Multi-species sequence comparison reveals conservation of ghrelin gene-derived splice variants encoding a truncated ghrelin peptide
- First Online:
- Cite this article as:
- Seim, I., Jeffery, P.L., Thomas, P.B. et al. Endocrine (2016) 52: 609. doi:10.1007/s12020-015-0848-7
- 738 Downloads
The peptide hormone ghrelin is a potent orexigen produced predominantly in the stomach. It has a number of other biological actions, including roles in appetite stimulation, energy balance, the stimulation of growth hormone release and the regulation of cell proliferation. Recently, several ghrelin gene splice variants have been described. Here, we attempted to identify conserved alternative splicing of the ghrelin gene by cross-species sequence comparisons. We identified a novel human exon 2-deleted variant and provide preliminary evidence that this splice variant and in1-ghrelin encode a C-terminally truncated form of the ghrelin peptide, termed minighrelin. These variants are expressed in humans and mice, demonstrating conservation of alternative splicing spanning 90 million years. Minighrelin appears to have similar actions to full-length ghrelin, as treatment with exogenous minighrelin peptide stimulates appetite and feeding in mice. Forced expression of the exon 2-deleted preproghrelin variant mirrors the effect of the canonical preproghrelin, stimulating cell proliferation and migration in the PC3 prostate cancer cell line. This is the first study to characterise an exon 2-deleted preproghrelin variant and to demonstrate sequence conservation of ghrelin gene-derived splice variants that encode a truncated ghrelin peptide. This adds further impetus for studies into the alternative splicing of the ghrelin gene and the function of novel ghrelin peptides in vertebrates.
KeywordsGhrelinPeptide hormoneEvolutionComparative endocrinologyAlternative splicing
Ghrelin is a 28-amino acid peptide hormone produced in the stomach and has potent appetite-stimulating effects [1, 2]. Ghrelin was initially described as a growth hormone-releasing peptide and is now recognised to have a diverse range of functions in a number of cell types and systems, including roles in energy balance, appetite regulation and food-seeking behaviour, insulin regulation and cell proliferation [1–4]. The ghrelin gene (GHRL) and the coding exons of the derived preprohormone, preproghrelin, are conserved in a wide range of species including fish , turtles , birds  and mammals . In humans, exons 1 to 4 encode the 117-amino acid preproghrelin, with exons 1 and 2 coding for the 28-amino acid peptide hormone ghrelin . After cleavage of a 23-amino acid signal peptide, proghrelin is processed to form ghrelin, and the third ghrelin residue (serine) is post-translationally octanoylated (acylated) by the enzyme ghrelin O-acyl transferase (GOAT, encoded by MBOAT4) [9, 10]. The C-terminus of proghrelin (encoded by exons 3 and 4) is further processed to give rise to the 23-amino acid peptide hormone obestatin , which has independent actions from ghrelin .
We and others have previously demonstrated that the ghrelin gene locus is complex and gives rise to numerous transcripts that could encode a wide range of peptides [13, 14]. A number of ghrelin variants have been characterised in humans  and transcripts containing intronic sequence, in1-ghrelin and in2c-ghrelin, have recently been described [15, 16]. Intron 2 cryptic (in2c) ghrelin is an insulin-regulated transcript that contains intron 2-derived exons and the coding region for ghrelin, but lacks the obestatin sequence . The in1-ghrelin transcript contains exon 1, intron 1 and exon 2, and the inclusion of intron 1 leads to a truncation of the ghrelin peptide sequence . We recently reported an exon 2-deleted splice variant in mouse and sheep . In this study, we sought to determine whether any ghrelin gene-derived splice variants are conserved across vertebrates and present preliminary findings on the function of splice variants encoding derived short, C-terminally truncated ghrelin peptides (here termed minighrelin).
Materials and methods
Ghrelin gene (GHRL) sequences (from genomes and transcriptomes) were derived from the UCSC multiway tool, NCBI GenBank , and from the NCBI Short Read Archive (SRA) . Putative GHRL sequences were interrogated using BLAST  in a local instance of the Ruby-based SequenceServer (http://www.sequenceserver.com), gmap v2013-06-27 (a genomic mapping and alignment program for mRNA and EST sequences) with the parameters --cross-species --align --direction=sense_force -Y , and custom Perl scripts with BioPerl modules . MUSCLE  was used for protein sequence alignments of ghrelin gene orthologs, using the human sequence as the reference.
Cell lines were originally obtained from the American Type Culture Collection (ATCC, Rockville, MD, USA). The PC3 (ATCC CRL-1435), DU145 (ATCC HTB-81), LNCaP (ATCC CRL-1740) and 22Rv1 (ATCC CRL-2505) prostate cancer cell lines were maintained in Roswell Park Memorial Institute (RPMI) 1640 medium (Invitrogen, Mulgrave, VIC, Australia) with 10 % New Zealand Cosmic Calf Serum (FCS, Thermo Fisher Scientific Australia, Scoresby, VIC, Australia) supplemented with 100 U/mL penicillin G and 100 µg/mL streptomycin (Invitrogen). The non-tumourigenic RWPE-1 (ATCC CRL-11609) and the transformed, tumourigenic RWPE-2 (ATCC CRL-11610) prostate epithelium-derived cell lines were cultured in keratinocyte serum-free medium (KSFM) (Invitrogen) supplemented with 50 µg/mL bovine pituitary extract and 5 ng/mL epidermal growth factor (Invitrogen). All cell lines were passaged at 2- to 3-day intervals at 70 % confluency using TrypLE Select (Invitrogen). Cell morphology and viability were monitored by microscopic observation and regular PCR testing was performed (Universal Mycoplasma Detection Kit, ATCC) to ensure that cells were not contaminated with Mycoplasma.
Identification of human exon 2-deleted preproghrelin
Human exon-2 deleted preproghrelin was cloned as follows: total RNA was harvested from cultured cells using an RNeasy Plus Mini Kit (QIAGEN, Germantown, MD, USA) according to the manufacturer’s instructions. Next, 2 μg total RNA was subjected to DNase I digestion (amplification grade, Invitrogen), immediately followed by cDNA synthesis with SuperScript III using oligo(dT)18 primers (Invitrogen) according to the manufacturer’s instructions. RT-PCR primers spanning exon 1 and 3 of preproghrelin (5′-CATGCTCTGGCTGGACTTGG-3′ and 5′-GACAGCTTGATTCCAACATCAAAGG-3′) were designed using PerlPrimer . RT-PCR using tissue (normal human tissue cDNA panel HMRT102 purchased from OriGene, Rockville, MD, USA) and cultured cells was performed in a total reaction volume of 30 μL containing 1× PCR buffer, 0.2 mM deoxynucleotide triphosphates, 1.5 mM MgCl2, 0.2 μM primers, 1 μL cDNA and 1 unit of Platinum Taq DNA Polymerase (Invitrogen) on a PTC-200 thermal cycler (MJ Research, Watertown, MA, USA) according to the manufacturer’s instructions, with an annealing temperature of 61 °C. Negative (no-template) controls were performed. RT-PCR products were purified using a MinElute (QIAGEN) PCR Purification Kit, cloned into pCR4-TOPO (Invitrogen), transformed into One Shot MAX Efficiency DH5α-T1R chemically competent cells (Invitrogen) and sequenced at the Australian Genome Research Facility (AGRF, Brisbane, Australia) using BigDye III (Applied Biosystems, Foster City, CA, USA).
Food intake as a measure of in vivo function of ghrelin peptides
Acylated (octanoylated) and desacyl 28-AA ghrelin peptides (H-GSSFLSPEHQRVQQRKESKKPPAKLQPR-OH) and 13-AA minighrelin peptides (H-GSSFLSPEHQRVQ-OH) were commercially synthesised (Mimotopes, Melbourne, VIC, Australia). Male 16-week-old C57BL/6J mice, purchased from the Animal Resources Centre (Perth, Western Australia), were housed separately and handled daily for 1 week with unrestricted access to standard chow and drinking water to acclimatise them to experimental conditions. Mice were then injected intraperitoneally with 2 nmol/mouse acyl ghrelin, desacyl ghrelin, acyl minighrelin, desacyl minighrelin, or saline (vehicle) (n = 6 per group except for the minighrelin treatment, where n = 5). This dose of acyl ghrelin (2 nmol/mouse) has previously been demonstrated to stimulate appetite in mice [25, 26]. Mice were injected at 09:00 (light phase) and given pre-weighed chow pellets. At 4 h post injection, the remaining chow pellets were weighed and the cumulative food intake per mouse was determined to the nearest 0.1 g (pre-injection pellet weight minus post-injection pellet weight). Experiments were carried out with approval of the Animal Ethics Committee, University of Queensland.
Forced overexpression of human exon 2-deleted preproghrelin
Coding regions of human exon 2-deleted preproghrelin and canonical preproghrelin were commercially synthesised and cloned into OriGene pCMV6-AC plasmid vectors (Blue Heron Biotechnology, Bothell, WA, USA). Constructs, and empty vector controls, were transformed into E. coli DH5α cells (Invitrogen) and purified using a QIAGEN plasmid purification kit, according to the manufacturer’s instructions. To produce a control cell line expressing the vector only, the PC3 prostate cancer cell line was transfected with pCMV6-AC plasmid DNA using Lipofectamine 2000 reagent (Invitrogen), according to the manufacturer’s instructions. Stably overexpressing PC3 prostate cancer cells were selected with 600 µg/mL G418 antibiotic (Invitrogen). Overexpression of ghrelin variants was confirmed by semi-quantitative RT-PCR (as described above).
Cell proliferation assays
Cell proliferation assays were performed using the xCELLigence Real-Time Cell Analyzer (RTCA) system (ACEA Biosciences, San Diego, CA, USA), using E plates, according to the manufacturer’s instructions. Briefly, cell lines overexpressing exon 2-deleted preproghrelin, canonical preproghrelin, or empty vector were cultured until they were 70 % confluent, detached from the cell culture flask using trypsin/EDTA (Invitrogen) and collected by centrifugation. Cells were then added to E plates at a density of 5000 cells/well in growth medium with 10 % New Zealand Cosmic Calf serum (FCS) (Thermo Fisher Scientific). Proliferation was measured for up to 72 h and compared to the rate of proliferation of cells expressing an empty vector. Each treatment (exon 2-deleted preproghrelin, canonical preproghrelin and empty vector overexpressing cells) was performed with three replicates, and the experiment was performed independently three times.
Cell migration assay
Cell migration assays were performed using the xCELLigence RTCA system and cell/invasion migration (CIM) plates with 8 µm pores (ACEA Biosciences). The lower well contained media with 10 % foetal calf serum as a chemoattractant, and the cells were added to the upper well at a density of 50,000 cells/insert in serum-free media. Migration was observed in real time for up to 48 h. Each ‘treatment’ (exon 2-deleted preproghrelin, canonical preproghrelin and empty vector overexpressing cells) was performed with three replicates, and the experiment was performed independently three times.
Statistical analyses were undertaken using GraphPad Prism v5.04 (La Jolla, CA, USA) and R statistical software v3.0.2 (http://www.r-project.org). If a data set did not differ significantly from a normal distribution (P ≤ 0.05, Shapiro–Wilk test), the data were considered to have a ‘normal’ distribution and a parametric one-way ANOVA with Tukey’s post hoc test was used to compare groups. If a data set differed significantly from a normal distribution (P ≥ 0.05, Shapiro–Wilk test), the Mann–Whitney U test (a general non-parametric test for two groups) was utilised to test for a difference between a treatment group and the control.
Results and discussion
Evidence of conserved exon skipping of the ghrelin gene
In addition to evidence of conserved exon 2 ‘deletion’ (exon skipping), we noted that in1-ghrelin, a splice variant previously cloned from the mouse , human and baboon , also harbours a minighrelin peptide sequence. In1-ghrelin contains exon 1, the short intron 1 and exon 2 . In contrast to all other described variants (e.g. canonical preproghrelin, in2c-ghrelin, exon 2-deleted preproghrelin and exon 3-deleted preproghrelin), in1-ghrelin lacks exons 3 and 4. In accordance with genome-wide transcriptome analyses , we speculate that inclusion of intronic (intron 1 in its entirety) sequence promotes an alternative polyadenylation site within exon 2 or the large intron 2 (~3 kb in human). Due to divergence in the C-terminal region, in1-ghrelin differs greatly in size between species, ranging from 45 AAs in black seabream, 46 AAs in sheep, 55 AAs in mouse, 117 AAs in human, to 124 AAs in the naked mole rat (Fig. 1b). The common feature of in1-ghrelin in all vertebrate species examined is the presence of C-terminally truncated ghrelin sequence, minighrelin.
Minighrelin mimics canonical ghrelin function
The splice variants described here harbour C-terminally truncated ghrelin peptide sequences that retain the first 13 amino acids of the peptide hormone, ghrelin, followed by obestatin sequence. We investigated whether the truncated ghrelin peptide derived from the exon 2-deleted splice variant could mimic the functional effects of canonical ghrelin. This region of the exon 2-deleted prepropeptide retains the active core (the first five amino acids) of ghrelin, which is required for the activation of its cognate receptor, GHSR1a [32, 33], and for octanoylation of the peptide by GOAT .
We next investigated the function of exon 2-deleted preproghrelin in our model assay system, cultured prostate cancer cell lines [16, 35–37]. We employed the PC3 prostate cancer cell line, which expresses exon 2-deleted preproghrelin (Fig. 2a), the processing enzymes necessary to produce mature, acylated ghrelin, the octanoylating enzyme GOAT  and the cognate ghrelin receptor, GHSR1a . A similar approach has been used to demonstrate that the in1-ghrelin variant increases proliferation in the MDA-MB-231 breast cancer cell line , mirroring the effect of synthetic 28-AA ghrelin peptide treatment . Overexpression of exon 2-deleted preproghrelin (which harbours a 13-AA ghrelin sequence) and full-length preproghrelin (28-AA ghrelin sequence) significantly increased cell proliferation and migration in the PC3 prostate cancer cell line (Fig. 3b, c) (P ≤ 0.05, ANOVA with Tukey’s post hoc test). These findings are consistent with our previous studies, which have demonstrated that treatment of prostate cancer cell lines with exogenous 28-AA ghrelin peptide stimulates cell proliferation [35, 37]. Ghrelin stimulates cell migration in cultured cell lines derived from cancers of the brain , colon  and pancreas . For the first time, we show that canonical preproghrelin and exon 2-deleted preproghrelin also induce cell migration in a prostate cancer cell line. These findings are in agreement with a very recent study showing that a 19-AA putative human in1-ghrelin-derived ghrelin peptide, which harbours the first 13 AAs of ghrelin, acts via the cognate ghrelin receptor, GHSR1a, and is able to mimic the effects of ghrelin . Taken together, these data suggest that naturally occurring ghrelin gene-derived variants coding for a C-terminally truncated ghrelin peptide sequence, minighrelin, are able to mimic the effect of ghrelin.
Implications of this study
Aided by comparative genomics analysis, we present preliminary evidence of the first conserved ghrelin gene-derived splice variants since the discovery of ghrelin in 1999 . Alternative splicing is a complex event where protein diversity is increased from a relatively small number of genes . Molecular innovation of the ghrelin gene, through alternative splicing, has largely been reported in the literature as species specific. This includes the recently described in2c-ghrelin in humans , the mouse-specific ghrelin gene-derived transcript (GGDT) with a novel first exon in intron 2 followed by exon 3 and 4 , and a variant with a 5′ truncated exon 3 lacking the N-terminus of obestatin in the red-eared slider turtle [6, 45]. An exon 3-deleted preproghrelin variant with a novel C-terminal region is present in mouse  and human , but absent in other mammals such as ruminants, where skipping of exon 3 alone would result in a prematurely truncated preproghrelin peptide (unpublished data). It has been proposed that highly conserved alternative splicing corresponds to molecular changes that are tolerated during evolution [47, 48]. Here, we show that skipping exon 2 of the ghrelin gene results in an open reading frame that is highly conserved in vertebrates. While exon 2-deleted preproghrelin produces a C-terminally truncated ghrelin peptide, minighrelin, it is otherwise identical to canonical preproghrelin. We also show that although vertebrate in1-ghrelin sequences have a highly divergent C-terminus, they all harbour an in-frame minighrelin peptide sequence.
The proteolytic cleavage of ghrelin gene-derived peptides remains largely unresolved; however, there are some commonly recognised preprohormone cleavage sites . The obestatin peptide is spanned by an Arg (R) or Gln (Q) in primates , and 10 years after its discovery the proteases responsible for its cleavage have not been firmly established. Similarly, minighrelin is spanned by Gln (Q) in all but two species (the common shrew and anole lizard, a discrepancy which may result from genome sequencing errors) and by mono- or dibasic (Arginine and Lysine) amino acid residues in all species except turtles (Online Resource 2). The small size of the putative minighrelin peptide (<1.5 kDa), and difficulty in distinguishing proteolytic cleavage products of canonical ~3.3 kDa ghrelin from minighrelin, has so far yielded mass spectrometry-based detection of minighrelin peptide unsuccessful. Moreover, it is possible that the human exon 2-deleted preproghrelin variant is not cleaved after 13-AA ghrelin peptide sequence and this would give rise to a chimaeric minighrelin–obestatin peptide. Ghrelin and obestatin are multifunctional peptide hormones [12, 50] and chimaeric peptide hormones can have improved, or novel, pharmacological properties compared to the individual peptides alone [51, 52].
It is reasonable to assume that minighrelin is acylated and functional regardless of its processed length. Human exon 2-deleted preproghrelin and in1-ghrelin lack the C-terminal 15 amino acids of the 28-amino acid canonical ghrelin peptide, but are able to mimic the function of ghrelin in vitro. Recent studies suggest that the C-terminal region of ghrelin may be particularly important for stabilising ghrelin in plasma [53, 54]. Therefore, in vivo, the minighrelin peptide might have a shorter half-life than canonical ghrelin and alternative splicing could provide a mechanism for regulating the bioavailability of circulating ghrelin. Nevertheless, even if short ghrelin peptides are less stable in the circulation, ghrelin produced in the stomach binds to its cognate receptor, GHSR1a, on vagal afferent neurons to signal centrally [55, 56] and ghrelin has local (autocrine and paracrine) effects in a number of cell types . Moreover, minighrelin retains the active core of ghrelin [32–34] and it is now firmly established that short synthetic ghrelin peptides have agonist activity [32, 53, 54, 57, 58], albeit with reduced potency compared to full-length ghrelin. Interestingly, short functional ghrelin peptides have been identified in the goldfish (Carassius auratus) [5, 59]. We speculate that these short ghrelin peptides in goldfish result from orthologous splice variants of exon 2-deleted preproghrelin or in1-ghrelin. Further proteomic and biochemical analyses are required to characterise variants with minighrelin peptide sequences.
More broadly, it is critical to appreciate the transcriptional complexity of the ghrelin gene locus in normal physiology and disease. Emerging technologies, such as RNA CaptureSeq which allows the detection of low-abundance gene expression in selected loci , are likely to greatly advance our understanding of the complex ghrelin gene locus. Since the discovery of ghrelin , approximately 8000 articles on ghrelin have been published, with many articles referring to the measurement of serum ghrelin levels and their association with physiology or disease. Discrepancies between studies  may be explained partly by the fact that most assays cannot currently discriminate between distinct ghrelin gene-derived peptides. In particular, several studies have relied on sandwich (two-site) ELISAs, with capture antibodies raised against the C-terminal region of the 28-amino acid ghrelin peptide [61, 62]. As these assays cannot detect minighrelin peptides, this would lead to an underestimation of total ghrelin levels. It may be necessary to conduct additional studies under many different physiological and pathophysiological conditions to ensure that the contribution of ghrelin peptides is reliably assessed.
In summary, we provide preliminary evidence that ghrelin gene-derived splice variants encoding C-terminally truncated peptides that retain the active core (the first five amino acids) of ghrelin, termed minighrelins, are conserved in vertebrates. Minighrelin peptide is able to stimulate appetite in mice, and preproghrelin splice variants that encode minighrelin are able to mimic the actions of canonical ghrelin in vitro. These findings add complexity to the study of ghrelin and further impetus for the study of alternative splicing of the ghrelin gene and the function of novel transcripts in diverse species.
We acknowledge the financial support from the National Health and Medical Research Council Australia (to IS, PLJ, ACH and LKC), a QUT Vice-Chancellor’s Senior Research Fellowship (to IS), the Movember Foundation and the Prostate Cancer Foundation of Australia through a Movember Revolutionary Team Award (to LKC and ACH), Professor Colleen Nelson and the Australian Prostate Cancer Research Center, Queensland, and QUT Bluebox.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.