Proton nuclear magnetic resonance study of human interleukin 6: Chemical modifications and partial spectral assignments for the aromatic residues

243

Biochimica et Biophysica Acta, 1041 (1990) 243-249

Elsevier BBAPRO 33777

Proton nuclear magnetic resonance study of human interleukin 6" Chemical modifications and partial spectral assignments for the aromatic residues Chiaki Nishimura 1, Hiroyuki Hanzawa 1, Shun-ichi Itoh 2 Kiyoshi Yasukawa Ichio Shimada 1, Tadamitsu Kishimoto 3 and Yoji Arata 1

2,

t Faculty of Pharmaceutical Sciences, The University of Tokyo, Bunkyo-ku, Tokyo, 2 Tosoh Corporation, Ayase-shi, Kanagawa and s Institute for Molecular and Cellular Biology, Osaka University, Suita City, Osaka (Japan)

(Received 1 March 1990)

Key words: NMR, ill-; Deuterium exchange; Iodination; Photo-CIDNP; Interleukin 6; Chemical modification; (Human)

Partial assignments for the IH-NMR resonances of the aromatic residues in human interleukin 6 (IL-6) are reported. The homonuclear Hartmann-Hahn spectrmn clearly shows all connectivities for the histidine, tyrosine and tryptophan residues that exist in IL-6. Using a deuterium exchange method, the imidazole proton resonances of His-16 and His-165 have been assigned, lodination of the tyrosine residues led to the assignment of Tyr-32. Photo-chemically induced dynamic nuclear polarization data have shown that His-16, Tyr-32 and Trp-158 are exposed to solvent, whereas His-165, Tyr-98 and Tyr-101 are buried. Iodination of Tyr-32 gave no significant effect on 11,-6 activity, suggesting that Tyr-32 is not responsible for 1I,-6 activity.

Introduction Interleukin 6 (IL-6) is a multifunctional cytokine which regulates the growth and differentiation of various cells, such as B cells [1,2], T cells [3,4], plasma° cytomas [5], hepatocytes [6], hematopoietic stem cells [7] and nerve cells [8]. IL-6 mediates pleiotropic function in various cells through IL-6-specific receptors [1], and plays a crucial role in the immune response and acute phase reactions [9]. A number of observations on patients with diseases, such as severe bums [10], renal transplant recipients [11], rheumatoid arthritis [12,13] and acquired immuno-deficiency syndrome [14], have shown that the unusual increase in the expression of

Abbreviations: IL-6, intedeukin 6; NMR, nuclear magnetic resonance; photo-CIDNP, photo-chemically induced dynamic nuclear polarization; RP-HPLC, reverse-phase high-performance liquid chromatography; DQF-COSY, double quantum filtered correlated spectroscopy; HOHAHA, homonuclear Hartmann-Hahn spectroscopy; ELISA, enzyme-linked immunosorbent assay; PBS, phosphatebuffered saline; CD, circular dichroism; NOE, nuclear Overhauser effect; SDS-PAGE, sodium dodecyl sulphate polyacrylamide gel dec_ trophoresis. Correspondence: C. Nishimura, Faculty of Pharmaceutical Sciences, The University of Tokyo, 7-3-1, Hongo, Bunkyo-ku, Tokyo 113,

Japan.

IL-6 in the serum is involved in inflammatory responses. Complementary DNAs encoding the human IL-6 and its receptor have been isolated by Hirano et al. [15] and Yamasaki et al. [16], respectively. Although the biological functions of IL-6 in vitro and in vivo have been studied extensively, very little information is available at present concerning the structure-function relationship of IL-6. The aim of the present investigation is to obtain information about the molecular mechanism of recognition of IL-6 by IL-6 receptor. The human IL-6, which consists of 185 amino acids ( M r 21000), has been recently expressed at a high level in Escherichm coli [17]. The recombinant IL-6 was indistinguishable from natural IL-6 as assessed by their biological activities [17]. In this paper, we report the result of partial assignments for the ~H-nuclear magnetic resonance (NMR) signals of the aromatic residues of the recombinant human IL-6, which contains two His, three Tyr, one Trp and seven Phe residues. Chemical modifications involving deuterium exchange and iodination will be used for the assignments of His and Tyr resonances. The result of assignments will be used along with photo-chemically induced dynamic nuclear polarization (photo-CIDNP) data to obtain information about the aromatic residues that exist on the surface of the IL-6 molecule.

0167-4838/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)

244 Materials and M e t h o d s

Materials Recombinant human IL-6 was expressed in Escherichia coli and purified as described by Asagoe et al. [17]. Purity of the final preparation was checked by reverse-phase-high-performance liquid chromatography (RP-HPLC) and SDS-PAGE. For 1H-NMR measurements, the purified IL-6 was dialyzed against 20 mM phosphate, 120 mM N a C I / D 2 0 (pH 7.4).

NMR measurements One-dimensional 1H-NMR spectra were recorded at 400 MHz with a Bruker AM-400 spectrometer with a probe temperature of 30 ° C. A spectral width of 5000 Hz with 8K data points was used for each measurement, and 1000-2000 scans were accumulated. Two-dimensional 1H-NMR spectra were recorded at 400 MHz with a JEOL JNM-GX400 spectrometer with a probe temperature of 30 ° C. Double quantum filtered correlated spectroscopy (DQF-COSY) [18] and homonuclear Hartmann-Hahn spectroscopy ( H O H A H A ) [19] spectra were obtained in the phase-sensitive mode as described by States et al. [20]. The water resonance was suppressed by selective irradiation during the relaxation delay. The 512 blocks were acquired with data points of 2048 and the obtained data matrix was zero-filled once along t 1 direction to the final data matrix of 1K × 1K. The 128-256 scans were accumulated for each t I with a relaxation delay of 1.5 s. H O H A H A spectra were recorded with a mixing time of 35 ms. For both DQFCOSY and H O H A H A spectra, a phase-shifted sine bell function was applied for both t 1 and t 2 dimensions. Photo-CIDNP spectra were recorded at 400 MHz on a Bruker AM-400 spectrometer with a probe temperature of 30 ° C [21,22]. A flavin dye (3-N-carboxylmethyllumiflavin) was added to the sample solutions at a final concentration of 0.2 mM. The sample was irradiated for 0.3 s using an argon laser (NEC-GLC-3300) followed by a pre-acquisition delay of 0.01 s. A spectral width of 5000 Hz with 8K data points was used for each measurement and 32 scans were accumulated. Alternating light and dark free-induction decays were collected, and light-minus-dark difference spectra were calculated.

pH titrations pH titration was performed with 0.1 M DCI or 0.1 M NaOD. All pH values given in this publication are direct p H meter readings of the N M R sample using a glass electrode adjusted with three reference buffer solutions, and not corrected for the deuterium isotope effect. The pH value at each step in the p H titration was checked before and after the N M R spectrum measurement, and the values agreed to within + 0.05 p H units.

Deuterium exchange of the imidazole C2 protons of His residues IL-6 (120 nmol) was concentrated into 0.4 ml of 20

mM phosphate, 120 mM N a C I / D 2 0 (pH 7.4), and then adjusted to p H 8.4 with 0.1 M NaOD. The sample was incubated at 37 ° C for 48 h or 96 h, and the pH value was readjusted to p H 7.4 with 0.1 M DC1 before the N M R measurements.

Iodination of Tyr residues IL-6 (250 nmol) was dissolved in 20 ml of 50 mM phosphate buffer (pH 7.6). Ten mM 12 solution in water containing 100 m M KI was gradually added to the IL-6 solution at 4 ° C until a molar ratio of 12 to IL-6 became a 3-fold excess. The iodinated IL-6 was dialyzed and concentrated into 20 m M phosphate, 120 m M N a C I / D 2 0 (pH 7.4) with an Am_icon Centriprep 10 cartridge for the N M R measurements or lysylendopeptidase-digestion.

Enzyme-linked immunosorbent assay (ELISA) for testing the binding activity of modified IL-6 Ninety-six-well microplates were coated with 100 #l of anti-IL-6 receptor antibody (2 /~g/ml) [23] at 4 ° C overnight in 0.1 M carbonate-hydrogen carbonate buffer (pH 9.6). The wells were blocked with 100 /~l of 1% bovine serum albumin phosphate-buffered saline (PBS) at room temperature for 2 h, washed and 100 /~1 of soluble IL-6 receptor (50 n g / m l ) [24] in PBS were added. After 2 h incubation at room temperature, the wells were washed, 100 /~1 of 125I-labelled IL-6 (200 cpm//~l) containing test samples were incubated at room temperature for 2 h. Following the last wash of the wells, radioactivities of each well were counted by a "r-counter (Aloka, Japan).

RP-HPLC of the fragments digested with lysylendopeptidase The chemically modified IL-6 (10 nmol) sample was digested with 0.05 nmol of lysylendopeptidase (Wako Pure Chemical, Japan) in 0.2 ml of 20 m M Tris-HC1 buffer (pH 8.5) in the presence of 2 M of urea at 3 0 ° C for 5 h. The resultant fragments (0.2 rnl) were separated by H P L C on an ODS-A-312 column (6 × 150 nun) (Yamamura Chemical Laboratories, Japan) with a linear gradient of 5-75% acetonitrile containing 0.1% trifluoroacetic acid (flow rate 1 ml/min), monitoring ultraviolet absorption at 230 nm. The partial amino acid sequence of each fragment was determined using a gas-phase protein sequencer (Applied Biosystems model 477A).

Circular dichroism (CD) measurements CD spectra were recorded on a Jasco 600 spectrophotometer in a 0.1 cm cell at 25°C. The reported spectra were obtained by employing blank samples containing all of the assay components except IL-6. The mean residue molecular weight of 185 was used for the calculation of mean residue ellipticities for IL-6. The content of the secondary structure was calculated by the method of Greenfield et al. [25].

245 Results One-dimensional N M R spectra and p H titration Fig. 1A shows the aromatic region of the ~H-NMR spectrum of IL-6 observed at p H 7.4. IL-6 possesses a total of 13 aromatic amino acid residues, i.e., two His at positions 16 and 165, three Tyr at positions 32, 98 and 101, one Trp at position 158 and seven Phe. pH titration curves of the imidazole C2 and C4 proton resonances of the two His residues are shown in Fig. lB. A significant difference exists between the titration curves for the two His residues, which will be designated as His A and His B. On the basis of the pK~ and H O H A H A data, C2 and C4 proton connectivities for His A and His B have been established as described in Fig. 1A. For the H O H A H A data, see Fig. 2B. The pK~ values of His A and His B are 6.30 and 5.15, respectively.

B @Oo • A

o o

8.5

o o

|

NI*A

o

8,0

ii

o o

I

7.5

O

OOOoo ° °Oo

ace e

,

7.0

~ I~ , ~ / / !

°e o

°Oooo o

e e ~ e~.~ee

8.0 7.8 7.8 ~.4 ?2 7.0 68 e.8 6.4 6.2 6.0

pH Fig. 1. (A) The 400-MHz tH-NMR spectrum of the aromatic region of IL-6 (0.3 mM) measured at 30°C, pH 7.4. (B) pH titration curves of the imidazoleprotons of the two His residues in IL-6. plom

Two-dimensional N M R spectra Fig. 2A shows the aromatic region of the DQF-COSY spectrum of IL-6. Three cross peaks are observed between C2,6 and C3,5 protons of the three Tyr residues. This result was further confirmed by spin-echo and nuclear Overhauser effect (NOE) experiments (data not shown). Fig. 2B shows the aromatic region of the 2DH O H A H A spectrum of IL-6. All connectivities for the His, Tyr and Trp residues can been clearly established. It was also possible to observe cross peaks originating from two of the seven Phe residues. On the basis of the H O H A H A experiment, we were able to identify in the DQF-COSY spectrum additional cross peaks that correspond to one Phe and one Trio residues (Fig. 2A). Deuterium exchange and the assignment of the His resonances The two His resonances were assigned by a deuterium exchange method [26]. Fig. 3 shows the aromatic region of the XH-NMR spectra of IL-6 incubated at pH 8.4, 37 ° C for 0, 48 and 96 h. With an increase in the incubation time, the intensity of the C2 proton resonance of His A decreased significantly, whereas virtually no change was observed in the case of His B. The IL-6 protein, which was incubated in D20, was subsequently digested with lysylendopeptidase, and the resultant fragments were separated by RP-HPLC. The ~H-NMR spectra of two fragments, segments 11-28 and 152-172, which contain His-16 and His-165 residues, respectively, are shown in Fig. 4. In monitoring the intensity of the C2 proton resonances, the C4 proton resonances were used as standard. As Fig. 4A and B show, the decrease in intensity of the C2 proton resonance in segment 11-28 is significantly larger than that

B

A

Phe

I

3

%f

i

o i

J~

•

i

r

u

¢9

'l

i

¢

r

u

,

u

,

,

,

,

,

,

,

,

,

•

7.8 7.6 7.4 7.2 7.0 6.8 6.6 6.4 6.2 6.0 ppm

i

718

i

u

i

i

i

i

i

,

,

i

,

,

~

,

7J6 7.'4 7.2 7.0 6.8 6.6 6.4 6 '~ 6JO' POre

Fig. 2. The aromatic re#on of (A) DQF-COSY and (B) HOHAHA spectra. Spectra of IL-6 (1.2 mM) were measured at 30 °C and pH 7.4.

246

=,=,

C

/

i

i

i

!

i

i

i

i

i

i

i

,

'r'.8 r.6 r.4 r.2 r:o 818 6~6 8:, 6:2 6:0

,

ppm

Fig. 3. The 400-MHz 1H-NMR spectra of the aromatic region of IL-6 modified with the deuterium exchange. The deuterium exchange of the imidazole C2 protons of the His residues of IL-6 (0.3 mM, pH 8.4) was performed by incubation for 0 (A), 48 (B) and 96 (C) h at 37 o C. Spectra were measured at 30 o C and pH 7.4.

residues, were identified by peptide-sequence analyses. See Fig. 6B. A comparison of H P L C profiles clearly indicates that iodination of IL-6 led to a significant decrease in intensity of only one peak (segments 29-42 and 122-129) resulting in the appearance of two new peaks. It was confirmed by peptide sequence analyses that each of the new peaks was derived from segments 29-42, which contains either mono- or di-iodinated Tyr-32. Furthermore, peptide-sequence analyses showed that Tyr-32 in the 29-42 peptide was iodinated by 90% (data not shown). By comparing the extent of the iodination of the Tyr residue with the decrease in intensity of the Tyr resonances, the Tyr A resonance was assigned to Tyr-32. After the iodination reaction, the remaining IL-6 activity was measured by a competitive inhibition of the binding of 12SI-labeled IL-6 to the soluble IL-6 receptor (Fig. 7). Iodinated IL-6 inhibited the binding of 125I-IL-6 to the soluble IL-6 receptor in the same dose-dependent manner when compared to intact IL-6. This result indi-

A

1-

3:

I

4

o

o

in the case of segment 152-172. On the basis of these experimental findings, His A and His B have been unambiguously assigned to His-16 and His-165, respectively.

Iodination and partial assignment of the Tyr resonances In order to assign the three Tyr resonances, iodination was performed for the IL-6 molecule [27]. Fig. 5 shows the aromatic region of the 1 H - N M R spectra of IL-6 before and after the modification. After the modification, the C3,5 proton resonance of Tyr A decreased significantly, whereas that of Tyr B or Tyr C was found almost unchanged. Concomitant with the decrease in intensity of the resonance for Tyr A, new resonances appeared at 7.2 and 7.5 ppm. These resonances are probably derived from the ring proton resonances of mono- and di-iodinated Tyr A, because iodination of the amino acid tyrosine gave the same resonance pattern (data not shown). It was also observed that two minor changes were accompanied with the chemical modification of the Tyr residue. One is a downfield shift from 6.07 to 6.10 p p m of the resonances assigned to the ring protons of a Phe residue (Fig. 5A and B ) a n d another is broadening and downfidd shift of several resonances observed in the region - 0 . 1 - 0 . 5 p p m (Fig. 5C and D). Iodinated and unmodified IL-6 were digested with Iysyl-endopeptidase, and the resultant fragments were separated by R P - H P L C (Fig. 6A and B). All fragments, which are composed of more than five amino acid

I-

B

1-

I

!

(j

(3 ¢o

u3 tO

~=

8.0

7.8

7.6

O~

h

7.4 7.2 ppm

7.0

Fig. 4. The 400-MHz 1H-NMR spectra of two fragments (segments 11-28 (A) and 152-172 (B)) modified with the deuterium exchange of the imidazolc C2 protons of the His residues. After the incubation for

48 h at 37 ° C, IL-6 was diBcsted with lysy|endopeptidase and the resultant fragments were separated by RP-HPLC. Spectra of two fragments (0.15 raM) (segments 11-28 and 152-172) were measured at 30 o C and pH 8.0.

247

"ryr C Tyr B

B

I

Tyr A

3,5

D

Phe

J

C

.

80

78

76

7.4

7

.

.

7.0

.

.

.

618

.

6~6

.

.

6.4

.

6.2

41o a:o 21o to o:o

6.0

ppm

ppm

Fig. 5. The 400-MHz 1H-NMR spectra of the aromatic region of IL-6 unmodified (A) and modified (B) with/odination. The 400 MHz ]H-NMR spectra of the alJphatic region of unmodified (C) and modified (D) IL-6. Iodination of IL-6 was performed by the addition of 3-fold molar ratio of 12. Spectra of IL-6 (0.3 mM) were measured at 30 °C and pH 7.4, Three Tyr residues were designated at Tyr A, Tyr B and Tyr C according to the connectivities for the Tyr residues as described in Fig. 2.

A

100(~ m

~

.-

E

-

o.

0 ,"

&

,

500

~ l

0

I

I

10

i

I

20

I

i

30

I

I

40

I

I

I

50

Time (min) Fig. 6. RP-HPLC profiles of the fragments modified (A) and unmodified (B) with iodination. After the iodination, IL-6 was digested with lysylendopeptidasc, and the resultant fragments (10 nmol) were separated by RP-HPLC, monitoring ultraviolet absorption at 230 nm. The figures on each peak indicate the residue number corresponding to each fragment.

0

98

39

156

625

2500

ng/ml Fig. 7. The competitive binding activity of iodinated (@) and unmodified (o) IL-6 to the soluble IL-6 receptor. The combination of ELISA and competitive binding assay was performed as described under Materials and Methods. Radioactivities of 125I-IL-6 binding to soluble IL-6 receptor were counted by a T-counter.

248 10

cates that iodinated IL-6 retains its activity of binding to the IL-6 receptor.

!

[

t

"

* 220

, 230

I

Photo-CIDNP

A photo-CIDNP difference spectrum of IL-6 is shown in Fig. 8. On the basis of the spectral assignments described in Figs. 3 and 4, two positive resonances observed at 7.77 and 6.98 ppm were assigned to the C2 and C4 protons of His-16, respectively. Furthermore, on the basis of HOHAHA experiment (Fig. 2B), four positive resonances observed at 7.60, 7.48, 7.21 and 7.13 ppm were found to originate from the indole C4, C7, C6 and C5 protons of Trp-158. The results of NOE experiments confirm that the resonance at 7.34 ppm is due to the indole C2 proton of Trp-158 (data not shown). The C7 and C5 protons were strongly cross-polarized, probably due to the long laser-irradiation time (0.3 s) or short pre-acquisition delay (0.01 s) [21,28]. On the basis of the spectral assignments performed in Figs. 5 and 6, the negative resonance at 6.80 ppm was assigned to the C3,5 protons of Tyr-32. These photo-CIDNP data show that His-16, Trp-158 and Tyr-32 are exposed to solvent, whereas His-165, Tyr-98 and Tyr-101 are buried.

significant difference in the CD spectra was observed between iodinated and unmodified IL-6.

CD spectra

Discussion

The CD spectra of iodinated and unmodified IL-6 were measured at pH 7.4 (Fig. 9). In both spectra, large negative ellipticities were observed at 208 and 222 nm, indicating the existence of a-helix. The helical content of IL-6 at pH 7.4 was estimated at about 35%. No

-r

Trp 158

'JLI

I

;

tO

81o' 7'.8' 71(5' 71,' ri2' r:o' 6'.8' e:6 's:4' 6~ "61o POre

Fig. 8. The aromatic region of the photo-CIDNP difference Sl~ctrum of IL-6. A flavin dye (0.2 m M ) was added to IL.6 and the sample tube was irradiated by a 488 n m line (1 V0 from an am,on ion laser. The p h o t o - C I D N P spectrum of IL-6 (0.2 raM) was measured at 30 o C and p H 7.2.

0

-10

-2( 200

, * 210

Wavelength

, 240

(nrn)

Fig. 9. C D spectra of iodinated ( - - - - - - ) and unmodified ( IL-6 (10/tM) measured at 25 o C and p H 7.4.

With proteins of the size of IL-6 ( M r 21000), it is generally difficult to assign all of the resonances to specific residues in the amino acid sequence due to the overlapping of resonances. Therefore, it becomes important to seek and assign amino acid residues that are closely related to the physiological functions of proreins. In this publication, we performed partial assignments of the aromatic resonances, and have shown that His-16, Tyr-32 and Trp-158 are exposed to solvent, whereas His-165, Tyr-98 and Tyr-101 are buried. The assigned resonances of these exposed amino acid residues may be used as probes for the analysis of the mode of the interactions between IL-6 and the IL-6 receptor. Very little information has so far been obtained concerning the structure-function relationship of IL-6. Brakenhoff et al. [29] reported the analyses of IL-6 analogs, in which various lengths of segments were deleted from the amino-terminal, and concluded that (1) the first 28 amino acid residues can be removed without significantly affecting the biologic activity of IL-6 and (2) the existence of segment Gln-29-Leu-34 is closely associated with the expression of the IL-6 functions. Ida et al. [30] prepared a monoclonal antibody that strongly neutralizes the IL-6 activity, and showed that the antibody recognizes the epitope comprizing segment Thr-150-Phe-174. It was concluded that segment Ala-154-Thr-163 is crucial for the recognition by the antibody. However, it is possible that the amino acid deletion or the antibody binding, which occurs at

249 sites remote from the receptor binding site, induces a significant degree of the conformation change of the receptor binding site. Therefore, it is still not clear which segment of IL-6 is actually involved in the receptor binding. The results of the present NMR studies suggest that Tyr-32 or Trp-158 can be used as probes to clarify whether segment Gln-29-Leu-34 or Ala-154Thr-163 binds directly to the IL-6 receptor. The NMR spectra of iodinated IL-6 and the RPHPLC profiles of digested fragments indicate that the iodination reaction of IL-6 occurred specifically to Tyr32 with no significant side reactions to Trp, His and Met residues. As Fig. 5 shows, the iodination of Tyr 32 results in the downfield shift of the ring proton resonances originating from one of the Phe residues. It is also to be noted that broadening and downfield shift are observed in several resonances in the methyl proton region. Since these resonances are most likely due to protons adjacent to aromatic residues in the folded structure, it is possible that iodination affects the interactions of Tyr-32 with the surrounding aromatic residues, resulting in a change in the upfield shift induced by the ring current. The CD spectra showed that upon iodination no significant change is induced in the secondary structure of IL-6. As described above, iodination of Tyr-32 gave no significant effect on IL-6 activity. These results suggest that Tyr-32 and its surroundings are not responsible for the IL-6 activity. However, from these results we cannot exclude the possibility that segment Gln-29-Leu-34 is directly involved in the receptor binding.

Acknowledgements This study was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science and Culture of Japan and a Grant-in-Aid from the Tokyo Biochemical Research Foundation.

References 1 Kishimoto, T. and Hirano, T. (1988) Annu. Rev. Immunol. 6, 485-512. 2 Hirano, T., Taga, T., Nakano, N., Yasukawa, K., Kashiwamura, S., Shimizu, K., Nakajima, K., Pyun, K.H. and Kishimoto, T. (1985) Proc. Natl. Acad. Sci. USA 82, 5490-5494. 3 Garman, R.D., Jacobs, K.A., Clark, S.C. and Raulet, D.H. (1987) Proc. Natl. Acad. SCi. USA 84, 7629-7633. 4 Lotz, M., Jirik, F., Kabouridis, R., Tsoukas, C., Hirano, T., Kishimoto, T. and Carson, D.A. (1988) J. Exp. Med. 167, 12531258.

5 Nordan, R.P. and Potter, M. (1986) Science 233, 566-569. 6 Nakamura, T., Arakaki, R. and Ichihara, A. (1987) Exp. Cell. Res. 179, 488-497. 7 Ikebuchi, K., Wong, G.C., Clark, S.C., Ihle, J.N., Hirai, Y. and Ogawa, M. (1987) Proc. Natl. Acad. Sci. USA 84, 9035-9039. 8 Satoh, T., Nakamura, S., Taga, T., Matsuda, T., Hirano, T., Kishimoto, T. and Kaziro, Y. (1988) Moi. Cell. Biol. 8, 3546-3549. 9 Gauldie, J., Richards, C., Harnish, D., Landsdorp, P. and Baumann, H. (1987) Proc. Natl. Acad. Sci. USA 84, 7251-7255. 10 Nijsten, M.W.N., De Groot, E.R., Ten Duis, H.J., Klasen, H.J., Hack, C.E. and Aarden, L.A. (1987) Lancet ii, 921. 11 Van Oers, M.H.J., Van der Heyden, A.A.P.A.M. and Aarden, L.A. (1988) Clin. Exp. Immunol. 71, 314-319. 12 Hirano, T., Matsuda, T., Turner, M., Miyasaka, N., Buchan, G., Tang, B., Sato, K., Shirnizu, M., Maini, R., Feldman, M. and Kishimoto, T. (1988) Eur. J. Immunol. 18, 1797-1801. 13 Houssiau, F.A., Devogelaer, J.P., Van Damme, J., Nagant de Deuxchaisnes, C. and Van Snick, J. (1988) Arth. Rheum. 31, 784-788. 14 Nakajima, K., Martines-Maza, O., Hirano, T., Brecn, E.C., Nishanian, P.G., Salazar-Gonzalez, J.F., Fahey, J.L. and Kishimoto, T. (1989) J. Immunol. 142, 531-536. 15 Hirano, T., Yasukawa, K., Harada, H., Taga, T., Watanabe, Y., Matsuda, T., Kashiwamura, S., Nakajima, K., Koyama, K., Iwamatsu, A., Tsunasawa, S., Sakiyama, F., Matsui, H., Takahara, Y., Taniguchi, T. and Kishimoto, T. (1986) Nature 324, 73-76. 16 Yamasaki, K., Taga, T., Hirata, Y., Yawata, H., Kawanishi, Y., Seed, B., Taniguchi, T., Hirano, T. and Kishimoto, T. (1988) Science 241, 825-828. 17 Asagoe, Y., Yasukawa, K., Saito, T., Maruo, N., Miyata, K., Kono, T., Miyake, T., Kato, T., Kakidani, H. and Mitani, M. (1988) Biotechnology 6, 806-809. 18 Rance, M., Sorensen, O.W., Bodenhausen, G., Wagner, G., Ernst, R.R. and Wuethrich, K. (1983) Biochem. Biophys. Res. Commun. 117, 479-485. 19 Bax, A. and Davis, D.G. (1985) J. Magn. Reson. 65, 355-360. 20 States, D.J., Haberkorn, R.A. and Ruben, D.J. (1982) J. Magn. Reson. 48, 286-292. 21 Kaptein, R. (1982) Biol. Magn. Reson. 4, 145-191. 22 Shirakawa, M., Lee, S.J., Akutsu, H., Kyogoku, Y., Kitano, I~., Shin, M., Ohtsuka, E. and Ikehara, M. (1985) FEBS Lett. 181, 286-290. 23 Hirata, Y., Taga, T., Hibi, M., Nakano, N., Hirano, T. and Kishimoto, T. (1989) J. Immunol. 143, 2900-2906. 24 Taga, T., Hibi, M., Hirata, Y., Yamasaki, K., Yasukawa, K., Matsuda, T., Hirano, T. and Kishimoto, T, (1989) Cell 58, 573-581. 25 Greenfield, N. and Fasman, G.D. (1969) Biochemistry 8, 41084116. 26 Arata, Y., Shimizu, A. and Matsuo, H. (1978) J. Am. Chem. Soc. 100, 3230-3232. 27 Perlman, R.L. and Edelhoch, H. (1967) J. Biol. Chem. 242, 24162422. 28 De Marco, A., Petros, A.M., Llinas, M., Kaptein, R. and Boelens, R. (1989) Biochim. Biophys. Acta 994, 121-137. 29 Brakenhoff, J.P.J., Hart, M. and Aarden, L.A. (1989) J. Immunol. 143, 1175-1182. 30 Ida, N., Sakurai, S., Hosaka, T., Hosoi, K., Kunitomo, T., Shimazu, T., Maruyama, T., Matsuura, Y. and Kohase, M. (1989) Biochem. Biophys. Res. Commun. 165, 728-73,1.