Proteomic Profiling of a Layered Tissue Reveals Unique Glycolytic Specializations of Photoreceptor Cells

The retina is a highly ordered tissue whose outermost layers are formed by subcellular compartments of photoreceptors generating light-evoked electrical responses. We studied protein distributions among individual photoreceptor compartments by separating the entire photoreceptor layer of a flat-mounted frozen retina into a series of thin tangential cryosections and analyzing protein compositions of each section by label-free quantitative mass spectrometry. Based on 5038 confidently identified peptides assigned to 896 protein database entries, we generated a quantitative proteomic database (a “map”) correlating the distribution profiles of identified proteins with the profiles of marker proteins representing individual compartments of photoreceptors and adjacent cells. We evaluated the applicability of several common peptide-to-protein quantification algorithms in the context of our database and found that the highest reliability was obtained by summing the intensities of all peptides representing a given protein, using at least the 5–6 most intense peptides when applicable. We used this proteome map to investigate the distribution of glycolytic enzymes, critical in fulfilling the extremely high metabolic demands of photoreceptor cells, and obtained two major findings. First, unlike the majority of neurons rich in hexokinase I, but similar to other highly metabolically active cells, photoreceptors express hexokinase II. Hexokinase II has a very high catalytic activity when associated with mitochondria, and indeed we found it colocalized with mitochondria in photoreceptors. Second, photoreceptors contain very little triosephosphate isomerase, an enzyme converting dihydroxyacetone phosphate into glyceraldehyde-3-phosphate. This may serve as a functional adaptation because dihydroxyacetone phosphate is a major precursor in phospholipid biosynthesis, a process particularly active in photoreceptors because of the constant renewal of their light-sensitive membrane disc stacks. Overall, our approach for proteomic profiling of very small tissue amounts at a resolution of a few microns, combining cryosectioning and liquid chromatography-tandem MS, can be applied for quantitative investigation of proteomes where spatial resolution is paramount. The retina is a highly ordered tissue whose outermost layers are formed by subcellular compartments of photoreceptors generating light-evoked electrical responses. We studied protein distributions among individual photoreceptor compartments by separating the entire photoreceptor layer of a flat-mounted frozen retina into a series of thin tangential cryosections and analyzing protein compositions of each section by label-free quantitative mass spectrometry. Based on 5038 confidently identified peptides assigned to 896 protein database entries, we generated a quantitative proteomic database (a “map”) correlating the distribution profiles of identified proteins with the profiles of marker proteins representing individual compartments of photoreceptors and adjacent cells. We evaluated the applicability of several common peptide-to-protein quantification algorithms in the context of our database and found that the highest reliability was obtained by summing the intensities of all peptides representing a given protein, using at least the 5–6 most intense peptides when applicable. We used this proteome map to investigate the distribution of glycolytic enzymes, critical in fulfilling the extremely high metabolic demands of photoreceptor cells, and obtained two major findings. First, unlike the majority of neurons rich in hexokinase I, but similar to other highly metabolically active cells, photoreceptors express hexokinase II. Hexokinase II has a very high catalytic activity when associated with mitochondria, and indeed we found it colocalized with mitochondria in photoreceptors. Second, photoreceptors contain very little triosephosphate isomerase, an enzyme converting dihydroxyacetone phosphate into glyceraldehyde-3-phosphate. This may serve as a functional adaptation because dihydroxyacetone phosphate is a major precursor in phospholipid biosynthesis, a process particularly active in photoreceptors because of the constant renewal of their light-sensitive membrane disc stacks. Overall, our approach for proteomic profiling of very small tissue amounts at a resolution of a few microns, combining cryosectioning and liquid chromatography-tandem MS, can be applied for quantitative investigation of proteomes where spatial resolution is paramount. The vertebrate retina is a layered tissue containing several types of neurons and supporting cells (1.Rodiek R.W. The Vertebrate Retina. Principles of Structure and Function. Freeman, San Francisco1973Google Scholar). The outermost half of the retina is formed by rod and cone photoreceptors, which produce electrical responses upon capturing photons. Photoreceptors are polarized neurons in which specific functions are carried out in individual highly specialized subcellular compartments. For example, the processes of light capturing and visual signaling are confined to the distal outer segment, whereas information transfer to the secondary retina neurons takes place at the synaptic terminal located at the opposite side of the cell. In this study, we analyzed the patterns of subcellular protein distribution in photoreceptors by label-free quantitative mass spectrometry. Proteomic analysis of layered or intricately structured tissues requires a difficult analytical combination of reproducible sample preparation with high spatial resolution, significant depth of proteome coverage, and accurate quantification. These challenges have been addressed to varying degrees by laser-capture microdissection along with MS analysis and matrix assisted laser desorption ionization (MALDI) 1The abbreviations used are:MALDImatrix-assisted laser desorption ionizationICCintraclass correlation coefficientDDAdata-dependent analysisLC-MS/MSliquid chromatography-tandem MSCRALBPcellular retinal aldehyde binding proteinPVDFpolyvinylidene fluoridePLGSprotein lynx global server. imaging. Laser capture microdissection is typically performed following immunostaining for specific markers to define the regions of interest (2.Umar A. Luider T.M. Foekens J.A. Pasa-Tolic L. NanoLC-FT-ICR MS improves proteome coverage attainable for approximately 3000 laser-microdissected breast carcinoma cells.Proteomics. 2007; 7: 323-329Crossref PubMed Scopus (58) Google Scholar, 3.Moulédous L. Hunt S. Harcourt R. Harry J. Williams K.L. Gutstein H.B. Navigated laser capture microdissection as an alternative to direct histological staining for proteomic analysis of brain samples.Proteomics. 2003; 3: 610-615Crossref PubMed Scopus (53) Google Scholar, 4.Ornstein D.K. Gillespie J.W. Paweletz C.P. Duray P.H. Herring J. Vocke C.D. Topalian S.L. Bostwick D.G. Linehan W.M. Petricoin 3rd, E.F. Emmert-Buck M.R. Proteomic analysis of laser capture microdissected human prostate cancer and in vitro prostate cell lines.Electrophoresis. 2000; 21: 2235-2242Crossref PubMed Scopus (221) Google Scholar, 5.Zang L. Palmer Toy D. Hancock W.S. Sgroi D.C. Karger B.L. Proteomic analysis of carcinoma of the breast using laser capture and 3: PubMed Scopus Google Scholar, Paweletz C.P. Petricoin 3rd, E.F. L. Proteomic to the and of 2003; PubMed Scopus Google Scholar, Paweletz C.P. Sgroi D.C. Petricoin 3rd, E.F. to proteomic analysis of breast PubMed Scopus Google this approach has very high resolution the of for MS analysis is small because of the of tissue has the of high and of sample but spatial resolution of in protein and depth of proteome coverage D. A. A. of on thin histological in proteome PubMed Scopus Google Scholar, A. A. S. L. A. by mass is a for in PubMed Scopus Google Scholar, R. J. D. R. and profiling for and of the is a cancer 2007; PubMed Scopus Google Scholar, R. A. R. analysis and of tissue 2007; PubMed Scopus Google Scholar, R. J. D. R. in for proteomic 2007; PubMed Scopus Google Scholar, using a of of PubMed Scopus Google Scholar). laser desorption ionization correlation analysis liquid chromatography-tandem MS retinal aldehyde binding protein protein lynx global server. quantitative label-free liquid chromatography-tandem MS is a of protein when the and approach for of peptides the E.F. using accurate mass and with mass 2003; PubMed Scopus Google Scholar, L. Proteomic using an accurate mass and PubMed Google Scholar, R. S. proteomic analysis by accurate mass PubMed Scopus Google Scholar, D.G. proteomic of the using an accurate mass and PubMed Scopus (53) Google Scholar, in analysis and using an accurate mass and PubMed Scopus Google Scholar, of for in using PubMed Scopus Google Scholar, for the of peptides for accurate mass and PubMed Scopus Google Scholar). This requires mass for and for and are to the a for PubMed Scopus Google Scholar, a for proteomic PubMed Scopus Google Scholar, J. high mass and protein PubMed Scopus Google Scholar, Hunt of protein by mass 7: PubMed Scopus Google Scholar). The label-free approach has the of with to of or of that for sample preparation are quantitative and for is of the highly ordered retina we obtained thin tangential the outer which proteins of the photoreceptor cells. the retina is by rod photoreceptors by A. cone types of retina by PubMed Scopus Google the majority of proteins in In the we this with of individual proteins in which an alternative to to protein distributions in cells of the two major compartments of rod a of light PubMed Scopus Google Scholar). This several the subcellular of which of of the two major compartments of rod a of light PubMed Scopus Google Scholar, in rod photoreceptors. the PubMed Scopus Google Scholar, J. in photoreceptor and the of 2003; PubMed Scopus Google Scholar). We this to distribution profiles of of proteins by combining of the retina with quantitative mass spectrometry. The are in a database we a proteome correlating the of all identified proteins in individual with of marker proteins representing specific compartments of photoreceptors and cells. We used this map to the subcellular distribution of all glycolytic and patterns for the two hexokinase and triosephosphate isomerase, each the high metabolic activity of photoreceptor cells. for at least and light or following to light on the as in S. in is by of the 2007; PubMed Scopus Google Scholar). was performed as in of the two major compartments of rod a of light PubMed Scopus Google with in S. in is by of the 2007; PubMed Scopus Google and is at a critical of visual signaling and is to PubMed Scopus Google Scholar). an and in retina in was the with a to the disc that the the analyzed membrane with the photoreceptor layer flat-mounted two by and frozen on tangential using a following the retina with the of the The of the retina and analysis was performed with two of retina obtained a and a as and the retina section was in II by and for with and with with at and II was by to the of and in and The of protein in a section as was as by in retina to analyzed by protein performed in used for MS in to the of for We this protein in to in and to quantitative of the obtained each of the each where analyzed using a to a mass of in was at for in with on a performed on a with using a of to with at a of and We two of each retina section for quantification and of obtained performed in and used and high to The quantitative by an in the analysis for each using a MS by on the with for each used an of a of and for For and label-free of individual peptides all sample we the with in a similar to several Hunt of protein by mass 7: PubMed Scopus Google Scholar, C.P. A. Hunt S. of an to markers in by high resolution mass PubMed Scopus Google Scholar, A. of PubMed Scopus Google Scholar, R. S. J. J.A. J.A. T.M. J. J. J. using PubMed Scopus Google Scholar, A. and proteomes in human human analysis by label-free mass PubMed Scopus Google Scholar, Paweletz C.P. analysis of by mass 7: PubMed Scopus Google Scholar, Paweletz C.P. analysis of using and mass 2007; PubMed Scopus Google Scholar, L. proteins using in and label-free quantitative PubMed Scopus Google Scholar). and intensities to the precursor mass in the used to the and protein on a analytical intensities for each by following and to for each of the photoreceptor layer We and to For in and to and the in an For was used to that to the The and of precursor and with PubMed Scopus Google Scholar, D. and of precursor and the analysis of and PubMed Scopus Google into global all using the and algorithms in using database A. R. to the of by and database PubMed Scopus Google Scholar, A. R. for proteins by mass 2003; PubMed Scopus Google Scholar). with this to a are in identified was to be assigned to a protein and by to the of and are in and the database with with database for and using The database mass was for and and was for and for was to for and for and a of two was was as a and and and of and was in peptides quantitative section of the upon visual to have where significant of the and this was for the was performed the by using the to the for and analysis to of The to was evaluated for The at all was with intensities for each sample and to a analysis to for analytical This analysis that the by the and in the each retina the reliability of several peptide-to-protein quantification algorithms we the correlation which is a used to correlation among of similar obtained by or in PubMed Scopus Google Scholar). The of among the of individual is by the correlation The and their with the using the The retinal is the sample to sample in the retina may in layer and small in tissue of the two major compartments of rod a of light PubMed Scopus Google Scholar). In the of the section be on the protein quantification obtained two an of protein distribution profiles in obtained each of at least two of and the section be This was using as a protein, because it is an protein the entire photoreceptor and the cell types in rod photoreceptors. the PubMed Scopus Google Scholar, C.D. C.D. The and of and protein in the Google Scholar). The distribution profiles of as the of the obtained each retina by in the and at the a of this S.L. for 3: PubMed Scopus Google Scholar). We the for the and the of this to the and We used of the to a to each we for the that the in retina to that in retina where is the of this at a the We the of and by the following that and is the The is because the the two section be a The is because the and sample and in our and their to and to are most to the The This to define the in as a the distal of the photoreceptor cell. and have been in the of to the database the with the following retina and retina retina in a of proteins in the retina was performed as R. at PubMed Scopus Google all performed in in and in and of the retina with a with and in with and with secondary with and using a hexokinase and hexokinase II by of and triosephosphate and in with the secondary the patterns of protein distribution along the of the photoreceptor we the entire photoreceptor layer a frozen flat-mounted retina into tangential and the of individual proteins in each section by is in The analysis of obtained two individual a of 5038 peptides to 896 protein database This for all peptides each which used for protein quantification. are several to information in intensities in into protein in each on the of peptides representing each protein in Paweletz C.P. analysis of using and mass 2007; PubMed Scopus Google Scholar, II R. analysis of membrane proteins in and using a label-free PubMed Scopus Google Scholar, R. D. and quantitative analysis of the a PubMed Scopus Google Scholar, protein quantification using or mass and with PubMed Scopus Google the most intense peptides quantification of proteins by a of MS PubMed Scopus Google or all peptides identified for a given protein S. quantitative using generated by liquid and mass PubMed Scopus Google Scholar, A. S. S. an for on proteome and to the analysis of 2007; PubMed Scopus Google Scholar, S. R. for combining intensities to protein PubMed Scopus Google Scholar). evaluated the of protein quantification algorithms in to our the protein of each are several protein to a constant in all where are These are proteins confined to specific which constant of their of their in We applied quantification algorithms to the distribution profiles of proteins representing two the rod outer and the we the for the most intense each protein as distribution among all we the for the most intense peptides representing a given we the for all peptides a given we the of each in each section and the among all peptides representing a protein of that in algorithms and peptides intense to the with whereas in each an to the The reliability of protein quantification obtained using each was by the profiles of proteins representing each We used the correlation analysis in PubMed Scopus Google an in which the of among the individual protein profiles was by the to The protein proteins to rod outer and indeed each protein quantification distribution profiles in the several most distal to rod outer on the most intense a of among individual proteins the for each other we with the obtained with summing the intensities of all peptides a given protein and the other two of and for algorithms and this analysis for obtained the retina correlation analysis of protein distribution profiles obtained by peptide-to-protein quantification of and performed for proteins each retina using the as in retina of and performed for proteins each retina using the as in retina retina to rod outer upon was this retina proteins intense of most intense of all of all proteins intense of most intense of all of all of and performed for proteins each retina using the as in to rod outer upon was this in a We a of proteins distribution profiles in photoreceptors are the majority in the and a in the synaptic terminal by for the among the protein quantification algorithms was the the was obtained by summing the intensities of all algorithms and and the was obtained using the We addressed the reliability of the of protein quantification on the of peptides used in The of and proteins in each section and most intense peptides into and the as an with by a and and at which to be an for this at least for the of tissue and used in our These are with by S. R. for combining intensities to protein PubMed Scopus Google whose analysis and using a protein that the of accurate of protein the of approach this to an of a We that the to protein quantification algorithms by correlating profiles of proteins to in in is to of We used the of intensities to distribution profiles for all identified proteins in each retina we on the of peptides or we that the of protein profiles obtained on the of or few peptides particularly in reliability of this we the protein profiles two This for in tissue upon and of each two of tissue and of the section be This was by the distribution profiles of a marker protein on the of high in photoreceptors and distribution the entire photoreceptor in rod photoreceptors. the PubMed Scopus Google Scholar, C.D. C.D. The and of and protein in the Google Scholar). of each that the photoreceptor layer in retina was in retina and to a section for we profiles as and applied a to the as the the to the to other proteins very as in for two proteins and in of the photoreceptor cell. In all we protein profiles in the of retina We the of major photoreceptor compartments in the proteome map using protein markers highly In this approach is similar to that of by on a map by protein correlation PubMed Scopus Google Scholar). For example, we generated a for rod outer by the individual profiles of proteins in the that this is a of the distribution profiles for other proteins to be confined to outer We that a small of signaling most and their distributions the outer and the of the photoreceptor cell in a with into and out of the outer of signaling proteins in vertebrate PubMed Scopus Google Scholar). a analysis of this was a of the study, the distribution patterns of and in with obtained by other as an that the protein distribution the two similar approach was used to for proteins representing photoreceptor the layer and the synaptic generating and synaptic was a on the of proteins representing each generating a for the was because most proteins are in other of the cell. is a protein to the of the J. a protein, is a of the PubMed Scopus Google Scholar). The distribution of was similar to the of protein but unlike most proteins a at the synaptic terminal other proteins in our database have profiles similar to we the of the and profiles of all other database The proteins on the of at least two peptides and at least with are in most because of their in several other most the and The was to be in photoreceptors S. distribution of and in retina and retinal PubMed Scopus Google and the is to be a of the matrix is a membrane protein with for and by 2007; PubMed Scopus Google Scholar). the highly patterns in have been for the other the of was to be to the R. and polarized distribution of and in photoreceptor PubMed Scopus Google and this was indeed in example, hexokinase be analyzed in In to photoreceptors, the outer retina of two cell the cells and the retinal The processes of the into the rod and cone outer several proteins the found in the two of our map representing the outer cells their cell the entire retina to the photoreceptor and the outer their thin into the several proteins in our distribution profiles this protein, the retinal aldehyde binding protein is to be in cells and the of two proteins in vertebrate PubMed Scopus Google in the of but other proteins cells an small in the Photoreceptors are among the most metabolically active cells in the In the to the and for synaptic In the significant to by rod photoreceptors in and in PubMed Scopus Google Scholar, J. R. L. T.M. of in the outer retina in and in PubMed Scopus Google for has been that a critical in fulfilling in photoreceptor outer in and in PubMed Google Scholar). the outer retina the the layer of cells with the of is in cells of PubMed Scopus Google Scholar, of the in cells of the in the Google Scholar). The is used for into photoreceptor cells, where it is in and synaptic but outer as in our 3rd, is rod outer PubMed Scopus Google Scholar). We used the photoreceptor proteome map to investigate the of by analyzing distribution patterns for the entire of glycolytic was unlike the majority of photoreceptors express hexokinase II in to the found hexokinase A. R. J. D. and hexokinase and of in PubMed Scopus Google Scholar). the distribution profiles of each hexokinase for the majority of glycolytic The profiles an distribution the entire of photoreceptors and cells. the distribution of for a in containing significant amounts of cells, with proteins in adjacent cell types as In the profiles for each hexokinase a in the segment, their of of hexokinase in on the and retina of the PubMed Scopus Google in Hexokinase binding to a for PubMed Scopus Google Scholar, to to the and a for PubMed Scopus Google Hexokinase a in whereas hexokinase II was confined to the the of on we retina with each hexokinase and found that each staining proteomic Hexokinase was most the photoreceptor it was found in photoreceptor A. R. J. D. and hexokinase and of in PubMed Scopus Google In all hexokinase II was found in with a very small in photoreceptor and staining in other retinal cells For we retina with an the proteomic of which was of the majority of glycolytic The staining for hexokinase and was indeed with an distribution in photoreceptors and other retina cells. The of hexokinase II and the for the of in photoreceptors. the in of to produce is for because the Hexokinase is in whereas hexokinase II is typically found in and cancer cells to be particularly on in PubMed Google Scholar, A. and of the PubMed Scopus Google The is to the catalytic activity of hexokinase II and PubMed Google Scholar). that photoreceptors, but other cells in the express hexokinase II that this in to with the high of We that the of hexokinase in the outer outer are of the 3rd, is rod outer PubMed Scopus Google Scholar). by hexokinase is to the cell because the cell the direct into the outer of in the cell in the segment, where it is into by and can in the and outer The glycolytic enzyme with a the majority of this was triosephosphate proteomic a outer to that the of this enzyme in photoreceptors is very with other cells. a staining in the we staining in outer and little staining the outer membrane and the a with staining of cells photoreceptors The of triosephosphate in is to dihydroxyacetone phosphate into are upon the of by of triosephosphate in photoreceptors is to in an of dihydroxyacetone phosphate with other retina cells. we can the functional of a is that it to phospholipid biosynthesis, as dihydroxyacetone phosphate as a precursor of the phospholipid The dihydroxyacetone phosphate for in and PubMed Scopus Google Scholar). Photoreceptors at an high because of the constant renewal of their outer and the outer of vertebrate PubMed Scopus Google Scholar, J. The cell of J. Scholar). The of in a retina the of the by T.M. D. and outer membrane in the retina of the for and photoreceptor PubMed Scopus Google Scholar). This it that an dihydroxyacetone phosphate this a in to glycolytic in the outer we that our a that photoreceptors, retinal and cells R. of in retinal and J. PubMed Scopus Google Scholar). We identified peptides this protein, that distribution the majority of glycolytic and found that other is in the outer retina the of our The in this the of combining label-free quantitative with subcellular of a layered tissue by Based on this we generated a proteome map the distribution profiles of of proteins representing the photoreceptor layer of the We used this map to the subcellular of glycolytic and patterns of and all by photoreceptor cells to their high and metabolic This is to be the our protein which is in a that can be by other other of photoreceptor approach to the photoreceptor proteome that used in several on analyzing the protein of rod outer the the of the retina by The proteome of the photoreceptor 2007; PubMed Scopus Google Scholar, of photoreceptor outer a of and proteins in membrane and 7: PubMed Scopus Google Scholar, L. A. S. D. A. A. A. Proteomic analysis of the retinal rod outer 7: PubMed Scopus Google Scholar). to proteomes of the photoreceptor cell by of the retina layers J. S.L. for retinal the of proteins a photoreceptor PubMed Scopus Google tangential containing the entire photoreceptor layer L. S. of retinal by and proteome Google or by analyzing the proteomes of the and in which D. S. A. J.A. proteomic analysis of the the of proteins by retinal in the 2003; PubMed Scopus Google Scholar). of thin to protein and quantification in all the individual subcellular compartments of cells, each by of marker approach is proteomes of which contain to the proteins in an of interest and For example, each rod outer proteome of and their outer The proteome of the photoreceptor 2007; PubMed Scopus Google Scholar, of photoreceptor outer a of and proteins in membrane and 7: PubMed Scopus Google Scholar, L. A. S. D. A. A. A. Proteomic analysis of the retinal rod outer 7: PubMed Scopus Google Scholar). representing the outer in our of but unlike in the outer proteins and other compartments be on the of each distribution we to that the approach in this can serve as a for generating similar proteome protein distribution profiles in layered a of protein quantification and for the of tissue In the retina is among the layered tissues in our the approach to tissues is to protein coverage, a similar be very in the proteomic investigation of where quantification of proteins with high spatial resolution at a is for the of We are to A. for and S. for the with