TRANCE (tumor necrosis factor [TNF]–related activation-induced cytokine) is a new member of the TNF family that is induced upon T cell receptor engagement and activates c-Jun N-terminal kinase (JNK) after interaction with its putative receptor (TRANCE-R). In addition, TRANCE expression is restricted to lymphoid organs and T cells. Here, we show that high levels of TRANCE-R are detected on mature dendritic cells (DCs) but not on freshly isolated B cells, T cells, or macrophages. Signaling by TRANCE-R appears to be dependent on TNF receptor–associated factor 2 (TRAF2), since JNK induction is impaired in cells from transgenic mice overexpressing a dominant negative TRAF2 protein. TRANCE inhibits apoptosis of mouse bone marrow–derived DCs and human monocyte-derived DCs in vitro. The resulting increase in DC survival is accompanied by a proportional increase in DC-mediated T cell proliferation in a mixed leukocyte reaction. TRANCE upregulates Bcl-xL expression, suggesting a potential mechanism for enhanced DC survival. TRANCE does not induce the proliferation of or increase the survival of T or B cells. Therefore, TRANCE is a new DC-restricted survival factor that mediates T cell–DC communication and may provide a tool to selectively enhance DC activity.

Apoptosis plays a critical role in the development and maintenance of the immune system (13). Members of the TNF family can regulate apoptosis in addition to an array of other biological effects such as cell proliferation and differentiation (4). Despite the functional redundancy of this family, specificity may be accomplished by coordinating the spatial and temporal expression of TNF-related ligands and their receptors and by restricting the expression of signal transduction molecules to specific cell types. TNF receptors interact with a family of molecules called TRAFs (TNF receptor–associated factors) that act as adaptors for downstream signaling events (5). For example, TRAF2 activates NF-κB (6) and also c-Jun NH2-terminal kinase (JNK; references 79). The biochemical events leading to apoptosis involve the caspase family of cysteine proteases (10), whereas NF-κB appears to inhibit cell death (11). The TNF receptor family can also regulate apoptosis by modulating the expression of Bcl-2 and Bcl-2–related proteins (12, 13). Recent data indicates that the Bcl-2 family controls apoptosis by altering transmembrane conductance in mitochondria and by preventing the activation of caspases (1416).

An important role of TNF members in dendritic cell (DC) biology has recently emerged. DCs have several specializations that lead to the stimulation of naive T cells and play a role in the initiation of the immune response (17). TNF-α and CD40 ligand (L) are molecules involved in the differentiation of DC from CD34+ bone marrow or cord blood progenitors (1820). Moreover, CD40L increases DC survival, upregulates MHC and costimulatory molecule expression, and induces the expression of a variety of cytokines (e.g., IL-12) in DCs (21). Both CD40 and TNFR interact with TRAF2, suggesting that TRAF2 plays a role in DC function.

Recently, TRANCE (TNF-related activation-induced cytokine), a novel ligand of the TNF family, was cloned during a search for apoptosis regulatory genes (22). Remarkably, TRANCE expression is restricted to lymphoid-specific organs and is selectively expressed in T cells (22). In this study, we describe that TRANCE-R signals via TRAF2 in thymocytes and increases DC survival by upregulating Bcl-xL expression, a property shared with CD40L. However, unlike CD40L, TRANCE selectively acts on mature DCs but not on B cells. In addition, high levels of the TRANCE-R are only detected on DCs, suggesting that a major function of TRANCE in vivo is to modulate DC activity.

Materials And Methods

Expression and Purification of Soluble TRANCE.

A FLAG epitope-tagged TRANCE molecule (FLAG-TRANCE) was expressed in 293T cells and purified as previously described (22). To create a human CD8-TRANCE recombinant molecule (hCD8-TRANCE), the extracellular domain of murine TRANCE (amino acid 245– 316) was fused to human CD8α (amino acid 1–182) and produced in a baculovirus expression system according to the manufacturer's instructions (BaculoGold; PharMingen, San Diego, CA). hCD8-TRANCE was purified on cyanogen bromide (CNBR)- activated Sepharose gel conjugated to OKT8 following the manufacturer's protocol (Pharmacia Biotech, Piscataway, NJ). mCD8-CD40L in insect cell culture supernatant was provided by Dr. Randolph J. Noelle (Dartmouth Medical School, Hanover, NH).

Mice.

C57BL/6 (H-2b) and BALB/c (H-2d) mice were from Taconic Farms (Germantown, NY). Transgenic mice expressing a dominant negative form of TRAF2 (TRAF2.DN) were engineered as previously described (23).

Cells.

Bone marrow–derived DCs (BMDC) were generated as previously described (24) and were used on day 8 of culture. Enriched populations of fresh lymph node or splenic DCs were prepared by digesting organs with collagenase then selecting for low density cells via centrifugation on a Nycodenz column (14.5% wt/vol in PBS + 5 mM EDTA; Nycomed Pharmaceuticals, Oslo, Norway) for 15 min at 4°C. Mature spleen DCs were prepared by culturing freshly isolated spleen DCs overnight as previously described (25). The cytokine-induced generation of human DCs from PBMCs was performed as previously described (26). After 2 d in monocyte-conditioned medium, TRANCE or PBS was added to the DCs. Lymph node T cells (99% CD3+ as assessed by flow cytometry) were prepared by magnetic bead depletion (Dynal, Oslo, Norway) of class II, B220, NK1.1, and F4/80 positive cells. B cells were prepared by magnetic depletion of Thy1.2 positive cells (Dynal). Cell viability was assayed by trypan blue exclusion or by propidium iodide uptake.

Flow Cytometry.

DC phenotype was assessed by flow cytometry as previously described (27) using the following FITC- or PE-conjugated mAbs: H-2Kb, I-Ab, intracellular adhesion molecule (ICAM)-1, CD11b, CD11c, CD80, CD86, CD25, and CD40 (all from PharMingen). Other mAbs used were biotinylated α-Fas, CD3-FITC, B220-FITC (PharMingen), and NLDC-145–FITC. The expression of TRANCE-R was assessed using the hCD8-TRANCE fusion molecule at 10 μg/ml at 4°C followed by biotinylated OKT8 mAb and then streptavidin-PE (BioSource International, Camarillo, CA). Negative controls were performed by omitting hCD8-TRANCE. For analysis of TRANCE-R expression on resting B cells and fresh DCs, low density cells were stained with FITC-B220 or FITC-CD11c, respectively, and analyzed on a FACScan® (Becton Dickinson, Mountain View, CA).

Mixed Leukocyte Reaction.

BMDCs treated for 48 h in the presence or absence of recombinant TRANCE were cultured with 105 purified allogeneic T cells in flat-bottomed 96-well plates in a final volume of 200 μl for 3 d and then pulsed for 8 h with 0.5 μCi of [3H]thymidine (Dupont-NEN, Boston, MA). The cells were then harvested on glass fiber filters and [3H]thymidine incorporation was measured using a standard scintillation-detection procedure.

JNK Assays.

2–5 × 106 cells from TRAF2.DN transgenic mice or from control littermates were incubated for 1–2 h at 37°C on plates coated with OKT8 antibody (10 μg/ml). The cells were treated with either soluble TRANCE or an equal volume of PBS before being harvested at the indicated time points and frozen in a dry ice/ethanol bath. JNK was immunoprecipitated with α-JNK1 antibody (Santa Cruz Biotechnology, Santa Cruz, CA) and kinase activity was assessed as previously described (22).

Western Blot and Reverse Transcriptase–PCR Analysis of Bcl-xL and Bcl-2.

BMDCs (8 × 106/well) were cultured in RPMI in 6-well plates and treated with PBS, FLAG-TRANCE (1 μg/ml), or soluble CD40L for 0 or 24 h. The cells were lysed, and 50 μg of protein from each sample was resolved on a 12% SDS-PAGE gel and transferred to Immobilon-P membranes (Millipore, Bedford, MA). The blots were blocked in 5% skim milk, probed with α-Bcl-2 (4C11) or α-Bcl-xL (236; both provided by Dr. Gabriel Núñez, University of Michigan, Ann Arbor, MI) and detected with the appropriate horseradish peroxidase–conjugated secondary antibodies and enhanced chemiluminesence substrate (ECL; Amersham Corp., Arlington Heights, IL). For reverse transcriptase– PCR analysis of bcl-xL mRNA expression, BMDCs (2 × 106 cells/well) were cultured in 24-well plates, treated with the appropriate reagents, and quickly frozen in a dry ice/ethanol bath at the various time points. Total RNA was extracted (RNEasy; QIAGEN Inc., Chatsworth, CA), and cDNA was diluted to allow PCR amplification to occur as a linear function of starting concentrations. PCR was performed using the conditions and primers as previously described (13).

Results And Discussion

TRANCE-R Is Expressed at High Levels in DCs.

To identify cells that express TRANCE-R, hCD8-TRANCE was used as a molecular probe for FACS® analysis. TRANCE-R was detected on mature BMDCs, freshly isolated lymph node DCs, and freshly isolated spleen DCs (Fig. 1). TRANCE-R was greatly upregulated upon the maturation of spleen DCs induced by overnight culture. No expression could be detected on freshly isolated lymph node B cells, lymph node T cells, thymocytes, or peritoneal macrophages. Therefore, the highest levels of TRANCE-R expression are found on mature DCs and suggest that the major role of TRANCE is restricted to DCs.

TRANCE Is a DC Survival Factor.

The biological effects of TRANCE were further studied on mature DCs. TRANCE-treated DCs formed densely packed clusters whereas control, untreated cells exhibited relatively sparse aggregates (Fig. 2,A). In addition, mature BMDCs treated with FLAG-TRANCE were significantly protected from spontaneous cell death compared to untreated cells. This effect was dependent on the dose of TRANCE (Fig. 2,B). hCD8-TRANCE elicited similar results (data not shown). This effect was not due to increased cell proliferation since the total number of cells remained the same over time (data not shown). TRANCE significantly prevented DC cell death until day 6, whereas untreated cells were almost completely dead by day 3 (Fig. 2,C). A similar effect on DC survival was observed with human monocyte-derived DC (Fig. 2,D). Confirming previous data, CD40L also induced the clustering of DCs (data not shown; 28, 29) and enhanced DC survival comparably to TRANCE (Fig. 2 C).

CD40L upregulates the antiapoptotic molecule, Bcl-xL, in B cells and protects them from Ig receptor–mediated cell death (13). In addition, CD40L upregulates Bcl-2 in human DC derived from CD34+ progenitor cells, a phenomenon that was correlated with a resistance to Fas-mediated apoptosis (12). To determine whether TRANCE can influence Bcl-2 or Bcl-xL, we measured their expression in DCs stimulated with TRANCE or CD40L by Western blot analysis. BMDCs expressed relatively high levels of Bcl-2 and relatively low levels of Bcl-xL after reaching maturity in GM-CSF (Fig. 2,E, 0 h). FLAG-TRANCE and CD40L stimulation lead to increased Bcl-xL expression by 24 h. Bcl-xL expression was nearly absent in cells treated with medium alone. bcl-xL mRNA was upregulated in TRANCE-treated DCs, suggesting a transcriptional as opposed to posttranscriptional regulation (data not shown). In contrast, Bcl-2 levels were decreased in both the TRANCE-treated and untreated cells (Fig. 2 E). These results suggest that TRANCE, in addition to CD40L, upregulates Bcl-xL in DCs, which enhances their viability in vitro.

TRANCE Enhances DC-mediated T Cell Proliferation.

To examine the functional consequences of TRANCE on DCs we measured the MLR-stimulating ability of DCs treated with TRANCE. Increasing doses of FLAG-TRANCE enhanced DC survival at 48 h, which in turn led to a proportional increase in the stimulation of T cell proliferation (Fig. 3,A). When equivalent numbers of viable TRANCE-treated or untreated DCs were used in an MLR, there were no differences in T cell proliferation, suggesting that changes in the expression of costimulatory and antigen-presenting molecules did not account for the enhanced T cell proliferation (Fig. 3,B). To verify this, the levels of several surface markers were tested by FACS® to evaluate any TRANCE-mediated changes to the DC phenotype. There was a slight but reproducible downregulation of MHC class II expression and a slight upregulation of MHC class I expression (Fig. 3 C). There were no TRANCE-mediated perturbations in the expression of the costimulatory molecules CD80 (B7-1) or CD86 (B7-2), and no changes in the expression of the adhesion molecules intracellular adhesion molecule (ICAM)–1, CD11b, and CD11c. Interestingly, CD40 expression increased but Fas and TRANCE-R did not. In sum, TRANCE enhances DC-mediated T cell proliferation by increasing the survival of DCs.

TRANCE Does Not Affect B or T Cell Proliferation.

Expression of high levels of the TRANCE-R appeared restricted to DCs by FACS® analysis. However, we found that TRANCE could activate JNK in thymocytes (22), suggesting that FACS® analysis might lack the sensitivity to detect low levels of receptor. To further examine the specificity of TRANCE for DCs, we tested its ability to induce B cell proliferation or survival, two functions mediated by CD40L. Recombinant hCD8-TRANCE, tested for its antiapoptotic function in BMDCs, could not stimulate B cell proliferation (Fig. 4), nor could it activate JNK activation (22). In contrast, CD40L efficiently stimulated B cell proliferation in a dose-dependent manner (Fig. 4). Finally, TRANCE could not prevent the spontaneous apoptosis of B and T cells as assessed by propidium iodide uptake (data not shown). Therefore, functionally, TRANCE appears to exhibit different cellular specificities and functions when compared to CD40L.

TRANCE-mediated JNK Induction Requires Functional TRAF2.

Recruitment of TRAF2 to the TNFR complex or the CD40 receptor complex is necessary for JNK activation (79, 23). To test the possibility that TRANCE-R also signals via TRAF2, we analyzed TRANCE-mediated JNK activation in thymocytes from transgenic mice overexpressing a dominant negative form of TRAF2 (TRAF2. DN; reference 23). JNK activity peaked 2.5-fold over unstimulated cells at 5 min in control littermates, whereas JNK induction was significantly reduced in TRAF2.DN thymocytes (Fig. 5). These results suggest that signaling from the TRANCE-R requires TRAF2. TRANCE-mediated JNK induction in DCs could not be assayed since TRAF2.DN expression has been restricted to lymphocytes in the TRAF2.DN transgenic mice. In addition, JNK activity was constitutively high in mature DCs (22), which are also known to have high levels of activated NF-κB (30), thus confounding detection of increased JNK activity.

In summary, we have shown that TRANCE, in addition to CD40L, is a regulator of DC function. Similar to CD40L, TRANCE promotes the survival of mature DCs by regulating the expression of Bcl-xL. However, in contrast to CD40L, TRANCE does not act on other APCs such as B cells. The signal transduction pathways via TRANCE-R in DCs are unknown. TRANCE appears to signal via TRAF2, at least in thymocytes, suggesting that TRAF2 may play a critical role in mediating signals for differentiation, activation, and survival in DCs.

These findings complement our previous report describing the selective expression of this new TNF family member in T cells. The high level of expression of TRANCE-R on DCs suggests a specific role for TRANCE in T cell–DC communication during the primary immune response. Rapid upregulation of TRANCE upon TCR engagement on T cells (22) could specifically enhance the survival of DCs during antigen presentation. Both antigen-specific T cells and the antigen-presenting DCs would therefore depend on each other for activation and survival. Mature DCs that fail to present antigen to T cells would not receive T cell help and would therefore die of neglect. This T cell–DC interaction is likely to occur in the T cell area of lymphoid organs that contain DCs of mature phenotypes (31). DCs can only be detected in afferent lymph, not efferent lymph, suggesting that DCs are destined to die when they migrate to the lymph node. TRANCE may be important to maintain DC survival, perhaps acting before CD40L as TRANCE is an immediate early gene (22). Many experiments indicate that DCs pulsed ex vivo with antigen can be used to induce immunity to tumor or viral antigens in vivo (32). TRANCE could provide a tool to specifically enhance DC function by enhancing their survival in vivo. This hypothesis is currently being examined in a variety of viral and tumor models in mice.

Acknowledgments

We would like to thank Angela Santana and Yaneth Castellanos for their technical help.

This work was supported in part by the National Institutes of Health (NIH) MSTP grant GM-07739 (B.R. Wong) and NIH grants CA-525133 (Y. Choi), AI-13013 (R.M. Steinman), and AI-39672 (R.M. Steinman). Y. Choi is an assistant investigator of the Howard Hughes Medical Institute. R. Josien is supported by the Association pour la Recherche contre le Cancer. B. Sauter is supported by the Deutsche Forschungsgemeinschaft.

References

1
Wyllie
AH
Apoptosis and the regulation of cell numbers in normal and neoplastic tissues: an overview
Cancer Metastasis Rev
1992
11
95
103
[PubMed]
2
Russell
JH
Activation-induced death of mature T cells in the regulation of immune responses
Curr Opin Immunol
1995
7
382
388
[PubMed]
3
Lynch
DH
,
Ramsdell
F
,
Alderson
MR
Fas and FasL in the homeostatic regulation of immune responses
Immunol Today
1995
16
569
574
[PubMed]
4
Smith
CA
,
Farrah
T
,
Goodwin
RG
The TNF receptor superfamily of cellular and viral proteins: activation, costimulation, and death
Cell
1994
76
959
962
[PubMed]
5
Rothe
M
,
Wong
SC
,
Henzel
WJ
,
Goeddel
DV
A novel family of putative signal transducers associated with the cytoplasmic domain of the 75 kDa tumor necrosis factor receptor
Cell
1994
78
681
692
[PubMed]
6
Rothe
M
,
Sarma
V
,
Dixit
VM
,
Goeddel
DV
TRAF2-mediated activation of NF-κB by TNF receptor 2 and CD40
Science
1995
269
1424
1427
[PubMed]
7
Natoli
G
,
Costanzo
A
,
Ianni
A
,
Templeton
DJ
,
Woodgett
JR
,
Balsano
C
,
Levrero
M
Activation of SAPK/JNK by TNF receptor 1 through a noncytotoxic TRAF2-dependent pathway
Science
1997
275
200
203
[PubMed]
8
Reinhard
C
,
Shamoon
B
,
Shyamala
V
,
Williams
LT
Tumor necrosis factor α–induced activation of c-jun N-terminal kinase is mediated by TRAF2
EMBO (Eur Mol Biol Organ) J
1997
16
1080
1092
[PubMed]
9
Liu
ZG
,
Hsu
H
,
Goeddel
DV
,
Karin
M
Dissection of TNF receptor 1 effector functions: JNK activation is not linked to apoptosis while NF-κB activation prevents cell death
Cell
1996
87
565
576
[PubMed]
10
Darnay
BG
,
Aggarwal
BB
Early events in TNF signaling: a story of associations and dissociations
J Leukocyte Biol
1997
61
559
566
[PubMed]
11
Baeuerle
PA
,
Baltimore
D
NF-κB: Ten years after
Cell
1996
87
13
20
[PubMed]
12
Bjorck
P
,
Banchereau
J
,
Flores
RL
CD40 ligation counteracts Fas-induced apoptosis of human dendritic cells
Int Immunol
1997
9
365
372
[PubMed]
13
Wang
Z
,
Karras
JG
,
Howard
RG
,
Rothstein
TL
Induction of bcl-x by CD40 engagement rescues sIg-induced apoptosis in murine B cells
J Immunol
1995
155
3722
3725
[PubMed]
14
Minn
AJ
,
Velez
P
,
Schendel
SL
,
Liang
H
,
Muchmore
SW
,
Fesik
SW
,
Fill
M
,
Thompson
CB
Bcl-xLforms an ion channel in synthetic lipid membranes
Nature
1997
385
353
357
[PubMed]
15
Chinnaiyan
AM
,
O'Rourke
K
,
Lane
BR
,
Dixit
VM
Interaction of CED-4 with CED-3 and CED-9: a molecular framework for cell death
Science
1997
275
1122
1126
[PubMed]
16
Chinnaiyan
AM
,
Orth
K
,
O'Rourke
K
,
Duan
H
,
Poirier
GG
,
Dixit
VM
Molecular ordering of the cell death pathway. Bcl-2 and Bcl-xLfunction upstream of the CED-3–like apoptotic proteases
J Biol Chem
1996
271
4573
4576
[PubMed]
17
Steinman
RM
The dendritic cell system and its role in immunogenicity
Annu Rev Immunol
1991
9
271
296
[PubMed]
18
Flores-Romo
L
,
Björck
P
,
Duvert
V
,
van Kooten
C
,
Saeland
S
,
Banchereau
J
CD40 ligation on human cord blood CD34+hematopoietic progenitors induces their proliferation and differentiation into functional dendritic cells
J Exp Med
1997
185
341
349
[PubMed]
19
Caux
C
,
Vanbervliet
B
,
Massacrier
C
,
Dezutter-Dambuyant
C
,
de Saint-Vis
B
,
Jacquet
C
,
Yoneda
K
,
Imamura
S
,
Schmitt
D
,
Banchereau
J
CD34+hematopoietic progenitors from human cord blood differentiate along two independent dendritic cell pathways in response to GM-CSF+ TNFα
J Exp Med
1996
184
695
706
[PubMed]
20
Young
JW
,
Szabolcs
P
,
Moore
MA
Identification of dendritic cell colony–forming units among normal human CD34+bone marrow progenitors that are expanded by c-kit-ligand and yield pure dendritic cell colonies in the presence of granulocyte/macrophage colony–stimulating factor and tumor necrosis factor α
J Exp Med
1995
182
1111
1119
[PubMed]
21
van Kooten
C
,
Banchereau
J
Functions of CD40 on B cells, dendritic cells and other cells
Curr Opin Immunol
1997
3
330
337
[PubMed]
22
Wong
B
,
Rho
J
,
Arron
J
,
Robinson
E
,
Orlinick
J
,
Chao
M
,
Kalachikov
S
,
Cayani
E
,
Bartlett
FS
III
,
Frankel
W
et al
TRANCE is a novel ligand of the tumor necrosis factor receptor family that activates c-Jun N-terminal kinase in T cells
J Biol Chem
1997
372
25190
25194
[PubMed]
23
Lee, S., A. Reichlin, A. Santana, K. Sokol, M. Nussenzweig, and Y. Choi. 1997. TRAF2 is essential for JNK but not NF-κB activation and regulates lymphocyte proliferation and survival. Immunity. In press.
24
Inaba
K
,
Inaba
M
,
Romani
N
,
Aya
H
,
Deguchi
M
,
Ikehara
S
,
Muramatsu
S
,
Steinman
RM
Generation of large numbers of dendritic cells from mouse bone marrow cultures supplemented with granulocyte/macrophage colony–stimulating factor
J Exp Med
1992
176
1693
1702
[PubMed]
25
Steinman
RM
,
Kaplan
G
,
Witmer
MD
,
Cohn
ZA
Identification of a novel cell type in peripheral lymphoid organs of mice. V. Purification of spleen dendritic cells, new surface markers, and maintenance in vitro
J Exp Med
1979
149
1
16
[PubMed]
26
Bender
A
,
Sapp
M
,
Schuler
G
,
Steinman
RM
,
Bhardwaj
N
Improved methods for the generation of dendritic cells from nonproliferating progenitors in human blood
J Immunol Methods
1996
196
121
135
[PubMed]
27
Josien
R
,
Heslan
M
,
Soulillou
JP
,
Cuturi
MC
Rat spleen dendritic cells express natural killer cell receptor protein 1 (NKR-P1) and have cytotoxic activity to select targets via a Ca2+- dependent mechanism
J Exp Med
1997
186
467
472
[PubMed]
28
Caux
C
,
Massacrier
C
,
Vanbervliet
B
,
Dubois
B
,
Kootan
C
,
Durand
I
,
Banchereau
J
Activation of human dendritic cells through CD40 cross-linking
J Exp Med
1994
180
1263
1272
[PubMed]
29
Ludewig
B
,
Graf
D
,
Gelderblom
H
,
Becker
Y
,
Kroczek
R
,
Pauli
G
Spontaneous apoptosis of dendritic cells is inhibited by TRAP (CD40-ligand) and TNF-α, but strongly enhanced by interleukin-10
Eur J Immunol
1995
25
1943
1950
[PubMed]
30
Granelli
PA
,
Pope
M
,
Inaba
K
,
Steinman
RM
Coexpression of NF-κB/Rel and Sp1 transcription factors in human immunodeficiency virus 1–induced, dendritic cell– T-cell syncytia
Proc Natl Acad Sci USA
1995
92
10944
10948
[PubMed]
31
Steinman
RM
,
Pack
M
,
Inaba
K
Dendritic cells in the T-cell areas of lymphoid organs
Immunol Rev
1997
156
25
37
[PubMed]
32
Young
JW
,
Inaba
K
Dendritic cells as adjuvants for class I major histocompatibility complex–restricted antitumor immunity
J Exp Med
1996
183
7
11
[PubMed]

Brian R. Wong and Régis Josien contributed equally to this report.

Author notes

Address correspondence to Dr. Yongwon Choi, HHMI, The Rockefeller University, 1230 York Ave., Box 295, New York, NY 10021. Phone: 212-327-7441; FAX: 212-327-7319; E-mail: choi@rockvax.rockefeller.edu