Peptidomics of the zebrafish Danio rerio: In search for neuropeptides

Highlights

• Peptidomics approach was undertaken to identify peptides from the zebrafish brain.

• We were able to sequence 62 peptides that belong to 34 different precursors.

• Several shortened forms could be identified, yielding 105 identifications in total.

Abstract

(Neuro)peptides are small messenger molecules that are derived from larger, inactive precursor proteins by the highly controlled action of processing enzymes. These biologically active peptides can be found in all metazoan species where they orchestrate a wide variety of physiological processes. Obviously, detailed knowledge on the actual peptide sequences, including the potential existence of truncated versions or presence of post-translation modifications, is of high importance when studying their function. A peptidomics approach therefore aims to identify and characterize the endogenously present peptide complement of a defined tissue or organism using liquid chromatography and mass spectrometry. While the zebrafish Danio rerio is considered as an important aquatic model for medical research, neuroscience, development and ecotoxicology, very little is known about their peptidergic signaling cascades. We therefore set out to biochemically characterize endogenously present (neuro)peptides from the zebrafish brain. This peptidomics setup yielded > 60 different peptides in addition to various truncated versions.

Significance

Though the zebrafish is a well-established model organism to study vertebrate biology and gene functions in either a medical or (eco)toxicological context, very little knowledge about neuropeptidergic signaling cascades is available. We therefore set out to characterize endogenously present peptides from the zebrafish brain using a peptidomics setup yielding a total number of 105 peptide identifications. To our knowledge, it is the first attempt to biochemically isolate and characterize neuropeptides from a fish species in a high-throughput manner. This archive of identified endogenous peptides is likely to aid further functional elucidation of defined neuropeptidergic signaling systems (e.g. characterization of cognate G-protein coupled receptors). Furthermore, our methodology allows studying the changes in peptide expression in response to changes in the organism or the environment using differential peptidomics.

Introduction

In virtually all Metazoan species, a broad range of diverse signaling molecules exist including small molecule neurotransmitters like acetylcholine (ACh), γ-aminobutyric acid (GABA), nitric oxide, excitatory amino acids like glutamate and biogenic amines such as octopamine, tyramine, serotonin (5-HT) and dopamine. In contrast to the small-molecule neurotransmitters, peptidergic signaling molecules are in vivo mostly derived from inactive preproproteins or peptide precursors in which one or multiple peptide sequences are contained. The bioactive peptides interact with cell surface receptors (mostly G-protein-coupled receptors (GPCRs)) to trigger an intracellular signaling pathway as to govern a diverse array of physiological processes and behaviors in fish (as in all other Metazoans) such as feeding, locomotion and reproduction. As they are structurally diverse, their signaling cascades are highly variable, hereby harboring a tremendous potential of different effects on living cells. Because of their critical signaling role, peptides, their processing enzymes or cognate receptors can be considered as attractive targets for pharmaceuticals [1], [2], [3], [4]. In order to obtain the biologically active entities, inactive preproproteins or peptide precursors have to undergo extensive posttranslational processing in the trans-Golgi network and dense core vesicles to produce the bioactive (neuro)peptides. After cleavage of the aminoterminal signal peptide, proprotein convertases (PCs) cleave the remaining part of the precursor at defined cleavage motifs containing basic amino acids (mainly KR and RR, while RK and KK are found in lower frequency); sometimes, the two basic residues are separated from each other by 2, 4, 6 or 8 other residues and were earlier described as “monobasic” cleavage sites [1], [4], [5]. In mammals, these cleavage motifs are specifically recognized by PC2 and PC1/3, reflecting their role in the processing of neuropeptide precursors [1]. The neuroendocrine protein 7B2 regulates the activity of PC2 [6], [7] whereas proSAAS inhibits PC1/3 activity [8]. After processing by the PCs, the resulting intermediate peptides still contain basic residues at the carboxyterminus which are cleft off specific carboxypeptidases (mainly carboxypeptidase E (CPE)) [9]. Finally, if a carboxyterminal glycine is present, this amino acid will be transformed into an amide functional group by the action of a bifunctional enzyme peptidylglycine α-amidating monooxygenase (PAM) [10], [11]. For some species (mostly the invertebrates) the two enzymatic activities of the PAM enzyme is contained in two separate enzymes: peptidylglycine α-hydroxylating monooxygenase (PHM) and peptidyl hydroxyglycine α-amidating lyase (PAL) [12].

Annotation of (neuro)peptide precursors can be quite challenging from genomic sequence information even from well-annotated protein databases. This is due largely to the absence of general discriminating features. Usually the conserved bioactive sequence is short and the only other “specific features” are the presence of a signal peptide and the existence of specific PC cleavage motifs. Even if a peptide precursor can be annotated, predicting the peptides that originate from the precursors can be difficult. Different mechanisms exist to generate peptide diversity, which is dependent on the actual need at a specific time and place. First of all, cell-specific expression of the respective peptide precursor genes and their processing enzymes can contribute to neuropeptide diversity. Next, resulting mRNAs can be alternatively spliced and alternative proteolytic processing of the resulting precursor proteins in addition to spatiotemporal regulation of post-translational modifications also contribute to peptide diversity. As a consequence, the endogenous peptide content of a cell, tissue or organism, is spatially and temporally dynamic, which has to be taken into account when monitoring peptide profiles. After processing, bioactive peptides can be stored in dense core vesicles prior to their release within the nervous system or peripheral organ systems where most of them will act through G-protein coupled receptors (GPCR) to govern physiological processes in response to internal and external stimuli. This emphasizes the importance of detailed knowledge of the full complement of the wide diversity of actually present neuropeptides. To this end, liquid chromatography and mass spectrometry (LC-MS)-based approaches have been used to identify these peptidergic signaling entities in a plethora of different tissues from different (model) species (see [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27] for some examples). While important as a model organism, however, no such high-throughput peptidomics analysis has been performed on the freshwater teleost zebrafish (Danio rerio) to date, though defined peptidergic signaling systems have been well studied in the zebrafish [28], [29], [30], [31], [32], [33], [34]. As we still lack a comprehensive overview of all (bioactive) peptides present, we set out to biochemically monitor and identify endogenous peptides from the brain of the zebrafish using a peptidomics workflow.