Philosophical Viewpoint: "Doing Cell Science"
J. Lippincott-Schwartz (J. Cell Science [2000] 113, 1499-1500)
As cell biologists, it's important to constantly remind ourselves of the purpose of cell organization and function and to show respect for evolution's ingenuity and subtlety. When faced with unsolved problems, I find it particularly useful to fall back on three basic lessons to help sort through the welter of data and conflicting theories: that cellular functions are historically selected for the purpose of cell reproduction, survival and/or adaptation; that these functions derive from many proteins acting together as functional units and cannot be understood merely by analyzing the details of individual proteins; and that progress and productivity in cell biology relies on overlappping and reinforcing data from many perspectives and not a single isolated line of inquiry.
Lessons from cellular evolution. Unlike inanimate systems, the functional systems of the cell are selected for a purpose and have an unbroken history stretching back to the origin of life. New functions arise only by tinkering with already existing biological machinery rather than being built from scratch. It is useful when studying any new process or phenomena, therefore, to ask what other functions it might have evolved from or be related to. One example that comes to mind is membrane resealing, a phenomenon which only recently has been characterized (McNeil and Steinhardt, 1997). The plasma membrane of cells is constantly being torn or disrupted and must be resealed. There is no doubt that membrane resealing is an ancient cellular function since without it cell content and integrity could not be maintained. Recent studies have shown that resealing involves Ca+- mediated fusion of endosomal membranes with the plasma membrane rather than plasma membrane expansion to fill in a perforated region (Terasaki et al., 1997). This finding raises the question of whether other Ca+ mediated processes of fusion with the plasma membrane, including synpatic vesicle fusion (and synaptic vesicle biogenesis from endosomes), have evolved as specialized adaptations of the more basic wound healing process. By asking how different processes are related to each other and to the historical evolution of cells, fundamental properties of a system often can be identified and used to generate testable hypotheses.
Emergent cellular properties. As cell biologists we have
been pre- occupied with assigning functions to proteins or genes
and with reducing biological phenomena to the behavior of individual
molecules. Despite the seeming success of this approach, it has
significant limitations for understanding many cell biological processes
(Hartwell et al., 1999). A discrete biological function can only
rarely be attributed to an individual molecule. Most biological
functions instead arise from interactions among many components
and often in a manner that can't be predicted from the activities
of individual components. An example is the functioning of the actin
cytoskeleton, which arises from interactions among an enormous array
of interacting molecules that regulate the kinetics of actin assembly/disassembly
at various cell locations (Theriot, 2000). In so doing, they allow
rapid transitions to occur between polmerized/depolymerized states
of actin, which in turn underlies directed cell or organelle movement.
The collective properties of this system, rather than any single
component, underlies its function. To understand the functioning
of the actin cytoskeleton, therefore, we need to know what drives
the rapid transitions between stable states of actin, its kinetic
parameters, and how any particular state is maintained. This effort
will require new methodologies for quantitative description and
modeling of the activities of large numbers of interacting molecules.
An appreciation of physio-chemical properties of cells is vital
for understanding how cell structure is organized and maintained.
The emergent properties of actin and tubulin as flexible, regulative
polymeric arrays, for example, likely underlies the evolutionary
selection of these proteins in so many activities of the cytoskeleton.
Cells exist far from equilibrium, harvesting energy from their environment.
Many cellular structures, therefore, are self-organizing systems
that are driven by changes in free energy. These include the mitotic
spindle apparatus, and polymerized arrays of actin and tubulin.
Less recognized as such is the Golgi complex, which serves as a
protein sorting and processing station in secretory traffic. Recent
studies characterizing Golgi disassembly/reassembly during mitosis
and in response to cellular perturbants in living cells (Storrie
et al., 1998; Shima et al., 1998; Zaal et al., 1999), however, are
consistent with a self-regulatory nature of this organelle. For
membrane organelles like the Golgi complex, it is the physical properties
of their membranes, in addition to protein-protein interactions,
that underlie their capacity for protein sorting and retention.
Such properties include membrane curvature induced by specific types
of lipids, lipid and protein partioning into microdomains, and tension-driven
membrane flow (Mui et al., 1995; Bretscher and Munro, 1996; Harder
et al., 1997; Sciaky et al., 1997). Because these properties as
a whole facilitate the Golgi's most fundamental functional property
(i.e., protein sorting and retention), better methods need to be
devised for characterizing such properties and their relationship
to the protein machineries required for secretory transport.
Multidisciplinary approaches. Because cells are enormously
complex with many interacting components, it is important to be
wary of conclusions or models drawn from a single line of inquiry.
Models for cell function must be reinforced by results from multiple
perspectives and using a wide repertoire of techniques, including
molecular biology, biochemical genetics and imaging at high and
low resolution. There's an understandable tendency for researchers
to be tempted to extrapolate to grand conclusions. Part of the work
of being a scientist, however, is to be disciplined enough to verify
hypotheses through a multitude of techniques and approaches. This
way a particular hypothesis can be linked to observations in other
systems, creating a net of related observations and proposed functions
that together explain large sets of data. The mere acquisition of
data from a single approach suggesting an alternative function,
in isolation from other avenues of investigation, can be misleading
and needs to be verified using a wide range of investigative tools.
Looking to the future, my sense is that making and testing quantitative
predictions about cell behavior will be crucial for verifying our
models of cell function. This will require integrating many experimental
approaches such as in vitro reconstitution, in situ imaging
and mathematical modeling in order to connect different levels of
analysis- from molecules, through functional molecular assembles,
to cells. In this way, emergent principles that govern the structure
and behavior of cells, and their evolutionary constraints, can be
understood and used for addressing the cell's role in development,
health and disease.