Synaptic proteome changes in the hypothalamus of mother rats

Accepted Manuscript Synaptic proteome changes in the hypothalamus of mother rats

Edina Brigitta Udvari, Katalin Völgyi, Péter Gulyássy, Diána Dimén, Viktor Kis, János Barna, Éva Rebeka Szabó, Gert Lubec, Gábor Juhász, Katalin Adrienna Kékesi, Árpád Dobolyi PII: DOI: Reference:

S1874-3919(17)30090-8 doi: 10.1016/j.jprot.2017.03.006 JPROT 2794

To appear in:

Journal of Proteomics

Received date: Revised date: Accepted date:

15 November 2016 1 February 2017 7 March 2017

Please cite this article as: Edina Brigitta Udvari, Katalin Völgyi, Péter Gulyássy, Diána Dimén, Viktor Kis, János Barna, Éva Rebeka Szabó, Gert Lubec, Gábor Juhász, Katalin Adrienna Kékesi, Árpád Dobolyi , Synaptic proteome changes in the hypothalamus of mother rats. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Jprot(2017), doi: 10.1016/j.jprot.2017.03.006

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

SYNAPTIC PROTEOME CHANGES IN THE HYPOTHALAMUS OF MOTHER RATS Edina Brigitta Udvari1,2, Katalin Völgyi1, Péter Gulyássy2,3,6, Diána Dimén1,4, Viktor Kis1,4, János Barna5, Éva Rebeka Szabó5, Gert Lubec3, Gábor Juhász2,6, Katalin Adrienna Kékesi2,7,

RI

1

PT

Árpád Dobolyi1,5*

MTA-ELTE NAP B Laboratory of Molecular and Systems Neurobiology, Department of

SC

Physiology and Neurobiology, Hungarian Academy of Sciences and Eötvös Loránd University, Budapest H-1117, Hungary

Laboratory of Proteomics, Institute of Biology, Eötvös Loránd University, Budapest H-

NU

2

1117, Hungary 4

Department of Pharmaceutical Chemistry, University of Vienna, Vienna A-1090, Austria Department of Anatomy, Cell and Developmental Biology, Eötvös Loránd University,

MA

3

Budapest H-1117, Hungary

Laboratory of Neuromorphology, Department of Anatomy, Histology and Embryology,

D

5

6

PT E

Semmelweis University, Budapest H-1094, Hungary MTA-TTK NAP MS Neuroproteomics Research Group, Hungarian Academy of Sciences,

Budapest H-1117, Hungary

Department of Physiology and Neurobiology, Eötvös Loránd University, Budapest H-1117,

Hungary *

CE

7

AC

Correspondence: Dr. Árpád Dobolyi

MTA-ELTE NAP B Laboratory of Molecular and Systems Neurobiology, Institute of Biology, Hungarian Academy of Sciences and Eötvös Loránd University, Budapest, Hungary Pázmány Péter sétány 1C, Budapest, H-1117, Hungary Email: [email protected] Tel.: +36-1-372-2500 /8775 Fax.: +36-1-218-1612

1

ACCEPTED MANUSCRIPT Abstract To establish synaptic proteome changes associated with motherhood, we isolated synaptosome fractions from the hypothalamus of mother rats and non-maternal control females at the 11th postpartum day. Proteomic analysis by two-dimensional differential gel electrophoresis combined with mass spectrometric protein identification established 26

PT

significant proteins, 7 increasing and 19 decreasing protein levels in the dams. The altered

RI

proteins are mainly involved in energy homeostasis, protein folding, and metabolic processes

SC

suggesting the involvement of these cellular processes in maternal adaptations. The decrease in a significantly altered protein, complement component 1q subcomponent-binding protein

NU

(C1qbp) was validated with Western blotting. Furthermore, immunohistochemistry showed its presence in hypothalamic fibers and terminals in agreement with its presence in

MA

synaptosomes. We also found the expression of C1qbp in different hypothalamic nuclei including the preoptic area and the paraventricular hypothalamic nucleus at the protein and at

D

the mRNA level using immunohistochemistry and in situ hybridization histochemistry,

PT E

respectively. Bioinformatical network analysis revealed that cytokines, growth factors, and protein kinases are common regulators, which indicates a complex regulation of the proteome

CE

change in mothers. The results suggest that maternal responsiveness is associated with synaptic proteins level changes in the hypothalamus, and that growth factors and cytokines

AC

may govern these alterations.

Biological significance The period of motherhood is accompanied with several behavioral, neuroendocrine, emotional and metabolic adaptations in the brain. Although it is established that various hypothalamic networks participate in the maternal adaptations of the rodent brain, our knowledge on the molecular background of these alterations remains seriously limited. In the

2

ACCEPTED MANUSCRIPT present study, we first determined that the functional alterations of the maternal brain can be detected at the level of the synaptic proteome in the hypothalamus. Independent confirmation of synaptic localization, and also the established decrease in the level of C1qbp protein suggest the validity of the data. Common regulators of altered proteins belonging to the growth factor and cytokine family suggest that the synaptic adaptation is governed by these

PT

extracellular signals and future studies should focus on their specific roles. Our study was also

RI

the first to describe the expression pattern of C1qbp in the hypothalamus, a protein potentially

SC

involved in mitochondrial and neuroimmunological regulations of synaptic plasticity. Its presence in the preoptic area responsible for maternal behaviors and also in the

NU

paraventricular hypothalamic and arcuate nuclei regulating hormonal levels suggests that the same proteins may be involved in different aspects of maternal adaptations. The conclusions

MA

of the present work contribute to establishing the molecular alterations that determine different maternal adaptations in the brain. Since maternal changes are models of neuronal

PT E

behavioral neuroscience.

D

plasticity in all social interactions, the reported results can affect a wide field of molecular and

Keywords: maternal behavior; lactation; synaptosome; 2-D DIGE; common regulator;

AC

CE

C1qbp

3

ACCEPTED MANUSCRIPT

Introduction The transformation from non-parental to maternal state requires a wide range of adaptation processes, which include physiological, emotional and behavioral changes. These adaptations facilitate the optimal conditions for taking care of the offspring. Around

PT

parturition, a shift can be observed in the mother’s behavior. Female rats avoid or even hurt

RI

neonates while newly parturient mothers nurse and protect their pups [1]. This behavior is

SC

driven by maternal motivation whose major control region is the medial preoptic area (mPOA) [2-7]. Damage of this region disrupts the maternal responsiveness toward infants

NU

[8,9].

The secretion of some reproductive hormones, such as estrogen, prolactin and oxytocin is

MA

increased at different phases of pregnancy, parturition and lactation. These hormones are known to contribute to the induction of maternal behavior [10-13] while neuronal inputs from

D

the pups are sufficient for its maintenance [13,14]. The hypothalamus also regulates other

PT E

aspects of the behavior altered throughout pregnancy and lactation. Increased metabolism for appropriate nurturing of the pups requires elevated food and water intake in mother rats

CE

[15,16]. In addition, the hypothalamo-pituitary-adrenal axis (HPA) undergoes adaptations during the time of pregnancy and lactation, which contributes to reduced anxiety and

AC

suppressed stress responsiveness in mothers [17]. Additional neuroendocrine adaptation also characterize the postpartum period, including suckling-induced prolactin release [18], and the inhibition of gonadotropin-releasing hormone (GnRH) secreting neurons for lactational anoestrous [19]. Hypothalamic brain regions are generally responsible for the maintenance of homeostasis by coordinating the sensory-motor apparatus and the neuroendocrine system for an appropriate response. Regarding behavior towards pups, the hypothalamus can switch

4

ACCEPTED MANUSCRIPT between two different attitudes toward infants: parental care and aggression [4,9]. Thus, it is conceivable that the hypothalamus can reorganize itself for the appropriate maternal responsiveness. For example, remarkable synaptic plasticity changes were observed in oxytocinerg neurons in the supraoptic and paraventricular nuclei during parturition and lactation [20,21]. Shams et al. suggest a more general, experience-dependent remodeling in

PT

the hypothalamus during motherhood [22]. Such restructuring processes must rely on changes

RI

at the proteome level as the underlying mechanism of maternal adaptations. Therefore, the

SC

proteomic alterations related to motherhood should be further investigated. In the present study, we isolated synaptosomes from the hypothalamus of mother rats and compared their

NU

proteome to that of non-maternal females without a litter. The results were analyzed bioinformatically to establish network connections of the altered proteins. For validation, the

MA

localization and protein level alteration of a protein with significant change, Complement component 1q subcomponent-binding protein (C1qbp) was further investigated using multiple

Animals

CE

Materials and methods

PT E

D

approaches.

AC

Animals obtained from Toxicoop, Hungary were kept under standard laboratory conditions with 12 hours light and 12 hours dark periods (light was on from 08.00 am to 08.00 pm). Food and water were supplied ad libitum. The care and experimentation of all animals conformed to Hungarian Act of Animal Care and Experimentation (1998, XXVIII) and to the guidelines approved by the European Communities Council Directive 86/609/EEC as well as with local regulations for the care and use of animals for research.

5

ACCEPTED MANUSCRIPT A total of 31 four-month old mother Wistar rats (12 for proteomics, 10 for Western blotting, 6 for immunohistochemistry, 3 for in situ hybridization histochemistry) were used for the experiments. In the mothers group, they were adjusted to rear 8 pups. In the control group, the pups were removed from the mothers immediately after parturition (pup-deprived mothers). The animals were investigated at the 11th postpartum day because the mother’s

PT

behavior is still strong and stable while the pup-deprived mothers do not show maternal

RI

behavior by this time [23]. For perfusions and dissections, rats were anesthetized with an

SC

intramuscular injection of anesthetic mix containing 0.2 ml/300g body weight ketamine (100 mg/ml) and 0.2 ml/300g body weight xylazine (20 mg/ml). The freshly removed brains were

NU

dissected for proteomics, Western blot, and in situ hybridization studies. For immunohistochemistry, the animals were transcardially perfused before removing the brains

MA

as described below.

D

Microdissection of brain tissue samples

PT E

Brains of 11 mothers with litter and 11 pup-deprived control rats were used, 6-6 for proteomics and 5-5 for Western blotting. The hypothalamus was dissected with razor blade

CE

cuts from the fresh brain. First a coronal cut was made immediately anterior to the optic chiasm (bregma level: +0.3 mm). Then, another coronal cut was made 4 mm caudal to the

AC

first cut (Figure 1A). Subsequently, the slide was placed with its posterior surface upside, and 2 lateral cuts were made above the optic tracts. Finally, a horizontal cut at the top of the third ventricle finalized the dissection (Figure 1B). The dissected tissue samples were quickly frozen on dry ice, and stored at -80oC.

Synaptosome preparation from the hypothalamus

6

ACCEPTED MANUSCRIPT Hypothalamic synaptosomes were isolated according to a published protocol with some minor modifications [24]. The hypothalamus samples (65-70 mg per animal) were homogenized in 1 ml of ice-cold isolation buffer (5 mM HEPES-KOH, pH 7.4, 320 mM sucrose, supplemented with protease and phosphatase inhibitor cocktails) using Dounce homogenizer (20 strokes Small Clearance). All steps of preparation were performed at 4oC.

PT

The homogenates were centrifuged at 1,000 g for 15 min. The supernatants were centrifuged

RI

at 13,300 g for 15 min and the resulting pellets resuspended in 1.25 ml isolation buffer. The

SC

resuspended pellets were loaded onto a discontinuous Ficoll gradient (13, 9 and 5% in isolation buffer), and the samples were centrifuged at 61,000 g for 40 min. After

NU

centrifugation, intact synaptosomes were present at the 9 and 13% interface. They were removed, supplemented with isolation buffer and centrifuged at 20,000 g for 32 min. The final

MA

synaptosome pellets were resuspended in 150 µl buffer (5 mM HEPES-KOH, pH 7.4) and

D

were precipitated overnight in ice cold acetone.

PT E

Synaptosome preparation validation with electron microscopy The quality of the synaptosome preparation is essential for a reliable proteomics data,

CE

therefore, the purity of the samples was verified with electron microscopy. The synaptosome pellets were fixed in 2% formaldehyde and 0.5% glutaraldehyde in

AC

0.1 M Na-cacodylate for 30 min at room temperature. The samples were washed in 0.1 M phosphate buffer (PB) and postfixed in 0.5% osmium-tetroxide and 0.75% potassiumferrocyanide for 45 min, followed by staining in half-saturated aqueous uranyl acetate for 30 min. Then, the samples were dehydrated and embedded in LR White resin according to the manufacturer’s instructions. Seventy nm ultrathin sections were cut and analyzed with JEOL JEM 1011 electron microscope operated at 60 kV. Images were taken with Olympus Morada 11 megapixel camera and iTEM software (Olympus).

7

ACCEPTED MANUSCRIPT

Proteomic analysis by two-dimensional differential gel electrophoresis (2-D DIGE) For proteomic analysis, 12 rats (6 mothers, 6 pup-deprived control rats) were used. Relative synaptosomal protein levels in the hypothalamus of mother and control rats were determined by using CyDye DIGE Fluor Saturation Labeling Kit. The detailed protocol was

PT

described in our previous study [25]. Briefly, the precipitated hypothalamus samples were

RI

resuspended in lysis buffer (7 M urea, 2 M thiourea, 4% CHAPS, 20 mM Tris and 5 mM

SC

magnesium-acetate). Five µg of the samples were labeled with CyDye DIGE Fluor Saturation Labeling Kit. The samples from the mother and pup-deprived rats were labeled with Cy5 and

NU

an internal standard (pool of equal amounts of all samples within the experiments) was labeled with Cy3. The mother or pup-deprived mother samples were combined with pool

MA

sample in case of each animal, and the twelve mixtures were run, six simultaneously. First, isoelectric focusing was performed for 24 hours on Ettan IPGphor instrument to attain a total

D

of 80 kVh. After the isoelectric focusing, the IPG strips (pH-3-10 NL, 24 cm) were

PT E

equilibrated in a reducing buffer containing mercaptoethanol for 20 min. Then, the strips were placed onto 10% polyacrylamide gels (24x20 cm). The SDS-PAGE gel electrophoresis was

CE

performed using an Ettan DALT System. Subsequently, the gels were scanned by Typhoon TRIO+ scanner using appropriate wavelengths. Gel images were analyzed by ImageQuant TL

AC

software and the differences in spot intensity were determined using DeCyder 2D software 7.0 with the Differential In-gel Analysis (DIA) and Biological Variance Analysis (BVA) modules. The DIA module of DeCyder can be used to identify and quantitate protein spots between different samples in the same gel. The cross gel comparison and statistical analysis of protein abundance changes can be carried out in the BVA module. The internal standard samples were the same across all gels and the fluorescent intensity of the proteins were compared to the suitable internal standard. The statistical analysis of the protein abundance

8

ACCEPTED MANUSCRIPT changes was performed by independent Student’s t-test. For further work, we selected those protein spots, which showed statistical significance of protein abundance changes at p < 0.05. For the identification of the proteins in the spots of interest, a preparative twodimensional gel electrophoresis was performed using a total of 800 µg proteins per gel. The gels were stained with Colloidal Coomassie Brilliant Blue G-250 (Merck, Darmstadt,

SC

Protein identification by mass spectrometry (LC/MS-MS)

RI

PT

Germany).

The spots of interest were manually cut from the gel with the same position as the CyDye

NU

labeled spots. Then the spots were tryptically cleaved according to a method published previously [26]. Liquid chromatography-mass spectrometry was performed using an Ultimate

MA

3000 RSLC Nanosystem (Dionex, Sunnyvale, CA, USA) coupled to a 3D high capacity ion trap mass spectrometer (AmaZon, Bruker Daltonics, Bremen, Germany) via a CaptiveSpray

D

ion source under the control of Hystar (Bruker Daltonics, Bremen, Germany). Samples of 10

PT E

μl were injected onto a PepMap100 C-18 (300 μm × 5 mm) trap column (Thermo Scientific, Sunnyvale, CA, USA) at a flow rate of 30 μl/min. After washing with 0.1% formic acid, the

CE

peptides were separated at a flow rate of 300 nl/min using an Acclaim PepMap RSLC (75 μm × 50 cm) separating column (Thermo Scientific, Sunnyvale, CA, USA). The gradient was 4%

AC

B from 0 to 4 min, 4–13% B from 4 to 10 min, 13–35% B from 10 to 45 min, 35%–90% B from 45 to 46 min, 90% B from 46 to 51 min, 90–94% B from 51–51.Five min followed by an equilibration step from 51.5 to 65 min (A: 0.1% formic acid in water, B: 0.08% formic acid in 80% acetonitrile). After each sample, a blank was run to avoid sample carryover. The drying nitrogen gas was heated to 150°C and the flow rate was 3 l/min, the capillary voltage was set to 1400 V. The peptide spectra was recorded in a positive mode over the mass range of m/z 400–1400, and MS/MS spectra in information-dependent data acquisition over the

9

ACCEPTED MANUSCRIPT mass range of m/z 50–3000. Only multiple charged peptides were chosen for MS/MS experiments. Up to 10 precursor ions above a threshold of 25,000 were selected per MS scan and an active exclusion of 0.2 min after 1 spectra was used. A target of 400,000 was set to ICC and maximal ion accumulation time allowed was 50 ms to prevent overfilling of the ion trap. The collision energy was set automatically according to the mass and charge state of the

PT

peptides chosen for fragmentation. Protein searching was conducted against the SwissProt

RI

sequence database with the taxonomy of Rattus norvegicus using Mascot Server v.2.2 (Matrix

SC

Science, London, UK). Trypsin was set as the enzyme, one missed cleavage was allowed, carbamidomethylation of cysteine was set as fixed modification and methionine oxidation was

NU

allowed as variable modification. The data were searched with 0.4 Da fragment and 0.35 Da parent ion mass tolerances. Positive protein identifications were based on a significant

MA

MOWSE score. Proteins with a minimum of two identified, unique peptides were accepted

PT E

Functional classification

D

and included in further analysis.

Significantly altered proteins were categorized manually according to the UniProt

CE

(http://www.uniprot.org/) and GeneOntology (http://geneontology.org/) databases. As most

AC

proteins have multiple roles, only the most relevant one was used for each protein.

Bioinformatical analysis of significant protein changes The linkage between significantly changed hypothalamic synaptic proteins was performed by Pathway Studio 11.0 software (Elsevier Life Science Solutions). We selected common regulators of the altered proteins in a way that only the regulator proteins with connections to at least 3 altered proteins were included. Potential mechanisms of the C1qbp protein in maternal adaptation were also examined with Pathway Studio 11.0 software. We

10

ACCEPTED MANUSCRIPT identified the neighbors of C1qbp to find connection with hormones participating in the adaptation of the brain to taking care of the offspring during motherhood.

Validation of C1qbp protein level change with Western blot (WB) technique For Western blot validation, hypothalamic synaptosomes from 5 mothers and 5 pup-

PT

deprived mothers (control group) were used. The acetone-precipitated samples were

RI

suspended in lysis buffer and 5-5 µg proteins from both maternal and pup-deprived samples

SC

were separated. Proteins were separated by 4% stacking and 15% resolving Tricine-SDSpolyacrylamide gel electrophoresis. Then, the proteins were transferred to Hybond-LFP

NU

PVDF membranes (GE Healthcare). Subsequently, blocking of non-specific labeling was performed with 5% bovine serum albumin (BSA) in Tris buffered saline and 0.05% Tween-20

MA

solution (TBS-T) for 1 hour. The membranes were incubated overnight in the primary antibody (rabbit anti-C1qbp at a dilution of 1:200 (sc-48795, Santa Cruz Biotechnology Inc.))

D

dissolved in TBS-T. After that, the membranes were washed in TBS-T followed by incubating

PT E

with ECL Plex CyDye conjugated anti-rabbit IgG secondary antibody (1:2,500, GE Healthcare). Detection of the proteins was performed using a Typhoon TRIO+ scanner with

CE

appropriate wavelengths. The fluorescent intensities of the bands were quantified by Image J software (NIH, Bethesda). Differences between samples from maternal and control animals

AC

were analyzed statistically using independent, two-sample t-test.

C1qbp immunohistochemistry For immunohistochemistry, pup-deprived and mother rats (n=3-3) were deeply anesthetized and perfused transcardially first with saline, and then with 4% paraformaldehyde (pH 7.4). Brains were removed and post-fixed in 4% paraformaldehyde for 24 hours, and then stored in 20% sucrose in PB buffer for 48 hours. Serial coronal brain sections were cut from

11

ACCEPTED MANUSCRIPT the hypothalamus at 40 µm with a sliding microtome. Sections were collected in PB containing 0.05% sodium-azide and stored at 4oC. Non-specific binding sites were blocked with 3% BSA, 0.5% Triton-X 100 and 0.05% sodium azide dissolved in PB for 1 hour. After the washing steps, the same anti-C1qbp antibody (at a dilution of 1:50) was used as for WB, followed by biotinylated goat anti-rabbit IgG secondary antibody (1:1,000, Jackson

PT

Immunoresearch, West Grove, PA). The visualization was performed with Avidin and

RI

Biotinylated horseradish peroxidase macromolecular Complex (ABC) kit (1:500 dilution,

SC

Vector Laboratories, Burlingame, CA, USA) and 3,3′-Diaminobenzidine (DAB) reaction.

NU

After that, sections were dehydrated, and coverslipped with DPX Mounting Medium.

In situ hybridization histochemistry for C1qbp

MA

Brains of 3 pup-deprived mother rats were dissected and frozen at 11 days postpartum. In situ hybridization histochemistry was performed, as described previously [27]. Briefly,

D

serial coronal sections (12 m) were cut, immediately mounted on positively charged slides

PT E

(Superfrost Plus, Fisher Scientific, Pittsburgh, PA, USA), dried, and stored at -80°C until use. A 222 bp long region of the rat C1qbp cDNA sequence (NCBI Reference Sequence: was

PCR

amplified

using

the

following

primers:

CE

NM_019259.2)

GGGCCTTGTATGACCACCTA and TGATGTCAAGGCAGCTTTTG. The PCR product

AC

was subcloned into a TOPO TA vector (Life Technologies) containing a T7 RNA polymerase recognition site. The T7 promoter was used to generate [35S]UTP-labeled riboprobes, with a MAXIscript transcription kit (Ambion, Austin, TX). Tissues were prepared using an mRNA-locator Kit (Ambion) according to manufacturer’s instructions. For hybridization, we used 80 µl hybridization buffer and 1 million DPM of labeled probe per slide. Washing procedures included a 30 min incubation in RNase A, followed by decreasing concentrations of sodium-citrate buffer (pH 7.4) at room

12

ACCEPTED MANUSCRIPT temperature, and then at 65°C. After drying, slides were dipped in NTB nuclear track emulsion (Eastman Kodak, Rochester, NY), stored for 3 weeks at 4°C for autoradiography, developed with Kodak Dektol developer, fixed with Kodak fixer, counterstained with Giemsa, dehydrated, and coverslipped with Cytoseal 60 (Stephens Scientific).

PT

Histological analysis

RI

Sections were examined using an Olympus BX60 light microscope equipped with

SC

fluorescent epi-illumination and bright- and dark-field condensers. Images were captured at a 2048 by 2048 pixel resolution with a SPOT Xplorer digital CCD camera (Diagnostic

NU

Instruments, Sterling Heights, MI) using 4-40 X objectives. The contrast and sharpness of the images were adjusted using the “levels” and “sharpness” commands in Adobe Photoshop CS

MA

8.0 (Adobe Systems). The full resolution of the images was maintained until the final versions

PT E

Electron microscopy

D

were adjusted to a resolution of 300 dpi.

Tissue preparation and low temperature embedding

CE

Low temperature embedding in acrylic resins and post-embedding immunolabeling were carried out as described earlier [5,25]. Rats (n=2) were perfused transcardially with

AC

saline (30 s) followed by fixative containing 2% paraformaldehyde (freshly depolymerized from paraformaldehyde), 0.5% glutaraldehyde dissolved in 0.1 M PB, pH 7.4 for 20 min at RT. After fixation, the brains were perfused again with saline (10 min) to avoid over fixation and then the brains were removed from the skull and sectioned at 50 µm on a vibratome. Brain sections containing the paraventricular hypothalamic nucleus (PVN) were processed for low temperature embedding. Free aldehydes were quenched with 50–50 mM ammonium chloride-glycine dissolved in TBS. After that, slices were postfixed in 0.1% tannic acid

13

ACCEPTED MANUSCRIPT (Malinckrodt) freshly prepared before use in TBS for 60 min at RT. Then, blocks were washed three times in maleate buffer (0.05 M, pH 5.5) before post-fixation in 1% uranyl acetate dissolved in maleate buffer for 180 min in the dark. This was followed by three washes in maleate buffer. Subsequently, blocks were dehydrated in graded series of acetone using progressive lowering temperature. Samples were then infiltrated with pure LR White

PT

alone (12 h, 20°C) followed by pure LR White containing 2% benzoyl peroxide as a

RI

catalyzator (3 h, 20°C). Curing was done with a homemade UV chamber using two DL-103

SC

2x6W UV lamp for 60 h at 20°C.

NU

Post-embedding immunolabeling and image acquisition

Small pieces from the PVN were reembedded and resectionned at 70 nm. Sections

MA

were collected on formvar coated 100 mesh nickel grids. All immunoreactions were carried out on humidified parafilm-coated 96 well plates. Tris buffer (0.05 M; pH 7.6) with 0.9%

D

sodium chloride (TBS) was used for all washes and dilutions. Briefly, the following procedure

PT E

was carried out: (1) 5% hydrogen-peroxide for 3 min; (2) wash in double distilled water (biDW; (3) 1% sodium borohydride and 50 mM glycine dissolved in TBS for 5 min; (4) two

CE

washes in TBS; (5) 2% BSA (Sigma, A7638) for 30 min; (6) rabbit anti-C1qbp in 1:20 in 1% BSA-TBS containing 0.05% sodium azide for 12–18 h; (7) washes in 0.5% BSA-TBS; (8) 10

AC

nm gold-conjugated goat anti-rabbit secondary antibody 1:50 in TBS with 1% ovalbumin for 4h; (9) wash in TBS; (10) 2% glutaraldehyde in TBS for 10 min; (11) wash in biDW; (12) air drying; (13) staining in 0.5% osmium tetroxide for 20 min followed by lead citrate for 30 s; (14) air drying. Electron micrographs were taken by a side-mounted Morada CCD camera (Olympus Soft Imaging Solutions) connected to a JEOL JEM 1011 electron microscope. In the digital images of immunogold labeling, brightness and contrast were adjusted when necessary using Adobe Photoshop CS2 (Adobe Systems). To calculate the C1qbp gold

14

ACCEPTED MANUSCRIPT particle densities 30 randomly selected synaptic terminals (asymmetric and symmetric synapses together) from the PVN with clearly visible synaptic specialization, containing mitochondria were photographed. Within a single bouton the area of the mitochondria and the synaptic cytoplasm area were determined with iTEM software (iTEM Software, Irvine, CA, USA) using the interpolated polygon algorithm. The gold particles were counted manually

SC

RI

PT

using the iTEM software. Data were analyzed using the Mann–Whitney U test.

Results

NU

Validation of the synaptosome sample purity with electron microscopy Electron microscopic analysis revealed that the samples consist of synaptosomes

MA

containing synaptic mitochondria and large pools of synaptic vesicles. In addition, postsynaptic element was attached to the sealed presynaptic terminals (Figure 2). Nucleus or

D

cell body related compartments (lysosomes and Golgi apparatus) were not present in the

PT E

samples suggesting the purity and enrichment of the synaptosome preparation.

CE

Proteomic identification of maternally altered synaptosome proteins The results of the proteomic analysis indicated 33 significantly altered protein spots.

AC

There were 10 protein spots, which showed higher, and 23 with lower fluorescence intensities in dams compared to the pup-deprived group. A representative el with significant protein changes, and representative Saturation labeling analytical gels of a mother and a pup-deprived synaptosome samples with the corresponding pool samples are shown in Figure 3. The fold changes of fluorescence intensities between spots from the two groups were in the range of 2.85 to +2.32 (Table 1). All of the identified spots with altered synaptic protein levels were

15

ACCEPTED MANUSCRIPT determined by mass spectrometry (MS). The MS analysis revealed 26 different proteins, of which 7 showed increased and 19 decreased levels in the maternal samples (Table 1).

Functional clusters of maternally altered synaptic proteins The synaptic proteins with altered levels belong to 13 functional clusters (Figure 4.).

PT

The clusters are electron transport chain / ATP metabolism (n=8), tricarboxylic acid cycle

RI

(n=4), protein transport, folding (n=2), amino acid metabolism (n=2), signal transduction

SC

(n=2), glucose metabolism (n=1), lipid metabolism (n=1), ion transport (n=1), immune regulation (n=1), oxidative stress (n=1), transcription regulation (n=1), neuron development

NU

(n=1), membrane formation, stabilization (n=1).

MA

Common regulators of maternally changed proteins

Common regulators, which can induce expressional or functional alterations in at least

D

3 of the altered proteins include Angiotensinogen (Agt), RAC-alpha serine/threonine-protein

PT E

kinase (Akt1), Cyclic AMP-responsive element-binding protein 1 (Creb1), Interferon gamma (Ifng), Insulin-like growth factor I (Igf1), Interleukin-10 (IL10), Interleukin-1 beta (IL1B),

CE

Interleukin-6 (IL6), Insulin (Ins), Mitogen-activated protein kinase 1 (Mapk1), Myc protooncogene protein (Myc), Beta-nerve growth factor (Ngf), Platelet-derived growth factor

AC

(Pdgf), Phosphatidylinositol 3-kinase (Pi3k), Protein kinase A/cAMP-dependent protein kinase (Pka), Protein kinase C (Pkc), Ras superfamily (Ras), Proto-oncogene tyrosine-protein kinase Src (Src), Transforming growth factor beta-1 (Tgfb1), Tumor necrosis factor (Tnf) as shown in Figure 5. The common regulators belong to the group of protein kinases (Akt1, Mapk1, Pka, Pkc, Pi3k, Src), growth factors (Igf1, Ngf, Pdgf, Ras, Tgfb1), cytokines (Ifng, IL10, IL1B, IL6, Tnf), peptide hormones (Agt, Ins), and transcription factors (Creb1, Myc).

16

ACCEPTED MANUSCRIPT Validation of decreased maternal level of C1qbp We selected C1qbp decreasing protein for validation by Western blot (WB). In the subsequent WB analysis using hypothalamic synaptosome preparation from 5 mothers and 5 pup-deprived control rats, the normalized protein value of C1qbp (-1.625 ± 0.2) was significantly decreased (p < 0.05) in dams compared to control rats (Figure 6). Therefore the

RI

PT

result of WB analysis confirmed the 2-D DIGE experimental data.

SC

Localization of C1qbp in the hypothalamus

C1qbp-immunoreactive (C1qbp-ir) fibers were present in the hypothalamus with

NU

generally higher density in the medial than the lateral zones of the hypothalamus. The highest density of C1qbp-ir fibers and terminals was found in the ventromedial hypothalamic (VMH)

MA

and adjacent arcuate nuclei (Arc) (Figure 7A,B). C1qbp-ir cell bodies were also present in the hypothalamus. The anteroventral periventricular nucleus (AVPV), the medial preoptic nucleus

D

(MPN) (Figure 8A1), the magnocellular preoptic nuclei (MCPO) (Figure 8A1), the supraoptic

PT E

nucleus (SON), the PVN (Figure 8B1), and the Arc (Figure 7Aa), as well as the ventromedial subdivision of the VMH (Figure 7Aa) contained significant densities of C1qbp-ir cells. The

CE

distributional pattern of C1qbp did not differ considerably between pup-deprived (Figure 7A) and mother rats (Figure 7B) except that the intensity of labeling seemed lower in mother

AC

resulting in less visibly labeled fibers and cell bodies. At the ultrastructural level, C1qbp-ir was present in mitochondria of the PVN as demonstrated by post-embedding immunogold electron microscopy. Within the presynaptic terminals, C1qbp immunoreactivity was predominantly located in mitochondria (Figure 7C,D) where the density of immunogold particles was significantly higher than that in the cytoplasm of the presynaptic terminal (Figure 7E).

17

ACCEPTED MANUSCRIPT C1qbp mRNA was also present in the hypothalamus as detected by in situ hybridization histochemistry. Labeled cells were present in all hypothalamic nuclei where C1qbp-ir cell bodies were found and they had very similar distribution to that of C1qbp-ir cell bodies (Figure 8A2,B2).

PT

The linkage between the C1qbp and maternal adaptation of the brain

RI

According to “shortest path” bioinformatical analysis between C1qbp and maternal

SC

adaptation related hormones, 43 different regulatory proteins were identified, which have relationship with C1qbp protein as well as estrogen, oxytocin, and prolactin hormones. These

NU

proteins are important ligands, protein kinases, receptors, transcription factors and transporters. Furthermore, their relations are colored by the effect (positive, negative,

MA

binding). Among C1qbp related proteins 32 proteins connect to estrogen, 17 proteins connect to oxytocin, and 27 proteins connect to prolactin. Among them 8 proteins have a strong

CE

PT E

maternal adaptation (Fig. 9).

D

relationship with C1qbp and all three maternal hormones suggesting their important role in

Discussion

AC

We present here the first proteomic analysis of the hypothalamic synaptic proteome of mother rats compared to pup-deprived females. The altered synaptic proteins are involved in a broad range of cellular processes, indicating complex cellular proteome changes, which will be discussed in relation to potential functions in maternal adaptation of the hypothalamus. In addition, maternal implications of C1qbp, a protein with validated decreased maternal level will be discussed.

18

ACCEPTED MANUSCRIPT Methodological considerations The 2-D DIGE gel can separate several thousands of proteins, still, only the portion of the proteins present in great amount can be detected with it [18]. Due to the separation efficiency insufficient to separate all proteins in the brain, one spot in the gel can contain two or more proteins, of which the one present with the highest amount based on the MS analysis

PT

was considered in the present study. Furthermore, some of the proteins are present in more

RI

than just one spot, which may be the consequence of their post-translational modifications or

SC

the existence of their isoforms. This fact explains why we detected 33 spots with changed fluorescence levels but identified only 26 different maternally altered proteins. Altogether,

NU

many maternally affected proteins may have remained undetermined with the proteomics method. Therefore, we cannot compare our data to previously published microarray studies,

MA

which identified hundreds of altered genes in the postpartum brain [28-30]. Still, it is relevant to mention that all of our altered proteins showed gene expression changes in the microarray

D

study of Gammie et al., 2016. Importantly, however, our study was performed at the protein

PT E

level, which may be more relevant than the mRNA level as far as the function of most genes. Furthermore, synaptic proteome examination permits the analysis of synapse related

CE

processes, which cannot be approached at the mRNA level. The preparation was carefully verified using electron microscopy. Indeed, all of the significantly changed proteins have

AC

synaptic localization according to the SynProt database (www.synprot.org), which further confirms the validity of our synaptosome preparation. The number of identified maternally altered proteins was large enough to permit systems biological analyses, which we performed in terms of establishing common regulators of the maternally altered proteins to describe new maternal adaptation related processes. Also importantly, the large number of proteins examined provides a good chance to identify maternally relevant, so far obscured proteins, as we present C1qbp in the present study.

19

ACCEPTED MANUSCRIPT

Interpretation of the maternally altered synaptic proteins The relatively high number of the functional clusters (n=13) to which the altered proteins belong suggests that proteins with several different synaptic roles are affected by maternal adaptation of the hypothalamus. The largest functional clusters were mitochondria

PT

related electron transport chain, ATP metabolism (32%) and the tricarboxylic acid cycle

RI

(TCA) (15%) suggesting alterations in synaptic energy metabolism, more specifically in the

SC

oxidative phosphorylation (OXPHOS) system. Almost all of these proteins showed reduced levels in mother rats as opposed to a member of the glycolysis pathway, which showed

NU

elevated level in dams. These findings suggest a shift towards glycolytic energy consumption in maternal synapses and a reduction in the activity of the OXPHOS system. A recent study

MA

revealed that in vivo estradiol treatment enhanced the level of several OXPHOS proteins and also brain mitochondrial efficiency [31,32]. Comparing our results with the findings from

D

Nilsen et al., 2007 reveals that 6 proteins (Aco2, Atp5a1, Dld, Glud1, Hspa60, and Mdh2) had

PT E

reverse changes in quantity. Thus, suppressed maternal estrogen level in the lactation period maybe the underlying reason why OXPHOS proteins changed in decreasing, opposite

CE

direction in mothers. Protein transport, folding and oxidative stress related proteins (Hspa9, Hspd1 and Sod1) were also attenuated in mothers. Since these proteins are upregulated as a

AC

cellular defense reaction [33], their maternal decrease suggests that the cells are less capable to counteract cellular stress during this period. A structural protein (Ina) was in turn increased in mothers suggesting that it might be involved in restructuring and growing mechanisms of the presynaptic terminals, which is in line with intensified synaptic remodeling during the postpartum period [20,22].

Involvement of the common regulators in maternal adaptation processes

20

ACCEPTED MANUSCRIPT Protein kinases, growth factors and cytokines were highly represented among common regulators suggesting significant impact of changed protein classes on the maternal adaptation process. Peptide hormones and transcription factors were less numerous than some other regulators but they had a large number of connections suggesting significant individual impact. Altogether, the main regulators were Myc, Tgfb1, Tnf, and Pkc suggesting their

PT

specifically important role in the neuronal network alterations in mother rats. The Myc protein

RI

promotes cell proliferation [34-36] and impacts mitochondrial biogenesis [37,38]. Tgfb1

SC

directs axon establishment and the formation of neuronal polarity during development [39]. In the adult brain, Tgfb1 is upregulated after brain injury and promotes neural regeneration. In

NU

addition, Tgfb1 was shown to synergize with other neurotrophic factors to support neuronal survival [40-44]. Tnf participates in immune defense, cellular homeostasis and protection

MA

against injury in brain [45]. In the hypothalamus, Tnf probably takes part in the mechanism of food intake regulation [46]. Tnf can also regulate synaptic plasticity and transmission [47,48].

D

Activation of the Pkc pathway can also enhance synaptic remodeling [49,50]. Thus, synaptic

PT E

plasticity in the maternal brain may be driven by the above listed regulators, which are first

CE

suggested to be involved in maternal regulations of the hypothalamus.

Complement component 1q subcomponent-binding protein (C1qbp) in the hypothalamus

AC

The C1qbp protein has some alternative names depending on the place of isolation or the binding partners: ASF/SF2-associated protein p32, Glycoprotein gC1qBP, Hyaluronicbinding protein 1 (Habp1), Mitochondrial matrix protein p32, gC1q-R protein and p33 (www.uniprot.org). We chose this protein for further analysis because it showed a significantly decreased maternal change and also because of its multifunctional roles potentially relevant for maternal adaptations. In addition, it showed the greatest level in synaptic mitochondria as compared to non-synaptic mitochondria in our previous study [25]

21

ACCEPTED MANUSCRIPT suggesting an unusually high degree of its plasticity in the synapse, which could be important for a maternally relevant protein. The distribution of C1qbp in the brain has not been described previously. Thus, we first established that C1qbp immunoreactivity is located in several hypothalamic areas related to maternal adaptations. In situ hybridization histochemistry established the same distribution

PT

pattern of C1qbp-expressing neurons, which confirms the specificity of the antibody. The

RI

preoptic area of the hypothalamus plays crucial roles in maternal motivation and behavior [2-

SC

5]. These, and other preoptic regions may also be involved in lactational anoestrous via suppression GnRH producing neurons [19]. In turn, the Arc plays a critically important role in

NU

the regulation of prolactin secretion via dopaminergic cells [15,16] and also in the regulation of food intake control via orexigenic and anorexigenic neurons located here [51]. The

MA

ventrolateral subdivision of the VMH in turn plays a role in sexual behaviors whose suppression characterizes the postpartum period [52]. The PVN also plays an outstanding role

D

in mothers. Oxytocin neurons located here are activated in mothers to result in the secretion of

PT E

oxytocin from the pituitary for milk ejection [53]. In addition, oxytocin released in the brain contributes to the induction of maternal behaviors [54]. Other magnocellular neurons of the

CE

PVN containing arginine vasopressin may also be involved in the modulation of maternal behaviors [55]. Finally, corticotropin-releasing hormone-containing neurons in the

AC

parvocellular subdivision of the PVN play a role in the stress response, which is suppressed in mothers [56]. Since C1qbp is present in all these brain areas, it is ideally located to affect their function for their adaptation to motherhood.

Potential mechanisms of C1qbp action in mothers C1qbp is a multicompartmental and multifunctional protein, which was first isolated from Raji cells as complement 1q (C1q) binding membrane protein [57] suggesting its role in

22

ACCEPTED MANUSCRIPT neuroimmunological processes. Recently, these processes are considered more and more important in the modification of dendritic spines, synaptic connections and synaptic scaling [58]. Thus, it is conceivable that C1qbp plays a role in the reorganization of hypothalamic connectivity in the maternal brain. C1qbp was found to be a cytoplasmic, cell surface, nuclear and mitochondrial protein

PT

in different studies [57,59-61]. The subcellular localization of C1qbp has even been addressed

RI

by immunogold electron microscopy in cell lines [62]. Labeling was found in mitochondria,

SC

but also in zymogene granules, condensing vacuoles and ER in pancreatic acinar cells, in nuclei of splenic lymphocytes, and on the cell surface of splenic lymphocytes and

NU

microvascular endothelial cells in the kidney. Thus, the amount of C1qbp in different compartments may depend on the type of tissue. Based on our data, the brain, specifically the

MA

PVN, contains C1qbp almost exclusively in the mitochondria, as first examined and established in our study. Recently, C1qbp as a mitochondrial protein has been proposed to be

D

a part of the mitochondrial permeability transition pore (mPTP) complex [63] and provide

The

interaction

of

PT E

protection against cell death due to oxidative stress and mitochondrial calcium overload [64]. C1qbp

Transcription

acetyltransferase

factor

component

A, of

mitochondrial pyruvate

(Tfam),

dehydrogenase

CE

Dihydrolipoyllysine-residue

with

complex, mitochondrial (Dlat), and Cyclophilin D (CypD) was suggested to stimulate

AC

pyruvate dehydrogenase (Pdh) thereby cellular energy metabolism [65-67]. For better understanding of its function, C1qbp-knockout mice were generated [68]. The disruption of C1qbp caused embryonic lethality around E11 due to severe organ development defect of the embryo. Fibroblasts isolated from the C1qbp-deficient mice had higher ATP levels than wild types due to increased glycolytic activity [68]. A similar shift in metabolism from OXPHOS to glycolysis was also observed in C1qbp-knockout human cancer cells [69]. These results are consistent with our findings of low level of OXPHOS proteins and suggest that the reduced

23

ACCEPTED MANUSCRIPT C1qbp level can contribute to a shift of using glycolysis as energy source in maternal presynaptic terminals. Numerous protein partners of C1qbp have a strong connection with estrogen, oxytocin, and prolactin hormones, which strongly influence both behavioral and physiological adaptation of the brain to take care of the offspring during motherhood [13]. These

PT

connections represent potential mechanisms how C1qbp level can be altered in the lactation

RI

period in response to hormonal changes in order to participate in the reorganization of

SC

hypothalamic connectivity or in a shift of using glycolysis as energy source in the maternal brain. Furthermore, the protein partners also provide potential molecular pathways how

MA

behavioral and physiological adaptations.

NU

C1qbp could exert its effect of the maternal hormones thereby influencing maternal

Conclusion

D

The high number of synaptic proteins with altered maternal levels suggests that hypothalamic

PT E

synapses are profoundly changed in the postpartum period. The proteins with altered levels may play a role in the hormonal, behavioral and metabolic adaptations of rat dams to

CE

motherhood. Protein level changes may be carried out with regulators belonging mostly to protein kinases, cytokines and growth factors. One of the altered proteins we identified is

AC

C1qbp, which is present in a number of maternally involved hypothalamic brain regions and could be involved in their maternal adaptations via neuroplasticity and metabolic regulations.

Acknowledgements This work was supported by the KTIA_NAP_B_13-2-2014-0004 program, the NKFIHOTKA K116538 research grant, as well as the KTIA_NAP_13-2-2015-0003 program.

24

ACCEPTED MANUSCRIPT References [1] Peters LC, Sist TC, Kristal MB. Maintenance and decline of the suppression of infanticide in mother rats. Physiol Behav 1991;50:451-6. [2] Numan M. Motivational systems and the neural circuitry of maternal behavior in the rat. Dev Psychobiol 2007;49:12-21.

PT

[3] Olazabal DE, Pereira M, Agrati D, Ferreira A, Fleming AS, Gonzalez-Mariscal G, Levy F,

RI

Lucion AB, Morrell JI, Numan M, Uriarte N. Flexibility and adaptation of the neural substrate

SC

that supports maternal behavior in mammals. Neurosci Biobehav Rev 2013;37:1875-92. [4] Dulac C, O'Connell LA, Wu Z. Neural control of maternal and paternal behaviors. Science

NU

2014;345:765-70.

[5] Cservenak M, Kis V, Keller D, Dimen D, Menyhart L, Olah S, Szabo ER, Barna J, Renner

MA

E, Usdin TB, Dobolyi A. Maternally involved galanin neurons in the preoptic area of the rat. Brain Struct Funct 2016;

D

[6] Scott N, Prigge M, Yizhar O, Kimchi T. A sexually dimorphic hypothalamic circuit

PT E

controls maternal care and oxytocin secretion. Nature 2015;525:519-22. [7] Renier N, Adams EL, Kirst C, et al. Mapping of Brain Activity by Automated Volume

CE

Analysis of Immediate Early Genes. Cell 2016;165:1789-802. [8] Miceli MO, Fleming AS, Malsbury CW. Disruption of maternal behaviour in virgin and

AC

postparturient rats following sagittal plane knife cuts in the preoptic area-hypothalamus. Behav Brain Res 1983;9:337-60. [9] Numan M. Hypothalamic neural circuits regulating maternal responsiveness toward infants. Behav Cogn Neurosci Rev 2006;5:163-90. [10] Rosenblatt JS, Mayer AD, Giordano AL. Hormonal basis during pregnancy for the onset of maternal behavior in the rat. Psychoneuroendocrinology 1988;13:29-46.

25

ACCEPTED MANUSCRIPT [11] Petraglia F, Imperatore A, Challis JR. Neuroendocrine mechanisms in pregnancy and parturition. Endocr Rev 2010;31:783-816. [12] Brunton PJ, Russell JA. Endocrine induced changes in brain function during pregnancy. Brain Res 2010;1364:198-215. [13] Bridges RS. Neuroendocrine regulation of maternal behavior. Front Neuroendocrinol

PT

2015;36:178-96.

RI

[14] Minden JS. DIGE: past and future. Methods Mol Biol 2012;854:3-8.

SC

[15] Li C, Chen P, Smith MS. Identification of neuronal input to the arcuate nucleus (ARH) activated during lactation: implications in the activation of neuropeptide Y neurons. Brain Res

NU

1999;824:267-76.

[16] Chen P, Smith MS. Suckling-induced activation of neuronal input to the dorsomedial

MA

nucleus of the hypothalamus: possible candidates for mediating the activation of DMH neuropeptide Y neurons during lactation. Brain Res 2003;984:11-20.

D

[17] Brunton PJ, Russell JA, Douglas AJ. Adaptive responses of the maternal hypothalamic-

PT E

pituitary-adrenal axis during pregnancy and lactation. J Neuroendocrinol 2008;20:764-76. [18] Diez R, Herbstreith M, Osorio C, Alzate O. 2-D Fluorescence Difference Gel

Raton (FL)2010.

CE

Electrophoresis (DIGE) in Neuroproteomics. In: Alzate O, editor. Neuroproteomics. Boca

AC

[19] Tsukamura H, Maeda K. Non-metabolic and metabolic factors causing lactational anestrus: rat models uncovering the neuroendocrine mechanism underlying the sucklinginduced changes in the mother. Prog Brain Res 2001;133:187-205. [20] Theodosis DT, Poulain DA. Maternity leads to morphological synaptic plasticity in the oxytocin system. Prog Brain Res 2001;133:49-58. [21] Marlin BJ, Mitre M, D'Amour J A, Chao MV, Froemke RC. Oxytocin enables maternal behaviour by balancing cortical inhibition. Nature 2015;520:499-504.

26

ACCEPTED MANUSCRIPT [22] Shams S, Pawluski JL, Chatterjee-Chakraborty M, Oatley H, Mastroianni A, Fleming AS. Dendritic morphology in the striatum and hypothalamus differentially exhibits experience-dependent changes in response to maternal care and early social isolation. Behav Brain Res 2012;233:79-89. [23] Orpen BG, Fleming AS. Experience with pups sustains maternal responding in

PT

postpartum rats. Physiol Behav 1987;40:47-54.

RI

[24] Tandon A, Bannykh S, Kowalchyk JA, Banerjee A, Martin TF, Balch WE. Differential

SC

regulation of exocytosis by calcium and CAPS in semi-intact synaptosomes. Neuron 1998;21:147-54.

NU

[25] Volgyi K, Gulyassy P, Haden K, Kis V, Badics K, Kekesi KA, Simor A, Gyorffy B, Toth EA, Lubec G, Juhasz G, Dobolyi A. Synaptic mitochondria: a brain mitochondria cluster with

MA

a specific proteome. J Proteomics 2015;120:142-57.

[26] Ghafari M, Whittle N, Miklosi AG, Kotlowski C, Schmuckermair C, Berger J, Bennett

D

KL, Singewald N, Lubec G. Dietary magnesium restriction reduces amygdala-hypothalamic

PT E

GluN1 receptor complex levels in mice. Brain Struct Funct 2015;220:2209-21. [27] Dobolyi A, Ueda H, Uchida H, Palkovits M, Usdin TB. Anatomical and physiological

CE

evidence for involvement of tuberoinfundibular peptide of 39 residues in nociception. Proc Natl Acad Sci U S A 2002;99:1651-6.

AC

[28] Dobolyi A. Central amylin expression and its induction in rat dams. J Neurochem 2009;111:1490-500.

[29] Gammie SC, Driessen TM, Zhao C, Saul MC, Eisinger BE. Genetic and neuroendocrine regulation of the postpartum brain. Front Neuroendocrinol 2016; [30] Ray S, Tzeng RY, DiCarlo LM, Bundy JL, Vied C, Tyson G, Nowakowski R, Arbeitman MN. An Examination of Dynamic Gene Expression Changes in the Mouse Brain During Pregnancy and the Postpartum Period. G3 (Bethesda) 2016;6:221-33.

27

ACCEPTED MANUSCRIPT [31] Nilsen J, Irwin RW, Gallaher TK, Brinton RD. Estradiol In Vivo regulation of brain mitochondrial proteome. Journal of Neuroscience 2007;27:14069-77. [32] Simpkins JW, Yang SH, Sarkar SN, Pearce V. Estrogen actions on mitochondria-physiological and pathological implications. Mol Cell Endocrinol 2008;290:51-9. [33] Kalmar B, Greensmith L. Induction of heat shock proteins for protection against

PT

oxidative stress. Adv Drug Deliv Rev 2009;61:310-8.

RI

[34] Zornig M, Evan GI. Cell cycle: on target with Myc. Curr Biol 1996;6:1553-6.

SC

[35] Bouchard C, Staller P, Eilers M. Control of cell proliferation by Myc. Trends Cell Biol 1998;8:202-6.

NU

[36] Nasi S, Ciarapica R, Jucker R, Rosati J, Soucek L. Making decisions through Myc. FEBS Lett 2001;490:153-62.

MA

[37] Le A, Dang CV. Studying Myc's role in metabolism regulation. Methods Mol Biol 2013;1012:213-9.

PT E

Perspect Med 2014;4

D

[38] Morrish F, Hockenbery D. MYC and mitochondrial biogenesis. Cold Spring Harb

[39] Yi JJ, Barnes AP, Hand R, Polleux F, Ehlers MD. TGF-beta signaling specifies axons

CE

during brain development. Cell 2010;142:144-57. [40] Abe K, Chu PJ, Ishihara A, Saito H. Transforming growth factor-beta 1 promotes re-

AC

elongation of injured axons of cultured rat hippocampal neurons. Brain Res 1996;723:206-9. [41] Brionne TC, Tesseur I, Masliah E, Wyss-Coray T. Loss of TGF-beta 1 leads to increased neuronal cell death and microgliosis in mouse brain. Neuron 2003;40:1133-45. [42] Unsicker K, Krieglstein K. Co-activation of TGF-ss and cytokine signaling pathways are required for neurotrophic functions. Cytokine Growth Factor Rev 2000;11:97-102. [43] Dobolyi A, Vincze C, Pal G, Lovas G. The neuroprotective functions of transforming growth factor beta proteins. Int J Mol Sci 2012;13:8219-58.

28

ACCEPTED MANUSCRIPT [44] Pal G, Vincze C, Renner E, Wappler EA, Nagy Z, Lovas G, Dobolyi A. Time Course, Distribution and Cell Types of Induction of Transforming Growth Factor Betas following Middle Cerebral Artery Occlusion in the Rat Brain. PLoS One 2012;7:e46731. [45] Sriram K, O'Callaghan JP. Divergent roles for tumor necrosis factor-alpha in the brain. J Neuroimmune Pharmacol 2007;2:140-53.

PT

[46] Plata-Salaman CR. Cytokines and Feeding. News Physiol Sci 1998;13:298-304.

RI

[47] Vitkovic L, Bockaert J, Jacque C. "Inflammatory" cytokines: neuromodulators in normal

SC

brain? J Neurochem 2000;74:457-71.

[48] Santello M, Volterra A. TNFalpha in synaptic function: switching gears. Trends Neurosci

NU

2012;35:638-47.

[49] Roberson ED, English JD, Adams JP, Selcher JC, Kondratick C, Sweatt JD. The

MA

mitogen-activated protein kinase cascade couples PKA and PKC to cAMP response element binding protein phosphorylation in area CA1 of hippocampus. J Neurosci 1999;19:4337-48.

D

[50] Nelson TJ, Sun MK, Hongpaisan J, Alkon DL. Insulin, PKC signaling pathways and

PT E

synaptic remodeling during memory storage and neuronal repair. Eur J Pharmacol 2008;585:76-87.

CE

[51] Amin T, Mercer JG. Hunger and Satiety Mechanisms and Their Potential Exploitation in the Regulation of Food Intake. Curr Obes Rep 2016;5:106-12.

AC

[52] Kow LM, Bogun M, Zhang Q, Pfaff DW. Hormonal induction of lordosis and ear wiggling in rat pups: gender and age differences. Endocrine 2007;32:287-96. [53] Crowley WR. Neuroendocrine regulation of lactation and milk production. Compr Physiol 2015;5:255-91. [54] Numan M, Young LJ. Neural mechanisms of mother-infant bonding and pair bonding: Similarities, differences, and broader implications. Horm Behav 2016;77:98-112.

29

ACCEPTED MANUSCRIPT [55] Fodor A, Klausz B, Pinter O, Daviu N, Rabasa C, Rotllant D, Balazsfi D, Kovacs KB, Nadal R, Zelena D. Maternal neglect with reduced depressive-like behavior and blunted c-fos activation in Brattleboro mothers, the role of central vasopressin. Horm Behav 2012;62:53951. [56] Slattery DA, Neumann ID. No stress please! Mechanisms of stress hyporesponsiveness

PT

of the maternal brain. J Physiol 2008;586:377-85.

RI

[57] Ghebrehiwet B, Lim BL, Peerschke EI, Willis AC, Reid KB. Isolation, cDNA cloning,

SC

and overexpression of a 33-kD cell surface glycoprotein that binds to the globular "heads" of C1q. J Exp Med 1994;179:1809-21.

NU

[58] Boulanger LM. Immune proteins in brain development and synaptic plasticity. Neuron 2009;64:93-109.

MA

[59] Peterson KL, Zhang W, Lu PD, Keilbaugh SA, Peerschke EI, Ghebrehiwet B. The C1qbinding cell membrane proteins cC1q-R and gC1q-R are released from activated cells:

D

subcellular distribution and immunochemical characterization. Clin Immunol Immunopathol

PT E

1997;84:17-26.

[60] Dedio J, Jahnen-Dechent W, Bachmann M, Muller-Esterl W. The multiligand-binding

42.

CE

protein gC1qR, putative C1q receptor, is a mitochondrial protein. J Immunol 1998;160:3534-

AC

[61] Majumdar M, Meenakshi J, Goswami SK, Datta K. Hyaluronan binding protein 1 (HABP1)/C1QBP/p32 is an endogenous substrate for MAP kinase and is translocated to the nucleus upon mitogenic stimulation. Biochem Biophys Res Commun 2002;291:829-37. [62] Soltys BJ, Kang D, Gupta RS. Localization of P32 protein (gC1q-R) in mitochondria and at specific extramitochondrial locations in normal tissues. Histochem Cell Biol 2000;114:24555.

30

ACCEPTED MANUSCRIPT [63] Starkov AA. The molecular identity of the mitochondrial Ca2+ sequestration system. FEBS J 2010;277:3652-63. [64] Halestrap AP. What is the mitochondrial permeability transition pore? J Mol Cell Cardiol 2009;46:821-31. [65] McGee AM, Baines CP. Complement 1q-binding protein inhibits the mitochondrial

PT

permeability transition pore and protects against oxidative stress-induced death. Biochem J

RI

2011;433:119-25.

SC

[66] Uchiumi T, Kang D. The role of TFAM-associated proteins in mitochondrial RNA metabolism. Biochim Biophys Acta 2012;1820:565-70.

NU

[67] Chen R, Xiao M, Gao H, Chen Y, Li Y, Liu Y, Zhang N. Identification of a novel mitochondrial interacting protein of C1QBP using subcellular fractionation coupled with

MA

CoIP-MS. Anal Bioanal Chem 2016;408:1557-64.

[68] Yagi M, Uchiumi T, Takazaki S, Okuno B, Nomura M, Yoshida S, Kanki T, Kang D.

D

p32/gC1qR is indispensable for fetal development and mitochondrial translation: importance

PT E

of its RNA-binding ability. Nucleic Acids Res 2012;40:9717-37. [69] Fogal V, Richardson AD, Karmali PP, Scheffler IE, Smith JW, Ruoslahti E.

CE

Mitochondrial p32 protein is a critical regulator of tumor metabolism via maintenance of

AC

oxidative phosphorylation. Mol Cell Biol 2010;30:1303-18.

31

ACCEPTED MANUSCRIPT Figure legends

Fig. 1. Dissection of the hypothalamic samples. A: The position of the coronal section cut as a first step of the dissection of the hypothalamus is shown by the 2 vertical lines on the schematic side view of the brain. B: The hypothalamic tissue sample was obtained from the

PT

coronal section by cuts at the solid lines. Further abbreviations: Am – amygdala, Cb –

RI

cerebellum, CP – caudate putamen, Cx – cerebral cortex, Hyp – hypothalamus, OB –

SC

olfactory bulb, Th – thalamus.

NU

Fig. 2. Electron microscopic validation of synaptosome fraction. Sealed presynaptic terminals are dominant (Pre) with small postsynaptic compartment (Post) attached to some of them. The

MA

presynaptic compartments are filled with synaptic vesicles (white arrowheads) and mitochondria (asterisks). Nucleus or other cell body related organelles are not present in the

PT E

D

sample. Scale bar is 500 nm.

Fig. 3. A: Image of the 2-D DIGE gel showing the locations of significantly altered synaptic

CE

proteins from the hypothalamus. Red and blue circles indicate increased and decreased changes in protein spots of mothers, respectively. The spot numbers appear next to the

AC

corresponding circle. Ba: Representative gel image of an internal standard, which was run simultaneously with the Mother-1 sample. Bb: Representative gel image of Mother-1 sample. Ca: Representative gel image of an internal standard, which was run simultaneously with the Pup-deprived-1 sample. Cb: Representative gel image of Pup-deprived-1 sample.

32

ACCEPTED MANUSCRIPT Fig. 4. Functional clustering of the maternally altered synaptic proteins (n=26). The clusters are listed on the right. The percentage of altered proteins participating in the particular cluster, labeled by colors, is shown on the left.

Fig. 5. Common regulators of maternally altered synaptic proteins. The abbreviated names of

PT

altered proteins are circled with orange. Their full names are present in Table 1. Proteins

RI

affected by several common regulators are arranged in the middle line, some other altered

SC

proteins are outside of the name of their regulators. Abbreviations of common regulators, arranged around the altered proteins, are surrounded with blue color. The shape of the blue

NU

surroundings indicates the type of the common regulator listed in the upper right corner. The different types of regulators are connected to the target with differently colored arrows.

MA

Stimulatory connections terminate with arrowheads while inhibitory with perpendicular lines. Abbreviations of common regulators include PI3K: Phosphatidylinositol 3-kinase, SRC:

D

Proto-oncogene tyrosine-protein kinase Src, PKA: Protein kinase A/cAMP-dependent protein

PT E

kinase, AKT1: RAC-alpha serine/threonine-protein kinase, PKC: Protein kinase C, MAPK1: Mitogen-activated protein kinase 1, TNF: Tumor necrosis factor, IFNG: Interferon gamma,

CE

IL10: Interleukin-10, IL6: Interleukin-6, IL1B: Interleukin-1 beta, Ras: Ras superfamily, IGF1: Insulin-like growth factor I, PDGF: Platelet-derived growth factor, TGFB1:

AC

Transforming growth factor beta-1, NGF: Beta-nerve growth factor, MYC: Myc protooncogene protein, CREB1: Cyclic AMP-responsive element-binding protein 1, AGT: Angiotensinogen, INS: Insulin.

Fig. 6. Western blot validation of C1qbp protein level alteration. The level of C1qbp protein is significantly decreased (*: p < 0.05) in hypothalamic synaptosomes from mother rats (n=5) as

33

ACCEPTED MANUSCRIPT compared to pup-deprived controls (n=5). The means and their standard deviations are shown in the graph while representative bands are shown in the picture of the gel.

Fig. 7. C1qbp protein in fibers of the hypothalamus. Aa: C1qbp-ir fibers are present in the arcuate (Arc) and ventromedial hypothalamic nuclei (VMH) of pup-deprived female rats. Cell

PT

bodies (some of then pointed to by the black arrowheads in the Arc and the ventrolateral

RI

subdivision of the VMH) are also labeled. Ab: The high magnification picture shows the

SC

enlarged image of the framed area in Aa. At this magnification, a fine C1qbp-ir fiber network can be demonstrated. Ba: C1qbp-ir fibers and cell bodies in the arcuate and hypothalamic

NU

ventromedial nuclei of a mother rat. The number of visibly labeled cell bodies seems less than in Aa. Bb: C1qbp-ir fibers are present in the arcuate nucleus of mother rat as well, However,

MA

the intensity of labeling appears lower than in Ab. C,D: Post-embedding immunolabeling of C1qbp protein in the paraventricular nucleus in mother rats. The C1qbp protein is almost

D

exclusively located in mitochondria, both in a terminal with asymmetric (C) and symmetric

PT E

(D) synapse. The black arrows show the gold particles in the presynaptic mitochondria, while the white arrows point to the synaptic specializations. E: The density of gold particles in the

CE

synaptic mitochondria is compared to that of the synaptic cytoplasm. ***: p < 0.001. Additional abbreviation: 3V – third ventricle. Scale bars: 400 µm for Aa and Ba, 40 µm for

AC

Ab and Bb, and 200 nm for C and D.

Fig. 8. Distribution of C1qbp-expressing cells in the hypothalamus. A1: C1qbp-ir cell bodies are present in the medial preoptic nucleus (MPN), and in the magnocellular preoptic nucleus (MCPO) as shown by black arrows. A2: A dark-field in situ hybridization histochemistry image demonstrates that C1qbp mRNA-expressing cells are located in the same positions as C1qbp-ir cell bodies in A1. B1: C1qbp-ir cell bodies are present in the paraventricular

34

ACCEPTED MANUSCRIPT hypothalamic nucleus, with the highest density in its magnocellular subdivision (PVNm). B2: C1qbp mRNA-expressing cells are located in the same positions as C1qbp-ir cell bodies in B1. Additional abbreviations: AH - anterior hypothalamic area, f – fornix, LPO – lateral preoptic area, MPA – medial preoptic area, och – optic chiasm, Tu – olfactory tubercle, 3V –

PT

third ventricle. Scale bars: 500 μm for A and B.

RI

Fig. 9. The relationship between C1qbp and hormones responsible for maternal adaptations.

SC

The large number of proteins connects C1qbp with one or more of the three major maternal hormones: estrogen, oxytocin (OXT) and prolactin (PRL). The relations are colored by the

NU

effect: positive (green), negative (red), binding (gray). Different directions of the connections, indicated by arrows, allow regulation of the hormones by C1qbp and also provide molecular

MA

pathways how C1qbp could be affected by the maternal hormones. Estrogen-regulated proteins: HABP2: Hyaluronan-binding protein 2, SRSF1: Serine/arginine-rich splicing factor

D

1, PPID: Peptidyl-prolyl cis-trans isomerase D, CHCHD2: Coiled-coil-helix-coiled-coil-helix

PT E

domain-containing protein 2, TFPI2: Tissue factor pathway inhibitor 2, GTF2B: Transcription initiation factor IIB, KLKB1: Plasma kallikrein, SERPING1: Plasma protease C1 inhibitor,

CE

ADRA1D: Alpha-1D adrenergic receptor, CD44: CD44 antigen, LRP1: Prolow-density lipoprotein receptor-related protein 1. Estrogen and oxytocin regulated proteins: ADRB1:

AC

Beta-1 adrenergic receptor, AVPR2: Vasopressin V2 receptor, TLR4: Toll-like receptor 4, MMP14: Matrix metalloproteinase-14, SERPINC1: Antithrombin-III. Estrogen, oxytocin and prolactin regulated proteins: CDH1: Cadherin-1, IFNG: Interferon gamma, TNF: Tumor necrosis factor, IL10: Interleukin-10, INS: Insulin, CXCL8: Interleukin-8, MAPK1: Mitogenactivated protein kinase 1, PRKCD: Protein kinase C delta type. Estrogen and prolactin regulated proteins: SOCS1: Suppressor of cytokine signaling 1, SOCS3: Suppressor of cytokine signaling 3, CALR: Calreticulin, EGF: Pro-epidermal growth factor, F2:

35

ACCEPTED MANUSCRIPT Prothrombin, IL4: Interleukin-4, PLAUR: Urokinase plasminogen activator surface receptor, ITGB1: Integrin beta-1. Oxytocin and prolactin regulated proteins: TP53: Cellular tumor antigen p53, STAT1: Signal transducer and activator of transcription 1-alpha/beta, IL6: Interleukin-6, INSR: Insulin receptor. Prolactin regulated proteins: CDKN1A: Cyclindependent kinase inhibitor 1, CFD: Complement factor D, IFNB1: Interferon beta, MAPK3:

PT

Mitogen-activated protein kinase 3, AKT1: RAC-alpha serine/threonine-protein kinase,

SC

RI

ADRA1B: Alpha-1B adrenergic receptor, EGFR: Epidermal growth factor receptor.

Table 1. The identity of the significantly altered synaptic proteins in the hypothalamus of

NU

mother rats arranged in functional clusters. Colors of fold change indicate the elevated (red) or reduced (blue) protein level changes in mother rats. The fold change indicates the ratio of

MA

mother rats/pup-deprived rats protein abundance. The sequence coverage is the percentage of the protein sequence covered by identified peptides. The number of unique peptides is the

D

number of peptide sequences that are unique to a protein group, what is important for protein

AC

CE

PT E

identification accuracy. The ρ-value is the statistical significance of the proteins identification.

36

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

AC

CE

PT E

D

MA

Figure 1

37

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

AC

CE

PT E

D

Figure 2

38

AC

CE

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

Figure 3

39

RI

PT

ACCEPTED MANUSCRIPT

AC

CE

PT E

D

MA

NU

SC

Figure 4

40

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

AC

CE

PT E

D

Figure 5

41

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

AC

CE

PT E

D

MA

Figure 6

42

AC

CE

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

Figure 7

43

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

AC

CE

PT E

D

Figure 8

44

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

AC

CE

PT E

D

MA

Figure 9

45

ACCEPTED MANUSCRIPT

AC

CE

PT E

D

MA

NU

SC

RI

PT

Table 1

46

PT

ACCEPTED MANUSCRIPT

AC

CE

PT E

D

MA

NU

SC

RI

Graphical abstract

47

ACCEPTED MANUSCRIPT Highlights We identified 26 protein changes in the hypothalamic synaptosomes of mother rats. Functional clustering suggests a shift from oxidative phosphorylation to glycolysis. Common regulators of the altered proteins are mainly growth factors and cytokines. A decrease in the maternal level of C1q binding protein (C1qbp) was confirmed.

AC

CE

PT E

D

MA

NU

SC

RI

PT

Neurons expressing C1qbp and its presence in synaptic mitochondria were determined.

48

ACCEPTED MANUSCRIPT CONFLICT OF INTEREST STATEMENT

AC

CE

PT E

D

MA

NU

SC

RI

PT

The authors of the manuscript have no conflict of interest.

49