Journal

of Leukocyte

Biology

48:549-556

(1990)

Bacterial Lipopolysaccharide Induces Release of Tumor Necrosis Factor-a From Bovine Peripheral Blood Monocytes and Alveolar Macrophages In Vitro Jeftrey Department

of Pathobiological

L. Adams Sciences,

and

Charles

School of Veterinary

J. Czuprynski Medicine,

University

of Wisconsin,

Madison

ln this study, we demonstrate that freshly adherent bovine monocytes release tumor necrosis factor-a (TNF-a) in response to stimulation with bacterial lipopolysaccharide (LPS). TNF-a was detected using actinomycin D-treated WEHI-164 murine fibrosarcoma cells as targets in an 18 hr cytotoxicity assay. Doses of LPS from 20 ng/mI to 20 g/ml were capable of inducing bovine TNF-a. The kinetics of TNF-a release from bovine monocytes demonstrated peak levels of cytotoxic activity at 1-3 hr post-LPS treatment, with a subsequent decline to background levels by 1 8 hr post-LPS treatment. A monoclonal antibody that neutralizes recombinant human TNF-a (rHuTNF-ca) significantly reduced the cytotoxicity of LPS-stimulated bovine monocyte culture supernatants. Size exclusion high-performance

and alveolar

liquid macrophage

chromatography

(HPLC)

analysis

of LPS-stimulated

monocyte

culture supernatants resulted in a molecular weight elution profile similar to that of recombinant human TNF-a. These elution profiles are consistent with the presence of multimers of TNF-ca. This is believed to be the first report of the in vitro production of bovine TNF-a.

key words: cachectin, cytotoxicity,

monocytes, alveolar macrophages,

INTRODUCTION Tumor necrosis factor-a (TNF-a) is a 17 kilodalton cytokine that is released primarily from monocytes and macrophages in response to a wide variety of inflammatory stimuli. TNF-a has been shown to mediate a wide variety of biological activities that play an important role in both the severity of and recovery from many bacterial, parasitic, and viral infections [14,21]. These activities include induction of fever; direct activation of neutrophils, T cells, and macrophages; direct tumor cell cytotoxicity, cachexia, induction of ther mediators (e.g., interleukin-l [IL-li, IL-6, and TNF itself), and suppression of lipoprotein lipase [14,21]. TNF-cx was first described as an endotoxin-induced serum factor, released from monocytes or macrophages, that caused the necrosis of transplanted tumors in mice and the selective lysis of tumor cells in vitro [5]. Subsequent studies demonstrated both in vivo and in vitro production of TNF-a by a variety of mammalian species including rabbits, nonhuman primates, and human beings in response to lipopolysaccharide (LPS) [2,12,17,20]. TNF-a has been shown to be an important mediator of endotoxin-induced shock in these animals. Administration of recombinant TNF-a in these animals mimics the shock-like state induced by endotoxin, whereas prior treatment with antibodies against TNF-a protects animals against the lethal effects of endotoxemia [3,13,16,24]. © 1990 Wiley-Liss,

Inc.

bovine

Calves are extremely sensitive to the presence of endotoxin in vivo [19]. Clinical signs of endotoxin shock in calves are similar to those observed upon LPS or TNF administration in other animal models. The roles that inflammatory mediators play in this response are unknown. Little is known about the production of TNF-a in ruminant species. We recently demonstrated the detection of circulating serum levels of bovine TNF-a in neonatal calves following the administration of endotoxin [1] . We were interested in determining whether bovine mononuclear phagocytes were capable of releasing TNFa in vitro. In this study, we demonstrate the in vitro release of bovine TNF-cx from peripheral blood monocytes and alveolar macrophages in response to LPS.

MATERIALS AND METHODS Isolation of Monocytes Venous blood was collected from the tail healthy adult Holstein cattle using .38% sodium an anticoagulant as described previously [28]

Received Reprint Sciences, Madison,

February

27,

1990:

accepted

May

veins of citrate as . Citrated

8, 1990.

requests: Ci. Czuprynski. Department School of Veterinary Medicine, 2015 WI 53706.

of Pathobiological Linden Drive West,

550

Adams

and Czuprynski

blood was centrifuged (300g for 20 mm at 22#{176}C), and the platelet-rich plasma was removed by aspiration. The remaining blood cells were recentrifuged ( 1 ,000g for 20 mm at 22#{176}C),and the buffy coat cells were collected. These cells were resuspended in 4 volumes of warm (37#{176}C) phosphate-buffered saline (PBS; pH 7.2), underlaid with Ficoll-Hypaque (density 1 .077, Sigma Chemical Co. , St. Louis, MO), and centrifuged at 400g for 30 mm at 22#{176}C. The mononuclear cells at the interface were removed and washed three times in cold Hanks’ balanced salt solution (HBSS; GIBCO, Grand Island, NY). The mononuclear cells were resuspended at 5 x 106 cells/ml in HBSS that contained 0.5% fetal bovine serum (FBS; lot 38P4084,GIBCO) and allowed to adhere to tissue culture flasks (Costar, Cambridge, MA) for 1-2 hr at 39#{176}C with 5% CO2. Incubation at 39#{176}C was used to simulate the normal body temperature of cattle. Following this, the nonadherent cells were removed with three vigorous washes with warm (37#{176}C)HBSS. Adherent cells were routinely 80-85% monocytes by nonspecific esterase positive staining and greater than 85% viable by trypan blue exclusion.

Collection

of Alveolar

Macrophages

Alveolar macrophages were collected by bronchial lavage of lungs, removed from freshly killed cows, with a saline ( 150 mM)-EDTA (9 mM) solution. The lung lavage cells were washed once in HBSS containing EDTA (9mM), then twice in HBSS, and then were resuspended in HBSS at 7 X 106 cells/mI. The resuspended lung lavage cells were seeded in tissue culture flasks (Costar). Lung lavage cells were routinely 90-95% alveolar macrophages and greater than 85% viable.

a1; lot 1942-33; specific activity 8 X lO6U/mg) and recombinant bovine IFN-y (rBoIFN-’y; lot 3229-38; specific activity 1 .8 x 105U/mg) were provided by Dr. M. Shepherd (Genentech, South San Francisco, CA) and Dr. R. Steiger (Ciba Geigy, Basel, Switzerland), respectively. Cell

Lines

WEHI-l64 murine sarcoma cells (ATCC CRL 1751) and L929 murine fibrosarcoma cells (ATCC CCL 1) were obtained from the American Type Culture Collection (Rockville, MD). Madin-Darby bovine kidney (MDBK) cells, bovine turbinate cells, and an adherent bovine lymphoblastoid cell line (BL-3 JY) were provided by Dr. R. Schultz (University of Wisconsin, Madison, WI). A bovine neurofibrosarcoma cell line (LMS) was obtained from Dr. G. Splitter (University of Wisconsin). All cells were maintained in Dulbecco’s modified Eagle tissue culture medium (DMEM, GIBCO) supplemented with 5% fetal bovine serum (FBS) (GIBCO), 20 mM HEPES, 2 mM L-glutamine, 100 U penicillin G (GIBCO), and 100 ig streptomycin (GIBCO) at 37#{176}C and 5% CO2. Murine hybridoma cells (B154.2. 1) that secrete neutralizing anti-rHuTNF-a antibodies (IgG 1 ) were a kind gift from Dr. Giorgio Trinchieri (Wistar Institute, Philadelphia, PA) and were maintained in DMEM with 5% FBS [8] . A control monoclonal antibody (MAb) directed against influenza hemagglutinin (IgG 1) was generously provided by Dr. V. Hinshaw (University of Wisconsin, Madison, WI).

Assay

for TNF-a

WEHI-l64 cells were grown to confluency in DMEM containing 5% FBS (GIBCO), 20 mM HEPES, 2 mM Stimulation of Monocytes and L-glutamine, 100 U penicillin G (GIBCO), and 100 pg Alveolar Macrophages streptomycin (GIBCO) at 37#{176}C with 5% CO2. Cells were Monocyte and alveolar macrophage monolayers were gently detached with a sterile cell scraper (American incubated in DMEM containing 20 mM HEPES, 4 mM Scientific Products, McGaw Park, IL) and washed in L-glutamine, and 100 U penicillin G (GIBCO) at 39#{176}C fresh DMEM. Cells were seeded in 96-well plates with 5% CO2. To stimulate TNF release, monocytes or (Costar) at 1 x l0 cells/well in 0.2 ml of fresh DMEM alveolar macrophages were cultured in the presence of 2 and allowed to adhere for 4-6 hr at 39#{176}C with 5% CO2. p.g/ml of Escherichia co/i LPS (055:B5; Difco, Detroit, The medium was then aspirated and 100 pA ( 1 p.g) of MI) for 2 hr at 39#{176}C and 5% CO,. Afterwards, the culture actinomycin D sulfate (Sigma Chemical Co.) was added medium was removed and centrifuged (300g for 10 mm), to each well. Culture supernatants to be tested were and the supernatants were transferred to freezer vials and serially diluted in complete DMEM, and 100 il of each frozen at -70#{176}C. dilution was added to triplicate or quadruplicate wells. rHuTNF-a ( 1 , 10, and 100 pg) and DMEM were Cytokines included as positive and negative controls, respectively, Recombinant human TNF-a (lot NP200B; specific in each plate. The plates were then incubated at 39#{176}C with activity 2.2 x l0 U/mg) was a kind gift of Dr. A. 5% CO2 for 18 hr. The medium was removed, and the Creasey (Cetus Corp. , Emeryville, CA). Recombinant monolayers were washed twice with warm (37#{176}C) PBS. human interleukin-la) (rHuIL-lci; lot IL-l 1/87; specific Monolayers were fixed with 40% formaldehyde and activity 2.5 x 109U/mg) was generously provided by stained for 10 mm with a 0.5% solution of crystal violet Drs. A. Stern and PT. Lomedico (Hoffman-La Roche, in 25% ethanol and 0.85% NaCI. Afterwards, the stained Nutley, NJ). Recombinant bovine interferon-a 1(rBoIFNcells were washed twice with a gentle stream of tap water

Bovine

LPS-Stimulatecl and allowed to dry. Stained monolayers were solubilized by the addition of 100 p.1 of a 0.5% sodium dodecyl sulfate (SDS)-.05 M acetic acid solution. The optical density (OD) of each well was read on a micro-ELISA plate reader (Dynatech Inc. , Chantilly, VA) at a wavelength of 550 nm. Data were expressed as percent cytotoxicity calculated as follows: OD

negative

control - OD test well/OD control x 100.

negative

Production

and Purification

of Anti-HuTNF-a

Hybridoma cells (B 154.2. 1) were grown in DMEM with 5% FBS. Five nu/nu BALB/c mice were treated with 1 ml 2,6, 10, 14-tetramethyl-pentadecane (Pristane; Sigma Chemical Co.) 2 weeks prior to intraperitoneal (i.p.) injection of 2 X 106 hybridoma cells. Ascites were collected and centrifuged to remove the particulate debris. The samples were delipidated using a lipid clearing solution (Beckman, Fullerton, CA), and the MAb were purified using high-performance anion exchange liquid chromatography (ABX 300, Beckman). The MAb was eluted from the column with a 0-400 mM gradient of sodium acetate buffer. Fractions were screened for the presence of murine IgG using an MAb isotyping kit (HyClone Laboratories, Logan, UT). Fractions that were positive by ELISA were then analyzed by SDS-PAGE to verify the purification of the antibody. Fractions that contained the MAb were dialyzed extensively against PBS, concentrated using an Amicon ultrafiltration apparatus (MW MA.), and

cut-off 50,000 daltons; Amicon, stored as aliquots at -70#{176}C.

Danvers,

Release

In Vitro

551

Fractions were lyophilized using a SpeedVac concentrator (Savant Instruments Inc. , Farmingdale, NY) and frozen at - 20#{176}C . As a control, 1 ng/ml of rHuTNF-a was tested under identical conditions except that fractions were collected every mm for 60 mm. The GPC-l00 column was calibrated with the following molecular weight standards (Sigma): blue dextran (MW 2,000000), albumin (MW 69,000), carbonic anhydrase (MW 29,000), cytochrome C (MW 1 2,500), and aprotinin

6,500).

(MW

The standard errors of the mean were routinely less than 10% of the mean within a given set of replicate wells. To verify that cytotoxicity was due to TNF-a, culture supernatants were diluted as previously described and added to triplicate wells that contained graded amounts of an MAb that neutralizes rHuTNF-a [8]. Control wells contained an irrelevant MAb. To assess TNF-a activity in HPLC fractions, lyophilized samples were reconstituted with 500 il of DMEM and assayed for cytotoxic activity.

TNF-a

RESULTS

WEHI-164 Cells Are Susceptible Bovine TNF-a

to

The standard assay to detect TNF-a is lysis of actinomycin D-treated L929 murine fibrosarcoma cells. Utilizing this assay, we were unable to detect TNF activity in culture supernatants from LPS-stimulated bovine monocytes (data not shown). We also examined a number of other cell lines for their susceptibility to LPS-stimulated bovine monocyte culture supernatants in the presence of actinomycin D. Bovine turbinate cells, MDBK cells, an adherent bovine lymphoid cell line (BL-3 JY), and a bovine neurofibrosarcoma (LMS) cell line were not susceptible to the cytotoxic activity of LPS-stimulated bovine monocyte culture supernatants. The murine fibrosarcoma cell line WEHI-l64, which has been shown to be more sensitive to murine TNF-a than the L929 cell line, was susceptible to the cytotoxic activity in bovine monocyte culture supernatants (Fig. 1). Although WEHI-164 cells were killed by LPS-stimulated

100 90

#{149}-#{149}Monocyt.s

+ LPS

0-

alon.

0

Monocyt.s

80 >, 0

x 0 0 >‘

0

70 60 50

-%.

40 30

20

Size-Exclusion

HPLC

Culture supernatants from LPS-stimulated monocytes and alveolar macrophages were concentrated 100-fold and 20-fold, respectively, using centriprep concentrator units (MW cut-off 10,000 daltons; Amicon). Concentrated supernatant was centrifuged (300g for 10 mm) to remove any particulate debris, and 500 l of the supernate was loaded onto a GPC- 100 size exclusion column (Alltech, Deerfield, IL). The column was eluted with 100 mM ammonium acetate (pH 6. 8) at a flow rate of 0.5 ml/min. Fractions were collected every 5 mm for 60 mm.

10 1

2 Reciprocal

Fig. 1 . Tumor

4

8 dilution

t.,

,

16

32

of culture



64

128

supernatant

necrosis factor (TNF) activity in culture supernatants from lipopolysaccharide (LPS)-stimulated monocytes. Freshly adherent bovine peripheral blood monocytes (4 x 10 cells/mI) were cultured in serum-free DMEM in the presence or absence of 2 .tg/ml E. coil LPS for 2 hr at 39#{176}C and 5% CO2. Culture supernatants were assayed for TNF activity at the indicated concentrations. Data expressed are the mean percent cytotoxicity of one representative experiment of four performed.

552

Adams

TABLE

and Czuprynski

1. LPS-Stlmulated

Bovin e Alveolar

Macrophages

TABLE 2. In Vitro Maturation of Freshly Monocyte 5 Results in the Loss of Their

Release Tumor Necrosis Factors

TNFaa

Dilution of culture supernatant Treatment

I :2

None 10 jg/ml LPS + 40 g Anti-rHU + 80 sg Anti-rHU aControl

and

17.5 72.8 8.8 0.8

TNF-a TNF-a

LPS-stimulated

alveolar

± ± ± ±

Monocytes 1: I 0

2.9 8.5 5.9 3.3

12.3 67.5

macrophage

Adherent Bovine Ability To Release

± 1.4 ± 0.7 NT’ NT

culture

super-

natants were diluted to final concentrations of I :2 and 1 : 10. Culture supernatants were added to actinomycin D-treated WEHI1 64 cells with or without a monoclonal antibody against rHuTNF-a and incubated for I 8 hr at 39#{176}C. Percent cytotoxicity was determined as described in Materials and Methods. Data are the mean ± SEM percent cytotoxicity of one representative experiment. bNot tested.

Percent

Fresh I Day 4 Days 7 Days

cytotoxicity

83.5 ± 2.6 6.8 ± 1.1 9.6 ± 4.5 15.6 ± 3.9

aFreshly

adherent

bovine

peripheral

blood

monocytes

were

either

stimulated with 10 g/ml LPS for 2 hr or incubated at 39#{176}Cfor the indicated times. After in vitro maturation, monocyte cultures were stimulated with 1 0 g/ml LPS for 2 hr. Culture supernatants were diluted to a final concentration of 1:2, added to actinomycin D-treated WEHI164 cells and incubated for 1 8 hr at 39#{176}C. Percent cytotoxicity was determined as described in Materials and Methods. Data are the mean ± SEM percent cytotoxicity ofone representative experiment.

80

70

0-01:2

70

#{149}-#{149}1:10 60 >

>‘ 4-,

.

0 0

.

40

4-,

0

30

0

0

20 10

0

20 pg

200

E. coli

Fig. 2. bovine

pg

2 ng

20 ng

(4 x 106 cells/mI)

concentrations(20

ng

2 sg

20 g

were

on release of TNF from bovine peripheral blood stimulated

pg.-20 g) ofE. coil LPSfor

of three

independent

30

60

90 Time

with

increasing

2 hr in serum-free

DMEM at 39#{176}C and 5% CO2. Culture supernatants were diluted to a final concentration of 1 :2 in DMEM and assayed for TNF activity. Data are expressed as the mean ± SEM percent cytotoxicity

0

Lipopolysaccharide/mI

Effect of LPS concentration monocytes. Freshly adherent

monocytes

200

experiments.

bovine monocyte culture supernatants, actinomycin D enhanced approximately twofold the susceptibility of WEHI- 164 cells to the low levels of cytotoxic activity in dilute culture supernatants (23.2% ± 4. 1% cytotoxicity on untreated WEHI-164 cells vs. 50.0% ±4.8% cytotoxicity on actinomycin D-treated WEHI-l64 cells at a 1:20 dilution of culture supernatant). Culture supernatants from unstimulated monocytes had little or no cytotoxic activity (Fig . 1). Culture supernatants from LPS-stimulated alveolar macrophages also demonstrated cytotoxic activity, although unstimulated alveolar macrophages released slightly higher background levels of cytotoxicity than did unstimulated monocytes (Table 1). Peripheral blood monocytes incubated in vitro for 1-7

120 post

150 LPS

180

18H

24H

30H

48H

treatment

Fig. 3. kinetics of the release of TNF from bovine monocytes. Freshly adherent bovine peripheral blood monocytes (4 x 106 cells/mi) were cultured in serum-free DMEM in the presence or absence of 2 ig/ml E. coil LPS for up to 48 hr at 39#{176}C and 5%

CO2. Culture medium was removed at the noted times, centrifuged, diluted to final concentrations of 1 :2 and I :10, and tested for TNF activity against WEHI-164 cells. Data expressed are the mean ± SEM percent cytotoxicity of three independent experiments.

days prior to LPS stimulation levels of TNF-a (Table 2).

failed

to release

Effects

of LPS Dose and Time on Release

of TNF

Activity

detectable

Bovine TNF-a was detected after stimulation of bovine peripheral blood monocytes with doses of E. coli LPS in the range of 20 ng/ml to 20 .tg/ml (Fig. 2). TNF-a was rapidly released from bovine monocytes, the peak activity being observed at 2-3 hr after the addition of LPS (Fig. 3). Cytotoxic activity then declined to background levels by 1 8 hr post-treatment. We observed no second

LPS-Stimulated TABLE 3. Freshly Absorb

Exogenously

Adherent Added

Bovine Monocytes rHuTNF-a Percent

Bovine

TNF-a

Release

In Vitro

553

Rapidly

cytotoxicity >‘

(pg)

Fresh monocytes (2 hr)

1 10 100

18.8 ± 4.9 29.4 ± 9.1 25.5 ± 12.3

19.3 20.0 25.3

± 3.0 ± 4.2 ± 0.9

ofrHuTNF-a

were

incubated

rHuTNF-a

Fresh monocytes (18 hr)

No monocytes (2 hr)

No monocytes (18 hr)

32.4 64.7 82.0

34.6 56.2 83.5

.4-,

0

x 0

0

aVarious

doses

± 2.7 ± 3.3 ± 1.9

with or without

± 4.5 ± 1.8 ± 2.7

>

0

freshly

adherent bovine monocytes for 2 or I 8 hr at 39#{176}C.Culture supernatants were diluted to a final concentration of I :2, added to actinomycin D-treated WEHI164 cells, and incubated for 1 8 hr at 39#{176}C. Percent cytotoxicity was determined asdescribed in Materials and Methods. Data are the mean ± SEM percent cytotoxicity of one representative experiment.

0

5.6

11.3

19

Fig. 4.

Neutralization

22.5

45

Anti-TNF-a

of the cytotoxic

activity

of bovine mono-

cyte culture supernatants

by a monoclonal

directed against rHuTNF-a. supernatants were diluted

LPS-stlmulated monocyte culture to a final concentration of I :10 in

antibody

(MAb)

DMEM. Various doses of MAb were added to each well concom-

peak of TNF-a activity in culture supernatants of monocytes cultured for up to 48 hr after the addition of LPS. Unstimulated monocytes exhibited little or no detectable release of TNF-cx activity at any time. We have previously shown that cytotoxic activity observed in the WEHI-l64 assay could not be attributed to the direct cytotoxic effects ofbovine serum, LPS, IL-la, IFN-a, or IFN-y, either individually or in combination [1]. The decline in cytotoxic activity observed in the WEHI- 164 cell assay might be attributed to monocyte binding of TNF-a, the presence of proteases or TNF-a inhibitors in the conditioned medium, or the nonspecific adsorption of TNF-a to tissue culture wells. We confirmed that TNF-ct (1 , 10,and 100 pg) diluted in DMEM was stable at 39#{176}C for 18 hr. We performed preliminary experiments to address the decline in TNF-a activity in culture supernatants over time. These experiments demonstrated that various doses of rHuTNF-a (1 , 10, and 100 pg) were readily absorbed within 2 hr by freshly adherent bovine monocytes (Table 3); extending the coincubation time to 18 hr did not result in further reductions of TNF activity (Table 3).

Neutralizing Cytotoxicity

Antibody to rHuTNF-a of Bovine TNF-a

Abrogates

the

An MAb that neutralizes rHuTNF-a was utilized to demonstrate that the cytotoxicity of bovine monocyte culture supernatants was due to TNF-a. Treatment of monocyte and alveolar macrophage culture supernatants with this anti-TNF-a MAb significantly reduced their cytotoxic activity as determined by the WEHI- 164 cell assay (Fig. 4, Table 1 ). We verified the effectiveness of this MAt, by demonstrating that 15 ig of MAb completely abrogated the cytotoxic activity of 100 pg of rHuTNF-a (92.3% ± 0.6% cytotoxicity for the control

itant with the culture supernatant. Preincubatlon of the culture supernatants with the MAb did not further enhance neutralization of cytotoxic activity. A control MAb against Influenza hemagglutinin had no effect on the cytotoxicity observed. Data

expressed are the mean representative experiment

SEM percent cytotoxicity of four performed.

±

of one

vs. 6.0% ± 1 .0% cytotoxicity with 15 ig of antibody). A control MAb of identical isotype (IgG 1 ) directed against an irrelevant epitope (influenza hemagglutinin) did not have any effect on the cytotoxic activity of culture supernatants from LPS-stimulated blood monocytes or alveolar macrophages (76.9% ± 5.7% cytotoxicity for an active LPS-stimulated monocyte culture supernatant and 72.5% ± 8.3% cytotoxicity for the same supernatant treated with 75 ig of control MAb).

Size-Exclusion

HPLC of Bovine

TNF-ot

Size-exclusion HPLC was utilized to estimate the size of TNF-a released from bovine mononuclear phagocytes. To obtain the large amounts of TNF needed for purification, concentrated ( 100-fold) culture supernatants from LPS-stimulated bovine monocytes were run over a GPC-lOO size exclusion column. Fractions were collected and assayed for cytotoxic activity in the WEHI164 cell assay (Fig. 5). The resulting molecular weight elution profile was similar to that observed when 1 ng/ml of rHuTNF-a was run over the same column. This MW elution profile is consistent with the presence of multimers of TNF-a. Similar results were obtained when concentrated (20-fold) culture supernatants from LPSstimulated alveolar macrophages were run over the same column (Fig. 5). Cytotoxicity of active fractions was neutralized by the addition of an MAb directed against rHuTNF-a (56. 1% ± 2.4% cytotoxicity for an active fraction versus 2.4% ± 1 .9% for the same fraction treated with 125 pg of antibody).

554

Adams

and Czuprynski 8Is,

!

A 70

50

I

I

60 LPS-StmuIat.d monocyte

eupematant

cutture

50

.

C)

4

0

!;:

30

0

Q

20 10

B 70

60 LPS-St)mulated >%

. C)

0

Alveolar

50

MO Culture

Supematant

4 30

0

20 10

C 100 90 80 rHuThF-e

.‘70 .

60

.

5

(1

ng/ml)

‘40 Q

30

20 10 5

10

15

20 Elution

25

30 Time

35

40

45

50

55

60

(Minutes)

Fig. 5. Size-exclusion HPLC analysis of TNF-a in culture supernatants from LPS-stimulated monocytes and alveolar macrophages. LPS-stimulated monocyte culture supernatants were concentrated 100-fold and run over a GPC-100 size exclusion

column

as described (A). LPS-stimulated

supernatants

column

were concentrated

alveolar macrophage

20-fold and run over the same

(B). Fractions were collected at 5 mm intervals for 60

mm. As a control, 1 ng/ml of rHuTNF-a was run over the same column (C). Fractions were collected at 1 mm intervals for 60 mm. All fractions were assayed for TNF activity at a final concentration of 1 :2. Data expressed are the mean ± SEM percent cytotoxicity of fractions from one representative experiment of four performed.

DISCUSSION These data are believed to represent stration of the release of TNF-a from

the first demonbovine monocytes

and alveolar macrophages in vitro. Using the standard L 929 cell assay for TNF-a, we were unable to detect bovine TNF-ct in culture supernatants from LPS-stimulated peripheral blood monocytes. A number of bovine cells were tested for their susceptibility to bovine TNF-a, none of which were susceptible to bovine TNF-a. Our inability to detect TNF-a activity with these cell lines could be due to the lack of susceptibility of the cell lines tested or to the low levels of TNF-a being assayed. Utilizing murine WEHI- 164 fibrosarcoma cells, which are more sensitive to the cytotoxic activity of murine TNF-a than are L929 cells [1 8], we were able to detect significant TNF-a activity in LPS-stimulated bovine monocyte and alveolar macrophage culture supernatants. Actinomycin D treatment of the WEHI-l64 cells was not required for the detection of bovine TNF-a, although treatment enhanced the cytotoxicity of dilute culture supernatants approximately two fold. Bovine TNF-a was released from bovine monocytes following stimulation with as little as 20 ng/ml of bacterial LPS . This dose of LPS is similar to doses used in other studies ofTNF-ct release in vitro [6,9]. Cytotoxic activity peaked at about 2 hr and declined to undetectable levels by 18 hr post-LPS treatment. No second peak of cytotoxic activity was observed over a 48 hr culture period after LPS addition. Maturation of monocytes in vitro for 1-7 days resulted in the loss of their ability to release TNF-a in response to LPS. TNF-ct has been demonstrated in culture supernatants from LPS-stimulated monocytes and macrophages from human beings, mice, and rabbits under conditions similar to those used in this study [6, 15, 17]. The kinetics of TNF-a release in this study are similar to those reported by Fischer and Rubinstein [10], who demonstrated the rapid release and subsequent decay of TNF-a activity in human monocyte culture supernatants. Our results are also consistent with studies of TNF-a mRNA transcription that demonstrated the rapid induction and subsequent decline of TNF-a mRNA [4] . In contrast, some studies with murine and rabbit monocytes and macrophages demonstrated elevated TNF-a levels in the conditioned medium for up to 24-48 hr after LPS stimulation [9, 1 1 1 . We performed preliminary experiments with rHuTNF-a suggesting that the loss of TNF-a activity over time could be attributed largely to binding by monocytes (Table 3). In addition, the release of proteases or TNF-a inhibitors from monocytes, or the nonspecific adsorption of TNF-a to tissue culture wells , may have also made minor contributions to the observed loss of TNF-a activity. Analysis of bovine TNF-a by size-exclusion HPLC resulted in an MW elution profile that was consistent with multimers of rHuTNF-a. TNF-a has been reported to occur as a monomer and to form dimers , trimers, and even larger aggregates [23] . The active form of human TNF has been demonstrated to be a trimer with a relative

LPS-Stimulated MW of approximately 50-55 kD [23]. Our HPLC analysis of rHuTNF-a, which yielded biologically active fractions in this molecular weight range, is consistent with this previous report of trimers of human TNF-a. Analysis of LPS-stimulated alveolar macrophage and monocyte culture supernatants by size-exclusion HPLC also yielded biologically active fractions with an MW range that is consistent with multimers of TNF-a. These data suggest that bovine TNF-a, like human TNF, spontaneously forms aggregates. Little is known about TNF-a production, or its biological activities, in ruminant and other food-animal species. Transcription of TNF-a mRNA has been demonstrated in LPS-stimulated porcine alveolar macrophages (M.J. Baarsch, and M.P. Murtaugh, Abstract 3 10, 70th annual Conference of Research Workers in Animal Disease, November, 1989, Chicago). The role that TNF-a release plays in endotoxemia in ruminants is being examined. We have previously demonstrated circulating levels of bovine TNF-a in sera obtained from endotoxemic neonatal calves [1] . Other investigators have reported that the response to rHuTNF-a in sheep mimicked the response that occurred after endotoxin administration [1 3]. These results suggest that TNF-a is an important inflammatory mediator that is released in response to endotoxin in ruminant species. In addition to LPS, many other bacterial, parasitic, and fungal agents have been shown to be capable of inducing TNF-a release [ 1 1 ,26] . The release of TNF-a has been described in vivo during the course of many different acute and chronic infections [25]. It is likely that the amount, timing, and duration of TNF-a that is released during the course of these infections may determine whether its effects are beneficial or detrimental. Chronic release or overproduction of TNF-a results in cachexia, shock, and in some cases death [27] . Prolonged release of TNF-a has been suggested to play a role in the cachexia observed in many chronic infections [27]. Two chronic infections in which a profound cachectic state is observed in cattle are trypanosomiasis (Trvpanosoma brucei brucei) and Johne’s disease (Mycobacterium paratubercu/osis) [7,22]. These are chronic infections in which infected cattle become profoundly cachectic, losing up to 50% of their lean body mass. It is likely that TNF-a release plays an important role in the cachexia seen in these animals. The results of this study will facilitate further investigations into the role of TNF-a in chronic infections in cattle.

ACKNOWLEDGMENTS We thank Dr. G. Trinchieri for generously providing the hybridoma cells that produce the anti-rHuTNF-a MAb. We also thank James Brown for his technical assistance in purifying the anti-rHuTNF-a monoclonal

Bovine

TNF-a

Release

In Vitro

555

antibody; and we also appreciate the helpful suggestions of Dr. R. Schultz and Dr. K. Schultz. This work was supported by funds from the Wisconsin Agriculture Experiment Station (WIS 0238), the USDA (87-CRCR1-2308), the U.S. Public Health Service (AI-21343), and the University of Wisconsin School of Veterinary Medicine.

REFERENCES 1. Adams, iL.. Semrad, S.D., and Czuprynski, Ci. Administration of bacterial lipopolysaccharide elicits circulating tumor necrosis factor-alpha in neonatal calves. J. Clin. Microbiol. 28,998,

1990.

2. Aggarwal. B.B. . Kohr. Wi. . Hass. P.E. . Moffat. B. . Spencer, S.A. . Henzel, Wi. . Bringman, T.S. , Nedwin, G.E. , Goeddel, DV. . and Harkins. RN. Human tumor necrosis factor. Production. purification. and characterization. J. Biol. Chem. 260,2345. 1985. 3. Beutler. B. , Milsark, 1W. . and Cerami, A. Passive immunization against cachectin/tumor necrosis factor (TNF) protects mice from the lethal effect of endotoxin. Science. 229,869, 1985. 4. Burchett, 5K. . Weaver, W.M. , Westall. iA. , Larsen. A.. Kronheim. S. . and Wilson, C.B. Regulation of tumor necrosis factor/cachectin and IL- I secretion in human mononuclear phagocytes. J. Immunol. 140.3473. 1988. 5. Carswell. E.A. . Old. Li.. Kassel. R.L. Green. S. . Fiore. N.. and Williamson. B. An endotoxin-induced serum factor that causes necrosis of tumors. Proc. NatI. Acad. Sci. USA 72,3666, 1975. 6. Chen. AR. . Mckinnon. K.P. . and Koren, H.S. Lipopolysaccharide (LPS) stimulates fresh human monocytes to lyse actinomycin D-treated WEHI-164 target cells via increased secretion of a monokine similar to tumor necrosis factor. J. Inimunol. 135. 3978. 1985. 7. Chiodini. Ri. , Van Kruiningen. Hi. . and Merkal. R.S. Ruminant paratuberculosis (Johne’s disease): The current status and future prospects. Cornell Vet. 74,217. 1984. 8. Cuturi. MC. , Murphy. M., Costa-Giomi, M.P. Weinmann, R.. Perussia. B. , and Trinchieri, G. Independent regulation of tumor necrosis factor and lymphotoxin production by human peripheral blood lymphocytes. J. Exp. Med. 165.1581. 1987. 9. Fisch. H. . and Gifford, G.E. In vitro production of rabbit macrophage tumor cell cytotoxin. Int. J. Cancer. 32,105. 1983. 10. Fischer. D.G. , and Rubinstein, M. Human nionocytes tumoricidal activity: The role of interferon-gamma and bacterial lipoplysaccharide in its stimulation. preservation. and decay. Immunobiol-

,

,

ogy

172.110.

1986.

GD. , and Flick, D.A. Natural production and release of tunior necrosis factor. In Tumor Necrosis Factor and Related Cytotoxins. (Bock. G.. and Marsh, J.. eds.). Ciba Foundation.

I I . Gifford.

Symp.

12.

131.3,

1987.

Hesse.D.G..Tracey,K.J..Fong.Y..Manogue.K.R..Palladino. MA. Jr. . Cerami. A. , Shires, G.T. , and Lowry. S.F. Cytokine appearance in human endotoxemia and primate bacteremia. Surg. Gynecol. Obstet. I 66. 147. 1988. 13. Johnson. J.. Myerick. B.. Jesmok. G.. and Brigham, K.L. Human recombinant tumor necrosis factor alpha infusion mimics endotoxemia in awake sheep. J. AppI. Physiol. 66. 1448, 1989. 14. Le. J., and Vilcek. J. Biology of disease. Tumor necrosis factor and interleukin 1: Cytokines with multiple overlapping biological activities. Lab. Invest. 56,234. 1987. 15. Mannel. D.N. . Moore. RN. . and Mergenhagen. SE. Macro-

556

Adams phages

and Czuprynski

as a source

factor). Infect. Immun.

16.

Mathison, iC. tumor necrosis lipopolysaccharide 81,1925,

.

,

of tumoricidal 30,523,

activity

(tumor

necrotizing

Wolfson, E. and Ulevitch, Ri. Participation of factor in the mediation of gram negative bacterial induced injury in rabbits. J. Clin. Invest.

,

,, , ,

,, , , , ,

triglyceride

removal.

Mol.

Biochem.

Parasitol.

2.31,

1980.

23. 24.

1988.

Matthews, N. Tumour necrosis factor from the rabbit. II. Production by monocytes. Br. J. Cancer. 38,3 10, 1978. 18. Meager, A. Leung. H. . and Woolley, J. Assays for tumour necrosis factor and related cytokines. J. Immunol. Methods 116,1, 1989. 19. Michaels, F.H. , and Banks, K.L. Contribution of various host factors to resistance to experimentally induced bacterial endotoxemia in calves. Am. J. Vet. Res. 49,557, 1988. 20. Michie, HR. Manogue, KR. Spriggs, DR. Revhaug, A., O’Dwyer, S. Dinarello, CA. Cerami, A. Wolff, SM. and Wilmore, D.W. Detection of circulating tumor necrosis factor after endotoxin administration. N. Eng. J. Med. 3 18, 148 1 1988. 21 . Rosenblum, MG. and Donato, N.J. Tumor necrosis factor-a: A multifaceted peptide hormone. Crit. Rev. Immunol. 9,21, 1989. 22. Rouzer. CA. and Cerami, A. Hypertriglyceridemia associated with Trvpanosoma brucei brucei infection in rabbits: role of 17.

defective

1980.

25.

1,454,

26. 27.

28.

, , , , , ,, ,

Smith, R.A. . and Baglioni, C. The active form of tumor necrosis factor is a trimer. J. Biol. Chem. 262.6951, 1987. Tracey. K.J. Beutler, B. , Lowry, S.F. . Merryweather, J., Wolpe. S. , Milsark, 1W. , Hariri, Ri. , Fahey, T.J. . III, Zentella, A. . Albert, J.D. Shires, G.T. . and Cerami, A. Shock and tissue injury induced by recombinant human cachectin. Science. 234,40, 1986. Tracey. K.J. and Cerami, A. Cachectin/tumor necrosis factor and other cytokines in infectious disease. Curr. Opinion Immunol. 1989.

,, , ,

Tracey, K.J. Vlassara, H. , and Cerami, A. Cachectin/tumor necrosis factor. Lancet. 1 I 122, 1989. Tracey. K.J. Wei, H. Manogue, KR. Fong, Y. Hesse, D.G., Nguyen, H.T. Kuo, G.C. , Beutler, B. Cotran, R.S. Cerami, A. . and Lowry. S.F. Cachectin/tumor necrosis factor induces cachexia, anemia. and inflammation. J. Exp. Med. 167,1211, 1988. Zurbrick, B.G. , Follett, D.M. and Czuprynski, C.J. Cytokine regulation of the intracellular growth of Mvcobacterium paratuberculosis in bovine monocytes. Infect. Immun. 5,1692, 1988.

,

Bacterial lipopolysaccharide induces release of tumor necrosis factor-alpha from bovine peripheral blood monocytes and alveolar macrophages in vitro.

In this study, we demonstrate that freshly adherent bovine monocytes release tumor necrosis factor-alpha (TNF-alpha) in response to stimulation with b...
1MB Sizes 0 Downloads 0 Views