infection. and body fluids during acute bacterial ...

Multiple forms of IFN-beta 2/IL-6 in serum and body fluids during acute bacterial infection. D C Helfgott, S B Tatter, U Santhanam, R H Clarick, N Bhardwaj, L T May and P B Sehgal J Immunol 1989; 142:948-953; ; http://www.jimmunol.org/content/142/3/948

Subscriptions Permissions Email Alerts

Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscriptions Submit copyright permission requests at: http://www.aai.org/ji/copyright.html Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/cgi/alerts/etoc

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 9650 Rockville Pike, Bethesda, MD 20814-3994. Copyright © 1989 by American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606.

Downloaded from http://www.jimmunol.org/ at New York Med College Health Sciences Library on October 12, 2016

This information is current as of October 12, 2016.

0022-1767/89/1423-0948$02.00/0 THEJounNAL OF IMMUNOLOGY Copyright 0 1989 by The American Association of lmmunologists

VOl. 142.948-953. No. 3. February 1. 1989

Printed in U.S.A.

MULTIPLEFORMS OF IFN-&/IL-6 IN SERUMANDBODY ACUTE BACTERIAL INFECTION’

FLUIDS DURING

DAVID C. HELFGOTT, STEPHEN B. TATTER, UMA SANTHANAM,ROBERT H. CLARICK, NINA BHARDWAJ,LESTER T. MAY, AND PRAVINKUMAR B. SEHGAL’ From the RockefellerUniversity, New York. NY 10021

These alterations are thought of as beneficial to the infected host and to play a role in limiting tissue damage (3-5). Theacutephasealterations in plasma protein composition, which are reflected in a n enhanced erythrocyte sedimentation rate (6-9), are elicited by cytokines released by damaged or infected tissues into the peripheral circulation which, in turn, modulate hepatic and non-hepatic plasma protein synthesis (10-13). It is now clear that the cytokine “IFN-P,/hepatocyte stimulating factor/IL-6” (IFN-P2/IL-6)directly elicits many of the acute phase alterations in hepatic plasma protein syntheses (14-16). These include increases in synthesisof C-reactive protein,fibrinogen,a,-antichymotrypsin, a,-antitrypsin, thiol-proteinase inhibitor, complement C3, serum amyloid A, and a,-acidglycoprotein and a decrease in the synthesisof albumin (14- 16). Thecytokines IL- 1 and TNF can also elicit increases in plasma levels of numerous acute phase proteins in animal experiments (17). However, the direct enhancement of acute phase plasma protein gene expression in hepatocytes by TNF or IL-1 is limited (12, 14, 18, 19) suggesting that IFN-P,/ IL-6, which is readily induced by IL-1 and TNF (20, 21), may largely account for the in vivo effects of these cytokines. Furthermore, it has proven difficult to detect the presence of biologically active IL-1 or TNF in the peripheral circulation duringthe “acute phase” (22). Expression of the human IFN-P,/IL-6 gene, located on chromosome 7p21 (23-25),is readily induced in different cell-types by a wide range of “noxious”stimuli. Bacterial products such as the LPS (endotoxin) strongly enhance IFN-P2/IL-6gene expression in cultures of human fibroblasts and monocytes/macrophages (10, 26, 27).IFN-P,/ IL-6 gene expression in human fibroblasts is also enTheplasmaproteinresponsecharacteristic of the hanced by infection with severaldifferent RNA- and DNA-containing viruses (28) andby inflammation-asso“acute phase” of bacterial infections includes increases platelet-derived in the levels of C-reactive protein, fibrinogen, other clot- ciated cytokines such as IL-1,TNF, growth factor, and other IFN (20, 21). Endothelial cells, ting factors, complement factors, several anti-proteinases, serumamyloid A, haptoglobin, hemopexin and nu- keratinocytes, andother epithelial and mesenchymal merous other proteins, and decreases in levels of proteins cells also express IFN-P2/IL-6 in response to many of such a s albumin, transcortin, and transthyretin (1-5). these stimuli (29). At least six forms of differentially modifiedIFN-P2/IL-6 phosphoglycoproteins in the size range from 2 3 to 30 kDa are secreted by induced fibroReceived for publication September8. 1988. Accepted for publication November 9, 1988. blasts andmonocytes (15, 16, 27-30).3 We have investiThe costs of publication of this article were defrayed in part by the gated the question whether serum and body fluids of payment of page charges. This article must therefore be hereby marked advertisement in accordance with18 U.S.C. Section 1734 solely to indi- patients with acute bacterial infections contain biologcate this fact. ically active IFN-P2/IL-6.We report the detection of bio’ This work was supported by Research Grants AI-16262 and CA- logically active IFN-@,/IL-6in patients with a variety of 44365 from The National Institutes of Health, by Medical Scientist Trainacute gram-positive and -negative bacterial infections in ing Program GM-07739 (S. B. T.), a contract from TheNational Founda-

tion for Cancer Research, and an Established Investigatorship from The American Heart Association(P. B. S . ) . * Address correspondence and reprint requestsDr.toP. B. Sehgal. The Rockefeller University, 1230York Avenue, New York. NY 10021.

Santhanam, U., J.Ghrayeb, P. B. Sehgal, andL. T. May. 1989. Posttranslationalandpost-secretorymodifications of human IFN-BzIIL-B. Submlttedfor publication.

948


Many of the major alterations in plasma proteins characteristic of the hepatic acute phase response are regulated by IFN+?,/IL-6. Usinga specific bioassay for IFN-B,/ILS, which relies on the induction of the hepatic acute phase plasma protein a,-antichymotrypsin in the human hepatoma cell line Hep3B clone 2 and its inhibition by anti-rIFN-B,/IL-6 antiserum, we have detected high levels of IFN-@,DL-6 in the body fluids of patients with acute bacterial infections. Cerebrospinal fluid from four patients with acute bacterial meningitis (Streptococcus pneumoniae, Staphylococcus aureus, two cases of Listeria monocytogenes)all had high levelsof IFNBzPL-6 (up to 500 ndml). Two of these patientswith concomitant bacteremia had lower concentrations ofIFN-&/IL-6 in the serum (5 to 70 ndml). Three additional patients with Escherichia coli,Pseudomonas aeruginosa, and Neisseria meningitidis bacteremia had high levelsof serum IFN-@,/IL-6,as did the ankle fluid of a patient with Streptococcus canus arthritis. Normal cerebrospinal fluid and serum had little detectable IFN-B,/ILS. A combination of immunoaffinity chromatography and immunoblotting procedures were usedto characterize the IFN-&/IL-6 species present in a representative sampling of serum and cerebrospinal fluids. Multiple immunoreactive species ofIFN-@,DL-6 in the size range 23 to30 kDa as well as immunoreactive complexes in the range 60 to 70 kDa were detected in human body fluids. This is thefirst demonstration that previous descriptions of heterogeneity in human IFN-@,/IL-6species produced in cell culture correspond to observations in the infected host.

IFN-PzIIL-6 SPECIES IN HUMANBODY

concentrations sufficient toelicit enhanced synthesisby hepatocytes of acute phase plasma proteins. We have also characterized IFN-B2/IL-6species present in human body fluids and report the detection of a complex array of IFN-P2/IL-6immunoreactive species in the size range 23 to 30 kDa in addition to apparentlypolymeric IFN-&/ IL-6 complexes.

949

FLUIDS RESULTS


The characteristics of the IFN-P2/IL-6 bioassay based upon its ability to enhance ACT synthesis andsecretion by human hepatoma Hep3B2 cells obtained directly from the ATCC (cat. no. 8064) aresummarized in Figure 1 (see Ref. 31 for a discussion of the regulation of acute phase plasma protein secretionin different hepatoma cell lines and different stocks of what can nominally be the same MATERIALS AND METHODS cell line). Figure 1A illustrates a n experiment in which Procurement of human bodyfluid and serum samples. Cerebrothe ability of different concentrations of IFN-P2/IL-6to spinal fluid samples from patientswith a n initial clinical presentaenhance ACT synthesis was determined. Figure 1B is a tion of meningitis were obtained from among the stored unused specimens in the Biochemistry Laboratoryof The New York Hospital representation of the densitometric data obtained from after determinationby the Microbiology Laboratory of The New York the autoradiogram in A. A linear relationship was obHospital that thefluid was culture-positive. Similarly,"control" cereserved between the absorbance(in arbitrary units) of the brospinal fluid samples wereobtained frompatientspresenting without acute infections. Joint fluid was obtained from discarded autoradiographicsignal of the [35S]methionine-labeled specimens in the course of repeated paracenteses for the therapy of ACT and thelogarithm of the concentration of IFN-P2/ILseptic arthritis. Serum samples from culture-positive blood were also obtained from among thestored unused specimens in theBiochem- 6 in this and in several additional experiments. Figure 1C shows that IL-1 does not enhance ACT synthesis in istry Laboratory of The New York Hospital. Similarly, control serum samples were obtained from patients presenting without acute in- this assay nor does it affect the ability of IFN-P2/IL-6 to fections and from healthy individuals. In replicate assays on aliquots do so (the IL-1 used was biologically active; it induces of the same sample of body fluids we observed little difference in IFN-P2/IL-6mRNA expression in human fibroblasts). In IFN-Bz/IL-6 levels whether the aliquots had been kept a t 4°C or a t -20°C. additional experiments we have determined that neither Assay forIFN-B,/IL-6 actiuity. The human hepatoma cell line IL- 1 at concentrations upto 50 ng/ml nor TNF at concenAmericanTypeCulture Hep3B clone 2 was purchased from the Collection (Rockville, MD; Cat. 8064) and grown to 80 to 90% con- trations up to 100 ng/ml, affect ACT synthesis in this fluency in 24-well Falcon tissue culture plates as described elseassay. It is already established that bacterial products where (1 6. 28). Appropriately diluted (usually 1/5)test sampleswere such as LPS do not affect plasma protein synthesis by cultures (0.5 ml medium/well) without or with added to theHep3B2 hepatocytes (10, 14); we have verified the fact that LPS the addition of rabbit polyclonal neutralizing antiserum (1/50 or 1/ 100 dilution) prepared against Escherichia coli-derived human IFN- at concentrations up to5 pgml does not alter ACT synBZ/IL-6(16). After incubation for 24 h a t 37°C the cultures were thesis in Hep3B2 cells. Figure 1D demonstrates that the washed with PBS and incubated for a n additional 24 h in medium containing 35S-methionine (100 pCi/ml. Du Pont-NEN. Boston, MA). culture medium from LPS-activated human monocytes The culture medium from each well was then collected and the enhances ACT synthesis and that this activity is comamount of labeled ACT4 secreted was quantitated by immunoprecippletely neutralized by the antiserum to E. coli-derived itation (rabbit antiserum to human ACT was purchasedfrom Calbihuman IFN-P,/IL-6. Inasmuch as these crude culturemeochem. La Jolla, CA).PAGE, autoradiography. and densitometry dium samples are expected to contain IFN-P2/IL-6,IL- 1, (LKB Ultroscan XL laser densitometer) using procedures described earlier (16, 28-30). Appropriate dilutions of a laboratory standard and TNF (expression of IL-1 in this experiment was dipreparation of IFN-P2/IL-6prepared from IL-1-induced human fibrorectly verified by immunochemical staining of the LPSblasts (16)were included in every assay. This preparation contains treated monocytes), the antibody neutralization data in a t least five forms of the lFN-B2/IL-6in the size range 2 3 to 30 kDa as judged by immunoblots. The total concentration of the different D further verify the specificity of the ACT assay. Furforms of IFN-&/IL-6 in this preparation was quantitated comparin thermore, the data in D , lanes l to 3 demonstrate that isonswith electrophoreticallyhomogeneous Escherichia coli-derived IFN-&/IL-6 in immunoblot assays using theanti-r1FN-B2/IL-6 human peripheral blood monocytes do not"constituantiserum. tively" secrete IFN-P2/IL-6even after isolation by plastic Induction of human monocytes byLPS. Plastic adherent human adherence, provided that theculture medium is LPS-free. monocyte cultures were prepared using standard procedures (16, 30).Bacterial lipopolysaccharide (Salmonella typhosa, strain 0901) Figure 2 presents a n overview of our observations on was purchased from Difco (Detroit, MI). the presence of IFN-P2/IL-6 in serum and body fluids lmmunoaffinitychromatographyandimmunoblottingproceduring acute bacterial infections. The right-most lanes dures. Rabbit Ig raised against purified E. coli-derived human IFNin each panel are assays of the IFN-P2/IL-6 standard &/IL-6 (16) was cross-linked to cyanogen bromide-activated Sepharose 4B (Sigma Chemical Co., St. Louis. MO). Body fluid samples (0.5 preparation at different concentrations.A 1/5 dilution of to 0.7 ml) were diluted 20-fold in PBS and passed through immueach body fluid or serum sample was assayed without noaffinity columns pre-equilibrated with PBS. The columns were washed with PBS and then eluted with 0.1 M glycine, pH 2.2. and with admixture of anti-rIFN-Pz/IL-6antiserum and the effects on ACT synthesis displayed in adjacent lanes Aliquots of appropriate flow-through and eluate fractions corresponding to approximately 50 pl body fluid were electrophoresed in the autoradiograms. Table I summarizes key clinical through a SDS-PAGE (17.5%) under reducing conditions, electrodata for each case studied. blotted and probed using the anti-rIFN-flz/IL-6 antiserum and the CSF (six samples) and serum(six samples)from indiABC Elite Vectastain kit (Vector Laboratories, Burlingame, CA). Other reagents. E. coli-derived human IFN-&/IL-6 and rabbit viduals without acute infections had little detectable IFNE. polyclonal antiserum to it were prepared as described earlier (1 6). p,/IL-6 by bioassay. CSF from four patients with acute coll-derived human I L - l a was a gift from Hoffman-LaRoche, Nutley. bacterial meningitis (Streptococcus pneurnoniae, N J ; E. coll-derived human TNF and a neutralizing mAb to it were gifts from The Suntory Institute for Biomedical Research, Osaka, Staphylococcus aureus, two cases of Listeria rnonocyJapan. Neutralizing mAb to humanI L - l a and -p were gifts fromDr. togenes) all contained high levels of IFN-P2/IL-6(in the A. C. Allison (Syntex Research, Palo Alto, CA).The biologic activities range from 70 to 500 ng/ml). Two of these patients with of the I L - l a and TNF preparations were verified by their ability to lower concentrations of IFNstrongly induce IFN-&/lL-6 mRNA in human fibroblasts and those concomitant bacteremia had and of the neutralizing Ig preparations were verified both before p,/IL-6 in the serum (5-8.5ng/ml). Three additional paafter experimental use in appropriate bioassays. tients with bacteremia (E. coli, Pseudomonas aerugiAbbreviation usedin this paper: ACT, u,-antichymotrypsin. Rosa and Neisseria rneningitidis) had serum levels of

950

IFN-&/IL-6 SPECIES IN HUMAN BODY FLUIDS

1 2

B

1234 56 7 8

I IO

4

ILr+.-

5

1234 56 7 8 6

2 4

I

1 2 3 4 5 6 7 8 91011

3 45 6 7 8 9 D l l u n # b ~

IO ABSORBANCE 5

4

15

IFN-P,/IL-6 in the range5 to 70 ng/ml. A clear fibrinclot or crusta inflammatoriawas observed when blood from the patientwith P . aeruginosa bacteremia (no. 6, serum A a - * IFN-P2/IL-6 level 70 ng/ml) was allowed to stand. High IFN-P,/IL-6 levels (190 ng/ml) were also observed in the 12345678 ankle fluid of a patient with Streptococcus canus arthritis. In every case where enhancement of ACT syntheBG 6 4 sis was observed in the assay used, it was blocked bythe anti-rIFN-@,/IL-6antiserum (Fig. 2). The addition of neutralizing antibodies to IL-l((u + 8) or a neutralizing anti1 2 3 4 5 6 7891011um4~& body to TNF did not affect the ACT synthesis observed (Fig. 20. lanes 3 to 6 and additional experiments; the cw b" 4 potency of the neutralizing Ig preparations was verified both before and after experimental use in appropriate bioassays). It is clear from Figure 2 and Table I that all 1 2 3 4 5678910lll2l.314 septic fluids that we have tested scored strongly positive D Y " 4 for IFN-P,/IL-6. IFN-P,/IL-6 species in infected human body fluids. Figure2. IFN-B,/IL-6 activity in body fluids and serum of patients with acute bacterial infections.Composite autoradiograms illustrate the One of the striking featuresof human IFN-@JIL-6 derived detection of IFN-Bz/IL-6 byits ability to enhance ACT synthesis inHep3B2 from a variety of cell-culture sources is its marked hetcells: each panel represents a separate assay: assays on each test sample erogeneity as judged by SDS-PAGE under reducing conare illustrated without or with the inclusion of anti-rlFN-fiz/lL-6 antiserum (1/50 dilution). Patient numbers in parenthesis (e.g. No. 1 ) refer to ditions (15, 16. 30) (see footnote 3). We have evaluated descriptions in Table I. Multiple ACT bands observed in assays with high whether this heterogeneity is also seen in vivo. IFN-P2/ IFN-fi2/IL-6activity represent differentially glycosylated forms of this IL-6 immunoreactive forms in serum from the patient plasma protein. A. Lane 1 . untreated Hep3B2 cells: lanes 2 to 3. CSFlate latent syphilis (no.9):lanes 4 to 5 and 6 to 7.CSF-S. pneumoniae with P . aeruginosa bacteremia (no. 6 in Table I) were meningitis on day 1 [no. 1): lanes 8 to 9.CSF from the same patient on evaluated by a combination of immunoaffinity chromaday 3 (no. 1): lanes 10 to 1 1 . CSF-S. aureus meningitis and sepsis (no. 2): lanes 12 to 13. serum from same patient [no. 2): lanes 14. 15. 16. tography and immunoblotting. Figure 3 shows clearly IFN-fi2/IL-6at IO.2. and 1 ng/ml. B.Lanes 1 to 2. CSF-L. monocytogenes that this serum contains high levels of IFN-PJIL-6 species meningitls (no. 3): lanes 3 to 4. ankle joint effusion-S. c a m s arthritis of molecular mass 23 to 30 kDa (lanes4 to 6).This serum and bacteremia [no. 8 ) :lanes 5 to 8.IFN-fi,/IL-6 at 10. 2. 1. and 0 ng/ml. also containsa n immunoreactive complex of M , 60 to 70 C . Lanes 1 to 2. CSF-L. monocytogenes meningitis (no. 4): lanes 3 to 4. CSF and lanes 5 to 6, serum from a patlent with E. coli bacteremia and kDa which binds weakly to the anti-rIFN-PJIL-6 column HIV infection (no. 7): lanes 7 to 8.serum-P. aeruginosa bacteremia (no. but which is readily detected after denaturation in the 6): lanes 9 to 16. IFN-B,/IL-B at 10. 2 . 1 , and 0 ng/ml in duplicate lanes. D.Lanes 1 to 2. CSF-inactive cerebral cysticercosis (no. 10): lanes 3 to immunoblot procedure (lanes 2 and 3).The generalityof 4. serum-P. aeruginosa (no. 6) and lanes 5 and 6 is the samewith anti- these observations was tested in an experiment in which TNF and anti-IL-l(n + 8) antibody: lanes 7 to 8.serum-acute meningoIFN-P2/IL-6 species were purified from three different coccemia (no. 5): lanes 9 to 14. IFN-B2/IL-6 at 2. 1. and 0 ngjmlin duplicate. cerebrospinal fluid samples (Fig. 4). In every instance multiple IFN-P2/IL-6species of M , 23 to 30 kDa (and high molecular mass immunoreactive complexes) were detected. The concentration of IFN-P2/IL-6detected by im-

-

--


Figure 1 . Assay of human IFN-fi2/IL-6 in Hep3B2 cultures. A. Autoradiogram showing the enhancing effect of different concentrations of fibroblast-derived IFNB2/IL-6 on the synthesis and secretion of ["'Slmethionine-labeled 68-kDa ACT (arrowhead).Lanes I . 2: 10 ng/ml: lanes 3. 4: 2 ng/ml: lanes 5. 6: I ngml: lanes 7. 8: untreated cells. B. Dose-response rela- A tionship derived from data in A by densitometryexpressedinarbitraryabsorbance units.C , Autoradiogram showingthe lack of effect of IL-1 on ACT synthesis. Lane 1 , untreated control: lanes 2 to 4. cultures treated withIFN-fiZ/1L-6at 10. 2. and 1 ng/ml: lane5. IL-1 at 1 n a m l : lanes 6 to 8.cultures treated with IL-1 at 1 ng/ ml and with IFN-fi2/1L-6at 10. 2. or 1 ng/ C ml. D. Autoradiogram showing the efficacy of anti-rlFN-fl,/IL-B antiserumin blocking enhancement ofACT synthesis in response to culturemedium from LPSInduced monocytes. Lane 1 is theuntreated Hep3B2 control: subsequent evennumbered lanes represent assays on culture medium[ 1/5 dilution) from untreated monocytes (lane 2 )or from monocytes in- D duced with LPS at 20pg/ml (lane 4 ) . 200 pg/ml (lane 6 ) .2 n a m l (lane8).o r 2 0 ng/ ml (lane 10): lanes 3. 5. 7. 9. and I I represent assays on monocytic medium in theeven-numberedlanes mixed with anti-rlFN-fi,/IL-6 ( 1 /50 dilution)

IFN-BJIL-6

951

SPECIES IN HUMAN BODY FLUIDS

TABLE I Description ojpatients investigated Dlagnosls Primary

Age

Sex

Dlsease Underlying

W Y Fluld

Organlsm

Acute meningitis

48

M

None

CSF"

S . pneumoniae

2

Acute meningitis. sepsis

48

M

None

CSF

S . aureus S . aureus

3

Acute meningitis.

30

M

Post-renal transplant

CSF

4

Acute meningitis

54

M

COPD". Prednisone therapy

5

Sepsis

46

M

6

Sepsis

65

7

Sepsis

38

Serum sepsis

a500 day 1 7.5day 3 70 8.5

L. monocytogenes

275

L. rnonocytogenes

7

CSF

L. monocytogenes

a500

None

Serum

N. meningitidis

F

Diabetes mellitus

Serum

P . aeruginosa

M

Asymptomatic HIV infection

Serum CSF

E. coli Sterile

(nu

2.5 70 5 None detected'

8

Septic arthritis

71

F

Diabetes mellitus

Ankle

S . canus

9

Late latent syph-

60

M

None

CSF

Sterile

None detected'

30

F

None

CSF

Sterile

None detected'

190

ills

cerebral 10 Inactive cysticercosis

Cerebrospinal fluid. bChronic obstructive pulmonary disease. CLimitof detection 51.5 ng/ml.

1 I

2

-7

4 5 "-

3 -7

6

kDa 31> 21, Figure 3. IFN-&/IL-6 species in serum of a patient withP . aeruginosa septicemia. Serum from patient 6 (Table I) was subjected to immunoaffinity chromatography usinga n anti-rlFN-&/IL-6 Ig column and aliquotsof the flow-through and eluate fractions corresponding to approximately 50 FI of serum were subjected to SDS-PACE (17.5%)and immunoblotting. Lane 1 . IL-la-induced fibroblast-derived IFN-&/IL-6 (2ng): lanes 2 and 3. flow-through and lanes 4 to 7.eluate fractions.

munoblotting is consistent with that estimated by the ACT bioassay. We have confirmed that the eluted immunoaffinity-purified IFN-@,/IL-6 species derived from serum and cerebrospinal fluid are biologically active in both the ACT assay and ina B cell differentiation assay (data not shown).

A s

a b c B 1 2 3 123123 s

c 1

kDa 31

,

21,

Ffgure 4. IFN-B2/IL-6 species in cerebrospinal fluid of patients with acute bacterial meningitides. Cerebrospinalfluid from patients 10 [a).2 ( b ) .and 1 ( c ) was evaluated for IFN-p2/IL-6 species by a combination of immunoaffinity chromatography and immunoblotting.A. Lanes. IL-lainducedfibroblast IFN-f12/IL-6 (2 ng): lanes I to 3 represent aliquots corresponding to 50 p l of CSF from the first three anti-rlFN-B2/IL-6 Ig column eluates. B illustrates a reanalysis of fraction 1 from patient c.

are readily available. We have alsodeveloped a combination of immunoaffinity chromatography and immunoblotting procedures using anti-rIFN-B2/IL-6Ig to characterize themolecular IFNp2/IL-6 species presentin human body fluids. It is striking that humanbody fluids containmultiple immunoreactive species of IFN-@JIL-6. DISCUSSION Body fluids of patients withlocal acute bacterial infecof patients with gram-negative and The biologic assay used in this study is based on the tionsandserum ability of IFN-&/IL-6 to enhance the synthesis of ACT by -positive bacteremiacontain elevated levels of biologthe humanhepatoma cell line Hep3B clone 2. This assay ically active IFN-P2/IL-6.The highest levels (2500ng/ml) in our hands when carried out using Hep3B2 cells ob- were observed in fluids in localized infected comparttained directly from the ATCC is specific forIFN-P2/IL-6; ments. Concentrations in the range from 5 to 70 ng/ml it is unaffected byLPS.IL-1, or TNF. This bioassay were observed in theperipheral circulation. These obsermeasures an effect of IFN-BJIL-6 on human hepatocytes vations indicate that IFN-@,/IL-6is a participant in the that is of intrinsic physiologic significance. The Hep3B2 host response toacute bacterial infections. In cell culture experiments crude bacterial extractsas cell line and antiserumto human ACT used in the assay


1

IFN-&/IL-6 mll

952

IFN-BzlIL-6 SPECIES IN HUMAN BODY FLUIDS

23. Sehgal, P. B., A. Zilberstein, R. Ruggieri. L. T. May, A. FergusonSmith, D.L. Slate, M. Revel, and F. H. Ruddle.1986.Human Acknowledgments. We thank Dr. Igor Tamm for his chromosome 7 carries the pz interferon gene. Proc. N a t l . Acad. Sci. U S A 83:5219. enthusiastic encouragement and Dr. Ralph Steinman for 24. Ferguson-Smith, A., Y-F. Chen. M. S. Newman. L. T. May, P. B. numerous helpful discussions. Sehgal, and F. H. Ruddle. Regional localisation of the &-interferon/ B-cell stimulatory factor 2/hepatocyte stimulating factor gene to human chromosome 7p15-p21. Genomics 2:203. REFERENCES 25. Bowcock, A. M.. J. R. Kidd, M. Lathrop, L. Daneshvar. L. T. May. A. Ray. P. B. Sehgal, K. K. Kidd, a n d L.L. Cavalli-Sforza. 1988. 1. Tillet. W. S . , a n d T. Francis. 1930. Serological reactions in pneuThe human “beta-2 interferon/hepatocyte stimulating factor/intermonia witha non-protein somatic fractionof pneumococcus. J . Exp. leukin-6 gene: DNA polymorphism studies andlocalization to chroMed. 52561. 3:8. mosome 7 ~ 2 1Genornics . 2. MacLeod, C. M.. and 0. T. Avery. 1941. The occurrence during acute 26. Helfgott. D.C., L. T. May, Z . Sthoeger, I. Tamm. and P. B. Sehgal. infections of a protein not normally present in the blood. 11. Isolation 1987. Bacterial lipopolysaccharide (endotoxin) enhances expression and properties of the reactive protein. J . Exp. Med. 73: 183. and secretion of p2 interferon by human fibroblasts. J . Exp. Med. 3. Kushner, I. 1982. The phenomenon of the acute phase response. 166: 1300. Ann. N . Y . Acad. Sei. 389:39. 27. Tosato. G . . K. B. Seamon, N. D. Goldman. P. B. Sehgal. L. T. May. of 4. Schreiber, G.. a n d G. Howlett. 1983. Synthesis and secretion G. C. Washington, K. D. Jones, andS. E. Pike. 1988.Identification acute phase proteins.In: Plasma Protein Secretion by the Liuer.H. of a monocyte-derived human B cell growth factor as interferon-& Glaumann, T. Peters. and C. Redman. eds. New York; Academic [BSF-2,lL-6).Science 239:502. Press, New York, p. 423. C. Helfgott, U. Santhanam. S. B. Tatter, R. H. 5. Fey, G. H.. a n d G. M. Fuller. 1987. Regulation of acute phase gene 28. Sehgal, P.B..D. Clarick, J. Ghrayeb, and L. T. May. Regulation of the acute phase expression by inflammatory mediators.Mol. Blol. Med.4:323.


6. Jones, W.H. S . 1923. Hippocrates [an English translation of the well as purified LPS preparations directly induce the original Greek). In: The Nature of Man, Val. 4. Harvard University expression of IFN-P2/IL-6mRNA and proteins in human Press. Cambridge, p. 1 . diploid fibroblasts and peripheral blood monocytes (10, 7 . Biernacki. E. 1894. W kwestyi wzajemnego stosunku czerwonych cialek osocza we krwi krazacej. (On mutual relation of erythrocytes 14.26.27.32). In a study in which LPS was administered to plasma in peripheralblood.) Gazeta Lek 14:274. to volunteers a n increase in plasma IFN-@,/IL-6levels 8. Fihraeus, R. 1921. The suspension-stabilityof the blood. Acta Med. Scand. 55: 1. was detected a s early as 45 min after LPS administration 9. Bedell, S. E., and B. T. Bush. 1985. Erythrocyte sedimentation rate: (33).In this study, peak IFN-PJIL-6 plasma levels were from folklore to facts. A m . J . Med. 78:1001. observed 2 to 4 h after LPS injection and were in the 10. Ritchie, D. G.. and G. M. Fuller. 1983.Hepatocyte-stimulatingfactor: a monocyte-derived acute-phase regulatory protein. Ann. N . Y . A c a d . range 4to 27 ng/ml(33). Sei. 408:490. The cytokinesTNF and IL- 1, which are also induced by 11. Koj, A., J. Gauldie. E. Regoeczi, D. N. Sauder. and G. D. Sweeney. LPS in monocytes, rapidly enhance IFN-P2/IL-6secretion 1984. The acute-phase response of cultured rat hepatocytes: system characterization and the effect of human cytokines. Biochem. J . by human fibroblasts (20, 21) and endothelial cells (D. 224:505. Stern, U. Santhanam, A. Ray, L. T. May, and P. B. Sehgal, 12. Koj. A., A. Kurdowska, D. Magielska-Zero, H. Rokita, J. D. S i p , J. M. Dayer, S. Demczuk, a n d J. Gauldie. 1987. Limited effects of unpublished data) andmay at least in part be responsible recombinant human and murine interleukin 1 and tumor necrosis for the elevated levels of IFN-P2/IL-6observed in patients factor on productionof acute phase proteinsby cultured rat hepatowith acute bacterial infections. cytes. Biochem. Int. 14:553. 13. Perlmutter. D. H.. F. S . Cole, P. Kolbridge, T. H.Rossing, and H. R. In patients with both meningitis and bacteremia, the Colten. 1985. Expression of the al-proteinase inhibitor gene in huconcentration of IFN-P2/IL-6in the CSF was 10- to 100man monocytes and macrophages. Proc. N a t l . Acad.Sei.USA 82: 795. fold higher than that in serum. This arguesstrongly for J.,C. Richards, D. Harnish. P. Lansdorp, andH. Baumann. considerable local production of IFN-P2/IL-6,perhaps by 14. Gauldie. 1987. lnterferon &/B-cell stimulatory factor type 2 shares identity epithelial cells, tissue macrophages, or the ubiquitous with monocyte derived hepatocyte stimulatory factor and regulates the major acute phase protein response in liver cells. Proc. Natl. stromal fibroblast.Although the elicitation of the hepatic Acad. Sei. USA84:7251. acute phase response appears to be a n important sys- 15. May, L. T..U. Santhanam, S. B. Tatter, J. Ghrayeb. a n d P. B. temic function of IFN-P,/IL-6, this cytokine clearly has Sehgal. 1988. Interferon &./B-cell differentiation factor BSF-2/hepatocyte stimulating factor: a major mediator of the ”acute phase additional local and systemic effects. In experiments on response.” In: TumorNecrosisFactorlCachectin.Lymphotoxins human cells inculture, phosphoglycoproteins derived Proceedings of a n International Conference, and Related Cytokines. Heidelberg, September 14-18, 1987. B. Bonavida and H. Kirchner. from this gene elicit a n anti-viral state (34). enhance eds. S. Karger. Basel, p. 259. major histocompatibility Ag geneexpression (35, 36). 16. May, L. T., J. Ghrayeb, U. Santhanam. S. B. Tatter. Z. Sthoeger. D. inhibit proliferation of fibroblasts and epithelial cells C. Helfgott. N. Chiorazzi, G. Grieninger. and P. B. Sehgal. 1988. Synthesis and secretion of multiple forms of @2-interferon/B-cell (37). enhance proliferation of and Ig secretion by B cells differentiation factor 2/hepatocyte stimulatory factor by human fi(27, 37), enhance proliferation of peripheral blood T cells broblasts andmonocytes. J . Blol. C h e m . 263:7760. and thymocytes (38, 39), and enhance cytotoxic T cell 17. Mortensen, R.F., J. Shapiro, B-F. Lin, S. Douches, and R. Neta. 1988. Interaction of recombinant IL-1 and recombinant tumor ne(40) function. In addition, circulating levels of IFN-@$ILcrasis factor in the induction of mouse acute phase proteins. J. 6 have been reported to correlate closely with the pyroImrnunol. 140:2260. genic responsein thermally injured patients (22); human 18. Evans, E., G. M. Courtois. P. L. Killan. G. M. Fuller, a n d G. R. Crabtree. 1987. Inductionof fibrinogen and a subset of acute phase IFN-&/ILB has been reported to be highly pyrogenic in response genes involves a novel monokine which is mimicked by the rabbit (22) (D. C . Helfgott, Y . Fong, S. Lowry, L. T. phorbol esters. J . Blol. Chern. 26210850. May, and P. E. Sehgal, unpublished data). Thus the pres-19.Goldman, N. D., and T-Y. Liu. 1987.Biosynthesis of human Creactive protein in cultured hepatoma cells is induced by a monocyte ence of IFN-Pz/IL-6 in local body fluids and in serum factor(s) other than interleukin-1. J . Biol. C h e m . 5:2363. during acute infections suggests that this cytokine is 20. Kohase, M., D. Henriksen-DeStefano. L. T.May, J. Vilccek. a n d P. B. Sehgal. 1986. Induction of &-interferon by tumor necrosis factor: likely to participate in a cascade of local and systemic a homeostatic mechanism in the control of cell proliferation. Cell events that help limit tissue damage. Detailed studies of 45:659. the local and systemic IFN-&/IL-6 response to acute in- 21. Kohase, M.. L. T. May, I. Tamm. J. Vilccek. a n d P. B. Sehgal. 1987. A cytokine network in human diploid fibroblasts: interactions of p fections of man, the role of IFN-Pz/IL-6 in experimental interferons, tumor necrosis factor. platelet-derived growth factor, animal infection and the relationshipof this cytokine to and interleukin-1. Mol. Cell. Biol. 7:273. specific clinical effects of infection will further helpde- 22. Nijsten, M. W.,E. R. DeGroot, H. J. TenDuis, H. J. Klesen. C. E. Hack, a n d L. A. Aarden. 1987. Serum levels of interleukin-6 and fine itsplace in the host response to infectious agents. acute phase responses.Lancet 2:921.

IFN-&/IL-6 SPECIES IN HUMAN BODY FLUIDS

29.

30.

31.

33.

34. Zilberstein. A.. R. Ruggieri, J. N. Korn. and M. Revel. 1986. Structure and expression of cDNA and genes for human interferon-betaby growth inhibitory cytokines. EMBO 2, a distinct species inducible J. 5:2529. 35. May, L.T.. D. C. Helfgott. and P.B. Sehgal. 1986. Anti-@-interferon of HLA-B7mRNA in antibodies inhibit the increased expression tumor necrosis factor-treated human fibroblasts: Structural studles 83:8957. of the p2 interferon involved. Proc. Natl. Acad. Sci. USA 36. Revel, M.. A. Zilberstein, R. M. Ruggieri, M. Rubinstein, and L. Chen. 1987. Autocrine interferons and interferon-@,.J. Interferon Res. 7:529. 37. Hirano, T., K. Yasukawa. H. Harada. T. Taga. Y. Watanabe. T. Matsuda, S . Kashiwamura, K. Nakajima, K.Koyama, K. Iwamatsu, S. Tsunasawa, F.Sakiyama. H.Matsui, Y.Takahara, T. Taniguchi. and T. Kishimoto. 1986. ComplementaryDNA for a novel interleukin (BSF-2)that inducesB lymphocytes to produce immunoglobulins. Nature 324:73. 38. Lotz. M.. F. Jirik, P. Kabourides. C. Tsoukas. T. Hirano, T. Kishimoto, and D. A. Carson. 1988. Bcell stimulatory factor Z/interleukin 6 is a costimulant for human thymocytes and T lymphocytes. J. Exp. Med. 167: 1253. 39. Tosato. G.. and S . E. Pike. Interferon-&/interleukin 6 is a costimulant for humanT lymphocytes. J. Irnrnunol. 141: 1556. 40. Wong. G . C.. and S.C. Clark. 1988. Multiple actions of interleukin 6 within a cytokine network.Irnrnunol. T o d a y 9:137.


32.

and immune responses in viral disease: enhanced expression of the "8,-interferonlhepatocyte stimulating factor/interleukin-6 gene in virus-infected human fibroblasts. J. Exp. Med. 167:1951. Santhanam, U.. S . B. Tatter, D. C. Helfgott, A. Ray, J. Ghrayeb, L. of T. May, and P.B. Sehgal. 1988. Structure, genetics and function human "@,-interferon/B-cell stimulatory factor-2/hepatocyte stimulatory factor" (interleukin-6). In: Monokines and Other Non-Lyrnp h o c y t f c C y t o k i n e s . M. C. Powanda. ed.Alan R. Liss. Inc.. NewYork. p. 29. May, L. T., U. Santhanam. S. B. Tatter, N. Bhardwaj, J. Ghrayeb. and P. B. Sehgal. Phosphorylation of secreted forms of human &interferon/hepatocytestimulatingfactor/interleukin-6.Biochern. B i o p h y s . Res. Cornrnun. 152:I 144. Ganapathi, M.K.. L. T. May, D. Schultz, A. Brabenec, J. Weinstein. P. B. Sehgal, and I. Kushner. 1988. Role of interleukln-6 in regulating synthesis of C-reactive protein an serum amyloid A in human hepatoma cell lines. Biochern. Biophys. Res. Cornrnun. 150:271. Aarden, L. A.. E. R. DeGroot. 0.L. Schapp, and P. M. Lansdorp. 1987. Productionof hybridoma growth factorby human monocytes. Eur. J. Irnrnunol. 17: 141 1. Fong. Y., L.L. Moldawer, M. Marano. H. Wei, S. B. Tatter, R. H. Clarick, U. Santhanam, D.Sherris, L. T. May, P. B. Sehgal, and S. F. Lowry. 1989. Endotoxemia elicits increased circulating@,-interferon/interleukin-6 in man.J. Irnrnunol. I n p r e s s .

953