Transcript Regulatory Molecular Biology

Regulatory Molecular Biology
Arthur B. Pardee
Dana-Farber Cancer Institute; Boston,
Massachusetts USA
INTRODUCTION
Numerous molecular mechanisms regulate normal and cancer
cells’ biological machinery.
These processes operate at multiple levels to produce
coordinated and economically functioning biological activities
and structures. The cells in a multi-cellular organism have
essentially the same genes but differ in functions, and their
genes are expressed differently.
Thus the genotype does not alone determine phenotype, and
life depends on both Nature and Nurture, interplay of heredity
with environment, selecting expressions of hereditary
information from genes and mRNAs, activities of enzymes, and
specificity of membrane transport. These regulations act by
different biochemistries and in different time frames. They
control transit between cell quiescence and proliferation, and
between stages of the cell cycle. The theme of this article is
briefly to summarize innovative discoveries that continue to
provide paradigms of regulatory processes.
Much of what we now take for granted was then
unknown. Methods were comparatively primitive.
Chromatography and spectrophotometry came in the
early 1940’s. Radioactive organic compounds became
available after World War II. There were no biochemical
supply houses and no kits. Nucleic acids were not in the
main picture; their status was like that of carbohydrates
and fats. Around 1950 major interlocking developments
of biochemistry with chemistry and genetics turned
research from metabolism and enzymes toward
macromolecules.
1 The field now called Molecular Biology was born.
Pinnacles are studies on organic structures and nature
of the chemical bond by Linus Pauling, the first
sequencing of a protein (insulin) by Frederick Sanger,
and of 3-dimensional protein structures by Max
Perutz and John Kendrew.
Molecular-biochemical regulation is an enormous
subject. It is summarized here historically as
discoveries and functions, as I remember them in
a scientific path that has led across unexplored
terrain and along byways toward the goal of
learning about the defects of molecular regulation
that lie at the heart of cancer. 2
References are limited to pioneering articles and
germinal reviews to indicate thinking at the time,
and to updating reviews. More can
be readily found by searching the Internet
(PubMed) for reviews on any topic.
MAJOR MECHANISMS OF REGULATION
Mutation. Genetic changes are now recognized to be the
origins of cancer. Although genetics and biochemistry were
separate disciplines in the 1950s, mutation was known to
change enzyme activities dramatically, per the one gene-one
enzyme model of George Beadle and Edward Tatum, And
added genetic material changes metabolism; nine enzyme
activities are quickly altered by additional genetic information
provided by infection of Escherichia coli with a DNA
bacteriophage. Strikingly, a completely novel enzyme
involved in synthesizing hydroxymethyl-cytosine appears,
discovered by Seymour Cohen in 1954. These include
deoxyribonuclease, consistent with a role of DNA in virus
replication, and many mutant progeny are produced after
replacement of thymidine by bromodeoxyuridine in the phage
DNA, as shown by Rose Litman in 1956. These experiments
are forerunners of genetic engineering, involving introduction
of normal or specifically modified DNAs.
Control of metabolic pathways. The great
achievement of biochemistry is to connect most
metabolites into the now familiar pathways catalyzed by
enzymes. Approaching its apex in the 1950s,
most biochemists were very busy successfully creating
this map. All its roads were of the same intensity, although
traffic along some is far greater than on others. Questions
about regulatory mechanisms were not posed. But it was
noticed that metabolism is precisely regulated
and is not wasteful; intermediary metabolites are not
overproduced and do not accumulate in the medium. 3,4
Living organisms usually produce their constituent
molecules in amounts only sufficient to meet their needs,
neither more nor less. It was also noticed that these
balanced internal events respond to extracellular
conditions. This tight control of metabolism is important
for efficient and economical cell functioning. This focuses
a cell’s resources.
Feedback inhibition. A mechanism for adjustments
to both environmental metabolites and to prevent
excessive intracellular end products is by economically
shutting down their synthesis when unneeded.
A breakthrough that established a ‘Root’ of molecular
biology was discovery of the general Feedback Inhibition
mechanism. The end product of a biosynthetic pathway
blocks production of an intermediate molecule in that
pathway by inhibiting an enzyme’s activity, see ref. 5.
Initial indications made in 1954 are rapid inhibition
by added tryptophan of biosynthesis of an intermediate
in its pathway, reported by Aaron Novick6 and Richard
Yates and Arthur Pardee.7 stated “ added uracil blocks
an enzyme step between aspartate and ureidosuccinate
formation”; “this block may be an important regulatory
mechanism in the cell” .
Feedback inhibition immediately created the problem if its
molecular basis. How can ATCase be inhibited by uracil that
is structurally very dissimilar from the substrates aspartate
and carbamyl phosphate? Enzyme catalysis was described
in the 1940’s as a three-step process in which substrate(s)
specifically bind to the catalytic site of an enzyme, are then
converted to product(s), and are released. The enzyme is
then free for another catalysis. Inhibitors were seen to
compete specifically for the enzyme’s catalytic site, thereby
excluding substrate. This was quantitatively described in
1913 by the equation of Lenor Michaelis and Maude Menten:
the velocity of the reaction (v) depends on the maximal rate
(VM), concentrations of enzyme (E), substrate (S), and
inhibitor (I), and their affinities (KM) and (KI).
v = VM E S______
S + [KM (I + KI) ]/ KI.
The key demonstration by John Gerhart and Pardee of
independent catalytic and regulatory sites came from an
unexpected observation made to establish the basis for
the feedback control of activity. Variable results of
inhibition of the pure enzyme by CTP were
repeatedly obtained. Frozen enzyme thawed at the
beginning of a week was strongly inhibited. But thereafter,
inhibition was lost during storage in the refrigerator.
Furthermore the activity actually increased, and kinetics
changed from the subunit-cooperative S-shape
to the classical Michaelis-Menten shape. Hypothesizing
that ATCase must change, even at zero degrees,
systematic warming showed that five minute exposure to
65° C abolishes its inhibition by CTP but not its catalytic
activity.
That enzymes are often complexes rather than single
proteins, as was then the general biochemical concept is
major development arising from feedback inhibition, now
well established. Hemoglobin and the b-galactosidase
repressor are tetramers of identical subunits,
ribonucleotide reductase has catalytic and regulatory
subunits, and there are many other multi-protein
complexes. Examples are cyclins that activate cdks. And
more than a dozen B proteins differently control properties
of the pleiotropically functioning and ubiquitous PP2A
phosphatase; one regulates degradation of oncogenic
myc.22 An early extreme example is the ribosome, a multiprotein complex that catalyzes protein synthesis. And DNA
synthesis is catalyzed by a Replitase complex that
contains enzymes for both precursor synthesis and
polymerase, as found by Prem Reddy.23
ACTIVATION BY COVALENT MODIFICATION.
After a protein is synthesized it may not have enzymatic
activity, which can be produced by a subsequent covalent
modification. A major mode of changing activity (plus or
minus) in higher organisms is produced by covalent
phosphorylation of proteins by the kinases, discovered by
Eugene Kennedy in 1954, which can be reversed by
phosphatases. Edwin Krebs and Edward Fisher in the 1950s
discovered that this covalent protein phosphorylation is a
mechanism for enzyme activity regulation; ATP level controls
glycogen phosphorylase which provides metabolic energy.24
The human genome contains 518 kinases (the kinome), each
of which is regulated to phosphorylate a distinct set of
substrates.25 Kinases, and also proteases, are often organized
into sequentially activating cascades that catalyze rapid,
exponential-like amplifications of downstream activity.
Examples are the kinase cascades activated by binding of
growth factors to their receptors on the mammalian cell’s
surface.
CONTROL OF GENE ACTIVITY
The rate of a reaction depends upon the amount of its
enzyme as well as upon its activity, as seen in the
Michealis-Menten equation. Amounts (maximal
activities) of some enzymes in bacteria known
before 1950 to be change by environmental molecules.
They “adapted” as a function of extracellular nutrients,
dramatically increasing in amount when their substrate
is provided. Jacques Monod, the outstanding
investigator of this problem, performed
elegant experiments on the dependence of bgalactosidase production in E. coli as a function of
availability of -galactoside sugars which were proposed
to act as ‘inducers ‘ of the gene. 26 This control of gene
expression acts relatively slowly as compared to
feedback inhibition of metabolic reactions.
The constitutive cells was concluded to lack a
repressor protein that is present in inducible bacteria
and is gradually produced in the mated cells after its
gene is introduced. This means that the repressor
specifically blocks gene expression; coding DNA is
shut down when repressor protein binds to an
upstream DNA repressor sequence. 27
The repressor is released when its other site binds a
low molecular weight inducer molecule. Specifically,
expression of -galactosidase (and two adjacent
genes) is inhibited when a lac repressor protein
binds to its upstream DNA operator region. and
mutant bacteria that cannot make repressor produce
the enzyme constitutively. The lac repressor protein
was isolated in 1966 by Walter Gilbert and
Benno Muller-Hill.
Major developments from PaJaMa.
i) Primarily, this experiment is the foundation of
transcriptional control of gene expression by
both bacteria and eukaryotes.
ii) Enzymes in synthetic pathways can be repressed by
low molecular weight compounds, as well as those
involved in catabolism; a metabolite can repress
transcription of its biosynthetic pathway. Examples
are the pathway of pyrimidine biosythesis by Richard
Yates and Arthur Pardee, and for arginine by Luigi
Gorini and Werner Maas (for a review see 30).
iii) The broad biological roles of functional sites
interacting with separate regulatory sites depends
upon these concepts of repressor and regulatory DNA
promoter sequences
Allostery. The two types of binding sites of proteins,
one functional and the other regulatory, permit many
types of biological reactions to be controlled by a
molecule that has no structural similarity to the
molecules acted upon. Jacques Monod combined
three lines of research to create this allosteric
concept,30 which he called “the second secret of life”:
i) feedback inhibition with its catalytic and regulatory
sites (see above),
ii) a site for binding galactosides
to the lac repressor modifies another functional site that
binds it to a DNA sequence, and
iii) cooperative binding of oxygen to the four protein
subunits of hemoglobin and which are modified by their
interactions with CO2. For an historical review see ref.
36.
Mathematics of multi-subunit interaction.
Allosteric activity depends on functional regulation by
alternative structures of multiprotein
complexes. An early example is the Hill equation which
mathematically describes interactions of the four
subunits of hemoglobin upon binding of O2. General
allosteric equations have been described by two
mathematical models, based upon alternative
active and inactive conformations of subunits controlled
by regulator binding. In one, the subunits conformations
change in a concerted, all-or-none, manner. 38 In the
other, each binding sequentially alters
the protein’s structure and changes the next binding
affinity; technical methods, such as 3D protein structure
determinations, are resolving this question of allosteric
changes. 39
REGULATION OF MEMBRANE FUNCTIONS.
Control by molecular location is seen at three levels, whose
amounts and activities are regulated both genetically and
environmentally First is extra-cellular vs. intra-cellular
location. Many molecules generally must pass into a cell to
metabolized. They cross the cell membrane via specific
transport mechanisms that permit either passive entry or
catalyze enzyme-like energy-dependent accumulation.
Second are systems that move molecules between
cytoplasm and organelles. For example, enzymes involved
in DNA synthesis accumulate in the nucleus before Sphase. Third, enzymes are often assembled, onto protein
scaffolds, into multi-protein complexes that perform
cooperative functions.40 As an example of such
interactions, compounds that specifically inhibit an
isolated enzyme also inhibit others that are in the replitase
complex.41
Individual mRNAs similarly have been localized in cells.42
Active transport of a molecule into the cell is a first step
in many metabolic pathways. Kinetics of substrate
uptake and enzymes are similar. It is therefore not
surprising that trans-membrane transport
is regulated and inhibited similarly to enzyme activity.
Transport of galactosides across the membrane of E.
coli is inducible;44 adjacent genes for -galactosidase
and galactoside transport (permease) are
co-induced by -galactosides, per the operon model
proposed by Jacob and Monod.30 Molecules catalyzing
transport were unknown in the 1950’s. One of the first
transport-related molecules to be purified
is a regulatory factor for sulfate transport.45 A transport
system was demonstrated for uptake of sulfate ion into
Salmonella typhimurium. Mutants that could not grow on
sulfate were isolated by applying toxic chromate ion;
they were defective in transport.
Cell surface membrane and transport are very important
ineukaryote metabolism and regulation, e.g., the coupled
transport of hydrogen ions across the mitochondrial
membrane that produces ATP discovered by Peter
Mitchell, or control of neuronal transmission
by regulated receptor-mediated transport of ions.
Density-dependent contact inhibition of cell growth
involves surface proteins such as cadherins, integrins,
etc. that make connections to other cells and to
the extracellular matrix. Membranes of eukaryotic cells
contain proteins with extra-cellular binding sites that are
specific receptors for protein growth factors. These
regulate these receptors’ intracellular tyrosine kinase
activity, as shown by Joseph Schlessinger.47
MORE MECHANISMS
Epigenetic controls can permit a cell to express only a
subset of its genes, for example differently in liver than
skin. Pioneering experiments by Werner Arber
demonstrated DNA methylation protects bacterial
DNA from hydrolysis of by restriction endonucleases,
and by Ruth Sager who found that methylation is the
basis of non-Mendelian inheritance of organelle genes in
the eukaryotic alga Chlamydomonas.
The effect of methylation then shifted from elimination of
DNA to blocking gene expression in higher organisms.
Methylation of DNA attracts enzymes that catalyze
acetylation of histones and thereby changes of
chromatin structure and activity. Mechanisms of histone
modifications and their effects on gene expression are
under vigorous investigation.48
Information about the classes and functions of RNAs are
increasing dramatically. Mechanisms are newly
discovered. that on the one hand regulate amounts and
functions of mRNA, and on the other regulations by
RNAs For an overview see ref. 49. Transcriptional
production of pre-mRNA is followed by its processing
and splicing, which produces hundreds of mRNAs and
then their corresponding proteins. 50 mRNA production
is also controlled by complex reactions such as transsplicing to their 5’ ends of short synthesis-regulating
leader RNAs in some organisms. 51 About 13,000 target
relationships have been identified as complimentary
seed sequences.52 Ribozymes catalyze molecular
reactions. siRNAs can block translational activity,
and importantly they activate specific mRNA
degradation, 53 which takes place in cellularly localized
P-bodies.
REGULATION OF CELL PROLIFERATION
Research shifted from bacteria toward higher organisms
in the 1960s, along with the rise of molecular biology.
Techniques had progressed sufficiently to make in vitro
culture of mammalian cells generally feasible. Functions
in eukaryotic cells are controlled by interplay of genetic
and environmental factors, and as with bacteria
these can regulate DNA and protein functions.
Processes that involve an entire cell homeostasis take
us into a new realm of regulation; these are at least an
order of magnitude more complex than is
gene expression or a metabolic pathway. The
mechanisms that control gene expression and enzyme
activity are applied in regulation of cell
proliferation.
Cell cycle control. Regulation in eukaryotic cells is at
a slower pace than controls in bacteria; completing the
cycle can require a day or longer vs. an hour.
The sequentially organized processes of cell
proliferation are described as the cell cycle. Early
research on the eukaryotic cycle is summarized,56 and
has since often been reviewed.57 To produce two cells
from one requires that all molecules, large and small,
must be duplicated precisely. These syntheses take
place at specific times, the most prominent example
being duplication of DNA in mid-cycle S-phase, shown
by Howard and Pelc in 1951.58 The cell cycle of bacteria
had begun to be investigated by1960. Its duration as
measured with synchronized E. coli depends on
carbon source, requiring over an hour on acetate and as
short as 15 min on glucose.
Emergence from quiescence and transit through G1 is
inhibited by density dependent physical-chemical
interactions between adjacent cell surfaces. It is
activated by proteins in serum, such as insulinderived
growth factor and epidermal growth factor. These bind to
external receptors on the cell membrane, which
activates intracellular auto-phosphorylation by the
receptor’s tyrosine kinase. Alternatively,
estrogen and androgen initiate proliferation of female
and male sexrelated cells, respectively, and these
relatively small molecules bind to receptors located to
the nucleus. Both activations initiate kinase
cascades that activate transcriptions. The cyclin
proteins increase and then decrease in a specific
sequence during the cycle, as discovered
by Tim Hunt and colleagues.62 They complex with and
regulate several cyclin-dependent kinases (cdks),
discovered by Paul Nurse.63
Other complicated mechanisms limit DNA replication to only
once per cycle, and yet others control mitosis and daughter
cell separation. Toward the end of G1 phase cyclin/cdk
activities phosphorylate the retinoblastoma protein, causing
its inactivation and release and activation of E2F-1, a
transcription factor for many enzymes required for DNA
synthesis. E2F-1 is also the (autocatalytic) factor for its own
transcription, and therefore it increases dramatically at the
G1/S boundary. But excessive E2F-1 is apoptotic, and so a
feedback control must exist; perhaps inhibition by the end
product dNTPs is responsible. Based upon this idea, a novel
chemotherapeutic principle for action of agents that deplete
dNTP pools has been suggested.64
Feedback loops between plus and minus balancing controls
are indicated, such that excess of one activates the opposite.
A major question that determines detailed investigations of
molecular mechanisms was at what point in the cell cycle
growth is regulated.
Production and removal are balanced at every biological
level. Specific multi-protein enzymatic machineries label
and then degrade metabolites, RNAs, proteins, and cells.
Such a major regulatory mechanism is proteolysis, the
most dramatic alteration of a protein’s structure. It was
discovered early to convert extracellular inactive
proteins (zymogens) to active enzymes; trypsinogen to
trypsin is an example. But proteolysis usually eliminates
intracellular activity,70 and the protein is in a steady
state determined by balance with its synthesis. This is
especially so for regulatory proteins including
cyclins that are produced transiently and are degraded
when their roles are complete. Another prominent
example is removal of proapoptotic P53, activated by its
ubiquitination involving Mdm271 and then degraded by
proteasomes.72 This control is a feedback loop
because P53 induces Mdm2.
Cancer and mis-regulation.
For regulatory mechanisms, ‘The pathological
illuminates the normal’. Defective controls created by
mutations and altering gene expressions are causal of
cancer. Genetic material introduced by viruses can also
cause cancer. These genetic level changes can produce
either gain of an activity of an oncogene such as ras
discovered by Ed Skolnik and shown by Robert
Weinberg and Geoffery Cooper to be mutated in cancers.
Mutational loss of a tumor suppressor gene such p53 or
pRb was proposed by Ruth Sager.74 The latter are more
frequent, and the more probable because the normal
phenotype is dominant in fused normal plus cancer
cells, as shown by Henry Harris and Boris Ephrussi.
Cancer cells require more oxygen, more energy, and
more active metabolism than do normal cells, which are
usually quiescent. A hallmark of cancer is deregulated
cell proliferation, and changes of controls through the
cell cycle are reported, particularly in G1 phase.
Numerous changes of kinases and phosphorylations
have been reported. Cyclin E is over expressed and
modified in advanced cancers, and it provides a clinical
marker.79 The tumor suppressing retinoblastoma
protein is very frequently inactive or absent, and the
E2F-1 protein that it negatively regulates is released to
activate S phase transcriptions. The critical Restriction
point control is relaxed or absent in cancers,80 the R
protein can be more stable, a difference that provides a
molecular basis for greater proliferative capacity.81
Of fundamental importance are the dynamic steady
states between production and removal. These are
active at all levels—molecular, cellular, and
biological. Regulatory interactions create off-on
switches between alternative pathways, negative
feedback loops for limiting pathways, positive
feedback loops that convert transient into sustained
signals, feed-forward loops and successive
activation pathways that amplify signals, as by MAP
kinases.84 Systems biology—mathematical
computer models are being developed to grasp
these interactions in large genetic and metabolic
networks.85
References