Arshavsky's Research Projects

Our laboratory studies the molecular and cellular mechanisms of signal transduction. Our favorite experimental model is the vertebrate photoreceptor cell, which is a sensory neuron responsible for detection and primary processing of information entering the eye in the form of photons of light. The basic functions of photoreceptors are to capture photons, to generate a second messenger signal, to translate this signal into a change in electrical activity and, finally, to transmit this information to the secondary neurons in the retina through synaptic release. Because the function of this cell is so well defined and because it is uniquely suitable for study using modern multi-disciplinary approaches, the photoreceptor is an almost unmatched model system for addressing fundamental issues in molecular and cellular neuroscience, as well as in cell signaling in general. At this time we are pursuing three major experimental directions.

I. REGULATION OF G PROTEIN SIGNALING

Our first direction is to understand how the duration of the light-evoked response is regulated on the molecular level. Photoresponse should not only be sensitive but also rapid, which allows it to respond to the next light event. As such, our cones can detect the flickering of a computer monitor operating at a 60-hertz frequency. The molecular mechanisms behind this phenomenon utilize the regulation of G protein signaling in the phototransduction cascade illustrated below. Vision begins when a molecule of a GPCR receptor rhodopsin becomes excited by light. Photoexcited rhodopsin activates multiple molecules of transducin, a photoreceptor specific member of the G protein family. Each activated transducin stimulates the activity of its effector, cGMP phosphodiesterase, which leads to the decrease in intracellular levels of cGMP. This ultimately leads to a reduction in the conductivity of ion channels on the photoreceptor plasma membrane defined as the photoresponse.


The G protein cascade theory of rod phototransduction faced a deep puzzle in the late 1980s. The molecular theory postulated that the hydrolysis of GTP bound to activated transducin was a requisite step in the inactivation of the cascade, but the numbers were way off the mark. The dim flash responses of mammalian rods reach their response peak in about 150 ms and subsequently recover with a time constant of about 200 ms (see Figure below). At the same time, numerous studies of transducin GTPase activity reported turnover numbers of only 1-2 turnovers per minute, nearly 100-fold too slow to account for the physiologically measured responses.

These flash responses were recorded from a rod of a wild type mouse (WT) and a rod of a mouse lacking the GTPase activating complex (R9AP knockout) by Marie Burns and her colleagues at the University of California-Davis.

A solution for this paradox has been found by demonstrating that photoreceptors contain several proteins engaged in activating the slow intrinsic GTPase activity of transducin to a physiologically fast rate. The first GTPase activating protein identified was the g-subunit of the cGMP phosphodiesterase, the effector protein for transducin (Arshavsky and Bownds, Nature 357, 416-417, 1992). However, evidence accumulated rapidly that PDE g alone is insufficient for activating transducin GTPase. Rather, PDE g potentiates the action of another photoreceptor-specific protein which was identified as the ninth member of Regulators of G protein Signaling family, RGS9, by Ted Wensel and his colleagues in 1998. Subsequent studies have shown that RGS9 in photoreceptors is present as a part of a large GTPase activating protein complex that includes a constitutive partner of RGS9, the type 5 b subunit of G proteins, and R9AP (RGS9 Anchoring Protein), which anchors the complex on the surface of the photoreceptor membrane.

The challenge of our current experiments is to define the role of each individual protein subunit of the GTPase activating complex in regulating its catalytic activity, substrate specificity as well as its delivery to photoreceptor outer segments. We are also interested in exploring several possibilities that novel mechanisms revealed in our studies of photoreceptors may reflect more general principles in cell signaling. For example, we developed the concept of “affinity adapters”, which are protein domains or subunits that direct regulatory proteins to their specific physiological targets in different signaling pathways, and found an example where such an adapter regulates the specificity of RGS-G protein recognition in the brain (Neuron, (2003), 38, 857-862. [PDF 202KB]). Another example regards the functional role of the N-terminal domain of RGS9, termed DEP. DEP homology domains are present in over 60 proteins found in the human genome, but in most cases their function remains unknown. We have shown that the binding of RGS9 to R9AP in photoreceptors occurs via the DEP domain and that this binding is essential for RGS9 delivery to photoreceptor outer segments. We are now exploring the hypothesis that the major general function of various DEP domains is to target signaling proteins to specific intracellular compartments.

The DEP domain of RGS9 is essential for its targeting to rod outer segments. The figure shows immunostaining of RGS9 in a wild type retina and in a retina of a transgenic mouse where RGS9 has been replaced by a mutant lacking the DEP domain. The cartoon to the right illustrates the relative positions of the rod cell compartments in the retina sections.

The significance of these studies for understanding general principles of G protein signaling in the nervous system is highlighted by our recent discovery of the brain-specific homolog of R9AP, R7BP (J. Biol. Chem. (2005), 280, 5133-5136. [PDF 198KB]). The studies of R7BP are now continued in the laboratory of our alumnus, Kirill Martemyanov at the University of Minnesota.

II. LIGHT-DEPENDENT TRANSLOCATION OF SIGNALING PROTEINS BETWEEN\
THE FUNCTIONAL COMPARTMENTS OF PHOTORECEPTORS

A fascinating property of photoreceptors is their ability to adjust the protein complement of their outer segment, the cellular compartment where phototransduction takes place, to optimize their ability to function at various levels of ambient illumination. Studies conducted over the past decade revealed that at least two signaling proteins, arrestin and transducin, translocate between the outer and inner segments in a light- dependent manner.

Recently, we introduced a novel experimental approach that allowed us to study protein translocation quantitatively (Neuron (2002) 33, 95-106. [PDF 448KB]). This approach takes advantage of the layered organization of the photoreceptors in the vertebrate retina. It is based on serial tangential sectioning of flat-mounted frozen retina followed by the Western blot analysis of proteins of interest in individual sections.

Schematic diagram depicting the method of serial tangential sectioning of the photoreceptor layer. Marker proteins, rhodopsin (confined to the outer segments), cytochrome C (present in high concentrations in the mitochondria-rich ellipsoid region of the inner segment) and synaptophysin (present in pre-synaptic vesicles), allow the assignment of proteins to specific subcellular compartments of the rod cell.

Applying serial tangential sectioning with Western blotting to the analysis of subcellular transducin distributions in the rods of dark- and light-adapted rats, we found that ~90% of transducin undergoes the light-dependent translocation within only ~ 10 minutes. Collaborating with the laboratory of Edward Pugh, we then found that transducin translocation is adaptive because the reductionin the outer segment transducin concentration is accompanied by a corresponding reduction in the signal amplification of the phototransduction cascade. Thismechanism extends the operating range of photoreceptor sensitivity upon exposures to bright background illumination.

Quantitative analysis of transducin distribution throughout the subcellular compartments of rat rods. A. Western blots of serial tangential sections from dark-adapted (Dark) and illuminated (Light) rat retinas. Rhodopsin (Rho), transducin subunits (G a t and G b t) and cytochrome C (Cyt) -specific protein bands were detected. B. Densitometric profiles of each of the proteins analyzed in the Western blots shown above in A expressed as a % of the total protein in all sections analyzed. C. Cartoon showing the distribution of transducin subunits in the rod photoreceptor cell in dark and light-adapted retinas.

Schematic Representation of Transducin, Arrestin, and Recoverin Subcellular Distribution in the Dark- and Light-Adapted Rod

One of our current directions is to identify the entire set of signaling proteins that undergo light-dependent translocation and to address the functional consequences of these events. We have confirmed previous reports that arrestin, a protein regulating the lifetime of photoexcited rhodopsin, moves to the outer segments upon light adaptation. Furthermore, our quantitative analysis has shown that the degree of arrestin translocation is even more significant than the degree of transducin translocation. The functional role of arrestin translocation is now under investigation using a combination of quantitative biochemical and electrophysiological techniques. We are conducting such experiments in both rods and cones which will allow us to explore whether significant differences in the ability of these cells to adapt to light may be explained by differences in the degree and kinetics of protein re-arrangements. Most recently, we discovered that another signaling protein, a Ca2+ binding protein recoverin, undergoes light-driven translocation as well.

Another direction is to understand the molecular and cellular mechanisms underlying protein translocation. We are interested in learning how these events are triggered by light, how steep concentration gradients of signaling proteins are maintained between the photoreceptor compartments, and whether molecular motors may be involved in each case. A significant progress in this direction has been made in our recent studies of phosducin, a photoreceptor- and pineal-specific representative of a protein family whose members interact with G protein bg subunits. Our data obtained with the phosducin knockout mouse indicate that phosducin facilitates light-dependent transducin translocation, most likely by reducing the membrane affinity of individual transducin subunits (J. Biol. Chem. (2004) 279: 19149 - 19156. [PDF 446KB]).

III. OCULAR PROTEOMICS

A recent acquisition of a state-of-the-art mass spectrometer, the 4700 Proteomics Analyzer from Applied Biosystems, has empowered us to use proteomics approaches to understanding photoreceptor functions. This enables us to rapidly identify unknown interacting partners for the signaling proteins we are studying (R9AP, phosducin, etc.). In the long run, we are interested in expanding this technology to build a proteome map of the photoreceptor cell that will describe protein compositions of all of its individual subcellular compartments in groups representing various functional ensembles (cf. intracellular trafficking, apoptosis, phototransduction, etc.). This map will be derived from exhaustive mass spec identification of proteins present in thin serial tangential sections obtained from the photoreceptor cell layer of the retina (see above). Obtaining such a proteome map would enable us to refine existing hypotheses and to formulate novel hypotheses regarding a broad array of photoreceptor functions and diseases.