CA2176114A1 - Cytokines from natural killer or t cells - Google Patents

Abstract

Cell-free preparations of a cytokine produced by natural killer (NK) cells activated with interleukin 2 (IL-2) or produced by CD4+ T cells activated with anti-CD3 monoclonal antibody are described. This cytokine induces antibody secretion by activated B cells. Methods for producing the cell-free supernatants containing the cytokine are also described.

Description

2 1 ~ 4 WO 95113294 -l- Pcr/uS94/12835 CYTOKINES FROM NATURAL KILLER OR T CELLS
2 I . Guv~ lh~ ' ~r~ ` l 3 This invention described herein may be 4 r-nllfAr~ red, 1 icr~n~ and used by or for the United 5 States G~Y-::L L without the payment of any royalties 6 to us thereon. The Federal G~ L L has a 7 nonexclusive nontransferable, irrevocable, paid-up 8 license to practice or have practiced for or on behalf g of the United States any subject invention LhLUU~ ,uL
l0 the world.

11 II . I~ECE~NICAT FJFT n OF TEIE INVENTIoN
12 The invention is in the f ield of cytokines . More 13 specif ically it is in the f ield of cytokines which 14 potently induce antibody secretion by activated animal 15 B cells.

16 III. BACRGROUND OF TU~ I~vENTIoN
17 Cytokines are prot~in~r~eo~lc substances or protein 18 factors ~rhose activity is derived from T cells or l9 Natural Killer tNK) cells. They serve as the 20 ~ ~r~lr~qir~l center piece cr source of origin ~or a Zl variety Or biological activities bec~use of their 22 ~oility to potently induce antibody secretion by 23 activated an~al B cells. As a natural c~n~q~r~nr~e of WO 95113294 2 ~ ~ ~ I l 4 PCT/US94/12835 this ability, derivative products can be made from 2 cytokines which further expand the scope o~ their 3 enormous utility in the medical and research community.
4 The s~ecif icity of the cytokines has raised great hopes for their use and products derived therefrom in 6 the prevention and therapy of abnormal health 7 conditions. They can be used individually or in 8 combination with other cytokines to stimulate antibody g production in order to modulate complex responses such as immune responses. Thus, there is a continuing need 11 and interest in the medical arts for new members of the 12 cytokine arsenal useful in the battle to prevent, treat 13 and/or ame:iorate the deleterious effects of infection 14 and disease state5 which affect the world community.
IV. SIIM~ARY OF T}IF I~V~NTION
16 It is ~ an object of this invention to provide a 17 novel cytokine. In accordance with this invention, and 18 as broadly described herein, a second object of this 19 inYention _s to provide a cytokine which can potently induce ant_body secretion by activated animal B cells.
21 These and other ' ' i- ts of the invention are 22 fulfilled by the following ` 'i- ~s, as illustrated 23 in a general discussion of cytokines derived from T
24 cells and NR cells. In one e~ho~i L a cell-free preparation of a novel active substance whose activity WO95/13294 2 1 7~ 1 1 4 PCr~Ss4/12835 .

is derived from T cell5 or NK cells is provided which 2 potently induces antibody secretion by activated animal 3 B cells; resist5 neutralization by monoclonal 4 antibodies against cytokine5 produced by NR cells 5 activated by IL-2 or neutralization by monoclonal 6 antibodies or receptor antagonists against cytokines 7 yr ~ ced by activated T cells; has Ig secretion 8 activity which is not mimicked by the addition of known g T cell cytokines or NK cell cytokines when added to 10 separate cell-free preparations in the absence of 11 applicants' novel cell-free preparations of cytokines.
12 Additionally, relative to the active substance whose 13 activity is derived from T cells, it has been 14 discovered that it is a protein; can be produced by 15 activation of the T cell through the T cell receptor 16 signalling pathway which induces cytokine production by 17 T cells; ha activity which is inhibited by treatment 18, with a proteolytic enzyme; has activity which is 19 optimal within 3 to 5 hours after activation of the T
20 cell; has detectable activity at dilution levels as low 21 as 1:3000 and operates as a late acting factor required 22 to induce proliferated B cells to mature into antibody 23 secreting cells. It appears that the similarity 24 between the cytokines discovered by applicants as being 25 derived from NK cells or T cells range from 26 substantially the same to identical.

WO95/13294 2 1 7 6 1 ~ 4 PCT/IIS94/12835 In another ~ho~ O~ the invention a method of 2 producing` a cytokine preparation is provided, 3 ccmprising: coating the bottom of a flask with anti-CD3 4 antibody;~ adding T cells into the flask and allow the T
5 cells to remain in the fla5k for a period of 24 hours;
6 harvesting the cell-free ~ L.,~t.~nt; inducing the 7 activation and proliferation of B cells in the cell-8 free supernatant through the B cell antigen receptor 9 mediated signalling pathway utilizing dextran lC conjugated anti-IgD antibodies in the presence of 11 Interleuki,n-l and Interleukin-2 wherein the B cells are 12 at least 98.5 percent purified through ~luorescence-13 activated cell sorting (FACS); adding activated T cell 14 supernatant; harvesting the B cell culture supernatant 7 5 6 days after culture; and measuring the concentration 16 of secreted IgN in the B cell cultured supernatant by 17 an enzyme-linked ~ ~ccrbent Assay (ELISA~.
18 These and other embodiments will be apparent to 19 those skilled in the art from the detailed disclosure 20 which f ollows .
21 Cytoklnes prepared in accordance with this 22 invention are derived from 1~ An cells, including 23 humans, and can be used generally to stimulate the 24 immune system in i ~ d hosts. Their 25 utility extends to such areas as adjuvants for 26 boostering -1 iAn immune le,~o.. ~es to vaccination, _5_ organ transplantation, chemotherapy, radiotherzpy, 2 genetic; -~-firic~n~ c, and AIDS (autoimmune 3 def iciency ~yndrome~ or ARC (AIDS related complex.
4 Additionally, an ant~gonist or monoclonal antibody to 5 neutralize the cytokine can be made or uniquely 6 identifying and isolating its receptor on the surface of the B cell would be useful in the modulation of 8 antibody production- Moreover, the identification of 9 the receptor on the surface of the B cell can be l0 utilized to create an antagonist (soluble receptor to ll prevent binding of the cytokine to the receptor.
12 Further, cytokines may be used alone or in combination 13 with other cytokines (i.e., as rh~ tical cocktail) 14 to stimulate antibody production.

15 V. pRTT F DESCRIPTION OF THE FIGTTR~
16 Figure l depicts total, RBC-lysed, spleen cells that 17 were stained with FITC-labelled anti-CD3 mAb (2Cll) and 18 ~1~y~o~:Ly~h~in-labelled polyclonal goat anti-mouse IgX
l9 (left upper panel [unsorted] ) . B cells are IgM+CD3~, T
20 cells are IgM'CD3+, and non-B, non-T cells are IgMCD3'.
21 Re-analyses of sorted cells are displayed in right 22 upper panel (B cells), left lower panel (T cells), and 23 right lower panel (non-B, non-T cells).

W095/13294 ~ 2 1 7 ~ 1 1 4 PCr/US94/12835 Figure 2 shows B' and B~ cells that were cultured at 2 1.2~ x 105; cells/ml with ~-dex ~3 ng/ml) and/or IL-2 3 (150 U/ml)~ in the absence or presence of 10% non-B, 4 non-T cells (IgMCD3 ) . siX day5 arter initiation of culture, SN was harvested for determination of secreted 6 IgM ~:Lce.lLLàtionS by ELISA.
7 Figure 3 lows B' cells that were treated for 30 min at 8 4C with a ;polyclonal r2bbit anti-AsGm-1 antisera (150 9 mg/ml) followed by a 30 min incubation at 37C in a complement-containin9 501ution in order to lyse AsGm-l~
11 cells. B' cells with or without anti-AsGm-1 antibody 12 treatment were cultured at 2 x 105 cells/ml in the 13 E,Lesen.~ or absence of ~-dex (3 ng/ml), and/or IL-1 14 (100 U/ml) plus IL-2 (150 U/ml). Six days after initiation of culture, SN was harvested for 16 determinat1on o~ secreted IgM ~llc~llLLations by ELISA.
17 Figure 4 depict5 a pure population of activated NR
18 cells were~established by IL-2 5timulation, in vitro, 19 of spleen cells derived from scid mice (see 20 methodologies). 7 days after initiation of scid spleen 21 cell cultures, cells were harvested for flow cytometriC
22 analysis. The following antibodies were utilized: (1) 23 polyclonal rabbit anti-AsGm-1 antisera + FITC-l~h~lled 24 polyclonal goat anti-rab~it IqG, (2) FITC-l~hr~llr~l PCI~US94/1~8:15 WO 95/13294 ~ l 7 6 1 ~ 4 anti-CD3 mAb t2cll~, (3) FITC-labelled anti-Thy-1 mAb 2 (30H121, (4) FITC-7~hallPd anti-MAC-l wAb (Ml/70), ~5) 3 FITC-labelled anti-Iad mAb (MKD6), and (6) 4 FITC-~h~ A antL-Ia~ mAb (Y3-P; control mAb-i.e. scid S mutation est:~hliph~l on BALB/c background [Iad]). Each 6 flow ~,y~ L ic tracing ~eyL-_~1L': the ~luorescence 7 pattern of 10, 000 viable cells.
8 Figure 5 shows B'P cellg that were cultured at l. 25 x 9 105 cells/ml with Q~-dex (3 ng/ml) and/or IL-l (100 U/ml) plus IL-2 (150 U/ml) in the absence or presence ll of 10% sort-purified splenic non-B, non-T cells 12 (IgMCD3 ), 10% sort-purified splenic T cells (IgMCD3~), 13 or 3% in vitro-generated pure NK cells. six days after 14 initiation of culture, SN wa5 harve5ted for determination of secreted IgM con.,éllL.aLions by ELISA.
16 Figure 6 shows B'P cell5 that were cultured at 1. 5 x 105 17 cells/ml with ~-dex (3 ng/ml), IL-1 (150 U/ml), plus 18 IL-2 (150 U/ml) in the presence of increasing numbers 19 ol~ fn vfL-u ~clleLated pure NK cells (0-80% of B'P cell numbers). 7 days after initiation of culture, SN was 21 harvested ~or A~t~nmin~tion of secreted IgM
22 cvl.- c~.L.~.Lions by ELISA.

PC~/US94/12~3S
WO9~113294 217~ 4 Figure 7 shows B'P cells that were cultured at 1. 5 x 105 2 Gells/ml with a~-dex ~3 ng/ml), IL-l (100 U/ml~ plus 3 IL-2 (150 U/ml) in the absence or presence of varying 4 .c".c~ ,a~ions of NK-3MF (10-50~L v/v). Six days after initiation ! f culture, SN was harvested for 6 determination o~ secreted IgM C.,~C~r~ations by ELISA.
7 Figure 8 shows B'P cells that were cultured at 1. 25 x 8 105 cells/ml with ~-dex (3 nq/ml), IL-l (150 U/ml) 9 plus IL-2 (I 50 U/ml) in the absence or presence of SN
obtained 24 h after activation, with plate-bound ll anti-CD3 mAb (2Cll), of the following CD4t Thl clones:
12 (l) RAl (1;0% SN v/v), (2) RA5 ~l.o~ SN v/v), ~3) RA8 13 (1. 0% SN v/v), (4) RC5 (l . 05~ SN v/v), and (5) RC9 (2 . 5%
14 SN v/v). SN was harvested for determination of secreted IgM co,~. ~"LL;ations by ELISA.

16 Figure 9 shows B'P cells that were cultured at l . 25 x 17 105 cells/ml with ~-dex (3 ng/ml), IL-l (150 U/ml), 18 plu5 IL-2 (150 U/ml) in the absence or presence of 19 varying, final .on.~"-Lations of SN ~0.01-1.0% v/v) 20 from anti-cD3-activated Thl clones (see Fig 8). SN was 21 harvested for determination of secreted IgN
22 uncer,~L ations by ELISA.

pCl~lUS94ll2835 WO 95113294 2 1 /' 6 1 ~ ~
g Figure 10 shows a 24 h SN from anti-CD3-activated RA5 2 Thl clone (T-BI~F) wa5 incubated at 37C in the pL~sence 3 tRA5-SN-PK) or absence (RA5-SN-Ctrl) of proteinase K.
4 3'P cells were cultured at 1. 25 x 10~ cells/ml with ~ dex (3 ng/ml), IL-1 (150 U/ml) plus IL-2 (150 U/ml) 6 in the absence or presence 0~ either RA5-SN-PK ( 1. 0%
7 v/v) or RA5-SN-Ctrl (1.0% v/v). As an additional 8 control, RA5-SN-P~ (1.0~ v/v) and RA5-SN-Ctrl (1.0 g v/v) were added to B'P cell cultures activated with ~-dex (3 ng/ml) plus IL-5 (150 U/ml). SN was harvested 11 for determination of secreted IgM concentrations by 12 ELISA.
13 Figure 11 shows a 24 h SN from the anti-cD3-activated 14 RA5 Thl clone (T-BMF) or SN from unactivated resting RA5 cells (Control SN) was tested in an IL-6-speci~ic 16 sandwich ELISA. Recombinant IL-6 was used to generate a 17 standard curve.
18 Figure 12 shows B'P cells that were cultured at 1. 25 x 19 10~ cells/ml with ~-dex (3 ng/ml), IL-l (~50 U/ml) plus IL-2 (150 U/ml) in the absence or presence of 21 varying . ~c~.-LLc~tions of anti-CD3-activated RA5 SN
22 (O.OOl-l.OS v/v) obtained at various time points after 23 anti-CD3 activation (0-24h). SN was harvested for 24 determination of secreted IgM c.,...e"~ .,tions by ELISA.

W095/13294 ~ 217~tt4 Figure 13~ shows B'P cells that were cultured at 1.25 x 2 105 cells~ml with ~r~-dex (3 ng/ml), IL-l (150 U/ml) 3 plus IL-2! (150 U/ml). T-BMF (1% v/v of RA5 SN) was 4 additionally added to replicate cultures on di~ferent days (dayi 0-6) . SN was harvested from all experimental 6 groups 6 days after initiation of culture for 7 determination of secreted IgM ,_vnc~l~LLations by ELISA.
' 8 VI. DEl'~TT.T~'n DF~CRIPTION OF TE~T~`. INV~NTION
9 A biological activity has been de~ined whose 10 dominate function is to potently induce antibody ll secretion by activated animal B cells. More precisely, 12 applicants have discovered:
13 1. ~ cell-free preparation of a novel active 14 substance whose activity is derived from T cells or NK
15 cells which potently induce antibody secretion by 16 activated animal B cells, said su~stance having Ig 17 secretion activity which is not mimicked by the 18 i~ddition of cytokine in the absence of said cell-~ree l9 preparation.
2. The cell-free preparation o~ Claim 1 21 wherein the~ active substance is ~ uduced by NK cells.
22 3. The cell-free preparation of Claim 1 23 wherein the ~active substance is resistant to WO95113294 2 ~ 76 ~ ~ PCT/US94/12835 neutrali2ation by ~ o~~lonAl antibodies against 2 cytokines ~L~duced by NK cells activated by IL-2 and 3 has Ig secretion activity which is not m;m~ d by the 4 addition of NK cell cytokine5 in the absence of said 5 cell-free ~ La~iOn.

6 4. The cell-free preparation of Claim l 7 wherein the active substance is a protein who~:e 8 activity is derived from T cells which potently induce g antibody secretion by activated animal B cells, said substance:
ll a. is resistant to neutralization by 12 -~no~ nAl antibodies or receptor antagonists against 13 cytokines produced by activated T cells;
14 b. has Ig secretion activity which is not mimi~ d by the addition of T cell cytokines in the 16 absence of said cell-free preparation;

17 c. can be produced by activation of the 18 T cell through the T cell receptor ~iqnAllin~ pathway l9 which induces cytokine production by T cells;
d. has activity which is inhibited by 21 LLe~ with a proteolytic enzyme;
22 e. has activity which is optimal within ~3 3 to 5 halr: ~ft~r ~ativat~o~ or th- ~ o~11;

WO95113294 2 l 75 ~ ~ 4 PCTrl3S94/1283S
f. has detectable activity at dilution 2 levels as low as 1:3000; and 3 ~ g. operates as a late acting factor 4 required to induce proliferated B cells to mature into antibody~ secreting cells.
6 ~ 5. The cell-free preparation of Claim 4 wherein the active proteinaceous substance is a 8 cytokine.
..
9 ~ 6. The cell-free preparation of Claim 5 which is Iproduced by T cells.

11 7. The cell-free preparation of Claim 1 1~ which is 'ree of mycoplasm.

Crl~JS94/12835 WO 95113294 2 1 7 ~ ~ 1 4 .

8. The cell-free preparatlon of Claim 6 2 wherein the prOtQi n~(~eo-lC substance can increase the 3 i n-lUl-tiorl of antibody secretion 10 to 40 fold.
4 9. The cell-free preparation of Claim 3 wherein the active substance can increase the induction 6 of antibody secretion lO to 30 fold.
7 10. The cell-~ree preparation of Claim 1 8 wherein the active substance whose Ig secretion 9 activity is not mimicked by the addition of T cell or NR cell cytokines comprising Interleukin-3;
11 Interleukin-4; Interleukin-6; Interleukin-lo; TNF-~;
lZ IFN-~; IFN-~y and TNF-c; IFN-2; IFN-~ and IFN-~y; IFN-~
13 and TNF-cl; IFN-~, IFN-y and TNF-~; GM-CSF; and TGF-~l.
14 11. The cell-free preparation of Claim 8 wherein the pro~i nAce~-C substance is resistant to 16 neutralizAtion by monoclonal antibodies or receptor 17 antagonists against cytokines comprising anti-IFN-~;
18 TNFRIg; anti-GM-CSF; anti-IL-3; anti-IL-4; anti-IL-5;
l9 anti-IL-6; anti-IL-10; and huCD40Fc.

12. The cell-free preparation of Claim 8 21 wherein the prot~ cemlC substance is produced by 22 activation of the T cell through the T cell receptor PCr/US94/12835 WO 95l13294 2t7~t t4 ~

pathway which induces cytokine production by T cells 2S .
2 shown in` Figure 12.
3 1 13. The cell-free preparation of Claim 9 4 wherein the active substance i5 resistant to neutralization by monoclonal antibodies against 6 cytokines comprising anti-TNF-~r; Anti-IFN-y; anti 7 IFNtr/~r; anti-IL-4; anti-IL-S; anti-IL-6; and anti-IL-8 10.
9 ~ 14. A method of producing a cell-free preparatisn containing the active pro~in~ces--c ll 6ubstance~0r Claim 4, comprising:
12 1 a . coating the bottom of a f lask with 13 anti-CD3 antibody;

14 ; b . adding T cells into the f lask and allow said T cells to remain in the flask for a period 16 of 24 hours;
17 c. harvesting the cell-free 18 supernatant;
l9 I d. inducing the activation and proli~eration of B cell5 in the cell-free supernatant 21 through the B cell antigen receptor mediated signalling 22 pathway utilizing dextran conjugated anti-IgD
23 antibodies ~ in the presence of Interleukin-l and 24 Interleukin-2 wherein said B cells are at least 98 . 5 WO 95113294 PCr/USs4/12835 21761 1~

percent purif ied through f luu- èSCe~lC~ ~ctivated cell 2 sorting (FACS~;
3 e. adding activated T cell supernatant;
4 f. harvesting the B cell culture 5 supernatant 6 days after culture; and 6 g. measuring the cullcellLLation of 7 secreted IgM in the B cell cultured supernatant by an 8 enzyme-linked ~ r~rbent Assay ( ELISA) .
9 15. The product produced in accordance with the lO method of Claim 14.
ll In sum, we have established a novel in vitro assay 12 which has allowed us to screen for potentially novel 13 factors which induce immunoglobulin (Ig) secretion by 14 murine B lymphocytes. We refer to such factors as B

15 cell maturation factors (BMF). Utilizing this asfiay we 16 have identif ied two such activities which we have shown 17 to be distinct from the well-described and 18 characterized cytokines which are known to play a role l9 in B cell activation and differentiation. The first 20 activity was identified in culture supernatants from 21 pure populations of in viL,., ~elleLated IL-2-activated 22 murine natural killer (NK) cells. The second activity 23 was identified in the culture supernatants of a panel 24 of anti-cD3-activated murine CD4+ T lymphocyte clones.

WO9~i/13294 ` 2 1 76 ~ 1 4 PCrlUS94112835 While these two activities are functionally similar, 2 they could represent distinct molecules. Studies to 3 identify the murine 3MF from one of the CD4~ T cell 4 clones ~re currently underway. A cytokine with the ability ~to induce high rate Ig sccretion, or a 6 selective antagonist of such a molecule, could be of 7 potentia1l value as a therapeutic in a diverse number of 8 clinical settings as well as an in v~ tro tool to 9 generatej antibody-forming cells.
VII. Methr~le ll ~. Female DBA/2 mice were obtained from the 12 National Cancer Institute (Frederick, MD) and were used 13 at 7-lO weeks of age. The experiments were conducted 14 according to the principles set forth in the Guide for lS the Care ` and Use of Laboratory Animals, Institute of 16 Animal Rlsources, National Research Council, Department 17 of Health, Education, and Welfare Publ No. (National 18 Institutes of Health~ 78-23.

19 Cul~re medium. R~MI 1640 ~3iofluids, Rockville, MD) supplemented with lO~ fetal bovine serum (Gibco 21 Laboratoriies, Grand Island, N.Y.), L-glutamine (2 mM), 22 2-mercaptoethanol (0.05 mM), p~nirillin (50 mg/ml), 23 :.L1 ~:~t l~in (50 ,ug/ml), and gentamicin (S0 mg~ml) were 24 used for culturing cells.
I

WO 95113294 2 1 7 6 1 ~ 4 PcrluS94/12835 Reaaents. ~-dex was prepared by conjugatiOn of 2 Hô~ Cl~nA l mouse IgG2b (b allotype) anti-mouse 3 IgD (a allotype) to a high molecular weight dextran ~2 4 x 106 M.W. ) . Approximately 6 H~'/1 were conjugated to 5 e~ch dextr~n molecule. LPS ~, extracted from 6 Pc~-h~richia ~ 0111:B4, was obtained from Difco 7 Labor~tories, Inc. (Detroit, NI) and was used at 20 8 mg/ml in all experiments. Affinity-purified, 9 phycoerythrin-labelled polyclonal goat anti-mouse IgM
10 was obtained from southern Biotechnology Associates 11 (Birmingham, AL). FITC-labelled monoclonal hamster IgG
12 anti-CD3e (2Cll) monoclonal antibody (mAb) was obtained 13 from Pharmingen (San Diego, CA). Polyclonal rabbit 14 anti-AsGm-l was obtained from Wako (Osaka, Japan).
15 Murine r~ ;n:~nt (r)IL-1 was a gift from Dr.
16 Stephanie Vogel (USUHS, Bethesda, MD). Murine rIL-2 was 17 a gift from-Dr. Maurice Gately (Hoffman LaRoche, 18 Nutley, NJ). Murine rIL-5 was a gift from Dr. Richard 19 Hodes (NIH, Be'rh~c-l~, MD).

Prel~aration and culture of B cells. Enriched 21 populations of B cells were obtained from spleen cells 22 from which T cells were eliminated by treatment with 23 monoclonal rat IgM anti-Thy-l (H013-4), rat IgG2b 24 anti-CD4 (GK1.5~, and rat IgG2b anti-CD8 (2.43), followed by mouse anti-rat Igk (MAR 18 . 5) and WO 9~/13294 2 1 7 6 ~ 1 4 PCTIUS9411~835 complement. Cells were then fractionated into high and 2 low density populations by centrifugation over a 3 Percoll gradient (Pharmacia, Piscataway, NJ) consisting 4 Or 70, 65, 60, and 50~ Percoll solutions (with densities of 1.086, 1-081, 1.074, and 1.062 g/ml, 6 respectively). The high density cells were collected 7 from the 70 to 66% interface and were used in all 3 experiments. The average percentage of Ig+ cells was 9 85-90% in ~ the 70 to 66% fraction. B cells obtained in this way were referred to as "B cell-enriched~ ~B') 11 cells. This cell population is also known to contain 12 small numbers of NR cells, macrophages, mast cells, and 13 cells of the gr~nulocytic series. To remove NR cells 14 from the B' cell preparation we further treated B' cells with polyclonal rabbit anti-AsGm-1 antibodies (Waco, 16 osaka, Japan) plus complement as follows: B' cells (107 17 cells/ml) were incubated in "cytotoxicity medium" (C~) 18 (RP~I + 2.5 mM Hepes + 0.37~ BSA) with anti-AsGm-1 19 antibody ~150 mg/ml final concentration) at 4C ~or 1 h. Cells were then washed once in cold Cl~ and 21 r~ n~lon in C~ containing a 1/10 dilution of rabbit 22 complement~ (Pel-Freeze, Brown Deer, Wisconsin) at 37C
23 ~or 1 h. Cells were then washed and ~ od in 24 medium for further use. Functional assays were carried out in 96-well flat-bottom Costar plates (Costar, ;

PCT/I~S94/12835 ~WO95/13294 2 ~ 76 1 1 4 --lg--Cambridge, Ma). Cultured cells were incubated at 37c 2 in a humidified atmosphere containing 6~c CO2.
3 ~C~Ahl i of NR cell rl- 1 tllres . Spleen cells 4 from C8-17 SCID mice, obtained from NCI (Frederick, MD), were cultured in medium at 1 x 10~/ml in the 6 presence of 500 U/ml of rhIL-2. Spleen cells were ~irst 7 treated with anti-Thy-l, anti-CD4, and anti-CD8 +
8 complement as a precaution against the possible 9 presence of small numbers of T cel~ls resulting from "leakiness" in the SCID mutation. NX cells were ll maintained by splitting them 1: 2 into fresh medium 12 500 U/ml of IL-2 every 2-3 days. Cells were used for 13 experiments beginning -7-10 days after establishment o~
14 culture . Such cells were monitored by f low cytometry to confirm the absence of CD3~ ~T) cells. NR cell cultures 16 were re-established every month, using fresh spleen 17 cells frcm SCID mice, since longer culture periods were 18 associated with deterioration of the NK cell line.
19 Cells derived from CB-17 mice, like those from D8A/2, express N-2d MHC class I molecules.
21 Prel~aration of suT~ernatAnts rSN~ from TT-2 22 activated. in-vit v ~ L~,I ed NR cell rl-ltl-r~2c 23 (NR-PlMF). NR-BMF was prepared in two basic ways: (1) In 24 vitro-maintained, IL-2-activated NR cells were washed WO9~/13294 ~ 2 ~ 76 t t 4 PCrNS941128~ ~

3x and recultured at 1- 0 x 10S cells/ml in the presence 2 of IL-2 (500 U/ml). 24 h later cell-free culture SN was 3 harvested, aliguotted, and stored at -20C until used.
4 t2~ In vitro-maintained, IL-2-activated NR cells were washed 3x and recultured at 1. 0 x 105 cells/ml in the 6 presence ~ of 5 x 105 B' cells/ml, but in the absence of other exogenouS stimuli- 24 h later cell-free culture 8 SN was h~rvested, aliquotted, and stored at -20c until g used.
PreDarz.tion o~ rnat~ntC (SN) ~rom activated 11 C~4~ T cell clQnes ~T-B~F~. Antigen-specific, ~HC class 12 I~-restricted, CD4~ T cell clones were established 13 elsewherel by standard methodologies. The following CD4t 14 T cell clones were assigned to the Thl subset on the basis of their secretion of IL-Z and IF~i-g, but not 16 IL-4: The ~rabbit gamma globulin-specific, 17 Ia~-restricted Thl clone, Dl.6, was established in the 18 laboratory of Dr. Abul Abbas (Harvard ~edical School, 19 Boston, MA) whereaS the RLH-specific Thl clones, RAl, RA5, RA8, RC5, and RC9 were established at Immunex 21 Corporation, Seattle, WA. These T cell clones were 22 maintainedj by weekly stimulation with antigen, spleen 23 cells (as a source of antigen-presenting cells [APCsl ), 24 and ,~ J~ t lc IL-2.
' i PCI'IUS94/12835 ~W09~113294 ~ ~ 7~ ~ 1 4 Cytokine-COntaining SN were obtaiAed from cultures 2 o~ these CD4~ T cell clones in the following manner:
3 Tissue culture well5 were incub2ted with anti-CD3 mAb 4 (2C11) at 10 mg/ml in PBS for 3h at 3 ~C 2nd then w2shed 3x in fresh PBS. T cell clones which were 6 211Owed to return to their resting state after 7 stimulation with antigen, APCs, and IL-2 by maintaining 8 them for -7 days in the absence of any ~urther g ~ nn~ stimuli were added to 2nti-CD3-co2ted plates at lxlO~/ml for various times, upon which cell-~ree SN
11 were obtained 2nd either stored 2t -20C or 4C. In the 12 12tter c2se, SN w2s used in cellul2r 2ss2ys within 1-2 13 weeks of h2ving been harvested.
14 Cvtoflu~ ic analvsic and cell sorf in~7, Spleen cells were stained for 30 min with FITC-~Ahe~
16 2nti-CD3 nAI~ + PE-labelled 2nti-IgM antibodies ( f inal 17 .n. ~--L,~.tion of 10 mg/ml each in the presence o~ a 18 5-fold excess of anti-Fc~yRII mAb to prevent cytophilic 19 antibody binding) at 107 cells/ml in cold clear HBSS
cn~-t~inin~ 3~ FBS and 50 mg/ml each of p~nici 11 in~
21 streptomycin, and gentamicin. Cells were then washed 22 and ~ L~ in staining buffer at 107 cells/ml in 23 prep2ration for flu~,re._~n. - an21ysis 2nd/or cell 24 sorting. For analysis, a FACStar Plus or FAcSCAN (y WO 95/13294 ~ 2 ~ 4 PCTillS94/1283~, Becton Di~l~incon~ Mountainview, CA) was used and 15,000 2 cells were collected using logarithmic amplif ication.
3 only viable cells were analyzed on the basis of their 4 characteristic for~ard and side scatter pro~iles. Cell ~i sorting was similarly carried out on a FACStar Plus, as 6 well as on an Epics Elite (Coulter Corp., Nialeah, FL), and sorteld cells were immediately reanalyzed to conf irm 8 their sta~ining profile. Only sorting purities of >98%
9 were acceptable for subsequent study. Sort-puriried B
10 cells (mIyMtCD3 ) were referred to as B'P cells. Non-3, 11 non-T cells (mIg~CD3~) and T cells (mIgMCD3~) were also 12 collected~ and macrophaqes were routinely eliminated, 13 during sorting, on the basis of their characteristic 14 forward and side scatter profile.

Ouantitation of secreted I~M. Ig~l concentratiOnS
16 were measured by ELI5A, with Immulon 4, 96-well 17 flat-bottomed ELTSA plates (Dynatech Laboratories, 18 Alexandria, VA). 3riefly, E:LISA plates were coated with 19 polyclonal goat anti-mouse IgM antibodies (Southern 20 BiotP~-hnn j~yy Associate5), followed by addition of 21 serial dilutions of samples and ~Ld~ dL~s~ Alk;~l inP
22 phosphatase-ConjUgated polyclonal goat anti-mouse Ig~
23 Ant iho~ ! were then added, upon which a flu-lr~.,cellL
24 product was generated from cleavage of 25 4-methyl; illiferyl phosphate (Sigma) by ~WO 95113194 -2 PCrlUS94/1283S

specifically-bound AlkAl in~ phosphatase-conjugated 2 Ant;hc~l~ie<:. FlUuLes~e.,c.e was measured on a 3~ 96 3 fl~lu~ r (l~ountainview, CA) and fluoL~5~ C:~ units 4 were converted to Ig ~_ull~ LatiOnS by extrapolation 5 from standard curves determined in each assay by using 6 purified myeloma Ig~ of known cu-.~e..LLation. IgM
7 measurements showed no significant cross-reactivity or 8 interference from the presence of other isotypes (IgD, 9 IgG3, IgGl, IgG2b, IgG2a, IgE, and IgA).

o~ titation of IL-6. Flat-bottomed, 96-well 11 microtiter plates (Nunc, Roskilde, Denmark) were coated 12 with 6 mg/ml of anti-mouse IL-6 mAb (MPS-20F3 ) in o .1 M
13 carbonate buffer (ph 9.6) for 3h at room tempera~ure, 14 blocked with PBS-1% BSA for lh at room temperature, then extensively washed with PBS-Tween. Supernatant or 16 r ~in~nt.purified lymrhl-l-in~ (PhA~min~n~ diluted in 17 PBS-1% BSA was incuhated on these plates f or lh .
18 Biotinylated anti-IL-6 mAb (MP5-32cll) was used at 1 19 mg/ml and avidin-conjugated alkaline phosph atase at a dilution of 1/2,000. Preliminary experiments showed 21 this assay to be sensitive to 0 . 22 ng/ml of rIL-6 .
22 Specificity was d~_ ~ed by the absence of binding by 23 1,000-fold higher ~ul.c~nLLàtions of IL-4, IL-5, and 24 IFN--y-W09~113294 ~ 217$~ t4 PCrlUS94/1283~

VIII . ~AMPT.F~
2 The1 following working examples are not intended to 3 limit the invention which i8 def ined by the claims 4 appended below.
A. Dextr;~n-ConiUqated i~nti-IaD ;~nt~ hodies 6 (~-dex~ ~ootentlv induce ~roliferation of r~rino B
7 col 1 s in ~ vitro . -We syntheSiZed a novel construct which 8 potently ~ activates murine B cells through the membrane 9 (m) Ig (antigen receptor) -mediated signal transduction pathway. ~This construct consi6ted of covalent linkage 11 of multiple anti-IgD - ~clort~l antibodies (mAbs) to a 12 high molecular weight dextran backbone. We henceforth 13 refer to this construct as R~-dex (Ref l). The high 14 valency of anti-IgD antibodies linked to dextran resulted ~in extensive 8 cell mIg crosslinkage at 16 extremely low concentration5 of anti-IgD. This resulted 17 in minimal modulation of mIgD from the B cell surface 18 and hence allowed for continuous B cell si~n~llin~.
l9 Given the small numbers of antigen-specific B cells in ZO the lln~ l mouse, the adYantages of ~-dex lied in 21 its abili~ty to polyclonally activate the vast majority 22 of matUre 8 cell5 through the mIg signal transduction 23 pathway. Further, c~-dex mimicked the repeating epitope 24 nature of~ polysaccharides, such as those found within I~acterial cell walls, and could be used as an in vitro PCTNS9~112835 ~W095113294 21 76 T ~ 4 model to study immune responses to that class of 2 antLgens. ~-dex induced resting murine splenic B cells 3 to proliferate but dia not, by it5elf, stimulate Ig 4 secretion. Compared to unconjugated anti-IgD, eô-dex 5 induced substantially higher maximal levels Or 6 proliferation at 1000-fold lower ~..,c~ ations of 7 anti-IgD (Ref 1).

B. InT-erl~kin ~IT-)-2 or rTI-5 in~1lce I~ secretion 9 in eô-dex-activa~ed E~ cell-enriched T cel]-de~leted 10 SDleen cell ~ re5-Addition of the cytokines IL-2 or 11 IL-5 to eô-dex-activated B cell cultures resulted in 12 the secretion of large amounts of IgM ~Ref 2). By 13 contrast, unconjugated anti-IgD was ineffective at 14 stimulating cytokine-mediated Ig 5ecretion. Thus, 15 45-dex represented the first efficient in vitro system 16 f or inducii~g cytokine-dependent polyclonal Ig secretion 17 through the mIg signal ~cransduction pathway. The 18 population of small, resting B cells utilized in these 19 studies were established by depleting T ly ~-_yLes 20 from splsen cells with a cocktail of anti-T cell 21 antibodies and l~ ~ followed by ~ractionation, 22 according to density, by centrifugation on a 23 rliccnr~in--n~C Percoll gradient. Cells were obtained 2~ from the high density Percoll fraction and represented 25 cells in the resting or G~ state of the cell cycle.

WO95/13294 2 1 7 6 ~ t ~ PcrluS94/12~
This population of sm~ll T cell-depleted spleen cells 2 consisted of 85-90% B cells with the r- in~lDr oS cells 3 consisting mostly of NK cells, macrophages, and 4 granulocytes. We henceforth refer to this B
5 cell-enriched spleen cell population as B' cells.

6 IL-2 st; lated IG 5ecretion bY ~r~-dex-activated 7 ~' cells but not bv hiGhlV ~urified B cells. To 8 determine whether IL-2 or IL-5 could act directly on 9 the ~-dex-actiVated B cell to induce Ig secretion we obtained a highly purif ied population of small B cells 11 through th~ use of a f 1UI escel~ce activated electronic 12 cell sorter (FACS) ~Fig 1~. Thus, small B' cells were 13 stained with phycoerythrin (PE)-labelled anti-IgM
14 antibody which selectively binds to m~gM~ B cells, plus FITC-anti-CD3 to identi~y and eliminate any residual T
16 cells and/or cells binding anti-IgM non-specif ically .
17 sort-purisied mIgM~ B cells were obtained at >99%
18 purity and are henceforth referred to as B'P cells.
19 Whereas LPS or ~-dex plus IL-5 stimulation resulted in a comparable level of Ig secretion by small B' and B'P
21 cells, activation with ~r~-dex plus IL-2 led to Ig 22 secretion only by B' cells (Table 1). This suggested 23 that a non-B, non-T cell removed during the FACS
24 ~)LUC~::dU-~ was critical for induction of Ig secretion by B'7 cells _n response to ~-dex plus IL-2 . B ell . .

WO gS113294 2 ~ 7 ~ 1 ~ 4 PC~n~S94112835 .

populations which were less than 98 . 59~ purity typically 2 contained suf~icient non-B, non-T cells to allow for 3 induction of Ig secretiOn in response to ~-dex plus 4 IL-2 . Hence FACS was required to obtain suf ~iciently 5 pure B cells to abrogate the ~r~-dex plUB IL-2-induced 6 Ig se.,L21 0Ly .~ e. We ' Llated that the Ig 7 secretory response of small B~ cells in response to 8 cr~-dex plus IL-2 could be restored by adding back the g splenic non-B, non-T cells which were removed by FACS
l~ ~F1g 2) .

11 The s~lpn; c non-B non-T ce~ l ~ res~oncihle ~or 12 induc~ion of Iq secretion i n 2~-dex ~ C IL-2-activated 13 ~'P cells are AsGm-1~-Depletion of macrophages from the 14 B' cell population, by exploiting the property of 15 macroph2ges to selectively adhere to plastic, had no 16 effect on I~ secretion in ~r~-dex plus IL-2 activated 3' 17 cells (data not shown). Thus, we postulated that the 18 non-B, non-T cell necessary for inrhlrtinn of Ig 19 ~ecretion in this system was an NK cell. Since the vast 20 majority of splenic NK cells, as well as some 21 macrophages, selectLvely express the m2rker Asialo Gm-l 22 (AsGm-l~ on their surface, we eliminated NR cells by 23 incubating small B' cells with anti-AsGm-l antibody 24 plus complement. This ~Lo~ eduLe abrogated the Ig secretory ~ of B' cells to ~-dex plus IL-2 W095113294 217~4 strongly 9uygesting that NK cells were responsible for 2 induction of Ig secretion in this system ( Fig 3 ~ .
3 Additional experiments d ~L ~ ted that depletion of 4 AsGm-l ' kells from the ~mall B' cell population had no S e~fect on the Ig secretory response to clô-dex plus IL-5 6 tdzta not shown), consistent with the ability of these 7 latter stimuli to act directly at the level of the B
8 cell.

9 A ~ure mo~ulation of ~ctlvated NK cel ~ ~ release an Ic~- j n~ i ncl activit~ To better def ine the role of NK
ll cells in~inducing Ig secretion by activated B'P cells we 12 established a pure population of NK cells in vitro.
13 Thus, we ~isolated spleen cells from mice which were 14 h: _y~ S for the severe combined; no~ ficiency (scid) mutation. Such mice genetically lack both B and 16 T cells bUt- contain functional NK cells. Culture of 17 scid spleen cells in relatively high l_o~ Lreltions of 18 IL-2 resulted, within 6 days, in a pure population o~
l9 activated1 NX cells as demonstrated by flow cytometric analysis (Fig 4). Thus, virtually all cells C~ IL ss~d 21 the NX cell marker AsGm-l as well as Thy-l, and lacked 22 markers fqr T cells (CD3), B cells (Ia) and macrophages 23 (MAC-l). The use of scid spleen cells were essential 24 for obtaining this population since high doses of IL-2 would als~ expand even small numbers o~ T cell~ which WO 95/132~4 2 ~ 7 6 1 1 4 PCT/US94/12835 might contaminate the B' cell population. Addition of 2 in vitro-generated pure NK cells to ~-dex plus 3 IL-2-activated B'P cells resulted in induction of Ig 4 secretion which was compar~ble to that observed when 5 sort-purified splenic non-B, non-T cells were added 6 (Fig 5). sort-purified 5mall naive splenic T cells, by 7 contrast, were ineffective at inducing Ig secretion in 8 this system. NK cells induced optimal Ig secretion when 9 present at 5-10% of the B cell population (Fig 6).
10 Addition of conditioned medium from these 11 IL-2-activated NK cell cultures also induced Ig 12 secretion in tr~-dex plus IL-2-activated B'P cells in the 13 absence of NK cells indicating that NX cells released a 14 B~F tTable 2). We henceforth refer to this BMF as 15 NR-BMF. optimal Ig induction by NK-aMF occured at 50%
16 flnal volume of NK-BMF with little if any activity 1~ observed at- 10~ (Fig 7) .

18 NK-BMF is not amonq the cYtoRines known to induce 19 B cell activation and maturation. To better 20 characterize the nature of NK-Br~F we performed two 21 types of experiments to ~t~rm; n~ whether NK-BMF was 22 among the cytokines known to induce B cell activation 23 and differentiation. In the first set of experiments we 24 added known cytokines, either singly or in combination, 25 at various c,,..- =.,L~tions, to cultures of B'P cells PCr/US94/1283~
WO95/13294 1 2 ~ 7 ~

stimulated with ~-dex plus IL-2. Neither IL-3, IL-~, 2 IL-10, TNF-~, IFN-y, IFN-~/15, GM-CSF, or TGF-~ was 3 capable of inducing Ig secretion in this system. In the 4 second &et of experiments we ' L- ated that 5 neutralizing antibodies to IL-4, IL-5, IL-10, TNF-~r, 6 IFN-7~, and IFN-~/3 failed to abrogate the Ig secretory 7 response tTable 3). In separate experiments, using 8 various concentrations of recombinant IL-12 and a g polyclonal goat anti-mouse IL-12 antisera, we also 10 demonstrated that NR-BMF was not IL-12 (data not 11 shown). As indicated below, several distinct 12 experiments were further carried out which indicated 13 that neither NK-BMF nor T-BMF was IL-6 (see Table 4, 14 Table 5, ~and Figure 11).

Anti-CD3-actiVated r-lr;n~ CD4~ T cell clQnes alsg 16 release a BMF. Studies with NK-BMF indicated that this 17 activity ~was present at relatively low titers as 18 indicated by a requirement for its presence at 50%
lg f inal volume to induce optimal Ig secretion . This has 20 made the ~further characterization and identification of 21 this activity technically dif~icult. Since no one has 22 yet been ~u~ rul at est~hli~hirq long-term murine NK
23 cell clones we were further hindered by an inability to 24 select for a high BMF-producing NX cell clone. We thus 25 determine :I whether other cell types were capable of W0 95/13294 2 1 7 6 1 1 ~ PCrlUS94112835 releasing an activity which could 5imilarly induce Ig secretion by ¢~-dex plus IL-2-activated B'P cells. Many 3 antigen-specific murine in vitro CD4+ T cell clones 4 have been es~hl ich.oA in numerous laboratories. It has 5 been vell-established that such clones secrete large 6 amounts of cytokinefi when cultured in tissue culture 7 wells that have been pre-coated with anti-CD3 mAb. CD3 8 is a ~ - t o~ the T cell antigen-receptor and its 9 crosslinkage induces T cell activation. Further, CD4~ T
10 cell clones have been subdivided into two broad 11 categories d~r~nAnt upon the profile of cytokines they 12 release upon activation. Thus Thl clones exclusively 13 release IL-2, IFN-~y and lymphotoxin upon activation 14 whereas Th2 clones exclusively release IL-4, IL-5, IL-6, IL-g, IL-10, and IL-13. 30th types of Th clones 16 share the ability to release TNF-¢, IL-3, and GM-CSF.

17 Since we d~ LL It~d that IL-5, present 18 exclusively in SN from anti-CD3-activated Th2 clones, l9 could induce Ig secretion by ¢~-dex or ~-dex plus 20 IL-2-activated B'P cells we evaluated Thl clones ~or 21 their potential release of a novel BMF.
22 Thus, SN from a single anti-CD3-activated Thl clone was 23 added to cultures of ¢~-dex plus IL-2-activated B'P
24 cells and was shown to stimulate large amounts of 25 secreted Ig tTab1e 2 ) . We similarly tested 5 additional PClfUS941128 WO9Sfl3291 ` 2 1 ~'~ t 1 4 dlstinct Thl clones for their ability to release a BMF
2 upon anti-CD3 activation. All Thl clones were found to 3 release ~BNF (Fig 8) . This BMF is henceforth referred to 4 as T-BMF. In contrast to NX-BMF which was required at a 5 25-50% final volume to induce Iq secretion, T-BMF
6 induced near-optimal Ig secretion at a f inal volume of 7 0.1% andj continued to have Ig-inducing activity at 8 O.Ol~ (Fig 9). Thus, BMF-containing T cell sl~pernatants 9 appeared to be at least 250-fold more potent in lO Ig-inducing activity than supernatants derived from ll activated NK cell cultures. T-BMF, in the absence of 12 ~-dex~IL-l+IL-2, had no Ig-inducing e~fect on small B`P
13 cells (dàta not shown).

14 T-8~F is a ~rot~;n and is not i ~ the cvtokines 15 known to ~induce B cell activation 2nd dif~erentiation.
16 That T-BMF is a protein was ~1 L.c,ted by the ability 17 o~ proteinase R LLe:a; to abrogate the Ig-inducing 18 activity of T-BMF (Fig lO). This effect was not due to l9 an inhibi~tory effect of residual, active proteinase R
20 upon addltion of proteinase X-treated T-BMF to the 21 activatedl B'P cell cultures since addition of proteinase 22 K-treated~ T-BMF to ~I~-dex plus I~-5-activated B'P
23 cultures did not reduce Ig secret~on.

PCT~S94/12835 ~WO9Y1329~ 2 1 76 ~ 1 4 To det~-rmi n~ whether T-BMF was among the cytokines 2 known to stimulate B cell activation and/or 3 dirferentiation, we stimulated B~ cells with ~-dex, . 4 IL-l, IL-2, and 0.1% T-BMF and added neutr~lizing 5 antibodies or antagonists to IL-3, IL-4, IL-5, IL-6, 6 IL-10, GM-CSF, TNF-, and IFN-~. none Or the 7 ;~ntihoAies tested inhibited Ig secretion (Table 4).
8 Since IL-6 has been shown to be a cytokine capable of 9 inducing B cell maturation to Ig secretion, applicants 10 further tested whether T-BMF was IL-6. T-BMF was not 11 IL-6 as indicated in two distinct ways: (1) direct 12 mea-uL~ - ~ of IL-6 in T-BMF, utilizing a highly 13 sensitive ELISA, indicated undetectable amounts Or IL-6 14 (c200 pg/ml) in undiluted T-BMF (Fig 11). Since T-BMF
15 induces Ig at 0.01% final volume, IL-6 if prese~t would 16 be <0 . 02 pg/ml . Since IL-6 has been reported to exhibit 17 B cell matu~ration effects only at ~;~.lc-:..L-ations in the 18 ng/ml range, it is highly unliXely that T-BNF was IL-6.
19 (2) Addition of IL-6 rrom 2-20,000 pg/ml to cultures of 20 ~-dex plus IL-2-activated B'P cells failed to induce Ig 21 secretion (Table 5). Finally, since TNF-~ and 22 lymphotoxin bind to the same receptor and exhibit 23 nearly overlapping functional efrects, it is highly 24 unlikely that T-BNP is lymphotoxin, since TNF-û is not 25 Active as ~ BMF in this system.

WO95/13294 2 1 7 ~ ~ 1 4 PCr/US9411283~

T-RhF is released ~:~rlv after ~r~ti-CD3-activation 2 of a Thl clone-Kinetic studies were performed to 3 determinè the time during which T-BMF was induced upon 4 activation with anti-CD3 mAb. Approximately 90~6 of 5 total T-~3MF activity was induced between 2-3 h a~ter 6 anti-CD3 1~ctivation (Fig 12). In the absence of 7 anti-CD3~ no detectable BMF wa8 induced in the Thl 8 clone. I
g T-BMF acts late in cul~re to s~im~late Iq secretion bY ~-dex ~lu5 IL-l+IL-2-activated ~3 cells.
11 To determine the time in culture when T-BMF was 12 required to induce Ig secretion by B cells activated 13 with ~ lex plus IL-l+IL-2, we established replicate 14 cultures ~ and added T-BMD on different days after initiation of culture- Culture SN was harvested from 16 all experimental groups 6 days after culture 17 initiation. T-BMF strons~ly induced IG secretion, even 18 when it ~as added as late as day 3 of the 6 day culture 19 and the ~evel of T-BMF-induced Ig secretion was comparable to that observed when T-BMF was added at 21 initiation of culture (Fig 13). Kinetic studies in our 22 lab have lindicated that the majority of Ig secretion in 23 this system occurs between days 3 to 5 (data not 24 shown). ~Addition of T-BMF 1 day after culture initiati~n induced a substantially higher Ig secretory .' , ~

-PC rlus94112835 ~WO 95113294 2 1 7 ~

L ~ ae relative to cultures in which T-BMF was added 2 at culture initation (Fig 13). Peak Ig secretion, 3 however, occured when T-BMF was added at day 2. Thus, 4 T-BMF acts late in culture to induce Ig secretion, 5 supporting the view that it functions as a true B cell 6 maturation f actor . The observation that T-BMF induced 7 opitmal Ig secretion when its addition to culture was 8 delayed by 2 days suggests that cither T-BMF or a 9 separate _ - -t in the T cell SN in inhibitory for lO Ig secretion when present early in culture, or that T-ll BMF activity progressively declines during the culture 12 period and is not available in optimal amounts, when 13 added at initiation of culture, for inducing Ig 14 secretion late in culture.

WO 95113294 1 '2 ~ 7 ~

Table l 2 ~ Ia21 secretion (n~/ml 3 1 B'P 8' 4 Medium ~ : 50 < So 5 ~rô-dex ~ 600 5zS
6 IL-2 < 50 70 7~-dex+IL-2 2,250 16,900 8c~-dex+IL-5 7, 500 7, 500 9LPS ~ ll9,000 90,600 lO ~able l. ~ B' and B'P cells were cultured at l. 25 X lOs ll cells/ml~ in the pre5ence of ~-dex ~3 ng/ml), IL-2 (150 12 U/mll, IL-5 (150 Ulml) and/or LPS (20 ~q~ml). six days 13 after initiation of culture, SN was harvested ~or 1~ ~iete min ~on or r Igll .~n~ ations by ~A.

') T/U594/12835 WO9S113294 ~ 1 7 6 1 ~ ~ PC

Table 2 2 IgM secretion (ng/ml) 3 B' B~P
4 Med 2 6 < 5 5 ~-dex 160 120 6 IL-l/IL--2 54 < 5 7 ~-dex +
8 IL-l/IL-2 7,500 220 g ~rS-dex + IL-5 23,100 15,000 -dex +
12 IL-l/IL-2 + NR-BMF 8, 750 13 T-BMF ~ 5 14 ~ô-dex + T-BMF 150 15 ~-dex + IL-l/IL-2+T-BMF 4, 500 16 Ta~le 2. B' and B~ cells were cultured at 1. 25 X 105 17 cells/ml in the presence of ~-dex (3 ng/ml), IL-l (150 18 U/ml), IL-2 (150 U/ml), IL-5 (150 U/ml), NR-BMF (25%
19 v/v), and/or T-BMF (25% v/v). Six days after initiation 20 of culture, SN was harvested for rl~te~minAtion 21 secreted IgM ._.,..,~.,LLc.tions by ELISA.

WO 95/13294 2 ~ t 4 PCT/US9 Table 3 2 AUqust 1-92 I~M secretion 3 (ng/ml) 4 B' cells 5 Medium < 125 6 a~-dex+IL-2 13, 700 7a~-dex+I;-2+anti-~NF-a 11,500 8a~-dex+IL-2+anti-IFNy l5,00o 5aô-dex+IL-2+anti-IFN2/,~ 10,250 a~-dex+IL-2+anti-IL-4 11, 000 11 a~-dex+IL-2+anti-IL-5 12,500 12 a~-dex+IL-2+anti-IL-6 11,250 13 a~-dex+IL-2+anti-IL-10 13,100 14 Sel~tembe~ 1992 B'P cells I~M 5ecretion 16 (nq/ml) 17 Medium ~ < 3 5 18 c~-dex+IL-2 1,250 19 a~-dex+IL-2+NK-BMF 35,000 a~-dex+Il-2+IL-3 2, 000 21 a~-dex+IL-2+IL-4 975 22 ~-dex+Il-2+IL-10 800 23 a~-dex+IL-2+TNF-a 1,250 24 a~-dex+IL-2+IFN-~ 2, 000 a~-dex+IL-2+IFNl~+TNFa 1,250 ~WO95/13294 ~ 6 ~ ~4 PCr/USs4/~2835 t~-dex+IL-2+IFN~ 1,875 2 ~-dex+IL-2+IFN~+IFN~ 1,250 3 ~-dex+IL-2+IFN~+TNFa 1, 375 4 a~-dex+IL-2+IFNa+IFNy+TNFa 1,000 ~-dex+IL-2+GM-CSF 1,313 6 ~r~-dex+IL-2+TGFB1 1,125 7 Ta_le 3 . (Au~ust l9g2 ~ . ~' cells were cultured at 1. O
3 x 105 cells/ml with ~r~-dex (3 nglml), IL-l (100 U/ml 9 and IL-2 (150 U/ml) in the absence or presence of the following antibodies: (1) anti-TNF-a mAb (XT22; 50 11 g/ml), (2) anti-IFNy mAb (XMG-6; 50 ILg/ml), (3) 12 polyclonal rabbit anti-IFNa/B anti-serum (1/50 v/v), 13 (4) anti-IL-4 mAb (llbll; 50 ~Lg/ml), (5) anti-IL-5 mAb 14 (TRFK-5; 50 llg/ml), (6) anti-IL-6 mAb (P7; 50 ~Lg/ml), and (7) anti-IL-10 mAb (SXC; 20 ~g/ml). Six days after 16 initiation ~f culture, SN was harvested for 17 ~jotPrmi n~tion of secreted IgM concentrations by ELISA.
18 (SeP~omhor 1992). B'P cells were cultured at 1.5 x 105 lg cells/ml with ~-dex (3 ng/ml), IL-l (100 U/ml) and IL-2 (150 U/ml~ in the absence or presence of the.
21 following cytokines: (1) IL-3 (t?EHI 3 SN 25% v/v), (2) 22 rIL-4 (lOQO U/ml), (3) rIL-10 (10 U/ml), 4) rT~F~ 0 23 U/ml), (5) rIFNa (10 U/ml), (6) rIFN~ (100 U/ml), (7) 24 rGM-CSF (100 U/ml), and/or (8) purified nztural TGF-~l (1. 0 ng/ml) . six days after initiation of culture, SN

WO 95/13294 2 1 7 ~ PCr/US94/1283j~

was harvested for determination of secreted IgM
2 ~ ce~L.I~ions by ELISA- (IL=interleukin, ~NF=tumor 3 necrosis factor, IFN=interferon, GM-CSF-granulocyte-4 macropha~-e colony stimulating factor, TGF=transforming 5 gro~th ~ctor~.

.

WO 95/13294 2 1 7 ~ PCT/US94/1283~i TABL~ 4 2A. IcM secretion fnq/lnl ) 3~-dex + IL-l + IL-2 82 4~-dex + IL-1 + IL-2 + RA5-SN 4, ooo s ~-dex + IL-1 1 IL-2 + RA5-SN + antiIFN-y 3, 250 6 + TNFRIg 3, 620 7 + anti-GM-CSF 2, 400 8 + anti-IL-3 2, 300 g + anti-IL-4 4, 750 + anti-IL-5 4, ooo 11 + anti-IL-10 2,750 12 + huCD40Fc 4, ooO
13 B.
14 ~-dex + IL-1 + IL-2 350 15 ~-dex + IL-l + ~L-2 + RA5-SN 15,000 16 cl~-dex + IL-l + IL-2 + RA5-SN + anti-IL-6 16,200 17Table 4 . B'~ cells were cultured at 1. 5 x 105 cells/ml 18 with ~-dex (3 ng/ml), IL-l tl50 U/ml), IL-2 (150 U/ml) 19 and/or T-BMF (0.1% v/v [24 hr SN from anti-CD3-actiYated RA5 Thl clone] ) in the absence or ~Lese.,ce o~
21 the following ~ntihorlies/antagonists: [A] (1) anti-IFN-22 y mAb (XMG-6; 10 ~lg/ml), (2) TNFRIg (10 llg/ml;
23 tTNFRIg=tumor necrosis factor receptor-Ig hybrid 24 molecule which binds and neutralizes free TNF-~]), (3) anti-GM-CSF mAb (10 ~lg/ml/ purchased from Genzyme), (4) 2 ~ 76 1 1 ~ 283~

anti-IL-3 mAb (8F8 l, lO ~g/ml), (5) anti-IL-4 mA~
2 (llBll; lO ~g/ml), (6) anti-IL-5 mAb (TRFK-5; lO
3 g/ml), (7) anti-IL-lO mAb (2A5; lO ,~g/ml), (8) huCD40Fc 4 (lO /lg/ml; [human recombinant CD40-lg hybrid with bind~
and neutrali2es ~ree CD40 ligand). [B] anti-IL-6 mAG
6 (MP 520 Pc; lO t~y/ml)- siX days after initiation of 7 culture, SN was harvested ~or determination o~ secreted I-l)l con~;ntr~t~ by ELISA.
' PCJIUS94/1 283~;
~WO9~/13294 2 ~ 7 6 ~ 1 4 2 IcrM secretion (nc~/ml 3 3Medium < lO0 4~-dex < 100 5~-dex+IL-l+IL-2 450 6 ~-dex+IL-l+IL-2+T-BNF 4000 7 ~2~-dex+IL-l+IL-2+IL-6 lU/ml 380 8 lO U/ml 320 g 100 U/ml 550 1000 U/ml 400 11 10, 000 U/ml 430 12 Table 5 . B'P cells were cultured at 1. 5 x 105 cells/ml 13 with cr~-dex (3 ng/ml), IL-l (150 U/ml), in the a~1sence 14 or presence of various concentrations of rIL-6 ( 1-10,000 U/ml) and/or T-BM~ (1.0% v/v [24 hr SN from 16 ~nti-CD3-activated RA5 Thl clone] ) . sic days after 17 initiation c)f culture, SN was harvested for 18 determination of secreted IgM Cv~ L-tions }~y ELISA.

Claims (15)

1. WHAT WE CLAIM IS:
   7. A cell-free preparation of a novel active substance whose activity is derived from T cells or NK
   cells which potently induce antibody secretion by activated animal B cells, said substance having Ig secretion activity which is not mimicked by the addition of cytokine in the absence of said cell-free preparation .
2. 2. The cell-free preparation of Claim 1 wherein the active substance is produced by NK cells.
3. 3. The cell-free preparation of Claim 1 wherein the active substance is resistant to neutralization by monoclonal antibodies against cytokines produced by NK cells activated by IL-2 and has Ig secretion activity which is not mimicked by the addition of NK cell cytokines in the absence of- said cell-free preparation.
4. 4. The cell-free preparation of Claim 1 wherein the active substance is a protein whose activity is derived from T cells which potently induce antibody secretion by activated animal B cells, said substance:
   
   a. is resistant to neutralization by monoclonal antibodies or receptor antagonists against cytokines produced by activated T cells;
   b. has Ig secretion activity which is not mimicked by the addition of T cell cytokines in the absence of said cell-free preparation;
   
   c. can be produced by activation of the T cell through the T cell receptor signalling pathway which induces cytokine production by T cells;
   d. has activity which is inhibited by treatment with a proteolytic enzyme;
   e. has activity which is optimal within 3 to 5 hours after activation of the T cell;
   f. has detectable activity at dilution levels as low as 1:3000; and g. operates as a late acting factor required to induce proliferated B cells to mature into antibody secreting cells.
5. 5. The cell-free preparation of Claim 4 wherein the active proteinaceous substance is a cytokine .
6. 6. The cell-free preparation of Claim 5 which is produced by T cells.
7. 7. The cell-free preparation of Claim 1 which is free of mycoplasm.
8. 8. The cell-free preparation of Claim 6 wherein the proteinaceous substance can increase the induction of antibody secretion 10 to 40 fold.
9. 9. The cell-free preparation of Claim 3 wherein the active substance can increase the induction of antibody secretion 10 to 30 fold.
10. 10. The cell-free preparation of Claim 1 wherein the active substance whose Ig secretion activity is not mimicked by the addition of T cell or NR cell cytokines comprising Interleukin-3;
    Interleukin-4; Interleukin-6; Interleukin-10; TNF-.alpha.;
    IFN-.gamma.; IFN-.gamma. and TNF-.alpha.; IFN-.alpha.; IFN-.alpha. and IFN-.gamma.; IFN-.alpha.
    and TNF-.alpha.; IFN-.alpha., IFN-.gamma. and TNF-.alpha.; GM-CSF; and TGF-.beta.1.
11. 11. The cell-free preparation of Claim 8 wherein the proteinaceous substance is resistant to neutralization by monoclonal antibodies or receptor antagonists against cytokines comprising anti-IFN-.gamma.;
    TNFRIg; anti-GN-CSF; anti-IL-3; anti-IL-4; anti-IL-5;
    anti-IL-6; anti-IL-10; and huCD40Fc.
12. 12. The cell-free preparation of Claim 8 wherein the proteinaceous substance is produced by activation of the T cell through the T cell receptor pathway which induces cytokine production by T cells as shown in Figure 12.
13. 13. The cell-free preparation of claim 9 wherein the active substance is resistant to neutralization by monoclonal antibodies against cytokines comprising anti-TNF-.alpha.; Anti-IFN-.gamma.; anti-IFN.alpha./.gamma.; anti-IL-4; anti-IL-5; anti-IL-6; and anti-IL-10.
14. 14. A method of producing a cell-free preparation containing the active proteninaceous substance of claim 4, comprising:
    a. coating the bottom of a flask with anti-CD3 antibody;
    b. adding T cells into the flask and allow sand T cells to remain in the flask for a period of 24 hours;
    c. harvesting the cell-free supernatant;
    d. inducing the activation and proliferation of B cells through the B cell antigen receptor mediated signalling pathway utilizing dextran conjugated anti-IgD antibodies in the presence of Interleukin-1 and Interleukin-2 wherein said B cells are at least 98.5 - 49 -percent purified through fluorescence-activated cell sorting (FACS);
    e. adding activated T cell supernatant;
    f. harvesting the B cell culture supernatant 6 days after culture; and g. measuring the concentration of secreted IgM in the B cell cultured supernatant by an enzyme-linked Immunosorbent Assay (ELISA).
    
    -49-a
15. 15. The product produced in accordance with the method of Claim 14.