhibitor, and carbachol were obtained from Sigma, St. Louis, Mo. Ficoll-hypaque (Ficoll-Paque) and Dextran 70 were obtained from Pharmacia, Piscataway" N. J. ; micropore filters (No. HA WP02500) from Millipore Corp., Bedford, Mass.; and Medium 199 with Hanks salts from Microbiological Ass ociates Inc., Bethesda, Md. Prostaglandins were obtained from Merck and Company, Rahway, N. J., and phorbol myrtstate acetate from Consolidated Midland, Brewster, N. Y. Zymosan was obtained from Schwartz-Mann, Orangeburg N. Y., and whole human complement from Cordis Laboratories, Miami, Fla. The E. coli bacterial chemotactic factor and zymosan-treated senlln were prepared as previously described (6, 7). Trypsinized human complement (8) was prepared by adding 0.2 mg trypsin to 150 CH50 units of whole human complement (1 bottle, lyophilized) reconstituted to 2 mI. After a 15 minute incubation at room temperature, O. 8 mg of soybean trypsin inhibitor was added. Preparation and incubation of cells. Leukocyte-rich plasma was obtained by allowing the erythrocytes in heparinized venous blood (10 U/ml) from normal adult donors to sediment over a 1 hour period. Addition of dextran 70 (final concentration of 1%) to the whole blood facilitated sedimen- 25 tation of erythrocytes and did not appear to alter cellular cyclic nucleotide concentrations. The presence of some erythrocytes (up to 20 per PMN) also had no noticeable effect on cyclic nucleotide concentrations. In addition, siliconizing of the culture tubes did not appear to affect cyclic nucleotide concentrations and was not performed. Leukocyte-rich plasma was layered undiluted on ficoll-hypaque and centrifuged at 400 g for 30 minutes. Ficoll-hypaque was carefully pipetted from cellular pellets and cells suspended at a final concentration of 5 X 106 PMNs per ml in Medium 199 to which Hepes buffer had been added to a concentration of 15 mM. The final pH of this solution was 7.4. This procedure yielded a leukocyte suspension containing approximately 98% PMNs and 2% mononuclear cells. All incubations were carried out using 0.5 ml of cell suspension in 15 X 100 mm glass culture tubes. The temperature of the water bath (Model G76, New Brunswick Sci. , o New Brunswick, N. J. ) was maintained at 37 C. Agitation was by slow radial motion (60 rpm). Cells added to culture tubes were left for one hour at room ten1perature, then preincubated at 370 C for 20 minutes prior to addition of stimulatory agents. The experiments without preincubation were performed by adding cell suspensions at room temperature to tubes o containing the stimulatory agent immediately before incubation at 37 C. Cellular incubations were terminated by exposing the glass tubes to a Bunsen burner flame for two seconds. The heat- kill method and subsequent radioimmunoassay of cyclic nucleotides are described in Part One. All cyclic nucleotide measurements in this paper are of the total cyclic nucleo- 26 tide content (cells plus lnedium) of cellular incubations. Statistics. A logit plot ana.lysis (Hewlett Packard program No. 9810A, Loveland, Co.) was used to derive sample cyclic nucleotide concen­trations from the standard curve. Results are presented in the tables as changes in cyclic nucleotide content between control and stimulated incuba­tions. The mean cyclic nucleotide change, the standard error of this change (n number of experiments), the significance of the change from control values, and the percent change are given. Values of p were derived by paired two-tailed t test with n equal to the number of experiments. Each experiment, from which a mean difference was determined, consisted of 2 to 5 incubations containing the stimulatory agent and 2 to 5 containing medium and cells only. The percent changes in cyclic nucleotide content were calculated using the mean basal cyclic nucleotide concentration given at the bottom of the tables. RESULTS Effects of chemoattractants on neutrophil cyclic nucleotide concentra­tions. Culture filtrates of E. coli (BF) , in chemotactically active concen­trations, caused a significant increase in cGMP levels in the PMNs (Table 1). Since many of our previous studies have employed BF as a chemo­attractant, the characteristics of the cGNIP elevation caused by this agent were explored further. Figure lA shows the effect of different incubation conditions on the time course of cGMP accumulation in response to B F. When the cells were not preincubated (a condition similar to that used in the in vitro assay of chemotaxis) a delayed peak in cGMP concentration was observed. In contrast, after one hour of preincubation, BF produced a marked rise in cGMP within 5 minutes. The dose-response behavior of BF in preincubated and non-preincubated cells is shown in Figures IB and lC. High concentrations of the BF (15% vol/vol) were less active in producing cGMP accumulation than Wet-e interme diate concentrations (10%). eycl ic AMP changes in response to BF could not be determine d because the BF contained large amounts of this cyclic nucleotide (107 nM). Small amounts of cGMP (15 nM) were also present in BF. Several experiments were conducted to determine if the cyclic nucleotides in the BF affected its ability to stilnulate cGIv1P accumulation. No significant change in either Table 1. The effect of chenloattractants on cyclic nucleotide content in preincubated neutrophils. No. of cell No. of 6 a A fmoles / 10 cells preparations experiments Chemoattractant added cGMP cAMP cGMP cAMP Cyclic GMP Cyclic AMP Bacterial chemotactic factor (5 % vol/vol) 6 6 +23. 8 :±:3. 3 c (53%) not done fMet-Ala (10-3 M) 4 3 10 9 + 8. 0 .:t 3.2b (36%) + 5.4 + 3. 4 T rypsinize d human b b complement (2% vol/vol) 5 5 13 13 +16.0-+6.8 (71%) -19.2 + 8. 6 (-14 %) Zymosan-treated serum c (5% vol/vol) 1 4 +35.0 + O. 9 (74%) not done Cells were incubated for 10 minutes after the addition of chemoattractants. Mean basal cyclic nucleotide 6 concentrations were 22.4.:t 4.7 (SE) fmoles cyclic GMP / 106 cells and 140.0 i: 21. 4 (SE) fmoles cyclic AMP/ 10 cells. Basal cyclic GMP concentrations in BF and zymosan-treated serum experiments was 45.8 + 5. 5 (SE) fmoles / 106 cells. - aMean ± SE of cyclic nucleotide changes caused by the chemoattractant. Percent changes in cyclic nucleotide concentrations are given in parentheses. bps; 0.05 cps; 0 .. 001 t....:l 00 29 1. Accumulation of cGMP in hUlnan neutrophils in response to E. coli chemotactic factor (BF). A. The effect of preincubation on the time course of cGMP accumulation: _, control cells; 0, cells plus 5% BF. BF was added to cells at room temperature (left) and to cells that had been preincubated for 1 hour at 37° C (right). B. Time course of cGMP accum­ulation in cells not preincubated using different concentrations of BF: ., control cells, 0, cells plus 5% BF; A, cells plus 10% BF; 0, cells plus 15% BF. C. Dose-response of BF-induced cGMP accumulation when cells were preincubated for 30 minutes. A 5-minute incubation followed the addition of BF. Each point represents the mean ± SE of 4 assay tubes from 2 incubations. 30 ~ ~ 70 A. 'l: to t ~ ..... Control ~ 60 0-0 5% BF ~ 1\ \ ~ ~ 50 '-..: (.~15~1 40 t/~ + -f-I Hour t-"-tf G + 30 No Preincubation Preincubation 10 20 30 40 50 60 70 80 90 Time (min) ~ 90 B. ~ ~ 80 (() C) ""'- "- 70 ~ ~ 60 ~ /t t ~ 50 40 9... ~ 4 30 .~~ 20r~ ~ 30 ~ 20 101 , , i 10 20 30 40 50 60 0 5% 100/0 15% Time (min) BF Concentration 31 cGMP or cAMP was detected after incubation with either cyclic nucleotide (alone or in combination) in concentrations similar to those found in BF. It was concluded, therefore, that the cyclic nucleotides in BF are not respoIEible, by themselves, for the cGMP elevating properties of this chemotactic factor. Since the chemoattractants of BF have not been completely isolated or characterized (9), we sougl1t a more basic compound for further studies. Schiffmann and coworkers (10) have indicated that the low molecular weight N-forlnylnlethionyl peptides are capable of initiating chemotaxis in PMNs. One of these agents, fMet-Ala, was tested at chemotactic concentrations. A number of experiments perfonned in different cell preparations showed significant (p::::;; 0.05) rises in cGMP in response to -3 -4 10 M and 10 M fMet- Ala (Tables 1 and 2). Cyclic AMP concentrations remained essentially unchanged. Figure 2 shows the time course of neutrophil cGMP accumulation produced by different concentrations of fMet-Ala under two conditions of incubation. As in the case of BF, a more rapid rise in cGMP concentrations occurred in the preincubated cells. In addition to BF and fMet-Ala, two chemoattractants containing the active complement fragments C3a and CSa were tested for their effects on neutrophil cyclic nucleotides. Trypsinized human complement (2% vol/vol) caused significant elevation of cGMP concentrations to mean values 71% above control (Table 1). In addition, this agent slightly lowered cAMP levels. Zymosan-treated serum (5% vol/vol) also elevated cGMP concen- Table 2. Changes in neutrophil cyclic GMP concentrations produced by chemoattractants and modulators when cells were not preincubated. No. of cell No. of 6 a Agent added preparations expe riments L\ fmoles cGMP / 10 cells Chemoattractants Bacterial chemotactic factor (5% vol/vol) 4 4 +31. 8 ± 6. 5b (82%) -4 fMet-Ala (10 M) 3 3 +15. 3 ± 2. 4c (39%) Enhancin,K. chemotactic modulators -4 Phenylephrine (10 M) 3 3 c + 8. 5 ± 1. 2 (22%) Phorbol myristate acetate (10 ng / ml) 3 3 c + 5. 3 .± O. 6 (14%) Prostaglandin F2a: (10-8 M) 3 3 + 7. 5 ± 1. 9b (19%) -5 Carbachol (10 M) 3 3 b + 7.5 .± 2. 4 (19%) Cells were added at room temperature to tubes containing the agents then incubated at 370 C for 20 minutes. Mean basal cGMP concentration was 39.0 + 6. 8 (SE) fmoles / 106 cells. aMean ± SE of cGMP changes caused by-the agent. Percent changes in cGMP concentrations are given in parentheses. bps: O. 05 cps: O. 025 W N 33 Figure 2. Time courses of cGMP accumulation in neutrophHs caused by N-formylmethionylalanine (fMet-Ala) in preincubated and non-preincubated cells. A. Neutrophils were preincubated for 20 minutes: ., control cells; 3 -4 o , cells plus 10- M fMet-Ala, 0, cells plus 10 M fMet-Ala. B. Time course of cGMP accumulation caused by fMet-Ala in cells that were not -3 preincubated: 0, unstimulated cells; 0, cells plus 10 M fMet-Ala; ., -4 -5 cells plus 10 M fMet-Ala; 0, cells plus 10 M fMet-Ala. Each point represents the mean of two incubations With 2 assay tubes per incubation. * Point is significantly elevated (p :::;; 0.05) above corresponding control value. 60-r- .... 10 Time (min) "*t o- 4M fMet-Ala [*J 10:- 5 M fMet-Ala ~ 2 ~;...--...--_:/o. /-Control Io ---- / *~n.Q 05 10 20 Time (min) 30 34 35 trations. In contrast, untreated human serum or whole human complement to which soybean trypsin inhibitor was added prior to the addition of trypsin, had no measurable effect on cyclic nucleotide concentration. Effects of chemotactic modulators on neutrophil cyclic nucleotide concentrations. Because the enhancing chemotactic modulators have been assumed to function by stimulating the accumulation of cGMP, it was hoped that an understanding of their effect on cellular cGMP might clarify the role of this cyclic nucleotide in chemotaxis. Phorbol myristate acetate, prostaglandin F2a, phenylephrine, and carbachol were therefore examined for their effect on cGMP levels in PMNs. Concentrations of these agents were chosen which enhanced the chern otactic response of PMNs to BF (1,2). Each chemotactic modulator, when added alone to incubations, produced a significant rise (p ~ ,0. 05) in cGMP levels. This elevation was generally small, however, with the mean ranging from 20-34% (Tables 2-3). Cyclic AMP concentrations remained unchanged after addition of these agents. Representative time courses of cGMP accumulation in PMNs following stimulation with phorbol myristate acetate and carbachol are shown in Figure 3. TIle time course of cGMP accumulation produced by the chemo­tactic modulators approximated that seen by fMet-Ala. Peak elevations in cGMP were usually seen between 5 and 20 minutes in preincubated cells and between 10 and 30 minutes in cells that were not preincubated. Tables 2 and 3 show that the magnitude of the cGMP rise that occurred in cells not preincubated was approxinlately the same as that seen in preincubated Table 3. The effect of chemo tactic modulators on cyclic nucleotide content in preincubated neutrophils No. of cell No. of 6 a A fmoles / 10 cells preparations experiments Nlodulator adde d cGMP cAMP cGlVIP cAMP Cyclic GlVIP Cyclic AMP --" Carbachol (10-5 M) 7 7 14 14 e + 7. 5 .± 1. 7 (34%) - 9. 8 + 7.4 -4 Phenylephrine (10 M) 5 3 10 8 b + 4. 8 :t 1. 6 (21%) - 1.2 + 3.4 Phorbol myristate acetate (10 ng/ml) 5 3 10 8 + 4. 6 .± 2. 0b (20%) - 8.2 + 4. 6 -8 Prostaglandin F 2a (10 M) 7 5 12 13- + 4. 9 ± 2. IJJ (22%) - 1. 5 + 4. 9 Isoproterenol (10 -3 M) 5 5 11 d e 11 +33. 6 ± 7. 8 (15C1%) +50. 7 ± 9. 1 (37%) Prostaglandin E 1 (10 -6 M) 2 2 4 4 +16.2 + 8. 8 c +80. 8 ± 15. 7 (58%) Cells were incubated for 10 minutes after the addition of modulators. Mean basal cyclic nucleotide concentrations were 22.4 + 4. 7 (SE) fmoles cyclic GMP / 106 cells and 140. 0 + 21. 4 (SE) fmoles cyclic AMP / 106 cells. aMean [SE of cyclic nucleotide changes caused by the chemoattractant. Percent changes in cyclic nucleotide concentrations are given in parentheses. b p ~ 0.05 c d p = 0.01 e p = o. 005 p ~ 0.001 UJ 0\ 37 80 Phorbol myristate acetate / (10 ng/ml) t- -~ Carbachol (I0-5MI I 50+-------~~------~--------__ 10 20 30 Time (min) Time courses ofGGNIP accumulation in neutrophils produced by carbachol and phorbol myristate acetate: 4), unstimulated neutrophils; 0, neutrophils stimulated with 10-5 M carbachol; 0, neutrophils plus 10 ng / ml of phorbol myristate acetate. Cells were preincubated for 20 minutes at 370 C before addition of the chemotactic modulators. Each point represents the mean ± SE of 3 incubations with 2 assay tubes per incubation. 38 cells. Prostaglandin E 1 and isoproterenol, agents which depress chemotaxis in PMNs, were also examined for their effects on cAMP and cGMP levels (Table 3). Both agents caused significant increases in cAMP (+50 and +80 fmoles / 10 6 cells). In addition, isoproterenol caused a significant increase in cellular cGMP (+33 fmoles /106 cells). A time course of cAMP accum- -3 ulation in response to 10 M isoproterenol is shown in figure 4 .. Effects of combined chemoattractants and chemotactic modulators on cGMP levels. The mechanism of enhancement of cell migration by chemo-tactic modulators was explored by determining cGMP changes in response to combinations of chemoattractants and modulators. Previous studies from our laboratory (1,2) showed an enhancement of chemotaxis with the modu-lators when a submaximal chemotactic stimulus (5% BF) was present on the attractant side of the Boyden chamber. Since dose response studies with BF (Figure 1) showed that 5% BF was also a submaximal stimulus for elevating cGMP in the PMN, we expected incubations containing 5% BF plus the chemotactic modulators to show higher cGMP levels than those seen with 5% BF alone. Table 4 shows, however, that 5% BF and the modulators did not exert synergistic or additive effects on peak cGIv1P concentrations. At a concentration of one percent, BF had less of an effect on cGMP levels, and in this case the presence of the chemotactic modulator usually resulted in a larger increase in cGMP than the chemoattractant alone. Experiments using trypsinized human complement and fMet-Ala did not show an additive or synergistic effect on cGMP accun1ulation (Table 4). 39 ~320 ~ C( (IO-3 M) ~ 300 Q "~ 280 r ~ ~ ~ 260 ~ ~ 240 .(.) ~ ~ ~ 220 Control + t I .200 5 10 15 Time (min) . Figure 4. Time course of cAMP accumulation in neutrophils produced by isoproterenoL Cyclic AMP concentrations of cells stin1ulated with 10-3 M isoproterenol, 0; and cells incubated in medium alone, 8; are shown. This experiment was performed at 240 C. Each point represents the mean .± SE for 3 incubations with 2 assay tubes per incubation. 40 Table 4. Cyclic GMP effects of combined chemoattractants and modulators. Peak elevation of c GMP (Percent of control) Chemoattractant Modulator Exp. Agents alone alone Combination l. BF (5%) + PMA * +25 +13 +21 BF (5%) + PGF 2a +10 +14 BF (5%) + Phenylephrine +14 +16 BF (5%) + Carbachol +25 +7 2. BF (1%) + PMA +11 +24 +32 BF (1%) + PGF2a +41 +43 BF (1%) + Phenylephrine +45 +32 BF (1%) + Carbachol +20 +15 3. TC (2%) +PMA +35 +29 +42 TC (2%) + PGF2a +15 +26 TC (2%) + Phenylephrine +5 +26 TC (2%) + Carbachol +13 +17 4. fMet-Ala (10-5 M) + Phenylephrine +27 +21 +29 Cells were incubated for 10 and 20 minutes after simultaneous addition of the agents. Peak cGMP elevations, which usually occurred after 10 minutes, are represented. Concentrations of the modulators and incubation conditions are the same as in Table 3, except for experiment 4 which was performed without preincubation. Each percent value is derived from the mean cGMP elevation above control of triplicate incubations. it: Abbreviations: PMA, phorbol myristate acetate; PGF2a, prostaglandin F 2a; TC, trypsinized human complement. DISCUSSION The present studies indicate that BF, fMet-Ala, and trypsinized human complement cause accumulation of cGNlP in neutrophil suspensions in concentrations similar to those which initiate chemotactic activity. Ignarro and George have shown (11), and we have confirmed, that zymosan-treated serum elevates neutrophil cGNlP. The fact that cGMP is elevated by all of the chemoattractants so far examined, indicates that an increase in the concentration of this cyclic nucleotide might be one of the necessary events in initiating cell movement. TIlat it is not the only essential event is indicated by the fact that the enhancing chemotactic modulators stimulate cGMP accumulation but do not initiate chemotaxis. Phorbol myristate acetate, prostaglandin F2a, phenylephrine, and carbachol, in concentrations which enhance the chemotactic response, caused small but significant increases in cG:MP in the absence of any added chemoattractant. No accompanying change in cAlvIP was seen in response to these agents. In findings similar to ours, Zurier and coworkers (5) have observed an elevation of neutrophil cGMP with carbachol alone (10- 8 and 6 . 10 M). These data suggest that the presence of a chemoattractant IS not required in order for the enhancing modulators to elevate cG:MP concen-trations. Zigmond and Hirsch (12), however, have shown that upon contact with glass, PMNs release factors which can attract other PMNs. 42 It may, therefore) be impossible to eliminate all chemoattractants from suspensions of these cells. Sandler et a1. (14) have reported that carbachol alone (10- 8 M) elevates cGMP in human mononuclear leukocytes but were unable to n1easure an effect of this agent on cGMP concentrations of PMNs. Ignarro and George (4) demonstrated a marked effect of acetylcholine (10-7 M), but only in the presence of zymosan-treated serum. It appears that PMNs purified by glass adherance, as in the method used by Ignarro and coworkers (4,7, 11), may have higher basal concentrations of both cyclic nucleotides than PMNs separated by the ficoll-hypaque technique (5,14, 15, 16). TIle magnitude of changes seen in response to stimulatory agents might be affected by the basal cyclic nucleotide content. We have observed low basal cyclic nucleotide concentrations as well as small cyclic nucleotide changes when compared with previous reports (4,5, 13). Zurier and coworkers have shown (5), and we have confirlned, that prostaglandin El and isoproterenol elevate cAMP concentrations in PMNs. Vie also found that cGNIP levels, in addition to cAMP levels were raised by concentrations of is oproterenol that are inhibitory to chemotaxis. The absolute rise (in fmoles) in cAMP levels in response to this agent was greater than the absolute rise in cGMP, however. Isoproterenol has also been shown to elevate cGMP concentrations in rat heart (17) and pineal (18), and in the mouse parotid gland (19). It is possible that cGMP effects are masked when a simultaneous rise in cAMP occurs. Further experiments will be necessary to verify this hypothesis. We have also found that the modulators can enhance the cGMP accum­ulation produced by low (1%) but not higher (5% vol/vol) concentrations of BF (Table 4). It seems likely that early in the chemotaxis assay the concentration of the chemoattractant at the surface of the migrating PMNs is low (due to the gradient establishe d) and, therefore, has a minimal 43 effect by itself on the cellular level of cGMP. h1 contrast, the enhancing modulators, which are present on the cell s ide of the Boyden chamber, increase the level of cGMP and enhance the chemotactic response. This idea is supported by the investigations of Zigmond and Hirsch (12),who reporte d that a concentration gradient of chemoattractant across the membrane of the Boyden chamber persists with little deterioration for at least 30 minutes. Other mechanisms for the enhancement of chemotaxis can also be advanced. The rate of increase or the duration of the cGMP response might be more important than the peak elevations. Smith and Ignarro (20) report an increased rate of cGNlP accumulation in incubations containing zymosan-treated serum and acetylcholine when compared to those containing zymosan-treated serum alone. The timing of the addition of the chemoattractants and the chemotactic modulators may also be critical. Ignarro and coworkers (4,20) and Zurier et al. (5) preincubated PMNs with modulators prior to adding zymosan-treated serum. We elected to add the agents simultaneously, because this procedure was used in our previous studies of the modulation of PMN chemotaxis (1). It should also be noted that the experiments reported here were performed on suspended 44 cells while the cells undergoing in vitro chemotaxi.s are in contact with a micropore filter. Becker et ale (21) showed that PMNs attached to micro­pore filters in the presence of chemoattractants release lysosomal enzymes. Large differences might therefore be expected between suspended and attached cells. Vie have sought to determine how cyclic nucleotide changes that occur in neutrophils might be involved in the behavior of these cells in the in vitro chemotaxis assay. Results presented here support the hypothesis that cGMP and cAMP exert opposing effects in the regulation of neutrophil chemotaxis. Agents previously studied which enhanced chemotaxis were found to increase cellular cG:rviP, while agents which depressed chemotaxis elevate cAMP concentrations. In addition, all of the chemoattractants we have studied elevate cGMP without affecting cAMP concentrations, suggesting that cGMP may be important in both initiating and modulating cell movement. REFERENCES 1. Hill, H. R., R. D. Estensen, P. G. Quie, N. A. Hogan, and N. D. Goldberg. 1975. Modulation of human neutrophil chemotactic responses by cyclic 3', 5'-monophosphate and cyclic 3',5' guanosine monophosphate. Metabolism 24:447 -456. 2. Estensen, R. D., H. R. Hill, P. G. Quie, N. A. Hogan, and N. D. Goldberg. 1973. Cyclic GMP and cell movement. Nature 245:458- 460. 3. Rivkin, 1., J. Rosenblatt, and E. L. Becker. 1975. The role of . cyclic AMP in the chemotactic responsiveness and spontaneous motility of rabbit peritoneal neutrophils. J. Immunol. 115:1126-1134. 4. Ignarro, L. J., and W. J. George. 1974. Mediation of immunologic discharge of lysosomal enzymes from human neutrophils by guanosine 3',5' -monophosphate. J. Exp. Med. 140:225-238. 5. Zurier, R. B., G. Weissmann, S. Hoffstein, S. Kammerman, and H. H. Tai. 1974. Mechanisms of lysosomal enzyme release from human leukocytes. II. Effects of cyclic AMP and cyclic GMP, autonomic agonists, and agents which affect microtubule function. J. Clin. Invest. 53:297-309. 6. Ward, P.A., I.H. Lepow, and L.J. Newman. 1968. Bacterial factors chemotactic for polymorphonuclear leukocytes. Am. J. Pathol. 52:725-736. 7. Ignarro, L. J., T. F. Lint, and W. J. George. 1974. Hormonal control of lys os omal enzyme release from human neutrophils. Effects of autonomic agents on enzyme release, phagocytos is, and cyclic nucleotide levels. J. Exp. Me d. 139: 1395-1414. 46 8. Becker, E. L. 1972. The relationship of the chemotactic behavior of the complement- derived factors C3a, C5a, and C567 and a bacterial factor to their ability to activate proesterase of rabbit polymorpho­nuclear leukocytes. J. Exp. Med. 135:376-387. 9. S chiffm ann , E., H. V. Showell, B. A. Corcoran, P. A. Ward, E. Smith, and E. L. Becker. 1975. TIle isolation and partial character­ization of neutrophil chemotactic factors from E. coli. J. Immunol. 114:1831-1837. 10. Schiffinann, E., B. A. Corcoran, and S. M. Wahl. 1975. N-formyl­methionyl peptides as chemoattractants for leukocytes. Proc. Natl. Acad. Sci. U. S. A. 72:1059-1062. 11. Ignarro, L. J. and W. J. George. 1974. Hormonal control of lysosomal enzyme release from human neutrophils: Elevation of cyclic nucleotide levels by autonon1ic neurohorm one s . Proc. Natl. Acad. Sci. U.S.A. 71:2027-2031. 47 12. Zigmond, S. H., and J. G. Hirsch. 1973. Leukocyte locomotion and chemotaxis. J. Exp. Med. 137:387-410. 13. Sandler, J. A., J. I. Gallin, and M. Vaughan. 1975. Effects of serotonm, carbamylcholine, and ascorbic acid on leukocyte cyclic G:MP and chemotaxis. J. Cell. Biol. 67:480-484. 14. Sandler, J. A., R. 1. Clyman, V. C. Manganiello, and M. Vaughan. 1975. 11le effect of serotonin (5 ... hydroxytryptamine) and derivitives on guanosine 3', 5'-monophosphate in human monocytes. J. Clin. Invest. 55:431-435. 15. Sandler, J. A., T. K. Smith, V. C. Manganiello, and C. H. Kirk­patrick. 1975. Stimulation of monocyte cyclic GMP by leukocyte dialysates. J. elin Invest. 56:1271-1279. 16. Boxer, L. A., A. M. Watanabe, M. Rister, H. R. Besch, J. Allen, and R. L. Baehner. 1976. Correction of leukocyte function in Chediak­Higashi syndrome by ascorbate. N. Engl. J. Med. 295:1041-1045. 17. Amer, M. S., and J. E. Byrne. 1975. Interchange of adenyl and guanyl cyclases as an explanation for transformation of beta to alpha adrenergic responses in the rat atrium. Nature 256:421-424. 18. OrDea, R. F., and M. Zatz. 1976. Catecholamine stimulated cyclic G:MP accumulation in the rat pineal: Apparent presynaptic site of action. Proc. Natl. Acad. Sci. U. S. A. 73:3398-3402. 48 19. Durham, J. P., R. Baserga, and F. R. Butcher. 1974. The effect of isoproterenol and its analogs upon adenosine 3',5' -monophosphate and guanosine 3',5' -monophosphate levels in mouse parotid gland in vivo. Biochim. Biophys. Acta 372:196-217. 20. Smith, R. J. and L. J. Ignarro. 1975. Bioregulation of lysosomal enzYlne secretion from human neutrophils: Roles of cyclic GMP and calcium in stimulus secretion coupling. Proc. Natl. Acad. Sci. U.S.A. 72:108-112. 21. Becker, E. L., H.J. Showell, P. M. Henson, and L. S. Hsu. 1976. TIle ability of chemotactic factors to induce lysosomal enzyme release. I. Characteristics of the release, the importance of surfaces and the relation of enzyme release to chemotactic responsiveness. J. Immuno!. 112:2047-2054.