Differences in Hippocampal Protein Expression at 3 Days, 3 Weeks, and 3 Months Following Induction of Perinatal Asphyxia in the Rat Rachel Weitzdörfer,†,§ Harald Höger,‡,§ Gudrun Burda,† Arnold Pollak,† and Gert Lubec*,† Department of Pediatrics, Medical University of Vienna, Waehringer Guertel 18, 1090 Vienna, Austria, and Division for Laboratory Animal Science and Genetics, Medical University of Vienna, Brauhausgasse 34, 2325 Himberg, Austria Received December 10, 2007
Perinatal asphyxia (PA) was induced in Sprague–Dawley rats; pups were sacrificed 3 days, 3 weeks, and 3 months following the asphyctic insult, and hippocampal protein levels were determined by a gel-based proteomic method. Levels of antioxidant, metabolic, cytoskeleton, signaling, channel, proteasomal, chaperone, splicing, and synaptosomal proteins were dysregulated depending on the age following induction of PA. These proteins are proposed to be potential markers or pharmacological targets for PA. Keywords: Perinatal Asphyxia • Hippocampus • Rat • Protein Dysregulation
Introduction A series of individual protein derangements have been reported in brain of animal models of perinatal asphyxia (PA), and a series of proteins from peripheral compartments were studied in Man and animals.1–9 Only one systematic study at the nucleic acid level has been carried out, however, in order to search for transcripts differentially expressed in brain of rodents with PA.10 Deficient protein synthetic machinery has been shown in brain of rats with PA in the early phase11 lasting for at least 8 days following the asphyctic insult.12 Deficient protein synthesis in terms of impaired RNA polymerase levels and activity would lead to expressional changes of brain proteins, thus, forming the rationale for designing a study to monitor protein levels at different time points following induction of PA. And indeed, brain protein derangement could be expected as a long-term study revealed longlasting effects of PA on brain protein levels in the rat still at the end of the life span.13 As brain protein expression varies enormously between brain areas, we selected the hippocampus as a brain area that was permanently affected and accompanied by cognitive and behavioral changes.14,15 The animal model of PA used herein was selected as this rat model is well-documented in terms of morphology, metabolism, and neurochemistry, and furthermore, asphyxiated pups can survive and be studied for the whole life span.16–20 The advent of high-throughput proteomics techniques allows generation of relatively large brain proteomes21–23 that can be reliably compared between groups and are enabling concomitant and unambiguous identification and characterization of * Corresponding author: Univ. Prof. Dr. Gert Lubec, Medical University of Vienna, Dept. of Pediatrics, Waehringer Guertel 18, A 1090 Vienna, Austria. Phone, +43 1 40 400 3215, fax, +43 1 40 400-6065, e-mail, gert.lubec@ meduniwien.ac.at. † Department of Pediatrics, Medical University of Vienna. ‡ Division for Laboratory Animal Science and Genetics, Medical University of Vienna. § These authors contributed equally to the project. 10.1021/pr700835y CCC: $40.75
2008 American Chemical Society
proteins. It was the aim of the study to search for differentially expressed hippocampal proteins that may be used as candidate markers for PA as representatives of individual pathways, cascades, and protein networks involved and representing pharmacological targets. Although a handful of markers for PA including S100,24 CKBB,25 interleukin-6, and neuron-specific enolase26,27 have been tested in humans, novel and more specific candidate markers for the detection of neuronal damage have to be developed; S100 is an astroglial marker, interleukin-6 is not directly related to neuronal structures, and neuron-specific enolase can be observed in many tissues using proteomic techniques.28,29 Likewise, many cascades have been incriminated to participate in pathomechanisms of PA30 including signaling, cytoskeleton, antioxidant, nitric oxide synthetising proteins, and apoptosisrelated proteins, but proteins of asphyxia-specific pathways remain to be identified. Finally, identification of tentative pharmaceutical targets has not been systematically envisaged. With the use of stringent statistical evaluation, proteins from the protein synthetic, chaperoning, and degradation machinery; oxygen and intermediary metabolism; cytoskeleton; signaling; and synaptic apparatus were differentially expressed in PA at different time points.
Experimental Section Animals. Female Sprague–Dawley rats (Institute of Animal Breeding, Medical University of Vienna, Himberg, Austria) were housed in groups of up to six per cage. Rats were maintained on 11/13 h light/dark cycle in a temperature (21 ( 1 °C) and humidity (50 ( 10%) controlled and well-ventilated room with access to food and drink ad libitum. The animals were bred and kept under specific pathogen free (SPF) conditions, and all of the experiments were carried out in accordance with the rules of the American Physiology Society.16 Experiments were approved by the Medical University of Vienna’s Committee for Animal Experiments and the Ministry of Science (certificate GZ66.009/80-BrGT/2003). Journal of Proteome Research 2008, 7, 1945–1952 1945 Published on Web 03/21/2008
research articles Induction of Perinatal Asphyxia. PA was induced as described elsewhere.31,32 Asphyxia was induced in pups delivered by Caesarian section on pregnant Sprague–Dawley rats. Within the last day of gestation as evaluated by estabularium protocols, animals were anesthetized with diethyl ether and hysterectomized. The uterus horns, still containing the fetuses, were extirpated and placed in a water bath at 37 °C for 20 min following removal of pups that served as normoxic controls. Following the asphyctic period, that is, incubation at 37 °C, the uterus horns were rapidly opened and pups removed. Pups were cleaned, the umbilical cord was ligated, and the animals were allowed to recover on a heating pad. Pups were readily accepted and nourished by the foster mothers. Only litters with pups weighing more than 4.5 g were used for the experiments. Animals from the offspring (n ) 10 per group), normoxic and asphyctic, at 3 days, 3 weeks, and 3 months (the rationale to select these three age groups was that these are widely used for biochemical, genetic, and pharmacological studies) were sacrificed by decapitation, brains were rapidly removed, and complete hippocampal tissue was taken within 1 min, snap-frozen, and stored at -80 °C until chemical analysis, and the freezing chain was never interrupted. Sample Preparation. Pooled left and right hippocampal tissue samples (n ) 10 hippocampi per group), from 10 rats that were not littermates, were homogenized and suspended in 1.8 mL of sample buffer consisting of 8 M urea (Merck, Darmstadt, Germany), 2 M thiourea (Sigma, St. Louis, MO), 4% CHAPS (3-[(3-cholamidopropyl) dimethylammonio]-1-propanesulfonate) (Sigma, St. Louis, MO), 65 mM 1,4-dithioerythritol (Merck, Germany), 1 mM PMSF, and 0.5% carrier ampholytes Resolyte 3.5–10 (BDH Laboratory Supplies, Electran, England). Samples were left at 21 °C for 1 h and then centrifuged at 14 000 × g for 60 min, and the supernatant was transferred into Ultrafree-4 centrifugal filter units (Millipore, Bedford, MA) for desalting and concentrating proteins.33 Protein content of the supernatant was quantified by the Bradford protein assay system.34 Two-Dimensional Gel Electrophoresis (2-DE). 2-DE was performed as reported.35 Samples of 800 µg of protein were applied on immobilized pH 3–10 nonlinear gradient strips in sample cups at their basic and acidic ends. Focusing was started at 200 V, and the voltage was gradually increased to 8000 at 4 V/min and then kept constant for a further 3 h (approximately 150 000 Vh totally). After the first dimension, strips (18 cm) were equilibrated for 15 min in the buffer containing 6 M urea, 20% glycerol, 2% SDS, and 2% DTT, and then for 15 min in the same buffer containing 2.5% iodoacetamide instead of DTT. After equilibration, strips were loaded on 9–16% gradient sodium dodecyl sulfate-polyacrylamide gels for second-dimensional separation. The gels (180 × 200 × 1.5 mm) were run at 40 mA per gel. Immediately after the seconddimension run, gels were fixed for 12 h in 50% methanol, containing 10% acetic acid, and the gels were stained with Colloidal Coomassie Blue (Novex, San Diego, CA) for 12 h on a rocking shaker. Molecular masses were determined by running standard protein markers (Biorad Laboratories, Hercules, CA) covering the range 10–250 kDa. pI values were used as given by the supplier of the immobilized pH gradient strips (Amersham Bioscience, Uppsala, Sweden). Excess of dye was washed out from the gels with distilled water, and the gels were scanned with ImageScanner (Amersham Bioscience). 1946
Journal of Proteome Research • Vol. 7, No. 5, 2008
Weitzdörfer et al. Electronic images of the gels were recorded using Adobe Photoshop. Quantification of Protein Spots. Protein spots were outlined (first automatically and then manually) and quantified using the ImageMaster 2D Elite software (Amersham Biosciences, Uppsala, Sweden). The percentage of the volume of the spots representing a certain protein was determined in comparison with the total proteins present in the 2-DE gel.36 Matrix-Assisted Laser Desorption Ionization Mass Spectrometry. Spots were excised with a spot picker (PROTEINEER sp, Bruker Daltonics, Germany) and placed into 384-well microtiter plates, and in-gel digestion and sample preparation for MALDI analysis were performed by an automated procedure (PROTEINEER dp, Bruker Daltonics).37,38 Briefly, spots were excised and washed with 10 mM ammonium bicarbonate and 50% acetonitrile in 10 mM ammonium bicarbonate. After washing, gel plugs were shrunk by the addition of acetonitrile and dried by blowing out the liquid through the pierced well bottom. The dried gel pieces were reswollen with 40 ng/L of trypsin (Promega) in enzyme buffer (consisting of 5 mM octyl-D-glucopyranoside (OGP) and 10 mM ammonium bicarbonate) and incubated for 4 h at 30 °C. Peptide extraction was performed with 10 µL of 1% TFA in 5 mM OGP. Extracted peptides were directly applied onto a target (AnchorChip, Bruker Daltonics) that was loaded with R-cyano-4-hydroxy-cinnamic acid (Bruker Daltonics) matrix thin layer. The mass spectrometer used in this work was an Ultraflex TOF/TOF (Bruker Daltonics) operated in the reflector mode for MALDI-TOF peptide mass fingerprint (PMF) or LIFT mode for MALDI-TOF/TOF fully automated using the FlexControl software. An accelerating voltage of 25 kV was used for PMF. Calibration of the instrument was performed externally with [M + H]+ ions of angiotensin I, angiotensin II, substance P, bombesin, and adrenocorticotropic hormones (clip 1–17 and clip 18–39). Each spectrum was produced by accumulating data from 200 consecutive laser shots. Those samples that were analyzed by PMF from MALDITOF and were significantly different between groups were additionally analyzed using LIFT-TOF/TOF MS/MS from the same target. A maximum of three precursor ions per sample were chosen for MS/MS analysis. In the TOF1 stage, all ions were accelerated to 8 kV under conditions promoting metastable fragmentation. After selection of jointly migrating parent and fragment ions in a timed ion gate, ions were lifted by 19 kV to high potential energy in the LIFT cell. After further acceleration of the fragment ions in the second ion source, their masses could be simultaneously analyzed in the reflector with the Mascot software (Matrix Science Ltd., London, U.K.). Database searches, through Mascot, using combined PMF and MS/MS data sets, were performed via BioTools2.2 software (Bruker). A mass tolerance of 25 ppm and 1 missing cleavage site for PMF and MS/MS tolerance of 0.5 Da but no missing cleavage site for MS/MS search were allowed, and oxidation of methionine residues was considered. The probability score calculated by the software was used as criterion for correct identification. The algorithm used for determining the probability of a false-positive match with a given mass spectrum is described elsewhere.39 Statistical Evaluation. Values from densitometry of spots from 60 gels are expressed as means ( standard deviation (SD) of the percentage of the spot volume in each particular gel after subtraction of the background values. Between-group differences were analyzed using the Kruskal–Wallis test or, when appropriate, the Fisher’s exact test was applied. GraphPad InStat version (GraphPad Software, San Diego, CA) was used for all statistical analysis, and the level of significance
research articles
Differences in Hippocampal Protein Expression was set at P < 0.001 at all instances, thus, warranting stringent statistical criteria.
Results A series of proteins, from several cascades and pathways has been identified (Tables 1 and 2; Supplemental Tables 3 and 4 in Supporting Information). A master map of proteins identified is shown in Figure 1. In Table 1, means and standard deviation, as well as statistical level of significance (P-values < 0.001), are listed. With the use of stringent criteria, the level of significance was set at this reliable level of statistical significance correcting for multiple testing.40 As shown in Table 1, proteins with levels significantly different between groups represented antioxidant, chaperone, cytoskeleton, miscellaneous, neuron-associated, nucleid acid binding, proteasome, and signaling proteins. Information on chemistry, biology, and medical relevance of the corresponding proteins is not given herein and can be tracked easily from the database (http://www.expasy.org/) as the Swiss-Prot numbers are provided. As shown in Table 1, findings were highly specific for individual time points following the asphyctic insult. Peroxiredoxin 6 was increased 3 months following the induction of PA and may indicate long-term sequelae and longlasting effects on the antioxidant response. Hsp90 co-chaperone Cdc37 was undetectably low in control rat pups and became transiently detectable at 3 days following PA. No levels of other heat shock proteins or chaperones out of the many detected (Supplemental Table 3 in Supporting Information) were deranged by PA probably indicating the specificity of this co-chaperone for hypoxic/ischemic states in PA. Only one of several fascin protein spots on 2-DE showed levels significantly dysregulated in PA, and this finding was observed in the early phase of PA. Dynamin-1 levels were significantly dysregulated at 3 weeks following PA, but due to technical reasons, quantification was not carried out because of insufficient separation of some dynamin-1 spots in the gel at the earlier period, and therefore, no conclusion can be drawn about the early time-point. An about 6-fold difference between 3-mercaptopyruvate sulfurtransferase levels between normoxic and asphyctic animals was shown at 3 days of PA probably indicating specific metabolic as well as redox derangement.41 Only one out of four detectable synapsin IIb expression forms, high-abundant proteins in hippocampus, showed levels different between normoxia and PA in the early phase of PA. The importance that the PA time point is studied is revealed by the long-term change of the voltage-dependent anionselective channel protein 1 expression form represented by spot no. 3. During the early phase of PA, levels of a major component of the protein synthetic machinery were deranged proposing a role for Heterogeneous nuclear ribonucleo-protein D-like in the early asphyctic pathology. An essential component of the proteasome, subunit beta type 4, presented with abnormal levels at 3 weeks following PA. Aberrant levels of signaling proteins were either observed at the early phase of PA (adenylyl cyclase-associated protein 1) or in the late phase exclusively (B-cell antigen receptor complex associated protein alpha-chain A).
It must be noticed that Fisher’s exact test was applied when numbers of present or absent spots were statistically analyzed (Table 1). Table 2 lists the protein identification criteria for proteins with different levels between groups. The list of nonsignificantly different protein levels is shown in Supplemental Table 3 in Supporting Information. The identification of proteins with comparable levels between groups is provided in Supplemental Table 4 in Supporting Information.
Discussion The major findings of this study reflect impairment of the brain’s protein machinery, deranged signaling, and altered glia maturation. The observations at the protein chemical level are mainly in agreement with biological findings of PA in the rat and, in particular, with pathobiochemical and histological findings described in the identical animal model. As shortly mentioned above, the protein machinery has been shown to be impaired in terms of protein levels and activity.11 Detailed information of the impairment, however, remains to be elucidated, and no specific proteins or cascades have been addressed so far. The findings also clearly show that different cascades are shown at different time points, and indeed, early changes were showing most of the derangements in protein levels while only two proteins from cytoskeleton and a proteasomal component were affected at 3 weeks following PA. Antioxidant, ion channel, and two signaling proteins were specific for late changes of protein levels. Heterogeneous nuclear ribonucleoproteins (hnRNPs) are key structures of the splicing apparatus, regulating mRNA-biogenesis and hnRNP D-like is a representative with increased levels in PA at the early (3d) phase following the asphyctic insult. Tsuchiya and co-workers reported a novel hnRNP-like protein and its expression in myeloid-leukemia cells,42 and Kamei and co-workers showed the presence of two expression forms.43 In the present study, one expression form was detected exclusively. Kamei and Yamada demonstrated domains with highaffinity binding sites that can be discriminated from hnRNPD.44 Kawamura et al. revealed that the dominant expression form shuttles between the nucleus and the cytoplasm and that the protein specifically accumulated upon polymerase transcription inhibition.45 It may well be that inhibition of polymerase I, as shown in the identical animal model of PA until at least day 8 following induction of PA,11 led to the observed increase of hnRNP D-like. We may propose that this hnRNP is reflecting impaired nuclear export in PA. Once proteins are synthetized, they are chaperoned by specific structures. Hsp90 co-chaperone cdc37(CDC) is a protein kinase-targeting molecular chaperone and is required in cooperation with Hsp90 for various signaling kinases.46 Harris and co-workers recently showed that CDC has a direct regulatory effect on endothelial NOS and may play an important role in mediating the eNOS protein complex formation as well as subsequent eNOS phosphorylation and activation.47 The intriguing finding that CDC was expressed in hippocampi of rats with PA but not controls may indicate a role in the nitric oxide pathways: eNOS and other NOS systems are key mediators in hypoxic-ischemic states in early life.48,49 In the applied rat model of PA, only nNOS was studied systematically,18 and it is not known whether CDC is interacting with this expression form as well. Journal of Proteome Research • Vol. 7, No. 5, 2008 1947
1948
Fascin (mouse)
Q61553-spot 11
Journal of Proteome Research • Vol. 7, No. 5, 2008
9QWV7 spot 12 9QWV7 spot 22 9QWV7 spot 32 9QWV7 total2 60932 spot 12
Adenylyl cyclase-associated protein 1 (mouse) B-cell antigen receptor complex associated protein alpha-chain [Precursor] A (mouse) Phosphatidylethanolaminebinding protein (rat)
P40124
0.443 ( 0.139 0.200 ( 0.035
0.219 ( 0.082 0.310 ( 0.058 0.156 ( 0.084 0.654 ( 0.150 0.258 ( 0.082 0.360 ( 0.187 0.401 ( 0.181 1.020 ( 0.369 0.163 ( 0.051
0.076 ( 0.015 (8/8) 0.362 ( 0.094
0.520 ( 0.093 1.770 ( 0.564
0.215 ( 0.083 0.083 ( 0.042 0.161 ( 0.023 0.441 ( 0.078 0.214 ( 0.098 0.176 ( 0.066 0.213 ( 0.090 0.536 ( 0.162 0.377 ( 0.045
0.000 ( 0.000 (0/9) 0.163 ( 0.052
x
x
0.135 ( 0.052
0.028 ( 0.055
0.000 ( 0.000 (0/7) 0.571 ( 0.167 0.115 ( 0.068 0.686 ( 0.168 x x x 0.173 ( 0.057
0.729 ( 0.076
Cytoskeleton 0.069 ( 0.042 (7/8) 0.195 ( 0.058 0.126 ( 0.027 0.395 ( 0.053 0.157 ( 0.116 0.237 ( 0.151 0.377 ( 0.210 Metabolic 0.082 ( 0.043 Miscellaneous 0.270 ( 0.075
0.370 ( 0.128 x
Signaling 0.203 ( 0.079 0.103 ( 0.064
0.448 ( 0.094
x
0.882 ( 0.163
x
0.000 ( 0.000 (0/9)
0.535 ( 0.061
0.147 ( 0.057 0.167 ( 0.135 0.358 ( 0.203
Proteasome 0.084 ( 0.054 (7/7)
Nucleic Acid Binding 0.148 ( 0.039
0.428 ( 0.181 0.154 ( 0.067 0.717 ( 0.320
0.170 ( 0.114 0.043 ( 0.083 0.605 ( 0.232 0.751 ( 0.365 0.100 ( 0.038
0.031 ( 0.053 (3/10) 0.334 ( 0.084 0.085 ( 0.024 0.409 ( 0.169 0.112 ( 0.046 0.186 ( 0.076 0.298 ( 0.111
0.239 ( 0.099 (9/9) 0.522 ( 0.136 0.177 ( 0.072 0.938 ( 0.184 x x x
0.033 ( 0.027 (6/7)
0.125 ( 0.024
Neuron Associated 0.338 ( 0.168 x 0.293 ( 0.119 0.631 ( 0.204 0.244 ( 0.090
x
0.000 ( 0.000 (0/9)
Chaperone x
Antioxidant 0.279 ( 0.081
3d PA mean ( SD (Fisher’s exact)
0.226 ( 0.178
0.397 ( 0.076
0.117 ( 0.037
3m CO mean ( SD (Fisher’s exact)
0.162 ( 0.106
3w CO mean ( SD (Fisher’s exact)
3d CO mean ( SD (Fisher’s exact)
a
1.018 ( 0.234
x
0.466 ( 0.135
0.000 ( 0.000 (0/8)
0.168 ( 0.048
0.427 ( 0.126 0.302 ( 0.166 0.974 ( 0.333
0.222 ( 0.073 0.265 ( 0.054 0.310 ( 0.176 0.709 ( 0.186 0.281 ( 0.148
0.192 ( 0.082
0.449 ( 0.103
0.165 ( 0.103
0.000 ( 0.000 (0/8) 0.290 ( 0.118 0.107 ( 0.074 0.370 ( 0.165 0.020 ( 0.035 0.124 ( 0.067 0.144 ( 0.091
x
0.395 ( 0.110
3w PA mean ( SD (Fisher’s exact)
1.062 ( 0.140
0.000 ( 0.000
0.227 ( 0.082
0.041 ( 0.027 (8/10)
0.123 ( 0.012
0.578 ( 0.082 0.424 ( 0.101 1.210 ( 0.438
0.466 ( 0.175 x 0.288 ( 0.126 0.710 ( 0.276 0.363 ( 0.136
0.162 ( 0.055
0.288 ( 0.074
0.057 ( 0.020
0.064 ( 0.058 (7/10) 0.192 ( 0.073 0.080 ( 0.021 0.331 ( 0.052 0.180 ( 0.121 0.283 ( 0.176 0.463 ( 0.292
x
0.670 ( 0.177
3m PA mean ( SD (Fisher’s exact)
x
x
0.0003
n.s.
0.0002
n.s. n.s. n.s.
n.s. n.s. 0.0002 n.s. n.s.
