RNA-binding is an ancient trait of the Annexin family.

Introduction: The regulation of intracellular functions in mammalian cells involves close coordination of cellular processes. During recent years it has become evident that the sorting, trafficking and distribution of transport vesicles and mRNA granules/complexes are closely coordinated to ensure effective simultaneous handling of all components required for a specific function, thereby minimizing the use of cellular energy. Identification of proteins acting at the crossroads of such coordinated transport events will ultimately provide mechanistic details of the processes. Annexins are multifunctional proteins involved in a variety of cellular processes associated with Ca2+-regulation and lipid binding, linked to the operation of both the endocytic and exocytic pathways. Furthermore, certain Annexins have been implicated in the regulation of mRNA transport and translation. Since Annexin A2 binds specific mRNAs via its core structure and is also present in mRNP complexes, we speculated whether direct association with RNA could be a common property of the mammalian Annexin family sharing a highly similar core structure. Methods and results: Therefore, we performed spot blot and UV-crosslinking experiments to assess the mRNA binding abilities of the different Annexins, using annexin A2 and c-myc 3'UTRs as well as c-myc 5'UTR as baits. We supplemented the data with immunoblot detection of selected Annexins in mRNP complexes derived from the neuroendocrine rat PC12 cells. Furthermore, biolayer interferometry was used to determine the KD of selected Annexin-RNA interactions, which indicated distinct affinities. Amongst these Annexins, Annexin A13 and the core structures of Annexin A7, Annexin A11 bind c-myc 3'UTR with KDs in the nanomolar range. Of the selected Annexins, only Annexin A2 binds the c-myc 5'UTR indicating some selectivity. Discussion: The oldest members of the mammalian Annexin family share the ability to associate with RNA, suggesting that RNA-binding is an ancient trait of this protein family. Thus, the combined RNA- and lipid-binding properties of the Annexins make them attractive candidates to participate in coordinated long-distance transport of membrane vesicles and mRNAs regulated by Ca2+. The present screening results can thus pave the way for studies of the multifunctional Annexins in a novel cellular context.

The flavagline FL3 interferes with the association of Annexin A2 with the eIF4F initiation complex and transiently stimulates the translation of annexin A2 mRNA.

Introduction: Annexin A2 (AnxA2) plays a critical role in cell transformation, immune response, and resistance to cancer therapy. Besides functioning as a calcium- and lipidbinding protein, AnxA2 also acts as an mRNA-binding protein, for instance, by interacting with regulatory regions of specific cytoskeleton-associated mRNAs. Methods and Results: Nanomolar concentrations of FL3, an inhibitor of the translation factor eIF4A, transiently increases the expression of AnxA2 in PC12 cells and stimulates shortterm transcription/translation of anxA2 mRNA in the rabbit reticulocyte lysate. AnxA2 regulates the translation of its cognate mRNA by a feed-back mechanism, which can partly be relieved by FL3. Results obtained using the holdup chromatographic retention assay results suggest that AnxA2 interacts transiently with eIF4E (possibly eIF4G) and PABP in an RNA-independent manner while cap pulldown experiments indicate a more stable RNA-dependent interaction. Short-term (2 h) treatment of PC12 cells with FL3 increases the amount of eIF4A in cap pulldown complexes of total lysates, but not of the cytoskeletal fraction. AnxA2 is only present in cap analogue-purified initiation complexes from the cytoskeletal fraction and not total lysates confirming that AnxA2 binds to a specific subpopulation of mRNAs. Discussion: Thus, AnxA2 interacts with PABP1 and subunits of the initiation complex eIF4F, explaining its inhibitory effect on translation by preventing the formation of the full eIF4F complex. This interaction appears to be modulated by FL3. These novel findings shed light on the regulation of translation by AnxA2 and contribute to a better understanding of the mechanism of action of eIF4A inhibitors.

Annexin A2 binds the internal ribosomal entry site of c-myc mRNA and regulates its translation.

The expression and localization of the oncoprotein c-Myc is highly regulated at the level of transcription, mRNA transport, translation, as well as stability of the protein. We previously showed that Annexin A2 (AnxA2) binds to a specific localization element in the 3'untranslated region (UTR) of c-myc mRNA and is involved in its localization to the perinuclear region. In the present study, we demonstrate that AnxA2 binds in a Ca2+-dependent manner to the internal ribosomal entry site (IRES) containing two pseudo-knots in the 5´UTR of the c-myc mRNA. Here, we employ an in vitro rabbit reticulocyte lysate system with chimeric c-myc reporter mRNAs to demonstrate that binding of AnxA2 to the c-myc IRES modulates the expression of c-Myc. Notably, we show that low levels of AnxA2 appear to increase, while high levels of AnxA2 inhibits translation of the chimeric mRNA. However, when both the AnxA2-binding site and the ribosomal docking site in the c-myc IRES are deleted, AnxA2 has no effect on the translation of the reporter mRNA. Forskolin-treatment of PC12 cells results in upregulation of Ser25 phosphorylated AnxA2 expression while c-Myc expression is down-regulated. The effect of forskolin on c-Myc expression and the level of Ser25 phosphorylated AnxA2 was abolished in the presence of EGTA. These findings indicate that AnxA2 regulates both the transport and subsequent translation of the c-myc mRNA, possibly by silencing the mRNA during its transport. They also suggest that AnxA2 act as a switch to turn off the c-myc IRES activity in the presence of calcium.Abbreviations: AnxA2, Annexin A2; ß2--µglob, ß2-microglobulin; cpm, counts per minute; hnRNP, heterogenous nuclear ribonucleoprotein; IRES, internal ribosomal entry site; ITAF, IRES trans-acting factor; MM, multiple myeloma; PABP, poly(A)-binding protein; PCBP, poly(rC) binding protein; PSF, PTB-associated splicing factor; PTB, polypyrimidine tract binding protein; RRL, rabbit reticulocyte lysate; UTR, untranslated region; YB, Y-box binding protein.

Corrigendum: Dynamic expression of long noncoding RNAs and repeat elements in synaptic plasticity.

[This corrects the article on p. 351 in vol. 9, PMID: 26483626.].

Proteins linked to spatial memory formation of CD1 mice in the multiple T-maze.

In own previous work CD1 mice were tested in the Multiple T-maze (MTM), a robust land maze allowing determination of latency to reach the goal box with food reward and to evaluate correct decisions made on the way to the goal box. Herein, hippocampi of these animals were used for the current study with the aim to investigate differences in protein levels between trained and yoked mice and, moreover, to determine differences in protein levels between trained and yoked mice with and without memory formation in the MTM. Three training sessions were carried out for four training days each, followed by probe trials on Days 5 and 12. Good and no-performers in the MTM were separated based on means and median of latency to reach the goal box on probe trial Day 12. Six hours following the probe trial on Day 12, animals were sacrificed and hippocampi were taken. Proteins were extracted and run on two-dimensional gel electrophoresis, spots were quantified and differentially expressed proteins were identified by mass spectrometry using an ion trap. Levels of 17 proteins were significantly different in trained vs. yoked mice. Seven proteins were differentially expressed comparing trained vs. yoked mice from good and no-performers. A series of proteins were significantly correlated with latency and may link these proteins to spatial memory formation. Differential protein expression in trained vs. yoked mice and in good and no-performers may allow insight into spatial memory formation as well as represent tentative pharmacological targets.

Proteins linked to extinction in contextual fear conditioning in the C57BL/6J mouse.

Studying fear extinction is a major topic in neuroscience. No information on systematic studies on the linkage of contextual fear conditioning (cFC) with hippocampal protein levels is available and we were therefore interested in protein differences between animals with poor and good extinction. cFC was carried out in C57BL/6J mice, hippocampi were taken and proteins were run on two-dimensional gel electrophoresis with subsequent quantification of protein spots. In-gel digestion with trypsin and identification by ion trap MS/MS (high-capacity ion trap) was used for the identification of significantly different hippocampal proteins between mice with good and poor performance of extinction. Signaling protein ras-related protein rab-7A and septin 8 levels were significantly higher in hippocampus of poor extinguishers, whereas ubiquitin carboxyterminal hydrolase isozyme L1 showed higher levels in animals with good extinction performance. A series of additional proteins showed significantly different levels between groups but the abovementioned were confirmed by immunoblotting. The abovementioned proteins have never been reported to be linked to extinction, memory, or learning and herein evidence for the involvement of several proteins in extinction mechanism as well as probably representing pharmaceutical targets is provided. Moreover, it is intriguing to demonstrate the differences between good and poor extinction performance at the protein level.

NMDA-complexes linked to spatial memory performance in the Barnes maze in CD1 mice.

The N-methyl-d-aspartic acid receptor (NMDAR) is a well-documented key element in the formation of several memories including spatial, olfactory and contextual memory. Although receptor subunits have been linked to memory formation, data on the involvement of the NMDAR complexes is limited. In previous work CD1 mice were trained in the Barnes maze, a low-stress landmaze, and yoked controls were serving as controls. Hippocampal samples from this behavioural study were taken for comparing NMDAR complexes. Hippocampi were taken and stored until analysis at -80 °C. Membrane proteins were extracted from hippocampi using an ultracentrifugation step and applied on Blue Native gels that in turn were used for immunoblotting with antibodies against subunits NR1, NR2A and NR2B. The subunit content of the complexes was determined by denaturing two-dimensional gel electrophoresis and subsequent immunoblotting. An NMDAR complex with an apparent molecular weight between between 146 and 242 kDa, probably representing an NR1 dimer was the only complex that was significantly different between trained and yoked animals. A series of NMDAR complexes containing modulatory subunits NR2A or NR2B or both were detected. All complexes contained the NR1 subunit. The NR1 dimer complex level, increased in memory formation, may be directly or indirectly involved in the process of spatial memory formation in the CD1 mouse. The results are enabling and challenging further NMDAR studies, both, at the pharmacological and molecular level. Moreover, several NMDAR complexes in the CD1 mouse were shown to be mainly heteropolymers of subunits NR1, NR2A and NR2B, although other recently described subunits were not tested due to unavailability of specific antibodies. Determination of native receptor complexes rather than individual subunits is mandatory and provides the molecular basis for understanding mechanisms of spatial memory.

Differences in hippocampal protein levels between C57Bl/6J, PWD/PhJ, and Apodemus sylvaticus are paralleled by differences in spatial memory.

There is a significant strain-dependent performance in the Morris water maze (MWM), a paradigm for the evaluation of spatial memory. In contrast, information on molecular differences that may be responsible for differences in spatial memory performance is limited. The aim of the study was therefore to investigate differences in hippocampal protein levels in three groups with different performance in the MWM. C57BL/6J (inbred laboratory strain), PWD/PhJ (inbred strain derived from wild animals of Mus musculus), and Apodemus sylvaticus (AS, genus Apodemus) mice were used for the experiments. Proteins from hippocampi, obtained from a behavioral study on these animals, were extracted and run on two-dimensional gel electrophoresis. Proteins spots were quantified, and spots with significantly different levels were identified by mass spectrometry using an ion trap. A series of 49 proteins from different pathways and cascades (signaling, neuronal network, protein synthesis, secretion and degradation, and antioxidant system; intermediary, fat, and carbohydrate metabolism) were significantly different among hippocampi at the stringent statistical level of P ≤ 0.001. These findings are paralleled by differences in the spatial navigation abilities between the strains within the species Mus musculus (C57BL/6 vs. PWD/PhJ) and between the genera Mus and Apodemus. As shown previously, AS learned the task in the MWM and showed good memory retention when tested at the probe trial (day 12), whereas C57BL/6J learned the task, but failed at the probe trial at day 12 as well as PWD/PhJ that failed to learn the task and failed at the probe trial at day 12. A list of above-mentioned proteins were different between PWD/PhJ with bad and AS with good memory retention and may therefore be related to performance in the MWM and thus to spatial memory formation. The experimental approach, however, does not allow discriminating between differences in protein levels a priori and different protein levels induced by the MWM testing. © 2010 Wiley-Liss, Inc.

A serotonin receptor 1A containing complex in hippocampus of PWD/PhJ mice is linked to training effects in the Barnes maze.

Although the involvement of the serotonin 1A receptor (5-HT(1A) R) in memory has been shown by several groups there is no information about the 5-HT(1A) R complex but rather the monomeric form. Moreover, there is insufficient information on the characterization of the antigen, the 5-HT(1A) R. PWD/PhJ mice, a wild-caught strain that was inbred at The Jackson Laboratories were used for the experiments. The Barnes maze (BM) paradigm for the evaluation of spatial memory was chosen because of experience with this setting in our laboratory. Mice were sacrificed 6 h following the probe trial on day 12, hippocampi were extirpated, proteins extracted and run on blue native gels with subsequent immunoblotting with a specific antibody against the mouse 5-HT(1A) R. Densitometry of the single band from the immunoblots was carried out. The receptor from the complex was identified by mass spectrometry (nano-LC-ESI-MS/MS). A serotonin receptor 1A complex was identified by immunoblotting at the apparent molecular weight of 480 kDa indicating the presence of a homopolymer as denaturation only revealed a single band at approx. 50 kDa. 5-HT(1A) R levels were significantly higher in the trained group as compared to yoked controls. The hippocampal 5-HT(1A) R of the trained group was unambiguously identified. Taken together, a serotonin receptor 1A homopolymer is associated with memory training effects in the BM using PWD/PhJ mice. This observation shows that a specific complex rather than a receptor subunit as previously shown is involved in memory process and the receptor protein was characterized.

Linkage of hippocampal proteins to spatial memory formation and strain-dependence in Apodemus sylvaticus, C57BL/6J and PWD/PhJ mice.

Strain-dependence of performance in the Morris water maze (MWM) has been reported but information on underlying differences at the protein level as well as linkage of hippocampal proteins to memory is limited. No data are available on differences in hippocampal protein levels between (within-strain) "good" and "bad" performers in the MWM. Hippocampi of wild-caught mouse strain Apodemus sylvaticus (AS), laboratory inbred mouse strain C57BL/6J and wild-derived inbred mouse strain PWD/PhJ, that were tested in the MWM, were taken and postsynaptic density protein 95, neuronal marker protein NeuN, dendritic spine protein drebrin, synapsins 1a and 1b and synaptosomal protein syntaxin of the SNARE complex were determined by immunoblotting. These mouse strains with known different performance in the MWM showed different hippocampal protein patterns and by the use of yoked controls proteins could be identified as linked to memory formation. When mice, classified as good or bad performers by the median of time spent in the target quadrant on day 12 of the MWM, experiments, NeuN was discriminating good from bad performers in AS at the stringent statistical level of P<0.0001, an effect not seen in the corresponding yoked controls. Strain-dependence of performance in the MWM was reflected by different hippocampal protein levels. Most hippocampal proteins given above were linked to memory formation in the MWM. The finding that the neuronal protein NeuN is able to discriminate between good and bad performers in the MWM may be of major interest and may open a new area in the search for protein markers of spatial memory performance.

Involvement of individual hippocampal signaling protein levels in spatial memory formation is strain-dependent.

Although a series of signaling cascades involved in spatial memory have been identified, their link to spatial memory and strain-dependent expression has not been reported so far. Hippocampal levels of the abovementioned signaling proteins were determined in laboratory inbred strain C57BL/6J, the wild-derived inbred strain PWD/PhJ and the wild caught mouse Apodemus sylvaticus (AS) by immunoblotting. The resulting hippocampal protein levels were correlated with results from MWM. Hippocampal signaling protein (hSP) levels were tested also in yoked controls. Within-strain comparison between trained and yoked controls revealed significant differences between levels of Phospho-CaMKII (alpha), Phospho-CREB, Egr-1, c-Src, Phospho-ERK5, Phospho-MEK5 and NOS1 in all of the three strains tested. In addition, the three strains revealed different involvement of individual hSP levels clearly indicating that individual mouse strains were linked to individual hSPs in spatial memory. Phospho-ERK5 levels were not detectable in hippocampi of yoked controls of each strain. We learn from this study that a series of hSPs are associated with spatial memory and that different hSPs are linked to spatial memory in different strains that show different outcome in the MWM. Even correlational patterns in the individual hSPs differed between mouse strains. This is of importance for the interpretation of previous studies on the abovementioned signaling cascades as well as for the design of future studies on these hippocampal proteins. It is intriguing that individual mouse strains, laboratory or wild caught, may use different signaling pathways for spatial memory in the Morris water maze.

Hippocampal protein levels related to spatial memory are different in the Barnes maze and in the multiple T-maze.

The Multiple T-maze (MTM) and the Barnes maze (BM) are land mazes used for the evaluation of spatial memory. The observation that mice are performing differently in individual mazes made us test the hypothesis that differences in cognitive performances in the two land mazes would be accompanied by differences in hippocampal protein levels. C57BL/6J mice were tested in the BM and in the MTM, hippocampi were extirpated 6 h following the probe trials each, and proteins were extracted for gel-based proteomic analysis. Mice learned the task in both paradigms. Levels of hippocampal proteins from several pathways including signaling, chaperone, and metabolic cascades were significantly different between the two spatial memory tasks. Protein levels were linked to spatial memory specifically as yoked controls were used.

Evaluation of spatial memory of C57BL/6J and CD1 mice in the Barnes maze, the Multiple T-maze and in the Morris water maze.

Evaluation of spatial learning and memory is mainly carried out using the Morris water maze as a single paradigm. We intended to test whether mice in the Barnes maze and Multiple T-maze would lead to comparable results and to test two individual mouse strains with different anxiety levels. C57BL/6J and CD1 male mice were used in the experiments. During the acquisition phase, learning was measured using parameters latency, path length, errors in the BM and correct decisions in MTM. Mice were trained for 4 days and probe trials were performed on days 5 and 12. Latencies reduction over the training period indicated that both strains learned all tasks. During retention phase at days 5 and 12 C57BL/6J performed the Barnes maze and Multiple T-maze task better than CD1 mice while CD1 performed better than C57BL/6J in the Morris water maze. In the BM at day 12, C57BL/6J kept the level of visits to target observed at day 5 whereas CD1 performed worse. Strain- and task-dependent differences were observed using the three mazes. Therefore, fair evaluation of spatial memory demands application of (at least) two different test systems, a water- and a land maze. Different anxiety-related behaviour as well as stress-responses in the strains used may help to interpret the findings reported and again may propose the use of at least two mouse strains when robust evaluation of spatial memory is considered.

Apodemus sylvaticus (LOXT) is a suitable mouse strain for testing spatial memory retention in the Morris water maze.

Information on the difference in cognitive function between laboratory and wild-caught mice is anecdotal and this question has not been systematically studied. Moreover, studying a wild-caught mouse strain per se may add information to the repertoire of mouse strains available. We aimed to study spatial memory in a wild mouse strain (Apodemus sylvaticus, AS) as compared to two individual laboratory mouse strains. Male AS (n=20), CD1 (n=19) and C57BL/6J mice (n=19), 12-14 weeks old, were used in the experiments. The Morris water maze (MWM) was used for determination of spatial memory and time spent in the target quadrant at time points 5 (D5) and 12 days (D12) was evaluated. During the acquisition phase latency to reach the platform and path length to reach the platform was evaluated. Following four training days on day 5 (D5), time spent in the target quadrant was highest in AS>CD1>C57BL/6J (P<0.006). On day 12 (D12), time spent in the target quadrant was significantly higher in AS than in both other strains (P<0.001). All animals learned the task and during the acquisition phase, latency to reach the platform as well as path length decreased significantly in AS. It is concluded that the AS is the most suitable strain for the evaluation of spatial memory in the MWM and is presenting with memory retention superior to laboratory mouse strains CD1 and C57BL/6J.