Tumor Necrosis Factor-induced Protein Phosphorylation in Human Neutrophils Jeffrey J. Crowley and Thomas A. Raflin Division of Pulmonary and Critical Care Medicine, Department of Medicine, Stanford University School of Medicine, Stanford, California

Protein phosphorylation is central to multiple regulatory processes in cells. Tumor necrosis factor (TNF) , a cytokine synthesized by macrophages, effects polymorphonuclear leukocyte (neutrophil) chemotaxis, induces superoxide anion generation, and mediates neutrophil adhesion to endothelial cells. Although protein phosphorylation is almost certainly involved in many TNF-mediated neutrophil functions, little is known about TNF's impact on neutrophil protein phosphorylation. Therefore, we studied human recombinant TNF-ex-induced protein phosphorylation in human neutrophils. Neutrophils were preincubated with 32POa- and treated with a variety of stimulatory agents. One- and two-dimensional polyacrylamide gel electrophoresis was used to analyze phosphorylated proteins. Phosphoaminoacids were identified by two-dimensional thin layer chromatography electrophoresis. The findings were as follows: (1) TNF induces the phosphorylation oftwo 16-kD proteins (pI = 5.9 and 6.1) by 5- to 6-fold, and a 57-kD protein (pI = 5.8) by 3- to 4-fold compared with untreated neutrophils; (2) these proteins are phosphorylated as early as 15 min after stimulation with TNF, and phosphorylation is induced by concentrations ofTNF as low as 1 ng/ml (10 U/ml); (3) TNF induces the phosphorylation of proteins at either serine or threonine residues and not at tyrosine; (4) TNF-stimulated neutrophils show a unique pattern of protein phosphorylation when compared to neutrophils treated with formylmethionylleucylphenylalanine; (5) lipopolysaccharide does not induce protein phosphorylation in neutrophils; (6) a 16-kD protein is phosphorylated in response to TNF in neutrophils but not in mononuclear cells; and (7) protein kinase inhibitors appear to have no effect on TNF-induced protein phosphorylation. Thus, the mechanism of action of TNF on neutrophils may involve protein phosphorylation.

Human tumor necrosis factor-ex (TNF), a 17-kD protein produced by macrophages, was originally defined by its tumorkilling activity both in vitro and in vivo (1, 2). TNF is now identified as a cytokine that influences growth, differentiation, and cellular function in many cell types and has effects in vivo (2-5). Our current interest in TNF stems from work in guinea pigs showing that injection of TNF induces acute lung injury and a syndrome similar to the adult respiratory distress syndrome (ARDS) (6-8). An expanding body of data suggests that polymorphonuclear leukocytes (neutrophils) are involved in acute lung

(Received in originalform December 7, 1990 and in revisedform February 21, 1991) Address correspondence to: Thomas A. Raffin, M.D., Chief, Division of Pulmonary and Critical Care Medicine, Department of Medicine, Stanford University Medical Center, Stanford, CA 94305-5236. Abbreviations: adult respiratory distress syndrome, ARDS; cyclic adenosine monophosphate; cAMP; formylmethylleucylphenylalanine, FMLP; isoelectric focusing, IEF; lipopolysaccharide, LPS; polyacrylamide gel electrophoresis, PAGE; isoelectric point, pI; protein kinase C, PKC; phorbol myristate acetate, PMA; phosphoprotein, pp; sodium dodecyl sulfate, SDS; tumor necrosis factor, TNF. Am. J. Respir. Cell Mol. BioI. Vol. S. pp. 284-291, 1991

injury and ARDS (9, 10). Histologic studies in both animals and humans have demonstrated that neutrophils are present in high numbers in the pulmonary capillary bed and the alveoli during acute lung injury (11, 12). Further, neutrophils and their products were identified in increased numbers in the bronchoalveolar lavage fluid of patients with ARDS compared with control subjects (13, 14). Additionally, leukopenia prevents acute lung injury in TNF-treated guinea pigs (15). However, neutrophils do not solely mediate tissue injury; for example, leukopenic guinea pigs develop significant acute lung injury in response to sepsis, and ARDS can occur in neutropenic patients (16, 17). In the neutropenic patient, tissue injury is most likely due to humoral factors such as fibrin, fibrin degradation products, activated complement products (C5a) , leukotrienes, and platelet-activating factor (PAF) (11). Therefore, TNF is capable of inducing acute lung injury, and neutrophils may playa key role in mediating lung injury. Studies in isolated neutrophils show that TNF stimulates oxygen free radical production, effects chemotaxis, and stimulates neutrophil adherence to endothelial cells (18-21). TNF may also "prime" neutrophils for response to other stimuli. For example, neutrophils pretreated with TNF show a dramatically potentiated response to formylmethylleucyl-

Crowley and Raffin: TNF-induced Protein Phosphorylation in Human Neutrophils

phenylalanine (FMLP) (22). Additionally, TNF stimulates neutrophil-mediated damage to endothelial cells (23) and induces degranulation of primary and secondary granules in adherent neutrophils but not in neutrophils in solution (24, 25). Furthermore, TNF-induces hydrogen peroxide release in neutrophils bound to endothelial cells or to an artificial surface but not to neutrophils in solution (26). Thus, TNF appears to effect multiple neutrophil activities and its effects may be influenced by adhesion of the neutrophil to a surface. FMLP stimulates neutrophil chemotaxis, superoxide anion generation, and degranulation. The proposed mechanism of neutrophil activation by FMLP involves specific binding to an extracellular receptor and activation of a Gprotein that in turn activates phospholipase C (27-29). Phospholipase C then cleaves phosphatidylinositol into diacylglycerol and inositol triphosphate, which leads to the release of calcium from intracellular stores, activation of protein kinases, and protein phosphorylation (30). The functional end points of this signal transduction cascade are superoxide generation and degranulation of primary and secondary granules. The biochemical mechanisms involved in TNF signal transduction, however, are poorly understood. Several researchers have shown that TNF binds specifically to highaffinity receptors on the neutrophil (20, 31). After TNF receptor binding, there is an increase in membrane fluidity and permeability (32, 33) and a possible subsequent influx of calcium into the cytoplasm from outside the cell. TNF, however, does not appear to alter cytosolic free calcium concentrations in the neutrophil and does not lead to the release of calcium from intracellular stores (33). The mechanism by which TNF activates neutrophil functions is not known. In many hormone receptor systems, protein phosphorylation is an essential component of signal transduction. TNF effects the expression of regulatory genes in multiple cell types (34, 35). In one study, protein kinase inhibitors blocked TNFmediated induction of its own gene. In U937 cells, TNF induces the specific serine phosphorylation of a 26-kD protein that may be involved in cellular differentiation (36). Thus, protein kinases and protein phosphorylation may playa role in TNF-mediated signal transduction in the neutrophil. The goal of this study was to demonstrate and characterize TNFinduced protein phosphorylation in the neutrophil.

Materials and Methods Neutrophil Isolation Neutrophils were isolated from healthy, drug-free human donors. Informed consent was obtained from volunteer donors, and the project protocol was approved by the Stanford Hospital Human Subjects Committee. Whole blood was collected in heparin-treated syringes, diluted 1:1 with phosphate-buffered saline (PBS) (pH 7.4), and layered onto a Ficoll-Hypaque gradient (specific gravity, 1.077, 1.119; Sigma Chemical Co., St. Louis, MO). The cells were centrifuged at 1,800 X g for 30 min. The neutrophil-rich band was collected, pooled, and pelleted. The pellet was resuspended and subjected to a single hypotonic lysis to remove red blood cells (37). The cells were pelleted and resuspended in PBS. The neutrophils were counted using a hemocytometer and Diff-Quik'" (Sigma). All neutrophil preparations were> 98 % pure and > 99 % viable by trypan blue exclusion.

285

Mononuclear Cell Isolation Human mononuclear cells were isolated from blood simultaneously with neutrophil isolation. The mononuclear cell band was removed from the Ficoll-Hypaque gradient and washed with PBS (37). The cells then underwent hypotonic lysis and were resuspended in PBS. All mononuclear cell isolates contained < 1% granulocytes and were> 99% viable by trypan-blue exclusion. Materials Human recombinant TNF was received as a gift from Cetus Corp. (Emeryville, CA), lot no. NP-20m. This TNF is > 95% electrophoretically pure, contains < 0.2 ng lipopolysaccharide (LPS)/mg of protein, and the activity is 10 U/ng. Phorbol myristate acetate (PMA) , LPS (Escherichia coli 055:B5), FMLP, and W-7were purchased from Sigma. Antibody to TNF was purchased from Endogen Co. (Boston, MA). Protein kinase inhibitors (H-7, H-8, H-9, and HA1004) were purchased from Seikagaku (Tokyo, Japan; 38). Incubation with

32P~-

Cells were adjusted to 1 X 107/ml and suspended in phosphate-free medium (20 mM Hepes [pH 7.4], 120 mM NaCl, 5 mM KCl, 1 mM MgCI2, and 5 mM glucose; 36). The cells were then incubated with 32PO~- (20 j.tCi/ml; New England Nuclear-Dupont, Boston, MA) for 30 min at 37° C. The cells were then pelleted and washed with phosphate-free medium. Protein Isolation from Whole Cell Lysates 32PO~--labeled cells were incubated with TNF for 15 min unless otherwise stated. The reaction was stopped by addition of trichloroacetic acid to 12% (vol/vol) and incubated on ice for a minimum of 3 h for protein precipitation. The lysate was spun in a microfuge for 15 min, washed 4 times with ice-cold acetone, and dried. The pellet was resuspended in either polyacrylamide gel electrophoresis (PAGE) sample buffer (62.5 mM Tris [pH 6.8], 9% sodium dodecyl sulfate [SDS], 10% glycerol, 0.02% bromphenol blue) or isoelectric focusing (lEF) buffer (39) for two-dimensional analysis. The isolates were spun in a microfuge for 5 min to remove any particulate matter. Protein concentrations were determined by a spectrophotometric kit (Bio-Rad, Richmond, CA). Samples were then adjusted to identical protein concentrations prior to loading on gels for electrophoresis. Electrophoresis One-dimensional SDS-PAGEwas performed as described by Laemmli (40). Briefly, an equal amount of protein was loaded onto 16% polyacrylamide gels. All products for SDSPAGE were purchased from Bio-Rad. Standard proteins, in the range of 14 to 97 kD, were purchased from Sigma. Twodimensional PAGE was performed as described by O'Farrell (39) using IEF in the pH range of 3 to 10 in the first dimension (ampholines and other reagents from Bio-Rad) followed by SDS-PAGEon 14% gels in the second dimension. Protein isoelectric points (pI's) were determined by sectioning IEF gels in I-em slices, immersing them in distilled water, and measuring pH.

286

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 5 1991

a

b

- 97 - 66 - 431') I

0

- 31 x L.. I:

-22 -14 n

0 Z

... Xl 0

r-

...z ...z .......,z ...z ."

~

"""

-" -

."

""

0

" ""

- 97 - 66

• •

- 43

I

0

- 31 x L..

I:

-22

•

-14

...z ...z ...z ...z ...z n ... - '" a- ...a3 3 '" 3 3 3 '" 3

n

."

0 Z

0 0

Xl 0

-

r

" ""

."

."

"

"

."

"

."

."

0

0

"

"

VISualization After electrophoresis (both one- and two-dimensional), the gels were stained with 0.01% Coomassie blue (40). Gels were destained, and dried onto Whatrnan 3-mm paper using a Bio-Rad gel dryer. The gels were exposed to Kodak XAR film and developed. For quantification of 32Pcy'- incorporation into proteins, phosphoproteins, identified by autoradiography, were cut out of two-dimensional gels and counted in a liquid scintillation counter. Gel slices from TNF-treated neutrophils were compared with controls, background counts were subtracted, and the data were expressed as percentage of control. Phosphoaminoacid Analysis Phosphoaminoacid analysis was performed by the method of Cooper and associates (41). Phosphoproteins, localized by autoradiography, were excised from dried and exposed twodimensional gels and eluted from the gel using an Isco electrophoretic concentrator (Lincoln, NE) . The concentrated protein was then precipitated by addition of trichloroacetic acid to 12 % (vol/vol), incubated on ice for 3 h, washed with acetone, and dried. The pellet was then hydrolyzed in 7.5 N

- 97 - 66

- 43 - 31

'"0 )(

- 22 r:'-

- 14

!'"l

Figure 2. The effects of agents on neutrophil protein phosphorylation. In this experiment, human neutrophils were incubated with 100 ng/ml TNF, 10-6 M formylmethylleucylphenylalanine (FMLP), 10-7 M phorbol myristate acetate (PMA), and 10 JLg/rnl E. coli (LPS)for 15min. Proteins were isolated, concentrationsadjusted, andloadedonto a 16%gel forPAGE. TNF,FMLP,andPMA induced the phosphorylation of 16-, 32-, and 47-kD peptides.

0 Z 0

"

Figure 1. Concentration and time dependence of tumor necrosis factor (TNF)-induced protein phosphorylation. In panel a, human neutrophils preloaded with l2PO~- were incubated with 0.1 to 100 ng/rnl (1 to 1,000 UIrnl) of TNF. Reactions were stopped at 15 min with trichloroacetic acid, and proteinswere isolated. Equal amounts of protein were loaded onto each lane, and polyacrylamide gel electrophoresis (PAGE) was performed on 16% gels. The migration of standard proteins (14 to 97 kD) is shown vertically. TNF, at a concentrationof 1 ng/rnl or greater, inducedthe phosphorylation of at least three proteins (16, 32, and 47 kD). In panel b, neutrophils were incubated with 100 ng/rnl TNF for 1 to 60 min . The sample marked CONTROL was stopped immediately after addition of TNF, and the sample marked CONT 60mn was incubated without TNF for 60 min. Proteins were isolated similarly to above. TNF induced the phosphorylationof 16-, 32-, and 47-kD peptides after incubation for 5 to 15 min (see arrows).

hydrochloric acid for 1 hat 110° C in a vacuum oven. The HCI was removed by centrifugation in a Speed Vac (Savant Co.) over NaOH pellets. The pellet was suspended in pH 1.9 buffer (formic acid:acetic acid: water, 25:78:897, vol/vol). The amino acids were spotted onto cellulose thin layer chromatography plates (Sigma), and electrophoresis was performed at pH 1.9 (750 V, 90 min) in one dimension and at pH 3.5 (500 V, 60 min; pyridine:acetic acid:water, 1:10:89, vol/vol) in the other dimension. Standards containing 1 mglrnl each of phosphoserine, phosphotyrosine, and phosphothrconinc were added to each sample. Standard amino acids were visualized by reaction with ninhydrin, and radioactive amino acids by autoradiography (42).

Results Figure 1a is an autoradiograph of a one-dimensional polyacrylamide gel of cellular proteins pre incubated with 32PO:and then stimulated with TNF for 15 min. TNF, at concentranons from I to 100 nglrnl (10 to 1,000 U/ml) , increased the phosphorylation of 16-, 32-, and 47-kD proteins. Thus, TNF-induced protein phosphorylation was dose dependent and occurred at concentrations as low as 10 U/m!. Figure lb shows the time course of phosphorylation for 16-, 32-, and 47-kD proteins. Phosphorylation occurred as early as 5 min and continued for at least 60 min after addition of TNF. Background phosphorylation of these proteins in unstimulated cells incubated for 60 min was negligible, as demonstrated in Figure lb. Other stimuli also induced protein phosphorylation in the neutrophil. FMLP and PMA also stimulated the phosphorylation of 16-, 32-, and 47-kD proteins, as shown in Figure 2. Interestingly, LPS did not induce the phosphorylation of these proteins under these conditions. Two-dimensional analysis of control, 100 nglrnl TNFstimulated, and 10-6 M FMLP-stimulated neutrophils is shown in Figure 3. Two-dimensional analysis of proteins allows the separation of proteins based on molecular weight and pH . Therefore, two-dimensional gels allow one to ob-

Crowley and Raffin: TNF-induced Protein Phosphorylation in Human Neutrophils

- 97 - 66

287

- 97 - 66

Figure 4. Effectof proteinkinase inhibitors on TNF-induced neutrophil protein phosphorylation. Neutrophils were preincubated with 100 /LM and 200 /LM H-7 and W-7 or with bufferfor 10 min prior to the addition of 10 nglml TNF for an additional 15 min. Proteins were isolated, loaded onto a 16% gel, and PAGE was performed. W-7 and H-7 had no effect on the phosphorylation of ppl6.

- 43

- 31

-22

- 14

- 97

- 43

'",o

- 31 -

)(

-22 ~

- 14

- 66

- 43

,..., 1

s

- 31

x

,

1:

-22

-14

- 97 - 66

- 43 - 31

-22

ACIDIC

-- - --)

BAS IC

Figure 3. Two-dimensional PAGE of FMLP- and TNF-stimulated neutrophils. Neutrophils were incubated with 100 ng/ml TNF or 10-6 M FMLP for 15 min. Proteins were isolated, and twodimensional PAGE performed as described in MATERIALS AND METHODS. The horizontal axis presentspH (approximate pH, 4 to 8), and the vertical axis molecularmass. BothTNF and FMLP induced the phosphorylation of several proteins. The arrow on the FMLP gel represents a set of proteins not phosphorylated by TNF.

serve changes in the phosphorylation of individual proteins. Proteins were separated first by IEF and then by SDS-PAGE. TNF induced the phosphorylation of several proteins compared with control and is discussed fuTther below. FMLP and TNF both induced the phosphorylation of some similar proteins; however, FMLP induced the phosphorylation of several proteins that TNF did not (see arrow in Figure 3). Additionally, 10-6 M FMLP, compared with 100 ng/ml TNF, was a more potent stimulus for protein phosphorylation in the neutrophil. In Figure 4,neutrophils were preincubated with high concentrations of the protein kinase inhibitor H-7 and the calcium-calmodulin kinase inhibitor W-7 and then stimulated with TNF. Neither of these inhibitors affected the phosphorylation of phosphoprotein (pp) 16. When neutrophils were pretreated with 200 /LM W-7 or H-7, however, other

proteins, including a 32-kD and a 47-kD protein, showed a decrease in phosphorylation compared to cells treated with TNF alone . High concentrations of these inhibitors, however, may nonspecifically decrease neutrophil protein phosphorylation. Indeed, a decrease in total phosphorylation was seen in neutrophils pretreated with 200 /Lmol H-7 or W-7. In data not shown, the protein kinase inhibitors H-8, H-9, and HA-I004 at 100 /Lmol had no effect on TNF-induced phosphorylation of pp16, pp32, and pp47. Concentrations of all agents used in these experiments were well above those known to inhibit several protein kinases in cells (38, 43). Therefore, all these agents had little or no effect on TNFinduced neutrophil protein phosphorylation at concentrations known to inhibit PKC but may, at higher concentrations, effect the phosphorylation of pp32 and pp47. TNF-induced phosphorylation of neutrophils and mononuclear cells is compared in Figure 5. TNF-induced phosphorylation of pp16 was observed in neutrophils but not in mononuclear cells. Therefore, TNF-induced phosphorylation of this 16-kD protein is not present in all cells and may be specific to the neutrophil. Additionally, TNF did not induce or alter the phosphorylation of any protein in the molecular mass range of 14 to 35 kD in mononuclear cells. In order to further characterize the proteins phosphorylated in neutrophils stimulated by TNF, we performed twodimensional gel electrophoresis. Figures 6a and 6b are autoradiographs of two-dimensional gels of proteins from 32PO~- -loaded neutrophils alone and neutrophils incubated

- 97

Figure 5. Neutrophils (106 /ml) and mononuclear cells (lQ6/ml) wereincubated withbufferor 100 nglml TNF for 15 min. In contrast to previous experiments, samples were adjusted to reflect equal amounts of incorporated 32PO;- and then loaded onto a 16% gel for sodium dodecyl sulfate PAGE. TNF induced the phosphorylation of 16-, 32-, and 47-kD proteins in the neutrophil. TNF did not induce a change in the phosphorylation of a 16-kD protein of mononuclear cells.

- 66 - 43

'" I

0

- 31

)(

...

1:

- 22 ·· ··16 - 14 "0

"0

:l:

:l:

z

z

.... z "-

-.,

:l: 0

z0

:l: 0

~

.... z "-

-.,

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL.

288

a 66 57 ~5

36

';' 0

29

.

x

I: 2~

20

20

16

16

1~

~ .7

5.5

5 .9

6 .3

6.5

7 .0

7.4

6 .2

pH -~

b 66 pp 5 7

';'

o

29

..

x

I:

pp 20

5 1991

identified in the two gel systems. Exogenous recombinant TNF (a 17-kD protein) was run on companion gels and did not comigrate with pp16or with any phosphorylated proteins. Therefure, nonspecific phosphorylation of exogenous TNF was not present in these experiments. Four specific phosphoproteins, localized by two-dimensional electrophoresis, pp16a, pp16b, pp20, and pp57, were further characterized. pp16a and pp16b were phosphorylated 5- to 6-fold and pp57 3- to 4-fold over control levels when neutrophils were exposed to TNF for 15 min. pp20, in contrast, showed little change between control and TNF-stimulated neutrophils and demonstrates that TNF does not stimulate an increase in the phosphorylation of all proteins but induces specific phosphorylation of individual proteins. pp16a and pp16b may represent different stages of phosphorylation of the same molecular weight protein . In data not shown, neutrophils pretreated with neutralizing TNF antibody (10 U of neutralizing antibody/U TNF) and then stimulated with TNF showed phosphorylation patterns identical to unstimulated neutrophils. In order to further characterize TNF-induced protein phosphorylation, four individual proteins were isolated from two-dimensional gels as described in MATERIALS AND METHODS. These proteins were then subjected to acid hydrolysis and analyzed for phosphoprotein content by twodimensional thin layer chromatography. An autoradiograph of this separation is shown in Figure 7. TNF induced the specific phosphorylation of the amino acids serine and threonine but not tyrosine. All four proteins showed phosphorylation of a single type of amino acid . A summary of these findings is presented in Table 1.

20

Discussion pp 16 0

pp16 b 1~

~7

5.5

5.9

6.3

6 .5

7.0

7.4

6.2

pH - ~

Figure 6. Two-dimensional gel analysis of TNF-induced protein phosphorylation . Panels a and b represent untreated control neutrophils and neutrophils treated with 100 ng/ml TNF for 15 min, respectively. Proteins were isolated by trichloroacetic acid precipitation , protein concentration was adjusted, and two-dimensional PAGE was performed as described in MATERIALS AND METHODS. The horizontal axis corresponds to pH (4.7 to 8.2), and the vertical axis to molecular mass. TNF induced the phosphorylation of several proteins, including two l6-kD proteins and a 57-kD protein , which are labeled in panel b, with corresponding migration sites in control neutrophils shown in panel a.

with 100 U TNF/ml for 15 min. The specific phosphorylation of several proteins is induced by TNF including two l6-kD proteins (pp16a, pI 5.9; pp16b, pI 6.3) and a 57-kD protein (pp57, pI 5.8). Phosphorylation of 32- and 47-kD proteins, observed in one-dimensional PAGE, was not clearly observed in two-dimensional PAGE analysis. Two-dimensional gels separate only proteins with pI's in a given range, whereas one-dimensional gels separate all proteins. This technical limitation may explain why different phosphoproteins were

The phosphorylation of pp16, pp32, and pp47 begins by 5 min and continues up to 60 min after stimulation with TNF. The phosphorylation of these three proteins shares the same time course of activation and requires similar TNF concentrations. This time course is consistent with the generation of superoxide anions in neutrophils induced by TNF and in the priming ofneutrophils by TNF (18,44,45). The concentration of TNF that induces the phosphorylation of ppl6, pp32, and pp47 in the neutrophil (as low as 1 ng/ml) is roughly consistent with concentrations required to induce the production of oxygen free radicals by neutrophils, "prime" neutrophils for response to other stimuli, and induce changes in mRNA expression in other cell types (18, 20,35, 45). Thus, the time course and concentration dependence of phosphorylation of these proteins is consistent with their possible involvement in several functions induced by TNF in the neutrophil. In this study, TNF and FMLP both induce the phosphorylation of some similar proteins ; however, FMLP is a more potent stimulus and appears to induce the phosphorylation of several proteins that TNF does not. This is consistent with previous findings (44). This also correlates with the differing effects ofTNF and FMLP on neutrophil chemotaxis, degranulation, and the degree of oxygen burst activity (18,22) . This finding demonstrates that FMLP and TNF do not act on neutrophils through identical mechanisms. In this study, LPS did not induce protein phosphorylation. LPS from various

Crowley and Raffin: TNF-induced Protein Phosphorylation in Human Neutrophils

, , I

'_I,

PP166

PP16b

,'",

I

('~=' ..... ,,/

pp20

,

.... ,

I

'-I

~,/

\ ....

ppS7

Figure 7. Phosphoaminoacid identification. Phosphoprotein (Pp) 16a, ppl6b, pp20, and pp57 were isolated from two-dimensional gels and hydrolyzed in acid. Phosphoaminoacids were spotted onto cellulose thin layer chromatography plates, and electrophoresis was performed horizontally in pH 1.9 buffer and then vertically in pH 3.5 buffer as described in MATERIALS AND METHODS. Figure 7 is an autoradiograph of the thin layer chromatography plates. The circles represent the migration of phosphoserine (Ser), phosphotyrosine (Ty), and phosphothreonine (Tr), which were added to each sample as standards and visualized with ninhydrin. pp16a and pp16b are phosphorylated only at threonine residues, whereas pp20 and pp57 show phosphorylation only at serine residues.

sources stimulates neutrophil chemotaxis and adherence and may "prime" neutrophils fur response to other agents. It is controversial, however, ifLPS alone directly stimulates oxygen free radical production in neutrophils (46, 47). Interestingly, TNF, FMLP, and PMA all induce oxygen free radical generation in neutrophils. Thus, lack of protein phosphorylation in LPS-treated neutrophils may be correlated with the inability of LPS to induce a respiratory burst. pp16 is phosphorylated in response to TNF in neutrophils but not in mononuclear cells. Several effects of TNF have been shown to be cell specific including the induction of transcription of mRNA in various cells and the specific effects of TNF on tumor cell lines (35). For example, the TNF-induced phosphorylation of serine residues from a 26kD protein in U937 cells is found in only a few of the other cell lines tested (36). Thus, ppl6 phosphorylation in the neutrophil may represent a unique effect of TNF on neutrophil function. Others have studied protein phosphorylation in the neutrophil induced by FMLP, PMA, and other stimuli (22, 48-51). Comparisons of the proteins phosphorylated in these works is difficult due to variations in technique, duration of stimulation, concentrations studied, and differences in neutrophil isolation procedures. Nevertheless, a comparison is

TABLE 1

Summary of protein phosphorylation induced by tumor necrosis factor

Protein

Molecular Weight

pI

% Increase in Phosphorylation (SEM)*

pp16a pp16b pp20 ppS7

16,000 16,000 20,000 57,000

5.9 6.3 7.0 5.8

566 (67) 586 (50) 140 (25) 362 (36)

Residue Phosphorylated

Threonine Threonine Serine Serine

* Data are from four separate experiments and are expressed as percentage of neutrophil control 32POa- incorporation ± SEM .

289

in order. The TNF-induced phosphorylation of a 47-kD protein observed in one-dimensional gels in this study may be the same protein (or group of proteins) observed in neutrophils stimulated with PMA and FMLP (48, 52) . The 47-kD protein is part of the NADPH oxidase complex in neutrophils and its phosphorylation corresponds with, but is not necessarily required for, stimulation of the oxygen burst in neutrophils treated with PMA (52, 53). Phosphorylation of pp47 is absent in patients with chronic granulomatous disease who are incapable of mounting a respiratory burst (54). pp32 may also be a portion of the NADPH oxidase complex (55). ppl6 appears similar in molecular weight and pI to a protein phosphorylated at serine residues in HL-60 cells stimulated with PMA . In our experiments with ppl6, however, phosphorylation was observed exclusively at threonine residues (36). Several enzymes are known to phosphorylate proteins in cells, including the following enzyme families: protein kinase C (PKC) , cyclic adenosine monophosphate (cAMP)dependent protein kinase, and the calcium-calmodulin-dependent protein kinases (56, 57). These enzymes are a central part of signal transduction and regulatory processes in all cells. In the neutrophil, proteins are phosphorylated in response to stimulation by agents such as PMA and FMLP (49, 51,58,59) . PKC appears responsible for inducing the respiratory burst in neutrophils treated with PMA . PKC may also be involved in FMLP signal transduction in neutrophils (59). PKC, however, is not the only kinase active in neutrophils . Tyrosine phosphorylation of multiple proteins occurs as an early event in neutrophils stimulated with FMLP, PMA, and leukotriene B. (60) . cAMP-dependent protein phosphorylation has also been demonstrated in neutrophils (50). PKC induces the phosphorylation of serine and threonine but not tyrosine residues . In bovine neutrophils, a single isoenzyme of PKC was partially purified and was capable of serine and threonine but not tyrosine phosphorylation in vitro (61). Is PKC involved in TNF-induced protein phosphorylation in the neutrophil? Berkow and Dodson have previously demonstrated that TNF does not alter the activity or induce the translocation of PKC in neutrophils (44). The present study demonstrates specific TNF-induced serine and threonine phosphorylation of neutrophil proteins and suggests that TNF activates a serine and threonine kinase or inactivates a serine or threonine phosphorylase. Thus, the type of phosphorylation observed is consistent with PKC activity. H-7, H-8, and H-9 are reasonably specific inhibitors ofPKC. HA-1004 is less effective in the inhibition ofPKC but is a potent inhibitor of cAMP-dependent protein kinase (38). In this study, H-7, H-8, H-9, and HA-lO04 had no effect on TNFinduced phosphorylation of 16-, 32-, and 47-kD proteins . Additionally, the calcium-calmodulin-dependent protein kinase inhibitor W-7had little or no effect on the phosphorylation of these proteins (62). Thus, it appears that ppl6, pp32, and pp47 may not be phosphorylated by PKC, cAMPdependent protein kinase , or the calcium-calmodulin-dependent protein kinase in TNF-stimulated neutrophils. The effectsof these inhibitors should be interpreted with caution, however, as the precise protein kinase isoenzymes found in neutrophils may be resistant to these compounds. The biochemical mechanism of action of TNF on neutrophils, as well as in other cells, remains uncertain. The present study demonstrates that TNF induces the phosphoryla-

290

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 5 1991

tion of one 57- and two 16-kD proteins at threonine or serine residues. TNF-induced protein phosphorylation is both concentration and time dependent, and the phosphorylation of pp16 does not occur in mononuclear cells. Finally, PKC inhibitors have virtually no effect on protein phosphorylation at concentrations known to inhibit PKC and neutrophil functions. Acknowledgments: This work was generously supported by the Paul Lam Fund and the Clementina Ho Fund for research in pulmonary medicine.

References 1. Old, L. J. 1985. Tumor necrosis factor (TNF). Science 230:630-632. 2. Beutler, B., and A. Cerami. 1986. Tumor necrosis factor and the macrophage secreted factor cachectin. Blood 320:584-588. 3. Vilcek, J., V. J. Palombella, D. Henriksen-DeStefano et al. 1986. Fibroblast growth enhancing activity of TNF and its relationship to other polypeptide growth factors. J. Exp. Med. 163:632-643. 4. Klebanoff, S. J., M. A. Vadas, J. M. Harlan et al. 1986. Stimulation of neutrophils by tumor necrosis factor. J. Immunol. 136:4420-4225. 5. Scheurich, P., B. Thoma, U. Ucer, and K. Pfizenmaier. 1987. Immunoregulatory activity of recombinant tumor necrosis factor (TNF)alpha induction of TNF receptors on human T cells at TNF-alpha mediated enhancement of T cell responses. J. Immunol. 138:1786-1790. 6. Lilly, C. W., J. S. Sandhu, A. Ishizaka et al. 1989. Pentoxifylline prevents tumor necrosis factor-induced lung injury. Am. Rev. Respir. Dis. 139:1361-1368. 7. Stephens, K. E., A. Ishizaka, J. W. Larrick, and T. A. Raffin. 1988. Tumor necrosis factor causes increased pulmonary permeability and edema: comparison to septic acute lung injury. Am. Rev. Respir. Dis. 137: 1364-1370. 8. Tracey, K. J., B. Beutler, S. F. Lowry et al. 1986. Shock and tissue injury induced by recombinant human cachectin. Science 234:470-474. 9. Tate, R. M., andJ. E. Repine. 1983. Neutrophils and the adult respiratory distress syndrome. Am. Rev. Respir. Dis. 128:552-559. 10. Hogg, J. C., 1987. Neutrophil kinetics and lung injury. Physiol. Rev. 67:1249-1295. 11. Stevens, J. H., and T. A. Raffin. 1984. Adult respiratory disease syndrome. I. Aetiology and mechanisms. Postgrad. Med. 60:505-513. 12. Weiland, J. E., W. B. Davis, J. F. Holter, J. R. Mohammed, P. M. Dorinsky, and J. E. Gadek. 1986. Lung neutrophils in the adult respiratory distress syndrome. Am. Rev. Respir. Dis. 133:218-225. 13. McGuire, W. W., R. G. Spragg, A. B. Cohen, and C. G. Cochrane. 1982. Studies on the pathogenesis of the adult respiratory distress syndrome. J. Clin. Invest. 69:543-553. 14. Wewers, W. D., D. J. Herzyk, and J. E. Gadek. 1988. Alveolar fluid neutrophil elastase activity in the adult respiratory distress syndrome is complexed to alpha-2-rnacroglobulin. J. Clin. Invest. 82:1260-1267. 15. Mallick, A. A., A. Ishizaka, K. E. Stephens et al. 1989. Multiple organ damage caused by tumor necrosis factor and prevented by prior neutrophil depletion. Chest 95:1114-1120. 16. Stephens, K. E., A. Ishizaka, Z. Wu, J. W. Larrick, and T. A. Raffin. 1988. Granulocyte depletion prevents tumor necrosis factor-mediated acute lung injury in guinea pigs. Am. Rev. Respir. Dis. 138:1300-1307. 17. Ognibene, F. P., S. E. Martin, M. M. Parkeretal. 1986. Adult respiratory distress syndrome in patients with severe neutropenia. N. Engl. J. Med. 315:547-551. 18. Yonemaru, M., K. E. Stephens, A. Ishizaka et al. 1989. Effects of tumor necrosis factor on PMN chemotaxis, chemiluminescence, and elastase activity. J. Lab. Clin. Med. 114:674-681. 19. Gamble, J. R., J. M. Harlan, S. J. Klebanoff, and M. A. Vadas. 1985. Stimulation of the adherence of neutrophils to umbilical vein endothelium by human recombinant tumor necrosis factor. Proc. Natl. Acad. Sci. USA 82:8667-8671. 20. Larrick, J. W., D. Graham, K. Toy, L. S. Lin, G. Senyk, and B. M. Fendly. 1987. Recombinant tumor necrosis factor causes activation of human granulocytes. Blood 69:640-644. 21. Ozaki, Y., T. Ohashi, Y. Niwa, and S. Kume. 1988. Effects of recombinant DNA-produced tumor necrosis factor on various parameters of neutrophil function. Inflammation 12:297-309. 22. Berkow, R. L., D. Wang, J. W. Larrick, and R. W. Dodson. 1987. Enhancement of neutrophil superoxide production by pre-incubation with recombinant human tumor necrosis factor. J. Immunol. 139:3783-3791. 23. Zheng, H., J. J. Crowley, J. C. Chan et al. 1990. Attenuation of tumor necrosis factor-induced endothelial cell cytotoxicity and neutrophil chemiluminescence. Am. Rev. Respir. Dis. 142:1073-1078.

24. Richter, J., T. Anderson, and I. Olson. 1989. Effects ofTNF and granulocyte/macrophage stimulating factor on neutrophil degranulation. J. Immunol. 142:3199-3205. 25. Luedke, E. S., and J. L. Humes. 1989. Effect ofTNF on granule release and LTB-4 production in adherent human polymorphonuclear leukocytes. Agents Actions 27:451-454. 26. Nathan, C. F. 1987. Neutrophil activation of biological surfaces. J. Clin. Invest. 80: 1550-1560. 27. Becker, E. L. 1987. The formylpeptide receptor of the neutrophil: a search and conserve operation. Am. J. Pathol. 129:16-24. 28. Goldman, D. W., F. H. Chang, L. A. Gifford, E. J. Goetzl, and H. R. Bourne. 1985. Pertussis toxin inhibition of chemotactic factor-induced calcium mobilization and function in human polymorphonuclear leukocytes. J. Exp. Med. 162:145-156. 29. Lad, P. M., C. V. Olson, I. S. Grenwal, and S. J. Scott. 1985. A pertussis toxin sensitive GTP-binding protein in the human neutrophil regulates multiple receptors, calcium mobilization, and lectin-induced capping. Proc. Natl. Acad. Sci. USA 82:8643-8647. 30. Omann, G. M., R. A. Allen, G. M. Bokoch, R. G. Painter, A. E. Traynor, and L. A. Sklar. 1987. Signal transduction and cytoskeletal activation in the neutrophil. Physiol. Rev. 67:285-307. 31. Shalaby, M. R., M. A. Paladino, Jr., S. E. Hirabayashi etal. 1987. Receptor binding and activation of polymorphonuclear neutrophils by TNF alpha. J. Leukoc. Biol. 41:196-204. 32. Angheleri, L. J., and J. Robert. 1987. Effects of tumor necrosis factor on tumor cell plasma membrane permeability. Tumori 73:269-271. 33. Kobayashi, Y., N. Utsunomiya, M. Nakanishi, and T. Osawa. 1987. Early transmembrane events in TNF and lymphotoxin-induced cytotoxicity. ImmunoI. Lett. 15:53-57. 34. Torti, F. M., B. Diekmann, B. Beutler, A. Cerami, and C. M. Ringold. 1985. A macrophage factor inhibits adipocyte gene expression: an in vitro model of cachexia. Science 229:867-869. 35. Kronke, M., C. Schluter, and K. Pfizenmaier. 1987. TNF inhibits myc expression in HL-6Ocells at the level of mRNA transcription. Proc. Natl. Acad. Sci. USA 84:469-473. 36. Schutze, S., P. Scheurich, K. Pfizenmaier, and M. Kronke. 1989. Tumor necrosis factor signal transduction. J. Bioi. Chem. 264:3562-3567. 37. English, D., and B. R. Anderson. 1974. Single step preparation of red blood cells, granulocytes, and mononuclear leukocytes on discontinuous density gradient Ficoll-Hypaque. J. Immunol. Methods 5:249-252. 38. Hikada, H., M. Inagaki, S. Kawamoto, and Y. Sasaki. 1984. Isoquinolinesulfonamides, novel and potent inhibitors of cyclic nucleotide dependent protein kinase and protein kinase C. Biochemistry 25: 5036-5041. 39. O'Farrell, P. H. 1975. High-resolution two-dimensional electrophoresis of proteins. J. BioI. Chem. 250:4007-4021. 40. Laernrnli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685. 41. Cooper, J. A., B. M. Sefton, and T. Hunter. 1983. Detection and quantification of phosphotyrosine in proteins. Methods Enzymol. 99: 387-402. 42. Blackshear, P. J., L. A. Witters, P. R. Girard, J. F. Kuo, and S. N. Quamo. 1985. Growth factor-stimulated protein phosphorylation in 3T3L1 cells. J. Bioi. Chem. 260:13304-13315. 43. Wollner, S., J. J. Crowley, and T. A. Raffin. 1990. Attenuation ofTNFinduced chemiluminescence by protein kinase inhibitors. Am. Rev. Respir. Dis. 140(Suppl.):A638. (Abstr.) 44. Berkow, R. L., and R. W. Dodson. 1988. Biochemical mechanism involved in the priming of neutrophils by tumor necrosis factor. J. Leukoc. Bioi. 44:345-352. 45. Tsujimoto, M., S. Yokota, T. Vilcek, and G. Weissmann. 1986. Tumor necrosis factor provokes superoxide anion generation from neutrophils. Biochem. Biophys. Res. Commun. 137:1094-1100. 46. Nicotra, J., A. J. Orsini, and A. D. DeBari. 1985. Chemiluminescence response of the human polymorphonuclear neutrophil to lipopolysaccharides. Cell Biophys. 7:283-291. 47. Wilson, M. E. 1985. Effectsof bacterial endotoxins on neutrophil function. Rev. Infect. Dis. 7:404-418. 48. Hayakawa, T., K. Suzuki, S. Suzuki, P. C. Andrews, and B. M. Babior. 1986. A possible role for protein phosphorylation in the activation of the respiratory burst in human neutrophils. Evidence from studies with cells from patients with chronic granulomatous disease. J. Biol. Chem. 251 :9109-9115. 49. Feuerstein, N., and H. L. Cooper. 1983. Rapid protein phosphorylation induced by phorbol ester in HL-60 cells. J. Biol. Chem. 258:10786-10793. 50. Huang, C.-K., 1. M. Hill, B.-1. Bormann, W. M. Mackin, and E. L. Becker. 1983. Endogenous substrates for cyclic AMP-dependent and calcium-dependent protein phosphorylation in rabbit peritoneal neutrophils. Biochim. Biophys. Acta 760:126-135. 51. White, 1. R., C.-K. Huang, 1. M. Hill, Jr., P. H. Naccache, E. L. Becker, and R. I. Sha'afi. 1984. Effect of phorbol 12-myristate 13-acetate and its analogue alpha-phorbol 12,13-didecanoate on protein phosphorylation and lysosomal enzyme release in rabbit neutrophils. J. Bioi. Chem.

Crowley and Raffin: TNF-induced Protein Phosphorylation in Human Neutrophils

259:8605-8611. 52. Sha'afi, R. I., T. F. P. Molski, 1. Gomez-Cambronero, and C. K. Huang. 1988. Dissociation of the 47-kD protein phosphorylation from degranulation and superoxide production in neutrophils. l. Leukoc. Bioi. 43:18-27. 53. Okamura, N., B. M. Babior, L. A. Mayo, P. Pever, R. M. Smith, and 1. T. Curnutte. 1990. The p67-phox cytosolic peptide of the respiratory burst oxidase from human neutrophils. Functional aspects. J. Clin. Invest. 85:1583-1587. 54. Okamura, N., 1. T. Curnette, R. L. Roberts, and B. M. Babior. 1988. Relationship of protein phosphorylation to the activation of the respiratory burst in human neutrophils. Defects in the phosphorylation of a group of closely related 48-kD proteins in two forms of chronic granulomatous disease. J. Bioi. Chem. 263:6777-6782. 55. Papini, E., M. Grzeskowiak, P. Bellavite, and F. Rossi. 1985. Protein kinase C phosphorylates a component of NADPH oxidase of neutrophils. FEBS Leu. 190:204-208. 56. Nishizuka, Y. 1989. The family ofprotein kinase C for signal transduction.

291

lAMA 262:1826-1833. 57. Krebs, E. G. 1989. Role of cyclic AMP-dependent protein kinase in signal transduction. lAMA 262:1815-1818. 58. Andrews, P. c., and B. M. Babior. 1983. Endogenous protein phosphorylation by resting and activated human neutrophils. Blood 61:333-340. 59. Pontremoli, S., E. Melloni, F. Salamino et al. 1986. Phosphorylation of proteins in human neutrophils activated with phorbol myristate acetate or with chemotactic factor. Arch. Biochem. Biophys. 250:23-29. 60. Berkow, R. L., and R. W. Dodson. 1990. Tyrosine-specific protein phosphorylation during activation of human neutrophils. Blood 75:24452452. 61. Dianoux, A.-C., M.-1. Stasia, and P. V. Vignais. 1989. Purification and characterization of an isoform of protein kinase C from bovine neutrophils. Biochemistry 28:424-431. 62. Wright, C. D., and M. 0. Hoffman. 1987. Comparison of the roles of calmodulin and protein kinase C in activation of the human neutrophil respiratory burst. Biochem. Biophys. Res. Commun. 142:53-62.