Translation Efficiency of Synaptic Proteins and Its Coding Sequence

Determinants

Shelly Mahlab

, Itai Linial

and Michal Linial

School of Computer Science and Engineering, The Hebrew University of Jerusalem, 91904, Israel

The Racah Institute of Physics, The Hebrew University of Jerusalem, 91904, Israel

Department of Biological Chemistry, Institute of Life Sciences, The Hebrew University of Jerusalem, 91904, Israel

Keywords: Translation Rate, Codon Usage, Neuron, tRNA Adaptation, Endocytosis, Local Translation, Dendrites.

Abstract: The synapse is an organized structure that contains synaptic vesicles, mitochondria, receptors, transporters

and stored proteins. About 10% of the mRNAs that are express in mammalian neurons are delivered to

synaptic sites, where they are subjected to local translation. While neuronal plasticity, learning and memory

occur at the synapse, the mechanisms that regulate post-transcriptional events and local translation are

mostly unknown. We hypothesized that evolutional signals that govern translational efficiency are encoded

in the mRNA of synaptic proteins. Specifically, we applied a measure of tRNA adaptation index (tAI) as an

indirect proxy for translation rate and showed that ionic channels and ligand-binding receptors are specified

by a global low tAI values. In contrast, the genuine proteins of the synaptic vesicles exhibit significantly

higher tAI values. The expression of many of these proteins actually accompanied synaptic plasticity.

Furthermore, in human, the local tAI values for the initial segment of mRNA coding differs for synaptic

proteins in view of the rest of the human proteome. We propose that the translation of synaptic proteins is a

robust solution for compiling with the high metabolic demands of the synapse.

1 INTRODUCTION

Translation must be tightly controlled for coping

with the cell demand and its limited resources.

Energetically, translation is the most expensive

operation in dividing cells (Arava, et al., 2003;

Gingold and Pilpel, 2011; Ingolia, et al., 2009).

Thus, an appropriate regulation of the rate of

translation reduces the ribosomal drop-off, the

translation errors and improves the overall ribosomal

allocation (Zhang, et al., 2010).

In unicellular organisms, it has been shown that

the genomic tRNA copy number (CN) approximates

the levels of intracellular tRNA and thus the codon

usage. Moreover, the relative genomic abundance of

synonymous codons varies in all organisms from

bacteria to mammals (Sharp, et al., 1993), and codon

usage among genes tends to be related to their

expression levels (dos Reis, et al., 2004; Marais and

Duret, 2001; Plotkin and Kudla, 2010; Tuller, et al.,

2010). Specifically, highly expressed genes (e.g,

ribosomal proteins) tend to include codons that are

recognized by abundant tRNA molecules,

suggesting that the control of the translation process

is under a selective pressure.

In all organisms, decoding of mRNA to proteins

occurs by tRNAs. The tRNA anticodon recognizes

the complementary codon or the wobble-based

codon that encodes the same amino acid (Percudani,

2001). In bacteria and fungi, the genomic tRNA CN

correlates with the intracellular tRNA levels

(Ikemura, 1985; Lucks, et al., 2008). A similar trend

is detected in healthy and diseased tissues in human

(Mahlab, et al., 2012). Consequently, the tRNA

adaptation index (tAI) (dos Reis, et al., 2004) is

applied as a measure for ranking the adaptation of a

gene in term of translation elongation. The

assumption is that the availability of relevant tRNA

types has a strong effect on the efficiency and speed

of translation (Mahlab, et al., 2012; Tuller, et al.,

2010).

Synapses are autonomous structures at nerve

terminals that are specified by high metabolic

demands, and functional plasticity. Communication

across the synapse is mediated by neurotransmitters

(NT) and neuropeptides (NP) that are released from

synaptic vesicles (SV) as a result of neuronal

activity. A success coupling of the action potential to

151

Mahlab S., Linial I. and Linial M..

Translation Efﬁciency of Synaptic Proteins and Its Coding Sequence Determinants.

DOI: 10.5220/0004238401510157

In Proceedings of the International Conference on Bioinformatics Models, Methods and Algorithms (BIOINFORMATICS-2013), pages 151-157

ISBN: 978-989-8565-35-8

 2013 SCITEPRESS (Science and Technology Publications, Lda.)

exocytosis requires a coordinated action of priming,

targeting and docking (Ferro-Novick and Jahn,

1994). Actually, tens of proteins that belong to SVs

and secretory granules participate in storing,

docking, fusion and recovery in synapses (e.g.,

worm, fly, human) (Broadie and Richmond, 2002).

The intense energetic demand is maximized (Ames,

1992) upon extensive brain activity and experience,

which is the basis for synapse plasticity (Nestler,

2001).

Most synaptic proteins belong to the secretory

systems. From a cellular perspective, the main sub-

compartments include: (i) the trafficking organelles

such as the Endoplasmic Reticulum (ER), Golgi,

endosomes, lysosomes and secretory granules. (ii)

The plasma membrane (PM) with a partition to pre-

and post-synaptic sites. (iii) The extracellular space

and the synaptic cleft.

Neurons are unique with respect to their ability

for local translation (Martin, et al., 2000). A tight

regulation of translation is achieved by translation

inhibition (Richter and Sonenberg, 2005). Still, 5-

10% of the brain transcripts have the potential for a

local translation at synaptic sites (Gebauer and

Hentze, 2004).

Misfolding of proteins in neurons is the basis for

diseases such as Prion, Alzheimer’s (AD) and

Parkinson’s diseases (PD) (Chiti and Dobson, 2006).

Other conditions with memory loss are associated

with a failure in the balance of synaptic proteins and

their proper folding (Ross and Poirier, 2004).

Unlike vesicle trafficking, the SV fusion in

synaptic structures is tightly regulated in time and

space (Brachya, et al., 2006; Trimble, et al., 1991).

The synaptic protein catalogue (Pielot, et al., 2012;

Yanay, et al., 2008) allows testing the evolutionary

refinement on the translational capacity. In this

study, we examine whether the synapse homeostasis

is governed by managing a stable production of

proteins at the right quantities. We propose a

translational dependent strategy to handle the

extreme metabolic and proteomic demands of

synaptic proteins across model organisms.

In this study, we address the notion of sequence-

encoded component of ‘speed controls’ as shaped by

evolution. We hypothesized that the sequences of

synaptic proteins, especially those needed at high

amounts or under restricted conditions are prone to

production failure. We will not elaborate on

additional critical factors that directly alter

translation elongation such as mRNA folding or

translational initiation (Holcik, et al., 2000).

2 METHODS

2.1 Databases

We retrieved sequences from UniProtKB according

to the selected organisms and their subcellular

localization annotations (Barrell, et al., 2009). We

applied the terms “synapse”, “pre-synaptic” and

“post-synaptic” and “complete proteome”. We

extracted proteins that are marked as ‘fragment’. A

partition of proteins to non-disjoint groups was

performed using the UniProtKB Sequence Features

(FT). We tested features such as ‘Signal peptide’,

‘Transmembrane’ (TMD), ‘Disulfide-bond’ and

‘Coiled coil’.

2.2 Computing tAI

tRNA adaptation index (tAI) was computed

according to (dos Reis, et al., 2004). The adaptation

of tRNAs (tAI) is calculated from the genomic

tRNA CN, combined with thermodynamic

considerations of the codon-anticodon interaction.

While the tAI is associated with each codon, the tAI

of a gene is the average of its codons’ tAI. This

measure gauges the availability of tRNAs for each

codon along an mRNA. As codon-anticodon

coupling is not unique due to wobble interactions,

practically, several anticodons can recognize the

same codon, with somewhat different efficiency.

Formally, Let ni be the number of tRNA

isoacceptors recognizing codon i. Let tCGNij be the

copy number of the jth tRNA that recognizes the ith

codon, and let Sij be the selective constraint on the

efficiency of the codon-anticodon coupling. We

define the absolute adaptiveness, Wi, for each codon

i as:

(1)

From Wi we obtain wi, which is the relative

adaptiveness value of codon i, by normalizing the

Wi values (dividing them by the maximal of all 61

Wi).

2.3 Computing Segmental tAI

Local tAI is calculated by dividing each coding

sequence into several overlapping windows (window

of 30, overlapping by 15 codons). For sequences that

are shorter than 180 amino acids, only local

segmental tAI were calculated. This was applied to

Wi  (1  Si j )tCGNij

j 1



BIOINFORMATICS2013-InternationalConferenceonBioinformaticsModels,MethodsandAlgorithms

152

avoid overlap between N’ and C’ terminal windows.

The successive windows are marked N1, N2 and a

similar notation for C’-terminal. In this study we

only focus on the N’-terminal region.

2.4 Statistical Analysis

Statistical significance, correlations, variance and p-

values were according to standard MatLab suite. We

used the t-test and Kolmogorov–Smirnov (KS)-

statistics.

3 RESULTS

3.1 tAI Values in Worm, Fly and

Human

The major model organisms for studying neuronal

functions include human, worm and fly. These

organisms were used to identify the common

molecular apparatus for fast and slow transmission

in the CNS. We took advantage of the manually

compiled SVs and the synaptic proteins catalogue

from human, fly and worm in order to find the

evolutionary signals that impact their translation

efficiency.

Table 1: Correlations of tAI codons in model organisms.

Hs Dm Ce

H. sapiens 1 0.57 0.46

D. melanogaster 1 0.6

C. elegans 1

The tRNAs CN is subjected to evolutionary forces.

Thus, it is significantly different along the

evolutionary tree. For example, there are 87 tRNAs

in E. coli K12, 287 in S. cerevisiae (budding yeast)

and over 3600 in Bos taurus (cow). The CN for

tRNAs in D. melanogaster (Dm, fruitfly) and C.

elegans (Ce, worm) is 299 and 606, respectively.

The variation in tRNA composition for tRNA

isoacceptors and the fraction for each isoacceptors

from the number of tRNA active genes is converted

to tAI values (dos Reis, et al., 2004) (see Methods).

Table 1 displays the correlation between the tAI

values of 61 codons (excluding stop codons) of the

selected model organisms. For example, for human

and fruit fly it is only 0.57 (p-value=6.7e-7) and the

correlation between C. elegans and D. melanogaster

is 0.6 (p-value=1.9e-7). Thus, the tAI values per se

cannot explain any apparent similarity in translation

signals across these species.

3.2 Global tAI values in synapse

The synapse is an autonomous structure.

Schematically, the synaptic proteins can be assigned

to the following functionalities: (i) SNAREs, the

minimal set of proteins that function in SV docking

and fusion. (ii) Direct regulator of SNAREs (iii) Ion

channels and transporters. (iv) Enzymes and

modifiers (e.g., kinases, phosphatases). (v)

Organizers, mainly PDZ and cytoskeleton proteins.

To test whether the above division is

recapitulated by the calculated tAI values, we

compiled a set of 167 synaptic proteins from C.

elegans as a test case. This relatively small set is

manually curated. For consistency, we maintained

identical annotation protocol throughout the study

(see Methods). The majority of C. elegans synaptic

proteins are channels and transporters (class 1.A.9,

NT receptor, and Ligand-gated ion channel). The

largest group includes >30 LGC (Ligand-gated

channel) proteins. These are the homologs of

vertebrate Glycine receptor superfamily.

Fig. 1 shows the sorted tAI values of the entire

set. About 20% of the genes deviate from the global

tAI mean value by >1 s.d. A segregation of

functional groups with proteins with relatively high

global tAI values was noted (noted as HAT).

Specifically, this fraction in enriched with proteins

that are associated with SV biogenesis, SNAREs and

their direct regulators. The enrichment of low tAI

proteins (LATS) includes ligand gated ion channels

(13 genes, >1 s.d., Enrichment score p-

value=0.0014). None of the ion channel (total 103)

was included in this list of 20 proteins with the

highest tAI values.

We found that extremely low values of global tAI

are associated with ligand gated channels. The

possibility that very high sequence similarity explain

this observation was discard. Actually, the C.

elegans ligand gated channels share only 50%

similarity at the amino acids and less than 40% at

the nucleotide levels. Among the ionotropic NT

receptors that are specified by low global tAI values

are the Acetylcholine (ACh), Glycine, GABA and

Glutamate receptors. Importantly, there is no

difference in view of the tAI between cationic

channels (e.g., AChR) and the anionic channels

(GABA/ Glycine receptors). Note that the overall

topology (i.e., N-terminal region facing the

extracellular / SV lumen space is common to these

channels. A similar distribution in tAI values is

noted for all tested organisms (Fig. 1, human).

TranslationEfficiencyofSynapticProteinsandItsCodingSequenceDeterminants

153

Figura 1: Global tAI of synaptic proteins from C. elegans

(top) and human (bottom). LAT and HAT are proteins that

significantly deviate (>1 s.d.) from the calculated mean

tAI value and are associated with low and high values,

respectively. Relative tAI measures the deviation relative

to the mean which is defined as Relative tAI of 1.0.

3.2.1 tAI Values across Organisms

We repeated the analysis presented above for

synaptic proteins of the fly, worm and human. In

human, we collected 469 synaptic proteins according

to their ‘cellular location’ annotation. A striking

observation is the abundance of membranous

proteins in this set (total of 63%). Some proteins are

associated to the membrane through an indirect

protein-protein interaction (PPI) network, or via a

lipid modification moiety (e.g., GPI anchor).

Functional synapse is strongly dependent on proteins

localization, protein state in term of its modification

and the abundance of proteins. The membranous

proteins in the synapse that contain TMD (single or

multi-pass) comprise the majority of the

membranous synaptic proteins (71% in human). A

large fraction of which includes receptors, ion

channels and transporters that are located to PM. An

additional set comprises proteins that are secreted.

Many of them are short proteins.

We found high correspondence in the list of proteins

that have maximal global tAI values across

organisms. A good example is the Complexin family

(Sudhof and Rothman, 2009). This protein forms a

tight PPI for directly regulating the SNARE complex

formation. As such, it is a major component in

controlling SV exocytosis. Complexins in mammals

are composed of 3 related genes (a single gene in fly

and worm). Multiple sequence alignment for

Complexin from human to chicken and Xenopus

showed that the core of the proteins is highly

conserved (69.4% amino acid identity). More

surprising is the observation that the calculated tAI

is extremely high for all tested organisms and the

calculated tAI value is within the top 15% of the

synaptic proteins (469 in human, 167 in worm and

203 in fly).

The overlap in the list of protein with maximal tAI

along the organisms is very significant. These

proteins are also among the most abundant proteins

of SVs. This set includes the SNAREs (VAMP,

Syntaxin, SNAP-25) as well as synaptotagmin, and

synaptophysin. In addition we noted that key

proteins that participate in endocytosis such as AP-2,

Unc-13 and Endophilin share the property of

extreme tAI value across organisms. This is a highly

significant finding in view of the moderate

correlation in tAI codon values for the tested

organisms (Table 1).

Enrichment tests with respect to GO-Slim

annotations (Barrell, et al., 2009) was performed for

the high and low global tAI values quartiles (the

complete list of synaptic proteins is used as a

background). The enriched terms (p-value <0.01)

for the high tAI include SVs, protein transport,

membrane docking, membrane fusion, synaptic

plasticity. The quartile of the proteins with the

lowest tAI values are enriched with regulation of

small GTPase, and anion transport.

3.3 Local tAI Values in Synapse

The tAI is an indirect measure that affect the

allocation of ribosomes on the transcript. Extreme

values of tAI are associated with ion channels (low),

SNAREs and their primary regulators (high). The

high global tAI of key proteins is in accord of high

production. However, it was proposed that the initial

segment is a critical feature for ribosomal flow

management control.

We evaluate the synaptic proteins in view of the

notion of 'speed controls'. Effectively, it is reflected

by an unequal distribution of low (low adapted

tRNA segments, called LATS) or high tRNA-

adapted codons (HATS). The idea tested extensively

for yeast and E. coli (Tuller, et al., 2010). In

metazoa, the picture is somewhat more complex and

the tAI values strongly correlate with codon usage,

gene expression, protein expression and GC content

(Mahlab, et al., 2012).

Global tAI

HAT

Relative tAI

Human

C. elegans

BIOINFORMATICS2013-InternationalConferenceonBioinformaticsModels,MethodsandAlgorithms

154

We compared the properties of the entire human

proteome and those of the synaptic proteome. Table

2 summarizes the global (the entire coding

sequence) and local (a segment of the coding

sequence) measure of synaptic protein (tAI-local,

coined TAIL). The calculated TAIL for the first

segment of the coding sequence (N1-TAIL) is

reported.

For synaptic proteins we noted that N1-TAIL tAI

value is statistically lower than that of the rest of the

coding gene, for the tested model organisms.

Notably, the average tAI value for all synaptic

proteins was used as a reference. Thus, the inherent

bias for the synaptic proteins was avoided.

The cytosolic fraction in the human proteome

occupies about 70% of all proteins (Fig. 2A, 11,500

proteins). We repeated the test for the local property

to the entire human proteome (TAIL, tAI of a

selected window).

The cytosolic fraction in the human proteome

occupies about 70% of all proteins (Fig. 2A, 11,500

proteins). We repeated the test for the local property

to the entire human proteome (TAIL, tAI of a

selected window).

Table 2: A global measure of tAI for synaptic protein.

 #Syn



#Proteins

Syn

tAI

SynN1

TAIL

P‐value

TAIL/tAI

Hs 469 18,433

0.329

0.328 0.045

Ce 167 3187

0.364

0.348 3.42E‐8

Dm 203 3094

0.329

0.312 1.96E‐9

N1, TAIL for the 30-codon segment of the N-terminal. Syn, synaptic

proteins; Hs, Ce and Dm are the proteomes from human, worm and fly,

respectively.

Fig. 2A shows the distribution of the TAIL values

for cytosolic human proteins relative to the same

analysis for the synaptic proteome (Fig. 2B).

Plotting the results for N1, N2 and N3 (each

sequential segment is 30 codons, no overlap)

indicates that the N1 TAIL is higher than the

following segments (N1>N2≥N3). Thus, the human

proteome is signifies by a N1 that has a tendency for

tRNA-adapted codons relative to the tAI of these

genes. However, for the Synaptic proteins, the

opposite trend holds with a calculated TAIL values

in which N1<N2<N3. While the synaptic proteins

display higher global tAI values (vertical dashed

line, Fig. 2), the N1-segment uses codons that are

slightly less adapted. The statistical analysis between

the two sets (Fig. 2) reveals that the variance in the

values of the synaptic proteome is considerably

lower than that of the entire human proteome (mean,

median and statistical error).

Figura 2: TAIL values of human cytosolic and synaptic

proteomes. (A) Distribution of the TAIL for N1, N2 and

N3. Each window covers 30 non-overlapping codons. (B)

Distribution of the TAIL for N1, N2 and N3 for synaptic

proteins. The global tAI is marked as a dashed line. The x-

axis range is identical for Fig. 2A and 2B.

4 DISCUSSION

Based on the results from this study, we test whether

the synaptic proteome is uniquely marked with

translation ‘speed controls’ signals. Several

properties signify synaptic proteins: the high

tendency of membranous proteins, diverse

cytosleletal proteins, precise subcellular localization

and accurate allocation of proteins to organelles (e.g.

SVs, endosomes, recycling vesicles). Many of the

proteins of the synapse share properties such as the

abundance of disulfide bonds, coiled-coil and

membrane association.

The SV is an autonomous organelle that accounts

for a substantial fraction of the protein mass in the

synapse. For example, synaptophysin and VAMP

together, account for 5% of all synaptic proteins.

The SV anatomy supports such load. A pre-synaptic

structure in the human or mouse hippocampus

contains about 300 SVs. Each of the SVs is

composed of tens of proteins. The quantitative

composition of the SVs was revisited using mass

spectrometry (Takamori, et al., 2006) and

quantitative antibodies imaging tools (Mutch, et al.,

2011). Each of the key proteins (VAMP, Rab3, SV2

and Synaptophysin) appears with 5-30 copies per

TranslationEfficiencyofSynapticProteinsandItsCodingSequenceDeterminants

155

SV. Consequently, the mass and protein packing in

SV is maximal.

5 CONCEPTUAL REMARKS

We postulate that variants in N1-TAIL are attractive

to cope with changing metabolic and activity status

of the synapse.

We suggest that in addition to the regulation for

uORFs (upstream ORFs) and activation of

alternative splicing, the translation regulation is an

additional mode for fine-tuning the overall protein

production. Investigating translational signals along

the transcripts of synapse is only in its infancy. We

expect methods such as ribosomal profiling (Ingolia,

et al., 2009) to provide quantitative data for

translation speed and efficiency.

An open question that we began to explore

challenges the translational management with

respect to ribosomal subunits in dendrites (and other

neuronal compartment) (Sutton and Schuman,

2006). We postulate that local translation is critical

for fast and efficient translation under conditions of

restricted resources. Recently, it was shown that

hundreds of mRNAs are localized to neuronal

compartments such as synaptic neuropils (Cajigas, et

al., 2012). Translation of such mRNA must be

highly regulated at the levels of transcript

accessibility and the translation efficiency. Here, we

provide a glimpse on an overlooked evolutionary

encoded signal for managing translation of synaptic

proteins.

ACKNOWLEDGEMENTS

We thank Amos Stern for useful discussions. The

work is supported by Prospects EU FRVII

consortium.

REFERENCES

Ames, A., 3rd (1992) Energy requirements of CNS cells as

related to their function and to their vulnerability to

ischemia: a commentary based on studies on retina,

Can J Physiol Pharmacol, 70 Suppl, S158-164.

Arava, Y., et al. (2003) Genome-wide analysis of mRNA

translation profiles in Saccharomyces cerevisiae, Proc

Natl Acad Sci U S A, 100, 3889-3894.

Barrell, D., et al. (2009) The GOA database in 2009-an

integrated Gene Ontology Annotation resource,

Nucleic Acids Res, 37, D396-403.

Brachya, G., et al. (2006) Synaptic proteins as multi-

sensor devices of neurotransmission, BMC Neurosci,

7 Suppl 1, S4.

Broadie, K.S. and Richmond, J.E. (2002) Establishing and

sculpting the synapse in Drosophila and C. elegans,

Curr Opin Neurobiol, 12, 491-498.

Cajigas, I.J., et al. (2012) The local transcriptome in the

synaptic neuropil revealed by deep sequencing and

high-resolution imaging, Neuron, 74, 453-466.

Chiti, F. and Dobson, C.M. (2006) Protein misfolding,

functional amyloid, and human disease, Annu Rev

Biochem, 75, 333-366.

dos Reis, M., et al. (2004) Solving the riddle of codon

usage preferences: a test for translational selection,

Nucleic Acids Res, 32, 5036-5044.

Ferro-Novick, S. and Jahn, R. (1994) Vesicle fusion from

yeast to man, Nature, 370, 191-193.

Gebauer, F. and Hentze, M.W. (2004) Molecular

mechanisms of translational control, Nat Rev Mol Cell

Biol, 5, 827-835.

Gingold, H. and Pilpel, Y. (2011) Determinants of

translation efficiency and accuracy, Mol Syst Biol, 7,

481.

Holcik, M., et al. (2000) Internal ribosome initiation of

translation and the control of cell death, Trends Genet,

16, 469-473.

Ikemura, T. (1985) Codon usage and tRNA content in

unicellular and multicellular organisms, Mol Biol

Evol, 2, 13-34.

Ingolia, N.T., et al. (2009) Genome-wide analysis in vivo

of translation with nucleotide resolution using

ribosome profiling, Science, 324, 218-223.

Lucks, J.B., et al. (2008) Genome landscapes and

bacteriophage codon usage, PLoS Comput Biol, 4,

e1000001.

Mahlab, S., et al. (2012) Conservation of the relative

tRNA composition in healthy and cancerous tissues,

RNA. 18, 640-652.

Marais, G. and Duret, L. (2001) Synonymous codon

usage, accuracy of translation, and gene length in

Caenorhabditis elegans, J Mol Evol, 52, 275-280.

Martin, K.C., et al. (2000) Local protein synthesis and its

role in synapse-specific plasticity, Curr Opin

Neurobiol, 10, 587-592.

Mutch, S.A., et al. (2011) Protein quantification at the

single vesicle level reveals that a subset of synaptic

vesicle proteins are trafficked with high precision, J

Neurosci, 31, 1461-1470.

Nestler, E.J. (2001) Molecular basis of long-term plasticity

underlying addiction, Nat Rev Neurosci, 2, 119-128.

Percudani, R. (2001) Restricted wobble rules for

eukaryotic genomes, Trends Genet, 17, 133-135.

Pielot, R., et al. (2012) SynProt: A Database for Proteins

of Detergent-Resistant Synaptic Protein Preparations,

Front Synaptic Neurosci, 4, 1.

Plotkin, J.B. and Kudla, G. (2010) Synonymous but not

the same: the causes and consequences of codon bias,

Nat Rev Genet, 12, 32-42.

Richter, J.D. and Sonenberg, N. (2005) Regulation of cap-

dependent translation by eIF4E inhibitory proteins,

BIOINFORMATICS2013-InternationalConferenceonBioinformaticsModels,MethodsandAlgorithms

156

Nature, 433, 477-480.

Ross, C.A. and Poirier, M.A. (2004) Protein aggregation

and neurodegenerative disease, Nat Med, 10 Suppl,

S10-17.

Sharp, P.M., et al. (1993) Codon usage: mutational bias,

translational selection, or both? Biochem Soc Trans,

21, 835-841.

Sudhof, T.C. and Rothman, J.E. (2009) Membrane fusion:

grappling with SNARE and SM proteins, Science,

323, 474-477.

Sutton, M.A. and Schuman, E.M. (2006) Dendritic protein

synthesis, synaptic plasticity, and memory, Cell, 127,

49-58.

Takamori, S., et al. (2006) Molecular anatomy of a

trafficking organelle, Cell, 127, 831-846.

Trimble, W.S., et al. (1991) Cellular and molecular

biology of the presynaptic nerve terminal, Annu Rev

Neurosci, 14, 93-122.

Tuller, T., et al. (2010) An evolutionarily conserved

mechanism for controlling the efficiency of protein

translation, Cell, 141, 344-354.

Yanay, C., et al. (2008) Evolution of insect proteomes:

insights into synapse organization and synaptic vesicle

life cycle, Genome Biol, 9, R27.

Zhang, Z., et al. (2010) Nonsense-mediated decay targets

have multiple sequence-related features that can

inhibit translation, Mol Syst Biol, 6, 442.

TranslationEfficiencyofSynapticProteinsandItsCodingSequenceDeterminants

157