Absolute Quantification of Photoreceptor Outer Segment Proteins

Abstract

Photoreceptor cells generate neuronal signals in response to capturing light. This process, called phototransduction, takes place in a highly specialized outer segment organelle. There are significant discrepancies in the reported amounts of many proteins supporting this process, particularly those of low abundance, which limits our understanding of their molecular organization and function. In this study, we used quantitative mass spectrometry to simultaneously determine the abundances of 20 key structural and functional proteins residing in mouse rod outer segments. We computed the absolute number of molecules of each protein residing within an individual outer segment and the molar ratio among all 20 proteins. The molar ratios of proteins comprising three well-characterized constitutive complexes in outer segments differed from the established subunit stoichiometries of these complexes by less than 7%, highlighting the exceptional precision of our quantification. Overall, this study resolves multiple existing discrepancies regarding the outer segment abundances of these proteins, thereby advancing our understanding of how the phototransduction pathway functions as a single, well-coordinated molecular ensemble.

Keywords: absolute protein quantification; photoreceptor; phototransduction; proteomics; retina; vision.

Figure 1.. Cartoon illustrating localization and function of proteins analyzed in this study.

A. Schematic of a rod photoreceptor cell. B. Zoomed-in view of the portion of the outer segment boxed in panel A. Illustrated at the upper disc is the activation of the phototransduction cascade, including photoexcited rhodopsin (R*), transducin (Gα and Gβγ) and PDE6 (PDEαβγ). Two guanylyl cyclase isoforms (GC1 and GC2) responsible for cGMP synthesis are shown on the bottom membrane leaflet. Illustrated at the middle disc are the reactions responsible for cascade deactivation. R* is deactivated through phosphorylation by rhodopsin kinase (GRK1) and subsequent arrestin (Arr) binding; the activity of GRK1 is regulated by recoverin. Transducin deactivation occurs through GTP hydrolysis facilitated by the RGS9/Gβ5/R9AP GTPase activating complex; this returns PDE6 to its inactive state. The plasma membrane harbors the cGMP-gated (CNG) channel containing three α and one β subunits. This channel is open in the dark and closes upon cascade activation by light. The same membrane harbors the Na⁺/K⁺/Ca²⁺ exchanger (NCKX1). Illustrated at the lower disc are two tetraspanin proteins (peripherin-2 and ROM1) fortifying the disc rim and two lipid flippases (ABCA4 and ATP8A2). The drawing is modified with permission from .

Figure 2.. Experimental workflow for absolute quantification of outer segment proteins.

A. Coomassie-stained SDS-PAGE gel of E. coli lysates (50 μl) from uninduced (−) and IPTG-induced (+) cells transformed with the DNA construct encoding the chimeric protein. Induction with IPTG caused a high level of expression of the chimeric protein migrating at the predicted molecular mass of ~143 kDa (arrow). B. General experimental workflow for absolute quantification of outer segment proteins. A preparation of rod outer segments was combined with a known amount of BSA and digested with trypsin on SP3 beads. In parallel, the ~143 kDa chimeric protein band was cut out from the Coomassie-stained SDS-PAGE gel and underwent in-gel trypsin digestion. The two digests were combined and subjected to LC-MS/MS analysis.

Figure 3.. Peptide selection for final protein quantification.

For each peptide representing a protein shown in each panel, the fraction of ion intensity that this peptide contributes to the total ion intensity of all peptides representing this protein was calculated separately for unlabeled and heavy isotope labeled peptide sets. A. Relative abundances of unlabeled and labeled peptides representing RGS9 in a single experiment with technical duplicates. In this case, the relative abundances of each unlabeled and the corresponding labeled peptide matched closely and satisfied the inclusion criterion. B. Relative abundances of unlabeled and labeled peptides representing Gβ₁. The relative abundances of one unlabeled and labeled peptide pair (shown in red) were >2-fold different. The other four peptide pairs satisfied the inclusion criterion. C. Relative abundances of unlabeled and labeled peptides representing peripherin-2. The relative abundances of two unlabeled and labeled peptide pairs were >2-fold different. The other three peptide pairs satisfied the inclusion criterion.

Figure 4.. Cartoon illustrating the amounts of proteins in a section of the photoreceptor disc.

The drawing shows the exact number of molecules for membrane proteins analyzed in this study within the space of one quarter of a photoreceptor disc. Shown is one disc surface and the corresponding portion of the outer segment plasma membrane surrounding this disc. The symbols representing each protein are specified in the figure. Symbol sizes do not reflect the actual dimensions of the proteins they represent; rather, less abundant proteins are shown as larger symbols for ease of visualization. Whereas plasma membrane shares many proteins with discs, only the plasma membrane specific CNG channel and NCKX1 are shown at this membrane. The peripherin-2/ROM1 structural units are shown as dimers as suggested by recent ultrastructural analyses described in the text.