Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!husc6!hao!ames!necntc!linus!philabs!aecom!diaz
From: diaz@aecom.YU.EDU (Dizzy Dan Diaz)
Newsgroups: sci.bio
Subject: Protein Domains Symposium -- A Summary  (PART 1 of 2)
Message-ID: <1422@aecom.YU.EDU>
Date: Mon, 9-Nov-87 12:54:55 EST
Article-I.D.: aecom.1422
Posted: Mon Nov  9 12:54:55 1987
Date-Received: Fri, 13-Nov-87 05:39:28 EST
Organization: Graduate School of Hard Knocks
Lines: 86
Keywords: protein structure, molecular biology, molecular recognition


The following briefly summarizes some of the highlights of the recent
PROTEIN DOMAINS symposium held 2-3 November at the Waksman Institute for
Microbiology at Rutgers University.

FIRST DAY

Alan A.R. Fersht
Imperial College of Science & Technology, London UK
"Protein Engineering of Enzyme Structure and Activity"

Fersht spoke on his work with Bacillus stearothermophilus Tyr tRNA
synthetase.  An interesting enzyme because although it is composed
of two identical subunits (a homodimer), only one of the two active
sites appears to be catalytic under physiological conditions.  This is
called "half the sites" activity and is apparently observed in some
other homodimers.  The crystal structure of the enzyme reveals that one
domain is so mobile that the structure has not been determined.  This is
strong evidence against the criticism of many that crystals are totally
static structures and can therefore tell us little about solution
structures.  Adenylate is H-bonded by 8 sidechains of the enzyme, as
demonstrated by crystal structure and site-directed mutagenesis, that
beloved new baby of structural biochemists.  Fersht stated very directly
that the adenylate is not distorted by the enzyme, and that in general
he does not believe in enzyme distortion of substrates.  He was also
very emphatic about using mutagenesis only after one has the crystal
structure, a view I considered narrow.

Adrian Goldman
Waksman Institute, Rutgers University
"Deconstruction:  The Relationship of Protein Folding to 
                  Literary Theory"

The protein folding problem has asked "given a sequence, what will its
structure be?"  Deconstruction asks "given a structure, what will be its
sequence?"  The structures of a number of alpha, beta barrel structures
were compared in an effort to see whether the sequences followed the
patterns of the structures.  Muconate lactinizing enzyme (MLE) was
compared to triose phosphate isomerase (some call me TIM) and pyruvate
kinase (why are phosphorylating enzymes called kinases?). 

Goldman concluded that because the structural similarities have no
underlying sequence similarities, there is no homology.  He concluded
that these very similar structures have arisen as a result of convergent
evolution.

Don Wiley
Harvard University
"A 3D Structure for Influenza Virus-Cell Receptor Interaction"

Influenza virus hemagglutinin is a very pretty trimer about 135 A tall.
It apparently binds sialic acid in the membranes of the cellular host.
The sialic acid binding pocket lies in a crater, and is composed of
amino acids conserved in over 50 strains of virus.  The amino acids
surrounding the crater are quite variable; mutating at a very rapid
rate.  It is these mutating aas that are recognized by neutralizing
antibodies.  Their rapid rate of mutation makes antibodies for one flu
strain ineffective against others. 

John Rosenberg
University of Pittsburgh
"Structure and Recognition Mechanism of the EcoRI Endonuclease"

I think most of us have seen the pretty pictures of the EcoRI homodimer
surrounding a DNA duplex containing the cannonical recognition sequence,
published in Nature.  A recent TIBS review is also fun to read.

EcoRI binds to and hydrolyzes DNA of its cannonical sequence 10^7 times
more than non-cannonical sequence.  Despite its low Vmax, therefore, its
incredible Km makes it as good as any of the "lightning" enzymes.
Specificity is the result of the number of H-bonds between enzymic
sidechains and major groove nucleotide bases.  Crystals are
enzymatically active, as shown by the hydrolysis of co-crystallized DNA
when Mg++ is diffused into co-crystals.  This is more evidence that
protein crystals are not static.

Rosenberg has annoyed some by not releasing the coordinates of the EcoRI
structure to the PDB at Brookhaven.  If Dan Davison had his way, people
like Rosenberg wouldn't get any funding unless they released the
coordinates of their structures (right Dan?).  I don't know if it would
work, but just discussing the possibility would likely scare people like
Rosenberg, Alex Rich and others.

-- 
      dn/dx      Dept Molecular Biology   diaz@aecom.yu.edu
     Dan Diaz      Albert Finkelstein College of Medicine