BioSystems 83 (2006) 217–224
Spatiotemporal chemical dynamics in living cells:
From information trafficking to cell physiology
Howard R. Petty ∗
Departments of Ophthalmology and Visual Sciences and Microbiology and Immunology, The University of
Michigan Medical School, 1000 Wall Street, Ann Arbor, MI 48105, USA
Received 6 December 2004; received in revised form 15 April 2005; accepted 4 May 2005
Abstract
Biological thought in the 20th century was dominated by the study of structures at increasingly minute levels. For biology to
advance beyond structural reductionism and contribute its full measure to clinical care, living biological structures must be understood
in the context of their collective chemical processes at the relevant chemical time-scales. Using high-speed fluorescence microscopy,
we have studied intra- and inter-cellular signaling using shutter speeds (∼100 ns) that remove the effects of wave motion and diffusion
from optical images. By collecting a series of such images, stop-action movies of signal trafficking in living cells are created; these
have revealed a new level of spatiotemporal chemical organization within cells. Numerous types of chemical waves have been found
in living cells expressing a great variety of physical properties. In this article I will review some of these basic findings, discuss
these events in the context of information trafficking, and illustrate the potential implications of this work in medicine.
© 2005 Elsevier Ireland Ltd. All rights reserved.
Keywords: Cell signaling; Chemical waves; Calcium; Metabolism
1. Introduction
Molecular sciences, including molecular biology,
genomics, proteomics, and crystallography, have now
described life in unprecedented depth and breadth. Yet
these descriptions have not improved significantly the
rate of drug discovery. The molecular defect in sickle cell
anemia was discovered nearly a half-century ago, yet in
50 years what treatments has this knowledge yielded?
The defects in oncogenes were found in the 1970s, and
those in cystic fibrosis and Duchenne muscular dystrophy were discovered in the 1980s, but we have yet to
see significant improvements in clinical care. Whatever
scientific insights might have been gained, molecular sci-
∗
Tel.: +1 734 647 0384; fax: +1 734 936 3815.
E-mail address: [email protected].
ence has not yielded sufficient insight to provide better
care for patients.
So what have we missed? By so thoroughly embracing structural reductionism, we have learned a great deal
about the system’s parts without really understanding
how the system works. For example, a list of the capacitors, resistors, and the other functional parts found in a
television set are not sufficient to deduce how a television
works. As the parts list of the human genome is much
longer than that of a television set, and the behavior of its
components more subtle, it is not surprising that molecular biology has not yielded the anticipated cornucopia of
new drugs. Living cells require networks of enzymes and
receptors with large numbers of feedback loops under
conditions held far from thermodynamic equilibrium.
Therefore, the properties of individual isolated components can never adequately model the dynamic chemical
processes that underlie cell functions. The part cannot
0303-2647/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.biosystems.2005.05.018
218
H.R. Petty / BioSystems 83 (2006) 217–224
explain the whole; to understand how all of the parts
of a cell work in concert, the parts must be studied in
their cell biological context at a time-scale relevant to
the physico-chemical processes under study.
I propose a paradigm shift in our understanding of
how cells handle information from the present molecular/structural consensus to a more dynamic and integrative cellular approach. Using new imaging technology,
my group has experimentally observed that mutations,
drugs, leukocyte invasion of tissues and physical stresses
perturb chemical wave patterns in cells and tissues. This
approach, based upon biophysical dynamics of the chemical circuitry of cells, can and should be used to address
key problems in medicine.
2. Background and approach
Chemical waves of calcium ions were discovered by
Ridgway et al. (1977) during studies of oocyte fertilization. These waves have been confirmed in many cell
types (Jaffe, 2002), but the observations are restricted
to very large cells (or cell arrays) such as oocytes and
myocytes. This restriction is due to the fact that such
waves cross typical individual cells (10–20 ␮m in diameter) in periods of time that are short in comparison to
the shutter speed of the detector. Hence, conventional
imaging approaches have little to say about the existence of such waves in more representative cell types.
To address this limitation, my laboratory recently developed high-speed microscopy, which reveals dynamic
molecular signaling events in cells using very short
shutter speed times, so short (e.g., 100 ns.) that small
atoms and molecules cannot diffuse significant distances
within a cell while the shutter is open. By taking these
images in rapid succession (∼10 ms/frame), we obtain
movies of how signals, metabolites, membrane potentials, and other chemical instabilities behave within cells
(Petty, 2004a). High-speed microscopy sacrifices some
spatial resolution to obtain very high temporal resolution whereas the complementary technique of scanning confocal microscopy sacrifices temporal for spatial
information. This tool does for cell biology what patchclamping did for electrophysiology, but our method
retains far more spatial information without wounding
a cell.
Calcium waves can be detected using high-speed
fluorescence microscopy in combination with a dye such
as indo-1. As imaging ratio is not yet possible at highspeed, it is important to use a dye, such as indo-1, that
does not require ratio to detect calcium changes. Indo-1
has a Kd of 230 nm; it has very little dynamic range from
1 to 100 ␮M calcium, which is the range associated
with certain physiological events of neutrophils (e.g.,
Nusse et al., 1998; Smolen et al., 1986). Consequently,
the images often appear rather binary in nature. On the
other hand, an additional advantage of indo-1 is that
it has a favorable forward rate constant, which allows
quick detection of calcium entry into the cytosol of a
cell. Fig. 1 shows how high-speed imaging and indo-1
can be combined to reveal spatiotemporal waves in
a living cell. These waves constitute single temporal
calcium “spike” observed using other microscopy
methods. Calcium signals are seen to travel along the
cell surface and from the surface to the two internal
phagosomes.
By labeling cells with reporters and using endogenous autofluorescent molecules, we have observed
many kinds of chemical waves including calcium,
pH, metabolic (NADPH and flavoproteins), membrane
potentials, and others (Petty, 2005). These waves not
only vary in their composition, but also in their shape,
number, direction, velocity, and route (Fig. 2). Highspeed imaging allows us to follow signal circuits in living
cells. For example, we have observed calcium waves
traveling from the plasma membrane to the phagosome
(Fig. 1), from the plasma membrane to the nucleus, from
the nucleus to granules and along other routes. We have
seen some waves annihilate when they come into contact with one another, and others that do not, depending
upon the physiological context. There are an enormous
variety of chemical waves in living cells, whose behavior
suggests a novel biological code.
The key to deciphering this code is the selection of a
simple model system that can be experimentally manipulated. Cells that function semi-autonomously in vivo,
such as leukocytes and tumor cells, are particularly good
candidates for dynamic evaluation of their chemical
behavior in vitro because their behavior is not strongly
influenced by adjacent cells. To simplify the problem
further while retaining the relevant chemical circuitry,
we evaluate chemical waves of discrete, self-contained
subsystems within a cell. A good example of such a subsystem is the glycolytic apparatus, which is composed of
10 enzymes with multiple feedback loops. This cellular
subsystem has been shown to display temporal oscillations and traveling metabolic waves. One could study,
for example, how drug-mediated chemical perturbations
influence the propagation of chemical wavefronts of glycolytic activity. My laboratory has already shown that
conventional drugs, such as indomethacin and dexamethasone, influence these waves. In this way, we focus
on specific subsystems of certain semi-autonomous cells
to develop new ways to perturb their systems behavior
into a normal phenotype.
H.R. Petty / BioSystems 83 (2006) 217–224
219
Fig. 1. High-speed microscopy reveals calcium signal trafficking in neutrophils labeled with a calcium-sensitive dye. A neutrophil has internalized
two smaller cells (arrows). Note that the calcium wave travels about the perimeter of the cell. When the calcium wave passes by the region containing
the phagocytosed cells, a calcium signal travels around both internalized targets (150 ns, 5 ms interval; ×1400).
Fig. 2. Survey of waves in cells. A variety of chemical wave types have been found in eukaryotes; calcium and metabolic waves are illustrated.
Waves can be constrained to a specific organelle, such as a phagosome, or not be geometrically constrained.
3. Specific examples
As I have recently reviewed high-speed imaging
methods and chemical waves (Petty, 2004a,b, 2005),
only a few examples will be discussed below to illustrate
certain biological principles. We will begin by discussing
calcium waves. Calcium is a biologically important
cation, which increases in concentration during many
signaling events. For lack of a better model, most biologists envision this as a uniform rise and fall within the
cytoplasm. This is far from what actually occurs: many
cell functions are characterized by one or more calcium
waves (Fig. 2). It should be clear that during chemical
wave motion, such as a calcium wave, it is the perturbation in the excitable calcium matrix that is propagating
long distances, but an individual calcium ion may not be
traveling far. One may envision this as waves traveling
on a body of water: the wave travels long distances, but
a bobber just goes up and down. Similarly, an individual calcium ion may diffuse only a short distance before
binding to a target site to induce calcium-induced calcium release or to be pumped out by a calcium-ATPase,
but the wave (a chemical disturbance) may travel across
the entire cell or tissue to carry a message.
As neutrophils become morphologically polarized
during migration, a calcium wave is ignited at the front,
220
H.R. Petty / BioSystems 83 (2006) 217–224
or lamellipodium of a cell, which then travels unidirectionally about a cell’s perimeter (Kindzelskii and
Petty, 2003; Kindzelskii et al., 2004a,b). This wave is
associated with plasma membrane calcium channels,
as indicated by the ability of specific drugs to block
its propagation. It is the spatiotemporal counterpart of
the temporal calcium spike found for these cells using
microfluorometry. This wave may, for example, participate in the restructuring of the cytoskeleton necessary
for cell movement. This peripheral calcium wave is in
agreement with the theoretical expectations of Simon
and Llinas (1985), who predicted intense submembranous calcium release at submicrosecond time-scales. The
wave is repetitively initiated at this same site as long as
the cell is moving. Not only does the calcium ignition
site correlate with the direction of cell migration, but
also when a chemotactic signal is delivered to a cell in
a direction perpendicular to the direction of migration, a
new calcium wave ignition site is rapidly assembled at
this location, long before the cell actually turns. Thus,
the ignition site is linked with chemosensory perception
and remembering directional cues.
Phagocytosis also generates calcium waves (Kindzelskii and Petty, 2003; Worth et al., 2003). When a
polarized neutrophil has internalized a target particle, a
calcium wave splits off from the perimembrane wave
mentioned above then travels from the plasma membrane to the phagosome (e.g., Fig. 1). This calcium
wave promotes the fusion of phagosomes with lysosomes, leading to the destruction of the target. Using
site-directed mutagenesis and transfection techniques,
we found that a certain sequence of amino acids (L–T–L)
with the phagocytosis receptor Fc␥RIIA is required to
route calcium waves to the phagosome. We have built
so-called “Trojan” peptides that will deliver the L–T–L
sequence of amino acids to the inside of a cell. These
Trojan peptides bind to the endoplasmic reticulum and
block both calcium signal trafficking to the phagosome
and phagolysosome fusion. Although it is doubtful that
a drug designed to mimic this peptide’s action could
be systemically administered, certain clinical conditions
such as topical application in keratitis and intra-articular
injections in arthritis might benefit from such a drug.
Calcium waves are also found in migrating tumor
cells, which may be relevant to the process of metastasis. We have observed that carboxyamido-triazole, an
anti-metastatic drug currently in clinical trials, affects
the calcium wave properties of a fibrosarcoma cell line
and, in parallel, the migratory and invasive properties of
these cells (Huang et al., 2004). As several gene products are necessary to support the formation of a calcium
wave, we speculated that multiple drugs acting on differ-
ent proteins contributing to the same wave might enhance
the efficacy of the drugs. We found that multiple drugs
acting on a single wave through different proteins could
be used to reduce cell invasiveness. Hence, the performance of carboxyamido-triazole, might be improved by
combining it with a drug active against other calcium
channels. Indeed, it might be possible to examine numerous combinations of drugs to find more effective ways
of managing many types of disease.
A variety of calcium waves have been observed in
other cell types. For example, calcium waves have also
been observed in endothelial cells, mast cells, and lymphocytes (Petty, 2005 and unpublished data). We have
observed a chorus of at least four different calcium waves
during dendritic cell- and tumor cell-mediated antigen
presentation to CD4-positive T lymphocytes (Li et al.,
2005). Fig. 2 illustrates several of the calcium wave types
discussed above as well as others we have found. For
example, some waves are not geometrically constrained
and travel throughout a cell whereas others may be
limited to travel to or within a particular organelle. One
type of non-geometrically constrained wave is the IP3dependent calcium wave, which has been shown to travel
across a cell as a spherical wave. The physical properties
of the wave, such as velocity, are in general agreement
with theoretical expectations (Falcke, 2004; Goldbeter,
1996; Pencea and Hentschel, 2000). By studying
and manipulating intracellular chemical circuits, it
should be possible to devise better ways of treating
patients.
High-speed fluorescence microscopy has also been
employed to study several self-organization of metabolic
wave patterns in neutrophils. In this case, exogenous
labels are not required because the metabolite NAD(P)H
is fluorescent. To respond to a pathogen or other physiological perturbation, spherical neutrophils in the circulation must first adhere to blood vessel walls then migrate
into the interstitial tissues. During adherence, a bright
circular region of NAD(P)H forms in the neutrophil
at the site of substrate contact (Petty and Kindzelskii,
2000). After the metabolic region reaches a critical size
(r ∼1 ␮m), a wave propagates from this site in the form
of a target pattern inside the cell. The critical radius is
in reasonable agreement with the size anticipated by the
physical properties of the system (Petty and Kindzelskii,
2000; Tyson and Keener, 1988).
Morphologically polarized neutrophils exhibit a single longitudinal wave of NAD(P)H autofluorescence
propagating from the tail to the front of a cell (Petty
et al., 2000). Thus, the redox conditions within a cell
vary in a spatiotemporal fashion and are oriented in the
direction of cell migration. Moreover, the properties of
H.R. Petty / BioSystems 83 (2006) 217–224
these metabolic waves respond to receptor stimulation
(Petty and Kindzelskii, 2001). When activating factors,
such as bacterial lipopolysaccharide and N-formyl-metleu-phe, are incubated with neutrophils, two longitudinal
waves moving in opposite directions are observed. Thus,
metabolic patterns respond to cell activation.
One of the primary functions of neutrophils is to
produce superoxide anions, which participate in host
defense. Superoxide anions are produced by the NADPH
oxidase, whose substrate is NADPH. By including a
molecule in the buffer that becomes fluorescent upon
reaction with superoxide, we have shown that substantial quantities of superoxide are produced only when
the NAD(P)H wave (substrate) reaches the front of the
cell (Kindzelskii and Petty, 2002). Wave arrival at this
location provides NADPH (or electrons) to the NADPH
oxidase thereby promoting the formation of superoxide (product). Thus, the spatiotemporal dynamics of the
substrate-to-product transformation in living cells by the
NADPH oxidase have been revealed. Moreover, these
metabolic waves allow a neutrophil to aim at a target cell.
By orienting the release of superoxide, a cell maximizes
superoxide delivery to the target while minimizing collateral damage. By focusing superoxide delivery into a
brief period of time, maximal target damage is achieved
via very high local concentrations at a target; damaging chemical reactions likely occur before a target cell’s
systems have time to adapt.
High-speed microscopy studies have also revealed
that neutrophils from pregnant women do not undergo
changes in metabolic wave patterns upon stimulation
with various activating substances. As superoxide production is depressed in pregnancy neutrophils, this ties
the status of chemical waves to cell physiological states.
These changes are of clinical importance as obstetricians have known for many years that pregnancy is often
accompanied by remission of autoimmune disease and
increased risk of certain infectious diseases (Kindzelskii
et al., 2002, 2004a,b). About 70% of women with multiple sclerosis, arthritis, uveitis, and other autoimmune diseases driven by type 1 T helper cells experience disease
remission during pregnancy, but relapse after delivery.
Prior modeling studies have suggested that the intracellular disposition of metabolic enzymes may influence spatiotemporal metabolic properties (Marmillot
et al., 1992). Relying upon our dynamic methods as
well as conventional imaging techniques, we showed that
enzymes of the hexose monophosphate shunt, which are
found at the periphery of leukocytes from non-pregnant
women, underwent translocation to the center of cells in
pregnant women (Huang et al., 2005; Kindzelskii et al.,
2002, 2004a,b). In this way the shunt’s supply of
221
glucose-6-phosphate becomes unavailable to the hexose monophosphate shunt as it is metabolized by glycolytic enzymes at the cell periphery. Consequently, less
NADPH is available to support superoxide production
and, therefore, less robust autoimmune and host defense
responses may be found in pregnant women. However,
detailed computational modeling of metabolic compartmentalization in pregnancy has not yet been performed.
It seems likely that solving the dynamic regulatory chemical codes of pregnancy will provide new tools to treat
certain autoimmune and inflammatory diseases.
On the basis of the behavior of single cells, our new
paradigm anticipates phenomena such as intercellular
synchronization among populations of cells. For example, it has been known for many years that when two cells
collide on a tissue culture plate, the pathways they follow
as they migrate away from each other are mirror images.
Although this has never been adequately explained by
conventional biology, wave ignition and interference
properties account for this phenomenon (Petty, 2005).
I postulate that in vivo inflammatory cell populations
exhibit coherent biochemical reactions thereby acting as
a self-organizing community to benefit the host. One
can imagine neutrophils entering an inflammatory site,
touching one another, and then temporally organizing
their biochemical reaction dynamics (and consequently
their physiological processes) leading to dynamic intercellular cooperation to destroy targets, a clear physiological advantage for the host.
For clarity, I have focused on just one type of chemical wave in each paragraph above. The biological setting,
however, is quite different: there are likely dozens, if not
hundreds, of chemical waves darting about a cell, each
with embedded information or commands. Just as a symphony emerges from hundreds of musicians each playing
their scores, biology emerges from numerous chemical
waves traveling simultaneously to perform chemically
based calculations and distribute decisions. Since the
function of individual enzymes may depend upon numerous chemicals (pH, ATP, calcium, etc.), novel spatiotemporal structures, such as nodes, may emerge. Phenomena
such as wave annihilation have been observed, which is
anticipated by theory (Tyson and Keener, 1988). Other
more complex biological wave phenomena such as the
overt lack of annihilation (Kindzelskii and Petty, 2003),
wave cascades (Li et al., 2005), and wave hierarchies
(two waves colliding with only one being annihilated)
(unpublished data) represent fertile new ground for combined experimental and theoretical inquiry. As these
dynamic biological codes become understood, bewilderingly complex biological and clinical problems may
become astoundingly simple.
222
H.R. Petty / BioSystems 83 (2006) 217–224
4. Biological implications
Chemical waves and their interactions within cells
have revolutionary implications in both basic science and
clinical medicine. It should be possible to derive robust
mathematical models of spatiotemporal waves within
and among cells, as we and our colleagues have done
previously for temporal chemical oscillations (Olsen
et al., 2003). This approach will allow us to predict the
actions of drugs on waves and their downstream physiological responses, which would launch the field of
clinical systems biology. Moreover, our new paradigm
and its instrumental methods are generally applicable to
biomedical research. High-speed imaging can be used
at the tissue and organism level in addition to the cellular level. This allows for the evaluation of numerous
physiological pathways and disease states.
The great variety of chemical waves, their locations, and physical behaviors lead to the speculation
that cells and tissues perform various sorts of analog chemical computations. For example, depending
upon the underlying biochemical pathways and kinetics, chemical waves may lead to the null (1 + 1 = 0),
identity (1 + 1 = 1 + 1), and additive (1 + 1 = 2) solutions,
which in turn lead to physiological responses or the
lack thereof. Numerous theoretical and experimental
studies have explored the use of excitable chemical
media in unconventional computation (e.g., Adamatzky
et al., 2002; Hjelmfelt et al., 1991; Kuhnert et al., 1989;
Sielewiesiuk and Górecki, 2002; Steinbock et al., 1995).
Chemical dynamics are not only the language of signaling, but likely represent how cells process information, compute solutions and distribute decisions. This
is how cells think; it is the next biological code to be
broken.
Although these waves are the physico-chemical
embodiment of signaling and are required for cell functions, it may seem difficult to understand how such waves
could affect biological processes taking place at longer
time-scale. Let us begin with a familiar example of a
depolarization wave traveling along a neuron. In this
trivial case, the waves are driven by voltage changes;
hence, the waves certainly form at the time-scale of ion
channel conformational changes. It seems safe to say
that these rapidly propagating (∼10 m/s) depolarization
waves lead to important physiological outputs at longer
time-scale, like motor neuron function and consciousness. If voltage-gated waves are physiologically relevant,
it is perhaps not so difficult to believe that chemical
waves, traveling 10,000 times slower provide physiologically relevant information as well. Consider a wave
0.5 ␮m in width traveling near a membrane with a veloc-
ity of 50 ␮m/s. As such a wave would pass by a stationary
point in only 10 ms, how could this have any biological effect? One must consider the relevant time-scale
of protein conformational changes. NMR spectroscopy
has revealed that several signaling proteins participating
in protein–protein binding, phosphorylation and calcium
binding undergo conformational changes relevant to protein activation on a time-scale of ␮s-to-ms. (Feher and
Cavanagh, 1999; Malmendal et al., 1999; Volkman
et al., 2001; Wand, 2001). Hence, it is not unreasonable to
suggest that chemical waves might influence the behavior of nearby proteins. Indeed, it is possible that different
wave velocities select different protein functions on the
basis of the characteristic time-scale of their conformational changes. The transient or permanent activation of
proteins then leads to longer time-scale changes. For
example, the transient calcium wave of Fig. 1 leads to
the fusion of lysosomes with phagosomes resulting in
the permanent digestion of phagosome contents.
Although calcium waves have been previously
demonstrated in oocytes and myocytes (Engel et al.,
1995; Jaffe, 2002), calcium signaling in most cell types
is envisioned as a uniform signal emanating from the
site of receptor activation. Our high-speed microscopy
experiments have shown that signals are not distributed
uniformly within cells, but what biological advantages
might waves provide to cells? To begin with the example of Fig. 1, the calcium signal travels from the plasma
membrane to the phagosome, thus providing synapticlike specificity to the fusion of lysosomes with phagosomes (Table 1). This, of course, minimizes the chaos
and destruction that might ensue if such signals were
uniformly distributed within cells. In contrast, certain
other signals travel as waves uniformly throughout cells
in a rather endocrine-like fashion, as if no organelles
were present (Petty and Kindzelskii, 2001). In another
Table 1
Biological advantages of chemical waves vs. the uniform signaling
paradigm
1
2
3
4
5
6
7
Synaptic level specificity in information transfer between
organelles (e.g., plasma membrane–lysosome communication)
Longitudinal NAD(P)H waves allow neutrophils to aim
superoxide at targets; calcium waves of CD8+ lymphocytes aim
granules at tumor cells
Ignition sites constitute directional memory (e.g., chemotaxis)
System adaptability (e.g., different spatial patterns elicit
different patterns of cytokine production)
Perform mathematical logic operations
Multiple dynamic states allow multiple outputs from the same
number of genes (e.g., eukaryote complexity)
Explains how one signal such as calcium can perform so many
different operations
H.R. Petty / BioSystems 83 (2006) 217–224
example of information distribution, waves have clearly
defined directional properties. Longitudinal NAD(P)H
waves allow neutrophils to aim superoxide production
in the direction of target cells (Kindzelskii and Petty,
2002). In this case the cell surface enzyme NADPH
oxidase directly converted local NADPH into superoxide anions. Similarly, vectorial calcium waves in CD8+
T cells allow these cells to aim their granules at specific antigen-presenting tumor cells (unpublished data).
Again, dynamic wave features are linked to the locations
of the cell physiological outputs. Due to their physical
nature, waves are also characterized by ignition sites. In
neutrophils, one calcium ignition site has been found to
be the lamellipodium. As calcium waves are repetitively
triggered from this site, it displays a rudimentary form
of cell memory. Depending upon the latest direction cue,
this ignition site can reorganize at different locations on
the plasma membrane. Although E. coli also possesses
strategically clustered chemosensory receptors, these
structures do not reorganize at the surface. The adaptability of eukaryotic systems is also illustrated by the ability
of the phagosome-associated calcium wave to disappear
after prolonged incubation, thus ending phagolysosome
formation, and the ability of different calcium wave patterns to elicit different cytokine profiles in T helper cells
(unpublished data). In addition to information transduction and routing, we have proposed that chemical waves
may perform logic operations in living cells, a feature
inconsistent with a uniform signaling paradigm. The
presence or absence of a wave corresponds to logic states
of 1 and 0. As the waves may annihilate, have no effect on
one another, or create new waves of unique properties,
novel intracellular processing events are likely taking
place in certain cells. Thus, the processes underlying
cell behavior more closely resemble the decision-making
processes of a computer than the dynamics of a stirred
chemical reactor.
Chemical waves may also explain several biological
puzzles. As illustrated by our work, chemical waves can
be made of different chemicals, have different ignition
sites, numbers and shapes, travel to different organelles,
possess various speeds, etc. Thus, just one signal can be
described by an enormous number of dynamical states.
As these dynamical states are linked with cell function, one can get similarly large numbers of physiological outputs. Consequently, chemical waves may explain
why eukaryotes are far more complex that prokaryotes
whereas the numbers of genes may be comparable. In
addition, calcium has been linked with a great variety of
cell functions, which has led to the puzzle of how one
signal could perform so many different operations. The
heterogeneity of calcium waves and their spatiotemporal
223
dynamics can certainly account for the breadth of these
responses.
5. Conclusions
Biological thought has not advanced much in the past
50 years. In the 1950s and 1960s biologists were using
electron microscopy to discover new cellular structures
and organelles, such as microtubules, coated pits, endoplasmic reticulum, and the Golgi. Components were
named, studied, categorized and then pigeon-holed. In
the 1980s and 1990s scientists were finding new genes,
sequencing them, knocking them in and out, mutating
them, categorizing them, and still pigeon-holing their
findings. Resources have been spent on a vast scale in
these reductionist frameworks, and they have generated
an equally vast base of knowledge — but one that is
currently of little use to a patient in need of clinical care.
Cell biology’s next step is into the great void between
its current time-scale measured in seconds, and the physicist’s time-scale measured in nanoseconds. By understanding how cells make decisions based upon their
dynamic chemical circuitry, we may be able to trick
them into doing our bidding. For example, it seems
likely that a leukocyte can be fooled into thinking that
its host is pregnant and to disrupt elements of its signaling circuits using Trojan peptides, thereby modifying
host inflammatory programs. Imagine what such a revolutionary re-programming could do in diseases such as
cancer, multiple sclerosis, sepsis (e.g., bacterial infections including bioterrorism) and ischemia-reperfusion
injury (e.g., due to organ transplantation or heart attacks).
By watching intracellular chemistry evolve at the timescale of its characteristic chemical reactions, we may
provide the insight needed to unify the great reductionist
paradigms of the past, and more important, to illuminate
a route to improved drug development and patient care
for tomorrow.
Acknowledgements
This work was supported by NIH grants CA74102
and AI51789. The author wishes to thank Aaron Petty
for assistance in preparing this manuscript.
References
Adamatzky, A., De Lacy, Costello, B., 2002. Experimental logical
gates in a reaction-diffusion medium: the XOR gate and beyond.
Phys. Rev. E 66, 046112-1–046112-6.
Engel, J., Sowerby, A.J., Finch, S.A., Fechner, M., Stier, A., 1995. Temperature dependence of Ca2+ wave properties in cardiomyocytes:
224
H.R. Petty / BioSystems 83 (2006) 217–224
implications for the mechanism of autocatalytic Ca2+ release in
wave propagation. Biophys. J. 68, 40–45.
Falcke, M., 2004. Reading the patterns in living cells — the physics of
Ca2+ signaling. Adv. Phys. 53, 255–440.
Feher, V.A., Cavanagh, J., 1999. Millisecond-time-scale motions contribute to the function of the bacterial response regulator protein
Spoof. Nature 400, 289–293.
Goldbeter, A., 1996. Biochemical Oscillations and Cellular Rhythms.
Cambridge University Press, Cambridge, UK.
Hjelmfelt, A., Weinberger, E.D., Ross, J., 1991. Chemical implementation of neural networks and turing machines. Proc. Natl. Acad.
Sci. U.S.A. 88, 10983–10987.
Huang, J.-B., Kindzelskii, A.L., Clark, A.J., Petty, H.R., 2004. Identification of channels promoting calcium spikes and waves in HT1080
tumor cells: their apparent roles in cell motility and invasion. Cancer Res. 64, 2482–2489.
Huang, J.-B., Espinoza, J., Romero, R., Petty, H.R., 2005. Human
neutrophil transaldolase undergoes retrograde trafficking during
pregnancy, but anterograde trafficking in cells from non-pregnant
women. Metabolism, 54, 1027–1033.
Jaffe, L., 2002. On the conservation of fast calcium wave speeds. Cell
Calcium 32, 217–229.
Kuhnert, L., Agladze, K.L., Krinsky, V.I., 1989. Image processing
using light sensitive chemical waves. Nature 337, 244–247.
Li, Q.-J., Kindzelskii, A.L., Ebert, P.J.R., Zhou, J., Krogsgaard, M.,
Petty, H.R., Davis, M.M., 2005. Calcium signaling cascades during
activation of T cell antigen responses. Nature, Submitted.
Kindzelskii, A.L., Petty, H.R., 2002. Apparent role of traveling
metabolic waves in periodic oxidant release by living cells. Proc.
Natl. Acad. Sci. U.S.A. 99, 9207–9212.
Kindzelskii, A.L., Huang, J.-B., Chaiworapongsa, T., Kim, Y.M.,
Romero, R., Petty, H.R., 2002. Pregnancy alters glucose-6phosphate dehydrogenase trafficking, cell metabolism and oxidant
release of maternal neutrophils. J. Clin. Invest. 110, 1801–1811.
Kindzelskii, A.L., Petty, H.R., 2003. Intracellular calcium waves
accompany neutrophil polarization, fMet-Leu-Phe stimulation and
phagocytosis: a high-speed microscopy study. J. Immunol. 170,
64–72.
Kindzelskii, A.L., Sitrin, R.G., Petty, H.R., 2004a. Optical microspectrophotometry supports the existence of gel phase lipid rafts at the
lamellipodium of neutrophils: apparent role in calcuim signaling.
J. Immunol. 172, 4681–4685.
Kindzelskii, A.L., Ueki, T., Michibata, H., Chaiworapongsa, T.,
Romero, R., Petty, H.R., 2004b. 6-Phosphogluconate dehydrogenase and glucose-6-phosphate dehydrogenase form complexes in
human neutrophils and traffic to the centrosome in pregnant, but
not non-pregnant, women. J. Immunol. 172, 6373–6381.
Malmendal, A., Evenas, J., Forsen, S., Akke, M., 1999. Structural
dynamics in the C-terminal domain of calmodulin at low calcium
levels. J. Mol. Biol. 293, 883–899.
Marmillot, P., Hervagault, J.F., Welch, G.R., 1992. Patterns of spatiotemporal organization in an “ambiquitous” enzyme model. Proc.
Natl. Acad. Sci. U.S.A. 89, 12103–12107.
Nusse, O., Serrander, L., Lew, D.P., Krause, K.H., 1998. Ca2+ -induced
exocytosis in individual human neutrophils: high- and low-affinity
granule populations and submaximal responses. EMBO J. 17,
1279–1288.
Olsen, L.F., Kummer, U., Kindzelskii, A.L., Petty, H.R., 2003. A model
of the oscillatory metabolism of activated neutrophils. Biophys. J.
84, 69–81.
Pencea, C.S., Hentschel, H.G.E., 2000. Excitable calcium wave propagation in the presence of localized stores. Phys. Rev. E 62,
8420–8426.
Petty, H.R., Worth, R.G., Kindzelskii, A.L., 2000. Imaging sustained
dissipative patterns in the metabolism of individual living cells.
Phys. Rev. Lett. 84, 2754–2757.
Petty, H.R., Kindzelskii, A.L., 2000. High-speed imaging of sustained
metabolic target patterns in living neutrophils during adherence. J.
Phys. Chem. B 104, 10952–10955.
Petty, H.R., Kindzelskii, A.L., 2001. Dissipative metabolic patterns
respond during neutrophil transmembrane signaling. Proc. Natl.
Acad. Sci. U.S.A. 98, 3145–3149.
Petty, H.R., 2004a. Applications of high-speed microscopy in biomedical research. Opt. Photonics News, 34–40 (January).
Petty, H.R., 2004b. Dynamic chemical instabilities in living cells may
provide a novel route in drug development. ChemBioChem 5,
1359–1364.
Petty, H.R., 2005. High-speed fluorescence microscopy of cell signaling. In: Paisley, D. (Ed.), SPIE Proceedings on High-Speed
Photography and Photonics. International Society of Optical Engineering 5580, pp. 81–92.
Ridgway, E.B., Gilkey, J.C., Jaffe, L.F., 1977. Free calcium increases
explosively in activating medaka eggs. Proc. Natl. Acad. Sci.
U.S.A. 74, 623–627.
Simon, S.M., Llinas, R.R., 1985. Compartmentalization of the submembrane calcium activity during calcium influx and its significance in transmitter release. Biophys. J. 48, 485–498.
Smolen, J.E., Stoehr, S.J., Boxer, L.A., 1986. Human neutrophils permeabilized with digitonin respond with lysosomal enzyme release
when exposed to micromolar levels of free calcium. Biochim. Biophys. Acta 886, 1–17.
Steinbock, O., Tóth, A., Showalter, K., 1995. Navigating complex
labyrinths: optimal paths from chemical waves. Science 267,
868–871.
Sielewiesiuk, J., Górecki, J., 2002. Complex transformations of chemical signals passing through a passive barrier. Phys. Rev. E 66,
016212-1–016212-9.
Tyson, J.J., Keener, J.P., 1988. Singular perturbation theory of
traveling waves in excitable media (a review). Physica D 32,
327–361.
Volkman, B.F., Lipson, D., Wemmer, D.E., Kern, D., 2001. Two-state
allosteric behavior in a single-domain signaling protein. Science
291, 2429–2433.
Wand, A.J., 2001. Dynamic activation of protein function: a
view emerging from NMR spectroscopy. Nat. Struct. Biol. 8,
926–931.
Worth, R.G., Kim, M.-K., Kindzelskii, A.L., Petty, H.R., Schreiber,
A.D., 2003. Signal sequence within Fc␥RIIA controls calcium
wave propagation patterns: apparent role in phagolysosome fusion.
Proc. Natl. Acad. Sci. U.S.A. 100, 4533–4538.