Deeply Mining a Universe of Peptides Encoded by Long Noncoding RNAs

Abstract

•Complementary enrichment strategies combined with membrane filtration and C8 SPE.•A combined database with the comprehensive putative SEPs and canonical proteins used.•Seven hundred sixty-two novel SEPs identified from human cell lines, mouse cell lines, and mouse tissues.•Nineteen SEPs have been validated by fusion expression or synthetic peptides. Many small ORFs embedded in long noncoding RNA (lncRNA) transcripts have been shown to encode biologically functional polypeptides (small ORF-encoded polypeptides [SEPs]) in different organisms. Despite some novel SEPs have been found, the identification is still hampered by their poor predictability, diminutive size, and low relative abundance. Here, we take advantage of NONCODE, a repository containing the most complete collection and annotation of lncRNA transcripts from different species, to build a novel database that attempts to maximize a collection of SEPs from human and mouse lncRNA transcripts. In order to further improve SEP discovery, we implemented two effective and complementary polypeptide enrichment strategies using 30-kDa molecular weight cutoff filter and C8 solid-phase extraction column. These combined strategies enabled us to discover 353 SEPs from eight human cell lines and 409 SEPs from three mouse cell lines and eight mouse tissues. Importantly, 19 of them were then verified through in vitro expression, immunoblotting, parallel reaction monitoring, and synthetic peptides. Subsequent bioinformatics analysis revealed that some of the physical and chemical properties of these novel SEPs, including amino acid composition and codon usage, are different from those commonly found in canonical proteins. Intriguingly, nearly 65% of the identified SEPs were found to be initiated with non-AUG start codons. The 762 novel SEPs probably represent the largest number of SEPs detected by MS reported to date. These novel SEPs might not only provide new clues for the annotation of noncoding elements in the genome but also serve as a valuable resource for functional study. Many small ORFs embedded in long noncoding RNA (lncRNA) transcripts have been shown to encode biologically functional polypeptides (small ORF-encoded polypeptides [SEPs]) in different organisms. Despite some novel SEPs have been found, the identification is still hampered by their poor predictability, diminutive size, and low relative abundance. Here, we take advantage of NONCODE, a repository containing the most complete collection and annotation of lncRNA transcripts from different species, to build a novel database that attempts to maximize a collection of SEPs from human and mouse lncRNA transcripts. In order to further improve SEP discovery, we implemented two effective and complementary polypeptide enrichment strategies using 30-kDa molecular weight cutoff filter and C8 solid-phase extraction column. These combined strategies enabled us to discover 353 SEPs from eight human cell lines and 409 SEPs from three mouse cell lines and eight mouse tissues. Importantly, 19 of them were then verified through in vitro expression, immunoblotting, parallel reaction monitoring, and synthetic peptides. Subsequent bioinformatics analysis revealed that some of the physical and chemical properties of these novel SEPs, including amino acid composition and codon usage, are different from those commonly found in canonical proteins. Intriguingly, nearly 65% of the identified SEPs were found to be initiated with non-AUG start codons. The 762 novel SEPs probably represent the largest number of SEPs detected by MS reported to date. These novel SEPs might not only provide new clues for the annotation of noncoding elements in the genome but also serve as a valuable resource for functional study. Long noncoding RNAs (lncRNAs), a family of noncoding RNAs that are greater than 200 nucleotides in length and lack long or conserved ORFs, were formerly regarded as “junk RNAs.” Recently, however, a growing amount of evidence has demonstrated that many short or small ORFs (smORFs) embedded in lncRNA transcripts are able to encode functional polypeptides (smORFs-encoded polypeptides [SEPs]). These SEPs contain less than 100 amino acids in eukaryotes (50 amino acids in prokaryotes) and play vital regulatory roles in diverse physiological processes, including cancer growth (1Huang J.Z. Chen M. Chen fnm Gao X.C. Zhu S. Huang H. Hu M. Zhu H. Yan G.R. A peptide encoded by a putative lncRNA HOXB-AS3 suppresses colon cancer growth.Mol. Cell. 2017; 68: 171-171.e6Abstract Full Text Full Text PDF PubMed Scopus (294) Google Scholar), mucosal immunity (2Jackson R. Kroehling L. Khitun A. Bailis W. Jarret A. York A.G. Khan O.M. Brewer J.R. Skadow M.H. Duizer C. Harman C.C.D. Chang L. Bielecki P. Solis A.G. Steach H.R. et al.The translation of non-canonical open reading frames controls mucosal immunity.Nature. 2018; 564: 434-438Crossref PubMed Scopus (74) Google Scholar), and fatty acid β-oxidation (3Makarewich C.A. Baskin K.K. Munir A.Z. Bezprozvannaya S. Sharma G. Khemtong C. Shah A.M. McAnally J.R. Malloy C.R. Szweda L.I. Bassel-Duby R. Olson E.N. MOXI is a mitochondrial micropeptide that enhances fatty acid β-oxidation.Cell Rep. 2018; 23: 3701-3709Abstract Full Text Full Text PDF PubMed Scopus (64) Google Scholar). These findings have subverted our understanding of lncRNAs and expanded our knowledge of the coding potential of the genome. Moreover, the development of genomics and bioinformatics, in particular the advent of high-throughput sequencing technology, accelerated the discovery of thousands of additional lncRNA transcripts with smORFs. Considering such large numbers of lncRNAs and smORFs, it is expected that SEPs may represent a large albeit neglected portion of nonannotated peptides involved in diverse physiological process. Therefore, large-scale discovery and functional characterization of unknown SEPs might provide new clues for the annotation and functional analysis of noncoding elements in the genome and their effects on biological evolution. A variety of different methodologies, such as smORF predictions by computational sequence analysis, deep sequencing–based ribosome profiling, and MS-based proteomics, have been developed for the identification and characterization of SEPs across different biological samples. However, each of these strategies presents caveats. First of all, while bioinformatics analysis of lncRNA transcript sequences is a typical first step to predict the existence of smORFs, achieving high prediction sensitivity and specificity remains a significant challenge (4Pauli A. Valen E. Schier A.F. Identifying (non-)coding RNAs and small peptides: challenges and opportunities.Bioessays. 2015; 37: 103-112Crossref PubMed Scopus (68) Google Scholar, 5Cohen S.M. Everything old is new again: (linc)RNAs make proteins!.EMBO J. 2014; 33: 937-938Crossref PubMed Scopus (24) Google Scholar). Furthermore, despite the power of deep sequencing–based ribosome profiling for the identification of the region of active translation in lncRNA transcripts, it nevertheless only provides indirect evidence of translation (6Aspden J.L. Eyre-Walker Y.C. Phillips R.J. Amin U. Mumtaz M.A. Brocard M. translation of small open reading frames revealed by 2014; PubMed Scopus Google Scholar, M. P. profiling provides evidence that large noncoding RNAs not encode Full Text Full Text PDF PubMed Scopus Google Scholar, A. J.L. A. Schier A.F. Valen E. profiling long RNAs and of coding PubMed Scopus Google Scholar). while MS-based SEPs by the peptides from embedded in lncRNA transcripts A.G. J. J.Z. J.L. A. discovery of short open reading peptides in human PubMed Scopus Google Scholar, J. A.G. J. M. A. of human polypeptides in cell lines and 2014; PubMed Scopus Google Scholar, H. W. E. G. G. peptide analysis of mouse novel 2018; PubMed Scopus Google Scholar), the number of SEPs identified by MS from different biological is still small peptides the of the Full Text Full Text PDF PubMed Scopus Google Scholar). The low number of SEPs detected by MS is to the that of is still First of all, of the low and small of SEPs, and comprehensive be and by are to SEPs from different biological by or and large proteins to the physical and chemical properties of different SEPs are by different may their the identification of SEPs using MS is by them the of peptides in a sequence that the the of a database the identification of novel the most is translation of the genome. such a is to of large and the of unknown it is a database by RNA transcripts from or only RNA transcripts and on sequencing In the we the challenges by effective SEP enrichment through the of two complementary enrichment on 30-kDa filter and C8 solid-phase extraction column. us to build a SEP database containing putative from lncRNA transcripts in the a that the discovery of SEPs from different cell lines and tissues. to the existence of these human and were in with and were in with and were in with and and mouse cell were from the cell the of and of were in with C. were in with and amino and mouse and were from the the of of were by the and of the of of cell were with 100 extraction containing by for with a and then the for were small and with extraction containing The for with a and the by for SEP enrichment from cell with 30-kDa by in containing three of with a the for and then on for a to the to a of and on for for in a the through a 30-kDa filter and the through and to by The then in SEP enrichment from with 30-kDa by small and then in containing The were the as for SEP enrichment from In order to SEP extraction and enrichment from cell using C8 we containing and C8 the C8 W. J. G. H. M. M. J. of growth PubMed Scopus Google Scholar, J. C. J. M. J.R. A. identification and analysis of small open reading encoded PubMed Scopus (68) Google Scholar). were in (50 and containing for for the C8 were with of and of the SEPs were in with of and in The were then combined and to less than 100 by SEPs were with to and the in SEP enrichment from using C8 of small and in (50 and containing The were the as for SEP extraction from cell samples. of proteins with (50 and for the were H. PubMed Scopus Google and for The with and then with were with with in the for were with a of in less than acid to the with to the The peptide then using with The peptides were with of of and of The peptide were combined and to by peptides were by MS by with a A 100 with and a with were to the peptides with A in and in a to in to in to in to in and then for The as The in a with a of and a of were a high of with a of to a of and a of each MS the most were for with of and the with of were a of The with a of and a of lncRNA transcripts for human and mouse from the database The and translation were to we smORFs, were then putative SEP for human and mouse by putative SEPs with a length of to 100 amino The were with using the different were in study. The of these are as canonical from the on and of canonical from the on and of putative human SEP including and putative mouse SEP including identification of novel peptides from the were the database of canonical database and putative SEP The and peptides. were as peptides with a of two were the of and as and on as on as we a of and a discovery using a from identification of canonical were the database of were as parallel reaction were on the as In SEPs were from SEPs identified in for The and identified peptides in the SEPs were for with a in order to the identified using of the and identified peptides in the SEPs to the identified peptides and discover novel peptides in the The are to proteins or in the putative SEP were from the A of peptide to were in the further the identification of novel SEPs, peptides from SEPs were by and on the as In order to fusion for the SEP and we SEP sequences the using and them a in the start codon to A of the in is in were with the using to RNA from using the RNA RNA of and SEP were detected by A containing the in is in to The in were as and were and by were with SEP and for and and were on and then with with with with were with the two different SEP enrichment we and three using the cell or samples. were by a and as the of a of and for and the in the were as the from three the low and small of SEPs to their poor it is to and build putative SEP from MS-based analysis in order to improve SEP discovery from different biological samples. MS-based database is the step for MS-based SEP In order to build a SEP database that the putative SEPs in human and we lncRNA transcripts in the database repository that the most complete collection and annotation of noncoding lncRNA transcripts were by and translation to make it to smORFs, were then to represent putative SEPs in and polypeptides in the human and mouse putative SEP the of these two we reported functional SEPs, including Chang C.A. McAnally J.R. P. J. Bassel-Duby R. Olson E.N. A micropeptide encoded by a putative long noncoding RNA 2015; Full Text Full Text PDF PubMed Scopus Google Scholar), P. A. H. J. McAnally J.R. E. Bassel-Duby R. Olson E.N. of by the micropeptide 2017; PubMed Scopus Google Scholar), J. R. A. L. J. E. C. The controls cell fusion and 2017; PubMed Scopus Google Scholar), A. A. M. R. J. E. A. and are by the 2017; PubMed Scopus Google Scholar), HOXB-AS3 (1Huang J.Z. Chen M. Chen fnm Gao X.C. Zhu S. Huang H. Hu M. Zhu H. Yan G.R. A peptide encoded by a putative lncRNA HOXB-AS3 suppresses colon cancer growth.Mol. Cell. 2017; 68: 171-171.e6Abstract Full Text Full Text PDF PubMed Scopus (294) Google Scholar), J. L. J. A. A human that with the 2017; PubMed Scopus Google Scholar), and M. Yan S. P. H. J. H. Huang J. et peptide encoded by of suppresses in 2018; PubMed Scopus Google and them our these SEPs be found our putative SEP the high and of our The and enrichment of SEPs from biological is step for their have been for including 30-kDa C8 and or the 30-kDa filter and C8 are commonly albeit on different In the of 30-kDa SEPs are and on their molecular the and are to and SEPs using C8 Therefore, we these may represent two strategies for SEP and that their combined improve SEP our by SEPs from of cell using 30-kDa filter and C8 and then and analysis on our database and that 30-kDa filter and C8 are effective in for in the molecular weight and further these by that and of the identified proteins in the 30-kDa filter and C8 were low molecular weight in to only in enrichment Importantly, of and SEPs were identified from the 30-kDa filter and C8 are than the identified using enrichment and as the of the two are only a SEP identified with 30-kDa filter and C8 despite the enrichment low the two were in mouse and cell A and is the of in enrichment to SEP the two analysis that SEPs for and of the SEPs identified in with 30-kDa filter and C8 In the of J. A for the of a PubMed Scopus Google of SEPs identified in C8 with only for those detected using the 30-kDa filter Moreover, the of for SEPs identified by each were and for C8 and 30-kDa the for these are the low extraction of proteins in the 30-kDa filter of the lack of in the and the that proteins to be in the of C8 such as J. J. Huang G. H. analysis of mouse by extraction with a of 2015; PubMed Scopus Google Scholar). the MS-based analysis, we for the identification of SEPs, most of putative SEPs in our database were than amino in with that only of SEPs identified in human cell lines or are less than amino acids long J. A.G. J. M. A. of human polypeptides in cell lines and 2014; PubMed Scopus Google Scholar). Moreover, our also that a number of SEPs be detected in with with those with In order to for a diverse and comprehensive of novel SEPs, the MS-based to eight human cell lines, three mouse cell lines, and eight mouse tissues. In and to SEP SEPs were through sequence and A.G. J. J.Z. J.L. A. discovery of short open reading peptides in human PubMed Scopus Google Scholar, C. C. P. and novel in 2018; PubMed Scopus Google Scholar). The peptides of the were from the peptide sequences to proteins in the database and as peptides with less than eight amino acids in containing less than or and many with high and containing less than and the peptides were as novel peptides. in the identification of 353 SEPs from eight human cell lines and 409 SEPs from three mouse cell lines and eight mouse to the of our the largest number of SEPs identified by MS reported to number of SEPs in different cell lines and from human and of identified identified cell cell in a new we identified novel peptides from 353 human SEPs, with detected from eight human cell we novel peptides from 409 mouse SEPs, with detected or across cell lines and tissues. In we have also been able to the reported functional peptide the 762 identified SEPs, including SEP peptide and be found in and of identified peptides are in The including are with the Importantly, we have identified a of SEPs that are found in most human cell lines in our many SEPs are cell A of of the 353 identified human SEPs were in three different human cell lines, with found in mouse cell lines These that SEPs are in different cell lines and in different Moreover, it is the that biological and might the number of SEPs we analysis on of SEPs from mouse in order to many are to a relative for SEP identification from a biological that of SEPs detected with a and SEPs in each the number of novel SEPs detected from the mouse to were found for the cell in SEPs were identified in across the different biological and These findings are with J. A.G. J. M. A. of human polypeptides in cell lines and 2014; PubMed Scopus Google and the that is the of our bioinformatics and were implemented to our prediction of and expression analysis for lncRNA transcripts. expression of identification of additional peptides on a and of peptide is to the functional roles of we expected that to in the than the to by and the lncRNA transcripts to the identified SEPs using a for lncRNAs on a Huang The A for long RNAs on a 2018; PubMed Scopus Google Scholar). we found that than of lncRNAs are to in the less than are to in the were mouse SEPs are different from those found in a R. G. A. S. H. G. A. J. L. et al.The of human long noncoding of their and PubMed Scopus Google Scholar), detected that of lncRNAs are in the in the of are in the in the it that these lncRNAs are to to for active the of RNA transcripts the of expression of their proteins M. J. H. A.M. M. E. L. S. H. S. U. H. et of the human 2014; PubMed Scopus Google Scholar). MS-based SEP as of lncRNA expression may make it for the SEPs to be we the of expression of the lncRNAs identified and the cell in from we that the of expression of the identified lncRNAs were than those of the cell in In we found mouse SEPs these are not in with that the of expression of lncRNAs are to those of the of expression found these SEPs be detected in our study. that most SEPs were only detected with a of the low of SEPs in we implemented a on the or identified peptides and the MS in order to additional evidence to our we analysis on of the identified SEPs in and identified eight additional peptides for eight SEPs SEP sequence to further the of SEPs identified in we peptides from SEPs identified in cell lines than to as peptide then the of synthetic peptide by and the of synthetic and identified SEP peptides. that identified peptides from SEPs were with the synthetic peptides The including are in with the In we human SEPs as and for that were found in most of the human cell lines and the The RNA transcripts to these fusion coding sequences were identified by from and also expression of or SEP fusion proteins expression of found in with the in the start codon of to to translation The were further by analysis using and to provide evidence of SEP expression by SEPs from the and using their peptides that and in and their expected molecular and high specificity of these two Importantly, also the relative molecular weight in human cell it in length and in verified the existence of polypeptides in by our analysis validated the existence of SEPs through different strategies including expression of and In order to a and the properties of the identified SEPs, we them from including codon usage, length amino acid and were on their and long noncoding RNAs and C. J. H. M. G. Zhu W. W. Chen R. the of long RNA 2014; PubMed Scopus Google Scholar). it has been found that are in a to as are conserved than and transcripts. is with our that SEPs for of the identified human SEPs, and lncRNAs for and SEPs for of the identified mouse SEPs, with and for and is in a genome and provides for and The codon of the identified SEPs by the start and as reported J. C. J. M. J.R. A. identification and analysis of small open reading encoded PubMed Scopus (68) Google Scholar). or codon in a sequence to be a start In SEPs were to have unknown start the identified human SEPs, and were initiated with or a the unknown start codon and of the 409 identified mouse SEPs with or a with unknown start codon canonical proteins were initiated with start in with J. A.G. J. M. A. of human polypeptides in cell lines and 2014; PubMed Scopus Google Scholar). The that a of SEPs is initiated with unknown start codon their novel less the length of SEPs with a start codon by the In for SEPs with unknown start length as the two codons. a of the identified human and mouse SEPs were to have a length and with the only amino acids Importantly, that nearly of SEPs were to be than amino acids in to be a for SEPs using the MS-based it is not to from amino acid it is nevertheless to and by the properties of the amino acids that it P. and of small open reading 2017; PubMed Scopus Google Scholar). amino acid composition analysis revealed that identified human and mouse SEPs to amino acids and less amino acids and than canonical while using a amount of amino acids and in with J.L. Eyre-Walker Y.C. Phillips R.J. Amin U. Mumtaz M.A. Brocard M. translation of small open reading frames revealed by 2014; PubMed Scopus Google Scholar, S. J. L. Chen J. G. A human encoded by PubMed Scopus Google Scholar). Furthermore, it has been that proteins containing amino acids and a region were commonly found across the cell membrane and (6Aspden J.L. Eyre-Walker Y.C. Phillips R.J. Amin U. Mumtaz M.A. Brocard M. translation of small open reading frames revealed by 2014; PubMed Scopus Google Scholar, R. S.M. P. M.A. of peptide J. PubMed Scopus Google Scholar, to PubMed Scopus Google Scholar, A. and characterization of 2015; PubMed Scopus Google Scholar). the for SEPs to as peptides. is also that SEPs containing a of amino acids may play a in to or RNA H. A. P. A. to in PubMed Scopus Google Scholar, C. G. transcripts of encode regulatory 2015; PubMed Scopus Google Scholar, J. A. A human short open reading polypeptide that 2014; Full Text Full Text PDF PubMed Scopus Google Scholar). the of and amino acids in SEPs that might not be the for SEP discovery, it small peptides that are not for MS a of improve SEP discovery and sequence we using of a and A novel for in of a from PubMed Scopus Google Scholar, E. A. identification and analysis in the Google and found only the of the identified SEPs and canonical proteins and is in with P. G. P. L. J. in not of lncRNA translation 2017; PubMed Scopus Google Scholar). that SEPs might be as as canonical proteins. the have to the discovery of biologically SEPs from different it is expected that many are to be SEP is of their low and small In the we implemented a comprehensive for SEP discovery and characterization from human and mouse cell lines and mouse through MS-based by two effective and complementary polypeptide enrichment with the of a SEP enabled the discovery of 762 novel SEPs from different human and cell lines and to our represent the largest number of SEPs to be The SEP discovery reported be to our SEP database a number of putative from lncRNA transcripts in the database by the translation with the in the database may in discovery it us to a number of SEPs across cell lines and and to identification with Furthermore, the of two effective and complementary polypeptide enrichment 30-kDa filter and C8 SEP of different and SEP Moreover, the of and biological and further MS-based SEP Importantly, the number of SEPs us to their physical and chemical some of might have to the identified SEPs were from lncRNA transcripts with a of expression than the translation of SEPs might be than and the sensitivity of SEP In than of the identified SEPs from human or mouse are initiated by unknown start have with of translation by non-AUG in Cell. PubMed Scopus Google Scholar, A new start for in 2017; PubMed Scopus Google Scholar). Moreover, non-AUG may make it less for SEPs to be and in a database for MS-based Furthermore, SEPs are commonly in amino acid length and with are typical of coding MS evidence L. S. M. on and the human from the human 2018; PubMed Scopus Google Scholar). these that the SEPs identified in might represent the of the and that many low SEPs Importantly, nearly of identified human SEPs were in than two different human cell evidence from in vitro immunoblotting, and bioinformatics also the existence of 19 novel SEPs in Moreover, these SEPs are to be as as canonical proteins and in their in human cell lines, as demonstrated by These that the identified SEPs not from but are of biological despite SEPs still a complete functional In we that MS-based for comprehensive discovery of of novel SEPs from different human and mouse cell lines and not only provide new clues for the annotation of noncoding elements in the genome but might also serve as a valuable resource for the functional characterization of The MS and are in with the The SEP have been to the for of of with number The from the of of for on MS and from the of of of for their with the analysis and computational were by the for in of of by the of and and the and of the of and of and R. and J. L. R. and J. L. and L. J. and G. E. W. and with and with with with with

Topics & Concepts