From: <complex-science@necsi.org>
Sender: <y3list1@necsi.org> (Yaneer Bar-Yam)
To: complex-science
Date: Wed, 09 Jul 2008 23:54:14 -0400
Message-ID: <redirect-22128257@necsi.org>
Subject: What is a gene?   A  dynamic &  triadic definition of a gene
User-Agent: SquirrelMail/1.4.13
MIME-Version: 1.0
Content-Type: text/plain;charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Importance: Normal

(Yaneer, if it is not too late, please replace my previous post with this
one.  Thanks.  Sung)

The most widely accepted definition of a gene during the past four decades
has been a stretch of DNA that codes for a protein. Although this simple
definition of a gene served well for the 20th-century molecular biology
and genetics, the new data that have been emerging since the mid-1990's
(when DNA microarrays were invented) have made the protein-centered
definition of a gene obsolete [1,2,3]. A new definition proposed by
Gerstein and his coworkers at Yale now includes as a gene those DNA
regions that code for RNA as well [2]:

     "A gene is a union of genomic sequences encoding a
      coherent set of potentially overlapping functional
      products."                                           . . . . . (1)

The important phrase here is "functional products", by which the authors
mean proteins and RNA molecules that are biologically active.

The new definition of a gene given in (1) was motivated by the recent
unexpected finding [1,3] that a large portion of the human genome (about
30% of the DNA mass), although not coding for any proteins, nevertheless
code for RNA molecules whose functions have not yet all been characterized.

There are two aspects to the definition of a gene given in (1) that I
believe require revisions:

  i)  It is too static, being based solely on gene "products", i.e.,
proteins and RNA, which are "equilibrium structures".  According to
Prigogine (917-2003)[4], there are two fundamental classes of
structures in nature -- equilibrium (e.g., rocks, chairs, DNA double
helix, nucleotide or amino acid sequences) and dissipative structures
(e.g., the flame of a candle, all sorts of gradients, action potentials,
gene expression profiles). One convenient way to distinguish dissipative
structures from equilibrium structures is to remember that, when energy
input is stopped, the former disappears but the latter remains.
For example, when a computer is turned off, the primary memory (a
dissipative structure) in CPU disappears but the secondary memory (an
equilibrium structure) in the hard disk remains.

  ii) It excludes those DNA regions that regulate gene expression (called
promoters, enhancers, silencers, etc.) without producing any proteins or
RNA. In other words, Gerstein et al's definition of a gene excludes 
"dissipative structures" which would include all regulatory processes in
the living cell. This is what Gerstein et al state [2]:

   "Although regulatory regions are important for gene
    expression, we suggest that they should not be
    considered in deciding whether multiple products
    belong to the same gene. . . . "      . . . . . . . . . . .  (2)

To remedy these perceived shortcomings, I suggest that the concept of
"dissipative structures" [4] be incorporated into the definition of a gene
itself. One way to do this is as follows:

    "A gene is a DISSIPATIVE STRUCTURE that embodies (or
     stores) not only genetic information (in the form of a
     nucleotide sequence of DNA regions) but also mechanical
     energy (in the form of conformationally strained DNA
     regions) generated from chemical reactions catalyzed
     by enzymes."           . . . . . . . . . . . . . . . . . . . (3)

The fact that active regions of DNA carry mechanical energy, for example,
in the form of DNA supercoils, has been well established [5].  Such
mechanical energy stored in DNA has been variously referred to as
conformons [6] and "Stress-Induced Duplex Destabilizations" or SIDDS [5].

The definition of a gene given in (3) is tantamount to postulating that a
gene is a molecular machine composed of DNA segments and associated
proteins that stores mechanical energy generated from chemical reactions
and uses this energy to transcribe its sequence information into RNA
molecules whenever and wherever needed in the cell for a right duration of
time.

The definition of a gene given by (1) can be made compatible with the
definition given by (3) if we make the following two postulates:

      "The whole DNA carries three kinds of genes -- p-genes
      coding for proteins, r-genes coding for RNA, and
      d-genes coding for DNA molecules."  . . . . . . (4)

The existence of d-genes is self-evident, since DNA serves as the template
for its own replication and this ability of DNA is heritable from one cell
generation to the next.

      "DNA carries not only genetic/sequence information but
       also the mechanical energy (called conformons or SIDDS)
       to power gene expression.       . . . . . . . . . . . . . .  (5)

In other words, by combining the dissipative structure concept of
Prigogine [4] and the conformon concept introduced in molecular biology
more than three decades ago (reviewed in [6]), a new definition of a gene
can be
formulated in two parts as follows:

    i) "DNA carries three kinds of genes, each coding
       for proteins (p-genes), RNA molecules (r-genes),
       and DNA molecules (d-genes)." . . . . . . . . . . . . . . . .(6)

   ii) "DNA stores mechanical energy in the form of
        conformons or SIDDS that powers the
        spatiotemporally organized motions of chromatins
        in order to express p-, r- and d-genes in
        response to the signals received from the
        cytosol."                              . . . . . . . . . . . (7)

Statement (6) can be regarded as a definition of terms that are compatible
with facts, and what is original in the proposed 'triadic' definition of a
gene is contained in Statement (7) in the concept of conformons [6] or
SIDDS [5]. Conformons are defined as the sequence-specific conformational
strains of biopolymers that carry 'ordered energy' to power goal-directed
molecular motions [6].  The first direct experimental evidence for
conformons in DNA was provided by DNA supercoils [5] and for conformons in
proteins by the single-molecule measurements of myosin motions along actin
filament [7]. Also, Statement (6) deals with the informational aspects of
a gene, while Statement (7) is concerned primarily with the energetic
aspect of a gene, consistent with the information-energy complementarity
principle believed to underlie all self-orgnaizng processes in nature [8].

With all the best.

Sung

___________________________________________
Sungchul Ji, Ph.D.
Department of Pharmacology and Toxicology
Rutgers University
Piscataway, N.J., 08855


References:
   [1] Pearson, H. (20056). Genetics: What is a gene? Nature 441:398-401.
   [2] Gerstein, M. B. et al. (2007). What is a gene, post-ENCODE? History
and updated definition. Genome Research 17:669-681.
   [3] Greally, J. M. (2007). Genomics: Encyclopedia of human DNA. Nature
447: 782-783.
   [4] Prigogine, I. (1977).  Dissipative Structures and Biological Order.
 Adv. Biol. Med. Phys. 16:99-113.
   [5] Benham, C. J. (1996). Duplex Destabilization in Supercoiled DNA is
Predicted to Occur at Specific Transcriptional Regulatory Regions.  J.
Mol. Biol. 255:425-434.
   [6] Ji, S. (2000).  Free energy and information content of Conformons
in proteins and DNA. BioSystems 54: 107-130.
   [7] Ishijima, A., Kojima, H., Higuchi, H., Harada, Y., Funatsu, T. and
Yanagida, T. (1998).  Simultaneous measurement of chemical and
mechanical reaction.  Cell 70:161-171.
   [8] Ji, S. (2002). The Bhopalator: An Information/Energy Dual Model of
the Living Cell (II). Fundamenta Informaticae 49(1-3), 147-165.