expression of class II genes must be carefully 
regulated. 
To dissect this complex regulation, Dr. Peterlin 
first analyzed cis-acting sequences in a number of 
class II genes. This resulted in functional delinea- 
tion of transcriptional enhancer elements and con- 
served upstream and downstream promoter se- 
quences. Class II transcriptional enhancers are 
lymphoid specific. Upstream promoter elements are 
responsible for B cell-specific and interferon-7 
(IFN-7)-inducible expression, while downstream 
promoter elements position the site of initiation of 
DRA transcription. Identical cis-acting sequences 
are responsible for B cell-specific and IFN-7-induc- 
ible regulation. By creating a cassette-forming syn- 
thetic DRA promoter, Dr. Peterlin was able to multi- 
ply certain conserved elements and to exchange 
promoter elements between class II genes. Subtle 
differences in transcriptional regulation of DR, DP, 
and DQ genes, as well as defects in the expression of 
DOB, DQA2, and DQB2 genes, were revealed. 
Next, Dr. Peterlin turned his attention to trans- 
acting factors that interact with conserved upstream 
promoter sequences. These are composed of reiter- 
ated motifs. For example, consensus activator pro- 
tein- 1 (AP-1) and cAMP-responsive element (CRE) 
binding sites flank those for helix-loop-helix (HLH) 
proteins — regulatory factor-X (RF-X) and Ephrussi 
factors 12 and 47 (E12/E47). Not only are these 
proteins modified post-transcriptionally, but they 
interact with one another to direct tissue-specific 
and IFN-7-inducible regulation of class II genes. 
Moreover, RF-X is translated from two different me- 
thionine codons. Functions of full-length and trun- 
cated RF-X are under investigation. 
Other aspects of this complex transcriptional reg- 
ulation were revealed. For example, AP-1 , which is 
present in cells that do not express class II determi- 
nants, negatively regulates class II transcription. In 
B cells, c-Jun does not form heterodimers with 
c-Fos, and AP I does not bind to DNA. However, 
following B cell activation, c-Fos forms hetero- 
dimers with CRE-BPl , and this new heterodimer in- 
creases levels of class II transcription. In contrast, 
EI 2/E47 and RF-X are positive regulators in B cells. 
Moreover, E12/E47 might regulate the developmen- 
tal expression of class II genes. 
Finally, Dr. Peterlin studies genetic defects in BLS 
II where transcriptional regulators are congenitally 
absent. Minimal class II promoter elements and iso- 
lated trans-acting factors have facilitated these analy- 
ses. First, interactions between E12/E47 and RF-X 
are nonfunctional in BLS II cells. Second, since 
both of these proteins are present, either post- 
transcriptional processes or their coactivators are 
mutated or missing in this disease. Using biochemi- 
cal techniques and genetic approaches. Dr. Peterlin 
hopes to characterize these defective proteins and 
to isolate their gene(s). (The project described 
above is supported by a grant from the National Insti- 
tutes of Health.) 
Regulation of HIV Gene Expression 
HIV, the cause of acquired immune deficiency 
syndrome (AIDS) , requires the expression of trans- 
activators Tat and Rev for efficient viral replication 
and gene expression. Dr. Peterlin has been studying 
steps leading to the activation of HIV transcription 
and the mechanism of action of the virally encoded 
Tat protein. 
HIV transcription is initiated by the release of nu- 
clear factor kB (NF-kB) and of nuclear factor of acti- 
vated T cells (NEAT) from the cytoplasm to the nu- 
cleus of infected cells. These trans-acting factors 
interact with HIV enhancer sequences and result in 
increased loading of RNA polymerase II. For the ac- 
tivation of transcription, p50, which is the DNA- 
binding subunit of NF-/cB, interacts with p65. Re- 
cently Dr. Peterlin observed that the tyrosine kinase 
pathway plays an important role in the cytoplasmic- 
to-nuclear translocation of NF-kB. In particular, in T 
cells that lack a receptor tyrosine phosphatase 
(CD45), NE-zcB is constitutive ly active in the 
nucleus. 
However, although increased rates of initiation of 
HIV transcription are observed with NF-kB and 
NFAT, assembled transcription complexes do not 
elongate efficiently through viral coding sequences. 
For that. Tat has to interact with its trans-acting re- 
sponsive region (TAR). Tat binds to the TAR RNA 
stem-loop with the help of cellular RNA-binding 
proteins. In rodent cells, where lower levels of 
trans-activation are observed, these cellular RNA- 
binding proteins are missing. 
Further evidence for these conclusions came from 
experiments where Tat was fused to heterologous 
RNA-binding proteins — for example, the coat pro- 
tein of bacteriophage MS2 and the RNA-binding do- 
main of Tat of the equine infectious anemia virus 
(ElAV) . TAR was replaced by appropriate RNA tar- 
gets, the MS2 operator and TAR of ELAV. These chi- 
meric Tats functioned equivalently in human and 
rodent cells. 
Hybrid trans-activators also led to extensive map- 
ping of activation and Tat RNA-binding domains of 
HIV-1, HIV-2, and EIAV. In addition, a minimal len- 
tiviral TAT of only 25 amino acids was constructed, 
where 15 amino-terminal and 10 carboxyl-terminal 
residues represent activation and RNA-binding do- 
mains, respectively. Structural studies of this mini- 
IMMUNOLOGY 355 
