QuantiGlo

Human TNF- Chemiluminescent Immunoassay

Catalog Number QTA00

For the quantitative determination of human tumor necrosisfactor alpha (TNF- ) concentrations in cell culture supernates,serum, and plasma.

This package insert must be read in its entirety before using this product.

FOR RESEARCH USE ONLY.NOT FOR USE IN DIAGNOSTIC PROCEDURES.

TABLE OF CONTENTSContents Page

INTRODUCTION 2PRINCIPLE OF THE ASSAY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3TECHNICAL HINTS 4REAGENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4STORAGE 5OTHER SUPPLIES REQUIRED . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5SAMPLE COLLECTION AND STORAGE 5REAGENT PREPARATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6ASSAY PROCEDURE 7ASSAY PROCEDURE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8CALCULATION OF RESULTS 9TYPICAL DATA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9PRECISION 10RECOVERY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11LINEARITY 11SENSITIVITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12CALIBRATION 12SAMPLE VALUES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12SPECIFICITY 13REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14PLATE LAYOUT 15

MANUFACTURED AND DISTRIBUTED BY:R&D Systems, Inc. TELEPHONE: (800) 343-7475614 McKinley Place N.E. (612) 379-2956Minneapolis, MN 55413 FAX: (612) 379-6580United States of America E-MAIL: info@RnDSystems.com

DISTRIBUTED BY:R&D Systems Europe19 Barton Lane TELEPHONE: (0)1235 529449Abingdon Science Park FAX: (0)1235 533420Abingdon, Oxon OX14 3NB E-MAIL: info@RnDSystems.co.ukUnited Kingdom

R&D Systems GmbH FREEPHONE: (0)800 909 4455Borsigstrasse 7 TELEPHONE: (0)6122 9098065205 Wiesbaden-Nordenstadt FAX: (0)6122 909819Germany E-MAIL: infogmbh@RnDSystems.co.uk

INTRODUCTIONTumor necrosis factor (TNF- ) (1, 2), also known as cachectin, and tumor necrosis factor(TNF- ) (3, 4), also known as lymphotoxin, are two closely related proteins (about 34% aminoacid residue homology) that bind to the same cell surface receptors and produce a vast rangeof similar, but not identical, effects. In contrast to the similarity of their biological activities, theregulation of the expression and processing of the two factors is quite different (5, 6). TNF- isproduced by neutrophils, activated T and B lymphocytes, NK cells, LAK cells, astrocytes,endothelial cells, smooth muscle cells, and some transformed cells (5, 6). TNF- is producedby lymphocytes (5, 6). The properties and activities of the TNFs have been the subject ofnumerous reviews (5 - 11).

Mature human TNF- is a polypeptide of 157 amino acid residues (mouse, rat and rabbitTNF- are one amino acid shorter) (5). The apparent molecular weight of human TNF- underdenaturing conditions is approximately 17 kDa (12). Human TNF- , in contrast to TNF- ,shows no N-glycosylation (mouse TNF- is N-glycosylated) (5). The biologically active nativeforms of both TNF- and TNF- are trimers (13, 14).

TNF- , unlike TNF- , does not possess a typical signal peptide sequence. TNF- is, however,initially synthesized as a larger protein with the mature 17 kDa factor comprising the C-terminalportion of this precursor. The N-terminal sequence of the precursor contains both hydrophilicand hydrophobic domains and its presence results in the occurrence of TNF- as amembrane-bound form from which the mature factor is released by proteolytic cleavage(15 - 17). Evidence suggests that the membrane-anchored form of TNF- on the surface ofmacrophages and/or monocytes, in addition to serving as a reservoir for release of solubleTNF- , has lytic activity and may also have an important role in intercellular communication(15 - 17).

Two distinct receptor types have been identified that specifically bind TNF- and TNF- .Virtually all cell types studied show the presence of one or both of these receptor types. Onetype, TNF RII (Type A, Type , 75 kDa or utr antigen), is a transmembrane glycoprotein withan apparent molecular weight of 75 kDa (18). The other type, TNF RI (Type B, Type , 55 kDaor htr antigen), is a transmembrane glycoprotein with an apparent molecular weight of 55 kDa(19, 20). The two receptor types are distinct immunologically, but show similarities to eachother and to the NGF receptor in the pattern of cysteine residue locations in four domains intheir extracellular portions (5, 18). The intracellular domains of the two TNF receptor types areapparently unrelated, suggesting that the two receptor types employ different signaltransduction pathways (18). Each receptor type can bind TNF- or TNF- with high affinity andthere is no evidence that interaction between the two receptor types is necessary for signaltransduction (20 - 22). Soluble forms of both types of receptors have been found in humanserum and urine (23 - 25). These soluble receptors are capable of neutralizing the biologicalactivities of both TNF- and TNF- and may serve to modulate and localize the activities of theTNFs or may serve as a reservoir for the controlled release of the TNFs.

2

The two TNFs are extremely pleiotropic factors. That they are capable of producing such awide variety of effects is attributable to the ubiquity of their receptors, to their ability to activatemultiple signal transduction pathways, and to their ability to induce or suppress the expressionof a vast number of genes, including those for growth factors and cytokines, transcriptionfactors, receptors, inflammatory mediators and acute phase proteins, etc. (5, 26). TNFs play acritical role in normal host resistance to infections and to the growth of malignant tumors,serving as immunostimulants and as mediators of the inflammatory response. Many of theactions produced by the TNFs are functionally similar to the effects produced by IL-1.

On the other hand, over-production of TNF has been implicated as playing a role in a numberof pathological conditions, including cachexia (progressive wasting) (2, 27), septic shockfollowing infection with Gram-negative bacteria (28), autoimmune disorders (29), andmeningococcal septicemia (30). Two studies have found elevated levels of TNF- in thecerebrospinal fluid (CSF) of multiple sclerosis (MS) patients, particularly those with activerather than stable disease (31, 32). TNF- was also detected histologically in MS lesions (33).

Current bioassays used for the detection of TNF- are usually based on the cytolytic effects ofTNF- on responsive cell lines, such as L929. These bioassays are tedious and are notspecific for human TNF- . The QuantiGlo TNF- Immunoassay is a 6.5 hour solid phasechemiluminescent ELISA designed to measure TNF- in cell culture supernates, serum andplasma. It contains E. coli-derived recombinant human TNF- and antibodies raised againstthe recombinant factor. It has been shown to accurately quantitate recombinant human TNF- .Results obtained using naturally occurring TNF- samples showed linear curves that wereparallel to the standard curves obtained using the QuantiGlo kit standards. These resultsindicate that the QuantiGlo Immunoassay kit can be used to determine relative mass values fornatural TNF- . Since the measurement of TNF- is insensitive to the addition of recombinantforms of either of the two types of soluble receptors, it is probable that this measurementdetects the total amount of TNF- in samples, i.e., the total amount of free TNF- plus theamount of TNF- bound to soluble receptors.

PRINCIPLE OF THE ASSAYThis assay employs the quantitative sandwich enzyme immunoassay technique. A monoclonalantibody specific for TNF- has been pre-coated onto a microplate. Standards and samplesare pipetted into the wells and any TNF- present is bound by the immobilized antibody. Afterwashing away any unbound substances, an enzyme-linked polyclonal antibody specific forTNF- is added to the wells. Following a wash to remove any unbound antibody-enzymereagent, an enhanced luminol/peroxide substrate (34) solution is added to the wells and light isproduced in proportion to the amount of TNF- bound in the initial step. A microplateluminometer is used to measure the intensity of the light emitted.

3

TECHNICAL HINTSFOR RESEARCH USE ONLY. NOT FOR USE IN DIAGNOSTIC PROCEDURES.The kit should not be used beyond the expiration date on the kit label.If samples generate values higher than the highest standard, dilute the samples with theappropriate Calibrator Diluent and repeat the assay.Variation in pipetting technique, washing technique, luminometers, incubation time ortemperature can cause variations in binding.This assay is designed to eliminate interference by soluble receptors, binding proteins,and other factors present in biological samples. Until all factors have been tested in theQuantiGlo Immunoassay, the possibility of interference cannot be excluded.When mixing or reconstituting protein solutions, always avoid foaming.To avoid cross-contamination, change pipette tips between additions of each standardlevel, between sample additions, and between reagent additions. Also, use separatereservoirs for each reagent.To ensure accurate results, proper adhesion of plate sealers during incubation steps isnecessary.Relative light units (RLUs) may differ among luminometers. The QuantiGlo TNF-Immunoassay was optimized using a DYNEX TECHNOLOGIES MLX™ luminometer.Other instruments may require settings to be adjusted.Relative light units may vary within the 20 minute reading window.Due to limitations of many luminometers, it is recommended that the standards beassayed in duplicate from high to low beginning with the high standard in wells A1 and A2.

REAGENTSTNF- Microplate (Part 890385) - 96 well polystyrene microplate (12 strips of 8 wells) coated with amouse monoclonal antibody against TNF- .

TNF- Conjugate (Part 890386) - 21 mL of polyclonal antibody against TNF- conjugated tohorseradish peroxidase, with preservatives.

TNF- Standard (Part 890387) - 2 vials (35 ng/vial) of recombinant human TNF- in a buffered proteinbase with preservatives, lyophilized.

Assay Diluent QD1-27 (Part 895245) - 6 mL of a buffered protein base with preservatives.

Calibrator Diluent RD5P Concentrate (5X) (Part 895151) - 21 mL of a buffered protein base withpreservatives. For cell culture supernate samples.Calibrator Diluent RD6N (Part 895135) - 21 mL of a buffered animal serum with preservatives.For serum/plasma samples.Wash Buffer Concentrate (Part 895222) - 100 mL of a 10-fold concentrated solution of bufferedsurfactant with preservatives.

Substrate A (Part 895224) - 12.5 mL of stabilized enhanced luminol.

Substrate B (Part 895225) - 12.5 mL of stabilized hydrogen peroxide.

Plate Covers - 4 adhesive strips.

MLX is a trademark of DYNEX TECHNOLOGIES

4

STORAGEUnopened Kit Store at 2 - 8° C. Do not use past kit expiration date.

Opened/ReconstitutedReagents

Diluted Wash Buffer

May be stored for up to 1 month at2 - 8° C.*

Calibrator Diluent RD5P (1X)

Calibrator Diluent RD6N

Assay Diluent QD1-27

Conjugate

Unmixed Substrate A

Unmixed Substrate B

Standard

Microplate Wells

Return unused wells to the foil pouchcontaining the desiccant pack, resealalong entire edge of zip-seal. May bestored for up to 1 month at 2 - 8° C.*

*Provided this is within the expiration date of the kit.

OTHER SUPPLIES REQUIREDDYNEX TECHNOLOGIES MLX luminometer set with the following parameters: 1.0 minutelag time; 1 sec/well read time; summation mode; auto gain on; or the equivalent.Pipettes and pipette tips.100 mL and 1 liter graduated cylinders.Deionized or distilled water.Multi-channel pipette, squirt bottle, manifold dispenser, or automated microplate washer.Horizontal orbital microplate shaker (0.12" orbit) capable of maintaining a speed of500 50 rpm.Human TNF- Controls (optional; available from R&D Systems).

SAMPLE COLLECTION AND STORAGECell Culture Supernates - Remove particulates by centrifugation and assay immediately oraliquot and store samples at -20° C. Avoid repeated freeze-thaw cycles.

Serum - Use a serum separator tube (SST) and allow samples to clot for 30 minutes beforecentrifugation for 10 minutes at approximately 1000 x g. Remove serum and assayimmediately or aliquot and store samples at -20° C. Avoid repeated freeze-thaw cycles.

Plasma - Collect plasma using EDTA, heparin, or citrate as an anticoagulant. Centrifuge at1000 x g within 30 minutes of collection. Assay immediately or aliquot and store samples at

-20° C. Avoid repeated freeze-thaw cycles.

Note: Hemolyzed samples are not suitable for measurement of human TNF- with this assay.

5

REAGENT PREPARATIONBring all reagents to room temperature before use.

Wash Buffer - If crystals have formed in the concentrate, warm to room temperature and mixgently until the crystals have completely dissolved. Dilute 100 mL of Wash Buffer Concentrateinto deionized or distilled water to prepare 1000 mL of Wash Buffer.

Substrate Solution - Substrates A and B should be mixed together in equal volumes15 minutes to 4 hours before use. Store in a capped plastic container protected from light.200 L of the resultant mixture is required per well.

Calibrator Diluent RD5P (1X) - Dilute 20 mL of Calibrator Diluent RD5P (5X) into deionized ordistilled water to yield 100 mL of Calibrator Diluent RD5P (1X).

Standard - Reconstitute Standard with 0.5 mL of deionized or distilled water. Thisreconstitution produces a stock solution of 70,000 pg/mL. Allow the standard to sit for aminimum of 15 minutes with gentle agitation prior to making dilutions.

Pipette 900 L of Calibrator Diluent RD5P (1X) (for cell culture supernates) or CalibratorDiluent RD6N (for serum/plasma samples) into the 7000 pg/mL tube. Pipette 800 L of theappropriate Calibrator Diluent into the remaining tubes. Use the stock solution to produce a5-fold dilution series (below). Mix each tube thoroughly and change pipette tips between eachtransfer. The 7000 pg/mL standard serves as the high standard. The appropriate CalibratorDiluent serves as the zero standard (0 pg/mL).

6

ASSAY PROCEDUREBring all reagents and samples to room temperature before use. It is recommendedthat all samples, standards, and controls be assayed in duplicate.

1. Prepare all reagents, working standards, and samples as directed in the previoussections.

2. Remove excess microplate strips from the plate frame, return them to the foil pouchcontaining the desiccant pack, reseal.

3. Add 50 L of Assay Diluent QD1-27 to each well.

4. Add 200 L of Standard, sample, or control per well. Cover with the adhesive stripprovided. Incubate for 4 hours at room temperature on a horizontal orbital microplateshaker (0.12" orbit) set at 500 50 rpm. A plate layout is provided to record standards andsamples assayed.

5. Aspirate each well and wash, repeating the process three times for a total of four washes.Wash by filling each well with Wash Buffer (400 L) using a squirt bottle, multi-channelpipette, manifold dispenser or autowasher. Complete removal of liquid at each step isessential to good performance. After the last wash, remove any remaining Wash Buffer byaspirating or decanting. Invert the plate and blot it against clean paper towels.

Note: Excessive drying of the wells can lead to poor assay performance and imprecision.Subsequent reagents should be added immediately after washing the plate, and the wellsnot allowed to dry completely. Also avoid prolonged exposure of the wells to vacuumaspiration apparatus.

6. Add 200 L of TNF- Conjugate to each well. Cover with a new adhesive strip. Incubatefor 2 hours at room temperature on the shaker.

Note: Prepare Substrate Solution at this time.

7. Repeat the aspiration/wash as in step 5.

8. Add 200 L of Substrate Solution to each well. Incubate for 20 - 40 minutes at roomtemperature on the benchtop (do not shake).

9. Determine the RLU of each well using a luminometer set with the following parameters;1.0 min. lag time; 1 sec/well read time; summation mode; auto gain on.

7

ASSAY PROCEDURE SUMMARY

8

CALCULATION OF RESULTSAverage the duplicate readings for each standard, control, and sample and subtract theaverage zero standard RLU.

Create a standard curve by reducing the data using computer software capable of generating acubic-spline or quadratic curve fit. If samples have been diluted, the concentration read fromthe standard curve must be multiplied by the dilution factor.

TYPICAL DATAThese standard curves were generated using a DYNEX TECHNOLOGIES MLX luminometerand are provided for demonstration only. A standard curve should be generated for each set ofsamples assayed.

9

(pg/mL)

0

2.2

11.2

56

280

1400

7000

(RLU)18.8218.8625.9926.7168.3572.30271.8274.21199120650385056

1633316523

Average

18.84

26.35

70.32

273.0

1202

5047

16428

Corrected

___

7.51

51.49

254.1

1184

5028

16409

(pg/mL)

0

2.2

11.2

56

280

1400

7000

(RLU)14.5415.3422.8823.4858.4259.93224.0226.41023105046364733

1486215803

Average

14.94

23.18

59.17

225.2

1036

4684

15333

Corrected

___

8.24

44.23

210.3

1022

4669

15318

PRECISIONIntra-assay Precision (Precision within an assay)Four samples of known concentration were tested twenty times on one plate to assessintra-assay precision.

Inter-assay Precision (Precision between assays)Four samples of known concentration were tested in forty separate assays to assessinter-assay precision.

Cell Culture Supernate Assay

Intra-Assay Precision

Sample 1 2 3 4

n 20 20 20 20

Mean (pg/mL) 7.34 74.3 639 2390

Standarddeviation

0.26 1.5 36 36

CV (%) 3.5 2.0 5.6 1.5

Inter-Assay Precision

Sample 1 2 3 4

n 40 40 40 40

Mean (pg/mL) 7.8 77.3 728 2581

Standarddeviation

0.70 5.2 44 168

CV (%) 9.0 6.7 6.0 6.5

Serum/Plasma Assay

Intra-Assay Precision

Sample 1 2 3 4

n 20 20 20 20

Mean (pg/mL) 12.4 109 1056 3604

Standarddeviation

0.75 2.2 19 99

CV (%) 6.0 2.0 1.8 2.7

Inter-Assay Precision

Sample 1 2 3 4

n 40 40 40 40

Mean (pg/mL) 12.8 108 1028 3551

Standarddeviation

1.1 8.0 74 188

CV (%) 8.6 7.4 7.2 5.3

10

RECOVERYThe recovery of natural and recombinant TNF- spiked to three different levels in samplesthroughout the range of the assay in various matrices was evaluated.

Sample Type Average % Recovery Range

Cell culture media (n=4) 103 95 - 108%

Serum (n=5) 103 96 - 111%

EDTA plasma (n=5) 106 89 - 118%

Heparin plasma (n=5) 98 88 - 109%

Citrate plasma (n=5) 108 95 - 120%

LINEARITYTo assess the linearity of the assay, samples with high concentrations of TNF- in variousmatrices were diluted with the appropriate Calibrator Diluent to produce samples with valueswithin the dynamic range of the assay.

Cell culturemedia(n=4)

Serum*(n=5)

EDTAplasma*(n=5)

Heparinplasma*(n=5)

Citrateplasma*(n=5)

1:2Average % of Expected 99

96 - 10410598 - 112

10297 - 107

107100 - 113

10495 - 110Range (%)

1:4Average % of Expected 99

88 - 10610291 - 109

9889 - 105

10493 - 111

9890 - 104Range (%)

1:8Average % of Expected 97

86 - 1079987 - 109

9482 - 106

10091 - 109

9487 - 102Range (%)

1:16Average % of Expected 97

87 - 1069989 - 111

9485 - 106

10189 - 112

9487 - 99Range (%)

1:32Average % of Expected 92

89 - 979282 - 108

9576 - 132

10291 - 131

8980 - 96Range (%)

*Serum and plasma samples were incubated for 10 minutes with gentle agitation betweendilutions.

11

SENSITIVITYOne hundred eighteen assays were evaulated and the minimum detectable dose (MDD) ofTNF- ranged from 0.28 - 1.7 pg/mL. The mean MDD was 0.45 pg/mL.

The MDD was determined by adding two standard deviations to the mean RLU of twenty zerostandard replicates and calculating the corresponding concentration.

CALIBRATIONThis immunoassay is calibrated against a highly purified E. coli-expressed recombinant humanTNF- produced at R&D Systems.

The NIBSC/WHO First International Standard 87/650 (recombinant human TNF- expressed inE. coli) was evaluated in this kit. The dose response curve of this First International Standardparallels the QuantiGlo standard curve. To convert sample values obtained with the QuantiGloTNF- kit to equivalent NIBSC 87/650 International Units, use the equation below.

NIBSC (87/650) equivalent value (IU/mL) = 0.03 x QuantiGlo TNF- value (pg/mL).

SAMPLE VALUESSerum/plasma - One hundred serum and plasma samples were evaluated for the presence ofTNF- in this assay.

Sample TypeRange(pg/mL) % Detectable

Mean of Detectable(pg/mL)

Serum ND - 7.46 71 3.58

EDTA Plasma* ND - 7.96 74 4.20

Citrate Plasma* ND - 6.3 55 3.48

Heparin Plasma* ND - 8.79 60 3.78

ND = Non-detectable

*1 serum, 1 EDTA, 5 heparin and 8 citrate plasma samples measured unusuallyhigh and were not included in this range.

Cell culture supernates - Human peripheral blood mononuclear cells (1 x 106 cells/mL) werecultured in RPMI supplemented with 10% fetal calf serum, 50 M -mercaptoethanol,2 mM L-glutamine, 100 U/mL penicillin, and 100 g/mL streptomycin sulfate and stimulatedwith 10 g/mL PHA. Aliquots of the culture supernate were removed on days 1 and 5 andassayed for levels of natural TNF- .

Condition Day 1 (pg/mL) Day 5 (pg/mL)

Unstimulated 197 325

Stimulated 770 2735

12

SPECIFICITYThis assay recognizes both natural and recombinant human TNF- . The factors listed belowwere prepared at 70 ng/mL in both Calibrator Diluent RD5P (1X) and Calibrator Diluent RD6N,and assayed for cross-reactivity. Preparations of the following factors at 70 ng/mL in amid-range rhTNF- control were assayed for interference. No significant cross-reactivity orinterference was observed.

Factors related to or associated with TNF- :

Other factors:

13

rhTNF- rhsTNF RI rhsTNF RII rmTNF-

Recombinanthuman:ANGARCNTF-ECGF

EGFFGF acidicFGF basicFGF-4FGF-5FGF-6G-CSFGM-CSFGROGROGROHB-EGFHGFIFN-

IGF-IIGF-IIIL-1IL-1IL-1raIL-1 sRIIL-1 sRIIIL-2IL-2 sRIL-3IL-3 sRIL-4IL-4 sRIL-5IL-5 sRIL-5 sRIL-6IL-7IL-8IL-9

IL-10IL-11IL-12IL-13KGF (FGF-7)LAP (TGF- 1)LIFM-CSFMCP-1MIP-1MIP-1-NGF

OSMPD-ECGFPDGF-AAPDGF-ABPDGF-BBPTNRANTESSCF

SLPITGF-TGF- 1TGF- 2TGF- 3TGF- sRIIVEGF

Recombinantmouse:GM-CSFIL-1IL-1IL-3IL-4IL-5IL-7IL-9IL-10IL-13

LIFMIP-1MIP-1SCF

Other:bFGF acidicbFGF basichPDGFpPDGFhTGF- 1pTGF- 1raTGF- 5

REFERENCES1. Carswell, E.A. et al. (1975) Proc. Natl. Acad. Sci. USA 72:3666.

2. Beutler, B. et al. (1985) Nature 316:552.

3. Williams, T.W. and G.A. Granger (1968) Nature 219:1076.

4. Ruddle, N.H. and B.H. Waksman (1968) J. Exp. Med. 128:1267.

5. Vilcek, J. and T.H. Lee (1991) J. Biol. Chem. 266:7313.

6. Ruddle, N.H. (1992) Curr. Opinion Immunol. 4:327.

7. (1991) Tumor Necrosis Factor: Structure, Function and Mechanism of Action, Aggarwal, B.B.and J. Vilcek eds., Marcek Dekker, Inc.

8. Sherry, B. and A. Cerami (1988) J. Cell. Biol. 107:1269.

9. Beutler, B. and A. Cerami (1989) Annu. Rev. Immunol. 7:625.

10. Loetscher, H. et al. (1991) Cancer Cells - A Monthly Review 3:221.

11. Aggarwal, B.B. et al. (1984) J. Biol. Chem. 259:686.

12. Davis, J.M. et al. (1987) Biochem. 26:1322.

13. Jones, E.Y. et al. (1989) Nature 338:225.

14. Eck, M.J. and S.R. Sprang (1989) J. Biol. Chem. 264:17595.

15. Kriegler, M. et al. (1988) Cell 53:45.

16. Luettig, B. et al. (1989) J. Immunol. 143:4034.

17. Perez, C. et al. (1990) Cell 63:251.

18. Dembic, Z. et al. (1990) Cytokine 2:231.

19. Schall, T.J. et al. (1990) Cell 61:361.

20. Loetscher, H. et al. (1990) Cell 61:351.

21. Smith, C.A. et al. (1990) Science 248:1019.

22. Engelmann, H. et al. (1990) J. Biol. Chem. 265:14497.

23. Seckinger, P. et al. (1989) J. Biol. Chem. 264:11966.

24. Olsson, I. et al. (1989) Eur. J. Haematol. 42:270.

25. Engelmann, H. et al. (1990) J. Biol. Chem. 265:1531.

26. Kronke, M. et al. (1991) in Tumor Necrosis Factor: Structure, Function and Mechanism of Action,Aggarwal, B.B. and J. Vilcek eds., Marcek Dekker, Inc. p. 189.

27. Oliff, A. (1988) Cell 54:141.

28. Tracey, K.J. et al. (1987) Nature 330:662.

29. Pujol-Borrell, R. et al. (1987) Nature 326:304.

30. Waage, A. et al. (1987) Lancet 1:355.

31. Tsukada, N. et al. (1991) J. Neurol. Sci. 102:230.

32. Sharief, M.K. et al. (1991) New Engl. J. Med. 325:467.

33. Selmaj, K. et al. (1991) J. Clin. Invest. 87:949.

34. European Patent Application 84300725.3.

14

PLATE LAYOUTUse this plate layout to record standards and samples assayed.

10.97 750259.4 4/03

15