TRANSCRIPTIONAL CONTROL MECHANISMS COMMON TO B CELLS AND MYELOID CELLS 
Thomas R. Kadesch, Ph.D., Associate Investigator 
During the past several years, Dr. Kadesch's labo- 
ratory has studied the immunoglobulin heavy-chain 
(IgH) enhancer as a cell type-specific transcrip- 
tional control element. The goal of this work is not 
simply to understand transcriptional control in B 
cells but to establish paradigms that may apply to 
other cell types and other gene systems. Results ob- 
tained over the past year have demonstrated that B 
cells and myeloid cells utilize similar mechanisms 
to regulate their differentiation programs and to 
control expression of their genes. 
Immunoglobulin Gene Transcription: 
the Heavy-Chain Enhancer 
The activity of the IgH enhancer is restricted to B 
cells, and this is dictated by both positive and nega- 
tive control mechanisms. Positive control stems 
from B cell-specific proteins, such as those repre- 
sented by the helix-loop-helix family of DNA- 
binding proteins, and from DNA-binding proteins 
ubiquitously expressed. Negative control, function- 
ing in non-B cells, assures that the activity of the 
ubiquitously expressed proteins is repressed. 
Previous work from Dr. Kadesch's laboratory indi- 
cated that repression of the enhancer in non-B cells 
is mediated through a particular site, designated 
mE5. That the presence of the site represses en- 
hancer activity can be measured through a neighbor- 
ing site, ^E3- The /xE3 site binds the ubiquitous tran- 
scription factor TFE3, which only in the absence of 
the ;uE5 site is sufficient to activate the enhancer in 
non-B cells. The hypothesis is that in non-B cells 
the /xE5 site binds a repressor that inhibits the activ- 
ity of proteins acting through that site. In B cells the 
mE5 site functions in an opposite manner to stimu- 
late enhancer activity by binding its own positive 
transcription factor. This factor is related or identi- 
cal to the protein E2-5. When artificially overex- 
pressed in non-B cells, E2-5 can displace the re- 
pressor and activate the enhancer. This raises the 
possibility that the activator and repressor exist in 
the same cell and that enhancer activity (on or off) is 
dictated by the ultimate outcome of their competi- 
tive mE5 binding. 
In an effort to understand enhancer regulation 
mediated through the jLtE5 site, a clone representing 
a candidate repressor protein has been isolated from 
a HeLa cell cDNA library. Isolated through its ability 
to bind the mE5 site, this cDNA encodes a zinc finger 
protein ([C2H2]3 type) , and the protein was accord- 
ingly designated Zeb (for zinc finger E 5 -binding 
protein). Zeb possesses two distinct regions con- 
taining zinc finger motifs. Each of these motifs is 
sufficient for DNA binding, and each apparently has 
the same DNA-binding specificity. The discovery of 
Zeb is encouraging, because all the AtE5-binding 
proteins previously identified (including E2-5) be- 
long to a distinct family of transcription factors that 
bind DNA through a basic-helix-loop-helix (bHLH) 
motif. The presence of the zinc fingers suggests that 
ZEB's mode of action might be different. 
The ixE5 site of the IgH enhancer is a member of a 
family of related sites, collectively referred to as E 
boxes. These are defined by the DNA sequence 
CANNTG and are found associated with a variety of 
genes expressed in cell type-specific manners. 
They generally bind bHLH proteins in vitro, such as 
E2-5, but only those of a subset bind Zeb. When 
these various E boxes are substituted for the tiE5 site 
in the IgH enhancer, only those that bind Zeb in 
vitro are able to confer repression in vivo. This is 
the strongest argument that Zeb actually encodes 
the repressor; additional support for that contention 
is still being pursued. 
The IgH enhancer is subject to an additional type 
of negative regulation. In this case inhibition is me- 
diated by a class of helix-loop-helix proteins that 
lack the ability to bind DNA and, by interacting with 
positive-acting bHLH proteins, render them also un- 
able to bind DNA. The first reported member of this 
class of proteins is the protein Id. Id is down- 
regulated during development in a number of dif- 
ferent cell lines, and it has been suggested that Id 
may be a general inhibitor of the differentiated state, 
by blocking the action of bHLH proteins specifically 
required for development (such as MyoD). Given 
that bHLH proteins play a role in activating the IgH 
enhancer, it was reasoned that Id may function simi- 
larly during B lymphoid development. In this sce- 
nario. Id would be expressed at high levels only in 
immature B cells, in which the enhancer has not yet 
been activated. 
Results from Dr. Kadesch's laboratory suggest that 
Id is indeed involved in repressing the IgH enhancer 
during early stages of B lymphoid development. In 
collaboration with Drs. Stephen Desiderio (HHMI, 
Johns Hopkins University) and Robert Benezra (Me- 
morial Sloan-Kettering Cancer Center), the labora- 
tory demonstrated that Id is not expressed in most B 
lymphoid cell lines, including a large array of early 
204 
