Oriol Morales (oriol.morales01@campus.upf.edu) i Lucía Llorens (llucia_llorens01@campus.upf.edu)

Facultat de Ciències de la Salut i de la Vida
Universitat Pompeu Fabra

In this project is explained the exonic structure, homology, function and expression of TLS-CHOP. This human hybrid gene is found in myxoid liposarcomas and is due to a characteristic chromosomal translocation, t(12;16)(q13;p11). In order to study TLS-CHOP here are shown the characteristics for each original gene separately. It is formed by FUS and DDIT3 genes, and both are expressed in different tissues like the fat tissue. When the translocation occurs, this tissue expresses an aberrant protein (TLS-CHOP) that alters the normal molecular pathways and develops to the disease.

As we are going to see, TLS-CHOP has the upstream sequence of FUS so the study of its expression and the transcription factors that are able to bind its promoter in order to regulate its transcription is focused on this gene. It is also presented an special algorism in Perl language made by us that has been used to made the promoter study.

Go back

• GENE FUNCTION

  
  
  TLS-CHOP is an hybrid protein involved in myoxid liposarcoma. This disease is the most common subtype of the malignant adipose tumours, which its cells are histologically presented as monomorphic, undifferentiated fusiform and small signed-ring lipoblast (see the image below). This tumour is characterized by the presence of a reciprocal chromosomal translocation, t(12;16)(q13;p11), which results to a fusion gene consisting of the 5' part of the FUS, which encodes for TLS protein (also called FUS), and the complete coding region of the DDIT3, which encodes for CHOP.
  
  CHOP is a nuclear protein, member of C/EBP transcription factors family, that is involved in growth arrest, adipogenesis and erythropoiesis. It is encoded by DDIT3 and its expression is induced by DNA-damage (DNA damage-inducible transcript 3). The function of CHOP consists in stopping the DNA synthesis on the cell cycle in order to avoid the replication of the cell until the DNA damage has been repaired. CHOP acts as a dominant negative that forms heterodimers with other members of its family by its leucine zipper dimeritzation domain. As a result the C/EBP tranctription factors can't bind to their DNA locus.
  It is known that CHOP can be fused with EWS as a result of another type of chromosomal translocation, t(12;22;20), that also causes myoxid liposarcoma.
  Below these lines we expose the biological function of CHOP provided by Gene Onthology data base.

  Transcription corepressor activity
  Transcription factor activity
  mRNA transcription from RNA polymerase II promoter
  Regulation of cell redox homeostasis
  Regulation of progression through cell cycle
  Regulation of transcription, DNA-dependent
  Response to DNA damage stimulus

  
  

  FUS is a protein component of nuclear riboprotein complexes. Because of the fact that it has a RNA binding motif it may be implicated in mRNA metabolism. It acts like a mRNA chaperone in orther to mantain the genomic stability. It was firstly discovered at myoxid liposarcoma, and was named TLS (translocated in liposarcoma).

  Here we expose the biological function of FUS provided by Gene Onthology data base.

  Nucleus
  Protein binding
  RNA binding

  
  

  In the hybrid gene de RNA binding motif of FUS is repaced by de DNA-binding and the leucine zipper dimeritzation domain of CHOP. In this situation FUS is able to bind to DNA and activate the expression of some oncogenic genes like FUS-ERG oncoprotein. Furthermore, it is thought that the hybrid protein functions as abnormal transcription factor, causing deregulation of CHOP target genes and attenuation of some functions that are critical for the differentation and growth control. Because of that, myoxid liposarcoma cell aren't able to stop the DNA synthesis when there is DNA damage. In addition, as we have described before, CHOP is involved in adipose cell differentiation and the presence of aberrant transcripts can alter the molecular pathways that control the differentiation driving to the development of myoxid liposarcoma.

  
  
  Go back
  
  
  

• CHARACTERITZATION OF THE GENOMIC STRUCTURE

  
  
  We are going to search the chromosomic position of the hybrid gene TLS/CHOP and analyze its mRNA sequence in order to know how many coding and non-coding exons has, but first we must analyze the two-single genes TLS (FUS) and CHOP (DDIT3) separately.
  
  
  

  TLS GENE

  The FUS gene (fusion protein involved in translocation t(12;16) in malignant liposarcoma) is found at the human sixteenth chromosome on its short arm (p), 11.2 band, specifically located from the 31.098.973th to 31.110.397th position, as you can see on the following image extracted from UCSC Genome Browser Database.
  
  

  It has three different transcripts which encode for the different isoforms of the FUS protein, which they are showed on the next image extracted from Ensembl Genome Browser Database. Click on the Ensembl ID of FUS to see the Ensembl Gene Report for ENSG00000089280.

  
  

  FUS protein, transcript variant 1
  
  

  FUS protein, transcript variant 2
  
  

  FUS protein, transcript variant 3
  
  

  The first transcript (Ensembl ID: ENST00000254108) encodes for the largest version of the protein (526 residues) and has 15 exons in total. On the other hand, the second transcript (Ensembl ID: ENST00000354711) is formed by an alternative splicing process and has 14 exons in total(*), missing the original seventh exon of the first transcript. Due to this the second transcript misses 35 nucleotides and as a result, it changes the codon lecture on the traduction process that derives to a new stop-codon (TGA) apparition specifically at the seventh exon (the eighth of the first transcript), which finalizes on a shorter isoform of the FUS protein (only 263 residues). Finally we found no information on the NCBI Database about the third transcript, but on the Ensembl Genome Browser we characterised that it has 14 exons and encodes for the shortest isoform of the FUS protein, which only has 151 residues. However, this last transcript is not involved in the formation of the hybrid protein TLS/CHOP so we haven't studied it.
  

  (*)We found some information about the second transcript on the Ensembl Genome Browser Database that said that actually it had 13 exons only. But when we noticed that the first codon on the translated region wasn't an ATG (actually it was a TTT codon that encodes for a phenylalanine residue), we considered that the information was wrong or that some data was missing. In addition, the first exon of the transcript found in Ensembl coincided with the coding part of the first exon and with all the second exon from the second transcript found on the NCBI Database (we did a ClustalW in order to check it, it is presented clicking here). As a result, we thought that actually the second transcript only have missed the seventh exon of the first transcript and conserved the first and second original exons, as well as the starting codon ATG.
  

  Below these lines we have represented the protein sequence of both trasncripts. Here we realize that the second transcript is shorter than the first. We have marked in purple colour the aminoacids which are found equal in both transcripts.
  

  FUS Transcript 1 Protein Sequence:

  MASNDYTQQATQSYGAYPTQPGQGYSQQSSQPYGQQSYSGYSQSTDTSGYGQSSYSSYGQSQNTGYGTQSTPQGYGSTGGYGSSQSSQSSY GQQSSYPGYGQQPAPSSTSGSYGSSSQSSSYGQPQSGSYSQQPSYGGQQQSYGQQQSYNPPQGYGQQNQYNSSSGGGGGGGGGGNYGQDQ SSMSSGGGSGGGYGNQDQSGGGGSGGYGQQDRGGRGRGGSGGGGGGGGGGYNRSSGGYEPRGRGGGRGGRGGMGGSDRGGFNKFGGPR DQGSRHDSEQDNSDNNTIFVQGLGENVTIESVADYFKQIGIIKTNKKTGQPMINLYTDRETGKLKGEATVSFDDPPSAKAAIDWFDGKEFSGNPIK VSFATRRADFNRGGGNGRGGRGRGGPMGRGGYGGGGSGGGGRGGFPSGGGGGGGQQRAGDWKCPNPTCENMNFSWRNECNQCKAPKPD GPGGGPGGSHMGGNYGDDRRGGRGGYDRGGYRGRGGDRGGFRGGRGGGDRGGFGPGKMDSRGEHRQDRRERPY
  
  

  FUS Transcript 2 Protein Sequence:

  MASNDYTQQATQSYGAYPTQPGQGYSQQSSQPYGQQSYSGYSQSTDTSGYGQSSYSSYGQSQNTGYGTQSTPQGYGSTGGYGSSQSSQSSY GQQSSYPGYGQQPAPSSTSGSYGSSSQSSSYGQPQSGSYSQQPSYGGQQQSYGQQQSYNPPQGYGQQNQYNSSSGGGGGGGGGGNYGQDQ SSMSSGGGSGGGYGNQDQSGGGGSGGYGQQDRGGRGRGGSGGGGGGGGGGYNRSSGGYEPRGRGGGRGGRGGMGPSGPRITS
  

  On the following table we have the sequences of the FUS first and second transcripts extracted from the NCBI Database, each on the first and second column respectively. We have presented the processed mRNA sequences (without introns) indicating by the two colours black and blue the different exons they have. We can also identify in bolds the translated starting-codon ATG (at the first exon in both transcripts) and final untranslated stop-codon (TAA for the first transcript in the last exon and TGA for the second transcript in the eighth exon). We also have marked in bolds the seventh exon on the first transcript mRNA which is missing on the second's.

  
  

  FUS Transcript 1 mRNA Sequence

  FUS Transcript 2 mRNA Sequence

  CGGGGCTGCTCAGTCCTCCAGGCGTCGGTACTCAGCGGTGTTGGAACTTCGTTGCTTGCT TGCCTGTGCGCGCGTGCGCGGACATGGCCTCAAACGATTATACCCAACAAGCAACCCAAA GCTATGGGGCCTACCCCACCCAGCCCGGGCAGGGCTATTCCCAGCAGAGCAGTCAGCCCT ACGGACAGCAGAGTTACAGTGGTTATAGCCAGTCCACGGACACTTCAGGCTATGGCCAGA GCAGCTATTCTTCTTATGGCCAGAGCCAGAACACAGGCTATGGAACTCAGTCAACTCCCC AGGGATATGGCTCGACTGGCGGCTATGGCAGTAGCCAGAGCTCCCAATCGTCTTACGGGC AGCAGTCCTCCTACCCTGGCTATGGCCAGCAGCCAGCTCCCAGCAGCACCTCGGGAAGTT ACGGTAGCAGTTCTCAGAGCAGCAGCTATGGGCAGCCCCAGAGTGGGAGCTACAGCCAGC AGCCTAGCTATGGTGGACAGCAGCAAAGCTATGGACAGCAGCAAAGCTATAATCCCCCTC AGGGCTATGGACAGCAGAACCAGTACAACAGCAGCAGTGGTGGTGGAGGTGGAGGTGGAG GTGGAGGTAACTATGGCCAAGATCAATCCTCCATGAGTAGTGGTGGTGGCAGTGGTGGCG GTTATGGCAATCAAGACCAGAGTGGTGGAGGTGGCAGCGGTGGCTATGGACAGCAGGACC GTGGAGGCCGCGGCAGGGGTGGCAGTGGTGGCGGCGGCGGCGGCGGCGGTGGTGGTTACA ACCGCAGCAGTGGTGGCTATGAACCCAGAGGTCGTGGAGGTGGCCGTGGAGGCAGAGGTG GCATGGGCGGAAGTGACCGTGGTGGCTTCAATAAATTTGGTGGCCCTCGGGACCAAGGAT CACGTCATGACTCCGAACAGGATAATTCAGACAACAACACCATCTTTGTGCAAGGCCTGG GTGAGAATGTTACAATTGAGTCTGTGGCTGATTACTTCAAGCAGATTGGTATTATTAAGA CAAACAAGAAAACGGGACAGCCCATGATTAATTTGTACACAGACAGGGAAACTGGCAAGC TGAAGGGAGAGGCAACGGTCTCTTTTGATGACCCACCTTCAGCTAAAGCAGCTATTGACT GGTTTGATGGTAAAGAATTCTCCGGAAATCCTATCAAGGTCTCATTTGCTACTCGCCGGG CAGACTTTAATCGGGGTGGTGGCAATGGTCGTGGAGGCCGAGGGCGAGGAGGACCCATGG GCCGTGGAGGCTATGGAGGTGGTGGCAGTGGTGGTGGTGGCCGAGGAGGATTTCCCAGTG GAGGTGGTGGCGGTGGAGGACAGCAGCGAGCTGGTGACTGGAAGTGTCCTAATCCCACCT GTGAGAATATGAACTTCTCTTGGAGGAATGAATGCAACCAGTGTAAGGCCCCTAAACCAG ATGGCCCAGGAGGGGGACCAGGTGGCTCTCACATGGGGGGTAACTACGGGGATGATCGTC GTGGTGGCAGAGGAGGCTATGATCGAGGCGGCTACCGGGGCCGCGGCGGGGACCGTGGAG GCTTCCGAGGGGGCCGGGGTGGTGGGGACAGAGGTGGCTTTGGCCCTGGCAAGATGGATT CCAGGGGTGAGCACAGACAGGATCGCAGGGAGAGGCCGTATTAATTAGCCTGGCTCCCCA GGTTCTGGAACAGCTTTTTGTCCTGTACCCAGTGTTACCCTCGTTATTTTGTAACCTTCC AATTCCTGATCACCCAAGGGTTTTTTTGTGTCGGACTATGTAATTGTAACTATACCTCTG GTTCCCATTAAAAGTGACCATTTTAGTTAAATTTTGTTCCTCTTCCCCCTTTTCACTTTC CTGGAAGATCGATGTCCCGATCAGGAAGGTAGAGAGTTTTCCTGTTCAGATTACCCTGCC CAGCAGGAACTGGAATACAGTGTTCGGGGAGAAGGCCAAATGATATCCTTGAGAGCAGAG ATTAAACTTTTCTGTCATGGGGAAAAAAAA

  CGGGGCTGCTCAGTCCTCCAGGCGTCGGTACTCAGCGGTGTTGGAACTTCGTTGCTTGCT TGCCTGTGCGCGCGTGCGCGGACATGGCCTCAAACGATTATACCCAACAAGCAACCCAAA GCTATGGGGCCTACCCCACCCAGCCCGGGCAGGGCTATTCCCAGCAGAGCAGTCAGCCCT ACGGACAGCAGAGTTACAGTGGTTATAGCCAGTCCACGGACACTTCAGGCTATGGCCAGA GCAGCTATTCTTCTTATGGCCAGAGCCAGAACACAGGCTATGGAACTCAGTCAACTCCCC AGGGATATGGCTCGACTGGCGGCTATGGCAGTAGCCAGAGCTCCCAATCGTCTTACGGGC AGCAGTCCTCCTACCCTGGCTATGGCCAGCAGCCAGCTCCCAGCAGCACCTCGGGAAGTT ACGGTAGCAGTTCTCAGAGCAGCAGCTATGGGCAGCCCCAGAGTGGGAGCTACAGCCAGC AGCCTAGCTATGGTGGACAGCAGCAAAGCTATGGACAGCAGCAAAGCTATAATCCCCCTC AGGGCTATGGACAGCAGAACCAGTACAACAGCAGCAGTGGTGGTGGAGGTGGAGGTGGAG GTGGAGGTAACTATGGCCAAGATCAATCCTCCATGAGTAGTGGTGGTGGCAGTGGTGGCG GTTATGGCAATCAAGACCAGAGTGGTGGAGGTGGCAGCGGTGGCTATGGACAGCAGGACC GTGGAGGCCGCGGCAGGGGTGGCAGTGGTGGCGGCGGCGGCGGCGGCGGTGGTGGTTACA ACCGCAGCAGTGGTGGCTATGAACCCAGAGGTCGTGGAGGTGGCCGTGGAGGCAGAGGTG GCATGGGGCCCTCGGGACCAAGGATCACGTCATGACTCCGAACAGGATAATTCAGACAAC AACACCATCTTTGTGCAAGGCCTGGGTGAGAATGTTACAATTGAGTCTGTGGCTGATTAC TTCAAGCAGATTGGTATTATTAAGACAAACAAGAAAACGGGACAGCCCATGATTAATTTG TACACAGACAGGGAAACTGGCAAGCTGAAGGGAGAGGCAACGGTCTCTTTTGATGACCCA CCTTCAGCTAAAGCAGCTATTGACTGGTTTGATGGTAAAGAATTCTCCGGAAATCCTATC AAGGTCTCATTTGCTACTCGCCGGGCAGACTTTAATCGGGGTGGTGGCAATGGTCGTGGA GGCCGAGGGCGAGGAGGACCCATGGGCCGTGGAGGCTATGGAGGTGGTGGCAGTGGTGGT GGTGGCCGAGGAGGATTTCCCAGTGGAGGTGGTGGCGGTGGAGGACAGCAGCGAGCTGGT GACTGGAAGTGTCCTAATCCCACCTGTGAGAATATGAACTTCTCTTGGAGGAATGAATGC AACCAGTGTAAGGCCCCTAAACCAGATGGCCCAGGAGGGGGACCAGGTGGCTCTCACATG GGGGGTAACTACGGGGATGATCGTCGTGGTGGCAGAGGAGGCTATGATCGAGGCGGCTAC CGGGGCCGCGGCGGGGACCGTGGAGGCTTCCGAGGGGGCCGGGGTGGTGGGGACAGAGGT GGCTTTGGCCCTGGCAAGATGGATTCCAGGGGTGAGCACAGACAGGATCGCAGGGAGAGG CCGTATTAATTAGCCTGGCTCCCCAGGTTCTGGAACAGCTTTTTGTCCTGTACCCAGTGT TACCCTCGTTATTTTGTAACCTTCCAATTCCTGATCACCCAAGGGTTTTTTTGTGTCGGA CTATGTAATTGTAACTATACCTCTGGTTCCCATTAAAAGTGACCATTTTAGTTAAATTTT GTTCCTCTTCCCCCTTTTCACTTTCCTGGAAGATCGATGTCCCGATCAGGAAGGTAGAGA GTTTTCCTGTTCAGATTACCCTGCCCAGCAGGAACTGGAATACAGTGTTCGGGGAGAAGG CCAAATGATATCCTTGAGAGCAGAGATTAAACTTTTCTGTCATGGGGAAAAAAAAAA

  
  
  

  When we have analyzed the exonic structure of the mRNA of each FUS transcript and then we have relacionated it with the length of the resultant protein, we can resolve that the first transcript has more coding exons than the second, and the second transcript has more non-coding exons than the first. Actually, all the exons of the first FUS transcript are translated into protein (they are all coding) except the 5' region before the starting-codon ATG and the 3' region behind the last stop-codon TAA. On the other hand, the first seven exons of the second FUS transcript are translated into protein except the 5' region before the first codon ATG. However, the 3' region after the stop-codon TGA in the seventh exon and the rest of exons of the transcript are non-coding. It means that they exist in the processed mRNA but they won't be translated into protein.
  

  In conclusion, we can affirmate that the synthesis of the different FUS transcripts by using the alternative splicing mechanism is changing the codon lecture during the translation process, and this can affect on the final translated protein. In our case, the hybrid protein TLS/CHOP conservates the first part of the aminoacidic residues from FUS protein, so it has incorporated the first exons of its mRNA, which are the same in both transcripts, as we have seen before. Because of this, we think that the hybrid protein could also suffer alternative splicing like the FUS protein, and as a result become two different isoforms.

  
  

  DDIT3 GENE

  The DDIT3 gene (DNA-damage inducible transcript 3) encodes for the CHOP protein and it's found on the human twelveth chromosome on its long arm (q), 13,3 band. It is specifically located from the 56,196,640th position to the 56,200,567th as it's seen on the next image extracted from the UCSC Genome Browser Database.
  
  

  We have found on the Ensembl Genome Browser Database that the DDIT3 gene only has a single transcript (Ensembl ID:ENST00000346473) which is translated into the 169 aminoacidic residues of the final protein, which sequence is represented below. As we can see on the transcript image, the DDIT3 gene is transcribed reversely (take a look at the sense of the transcription marked by the black arrow). Click on the Ensembl ID of DDIT3 to see the Ensembl Gene Report for ENSG00000175197.

  
  

  DDIT3 transcript

  
  

  CHOP Protein Sequence:

  MAAESLPFSFGTLSSWELEAWYEDLQEVLSSDENGGTYVSPPGNEEEESKIFTTLDPASLAWLTEEEPEP
  AEVTSTSQSPHSPDSSQSSLAQEEEEEDQGRTRKRKQSGHSPARAGKQRMKEKEQENERKVAQLAEENE
  RLKQEIERLTREVEATRRALIDRMVNLHQA
  
  

  If we analyze the processed DDIT3 mRNA sequence which is showed below we will find that it has 4 exons in total, separated by black and blue colours. We also will realise that only the two last of them are codifying because the first translated codon ATG is on the start of the third exon and the TGA stop-codon is located in the middle of the fourth. We can see both codons marked in red bolds.
  

  DDIT3 mRNA Sequence:

  GAGGTCAGAGACTTAAGTCTAAGGCACTGAGCGTATCATGTTAAAGATGAGCGGGTGGCA
  GCGACAGAGCCAAAATCAGAGCTGGAACCTGAGGAGAGAGTGTTCAAGAAGGAAGTGTAT
  CTTCATACATCACCACACCTGAAAGCAGATGTGCTTTTCCAGACTGATCCAACTGCAGAG
  ATGGCAGCTGAGTCATTGCCTTTCTCCTTCGGGACACTGTCCAGCTGGGAGCTGGAAGCC
  TGGTATGAGGACCTGCAAGAGGTCCTGTCTTCAGATGAAAATGGGGGTACCTATGTTTCA
  CCTCCTGGAAATGAAGAGGAAGAATCAAAAATCTTCACCACTCTTGACCCTGCTTCTCTG
  GCTTGGCTGACTGAGGAGGAGCCAGAACCAGCAGAGGTCACAAGCACCTCCCAGAGCCCT
  CACTCTCCAGATTCCAGTCAGAGCTCCCTGGCTCAGGAGGAAGAGGAGGAAGACCAAGGG
  AGAACCAGGAAACGGAAACAGAGTGGTCATTCCCCAGCCCGGGCTGGAAAGCAGCGCATG
  AAGGAGAAAGAACAGGAGAATGAAAGGAAAGTGGCACAGCTAGCTGAAGAGAATGAACGG
  CTCAAGCAGGAAATCGAGCGCCTGACCAGGGAAGTAGAGGCGACTCGCCGAGCTCTGATT
  GACCGAATGGTGAATCTGCACCAAGCATGAACAATTGGGAGCATCAGTCCCCCACTTGGG
  CCACACTACCCACCTTTCCCAGAAGTGGCTACTGACTACCCTCTCACTAGTGCCAATGAT
  GTGACCCTCAATCCCACATACGCAGGGGGAAGGCTTGGAGTAGACAAAAGGAAAGGTCTC
  AGCTTGTATATAGAGATTGTACATTTATTTATTACTGTCCCTATCTATTAAAGTGACTTT
  CTATGAGCC
  
  

  In conclusion, we can affirmate that the total transcript has two non-coding exons (first and second) and two coding exons (third and fourth). And because of it only has one transcript on the Databases it only encodes for one isoform.

  
  

  TLS/CHOP GENE

  The TLS/CHOP gene is transcribed and translated to an hybrid protein that is made of the first part of FUS protein and the complete CHOP protein. We have analyzed more specifically it's mRNA in order to know which exons are conserved from the original genes and which aminoacids are derived from those exons.
  

  First of all we are going to analyze the protein sequence of TLS/CHOP which is showed below. We have coloured the different regions that takes coincidences with the original proteins . We have marked in purple the aminoacids that TLS/CHOP shares with the FUS protein and in orange the aminoacids that TLS/CHOP shares with the CHOP protein. We can check that while the whole CHOP protein is conserved in the hybrid potein, only the first half of FUS is present on it, until the region encoded by the seventh exon.
  

  TLS/CHOP Protein Sequence
  

  MASNDYTQQATQSYGAYPTQPGQGYSQQSSQPYGQQSYSGYSQSTDTSGYGQSSYSSYGQSQNTGYGTQS
  TPQGYGSTGGYGSSQSSQSSYGQQSSYPGYGQQPAPSSTSGSYGSSSQSSSYGQPQSGSYSQQPSYGGQQ
  QSYGQQQSYNPPQGYGQQNQYNSSSGGGGGGGGGGNYGQDQSSMSSGGGSGGGYGNQDQSGGGGSGGYGQ
  QDRGGRGRGGSGGGGGGGGGGYNRSSGGYEPRGRGGGRGGRGGMGGSDRGGFNKFGVFKKEVYLHTSPHL
  KADVLFQTDPTAEMAAESLPFSFGTLSSWELEAWYEDLQEVLSSDENGGTYVSPPGNEEEESKIFTTLDP
  ASLAWLTEEEPEPAEVTSTSQSPHSPDSSQSSLAQEEEEEDQGRTRKRKQSGHSPARAGKQRMKEKEQEN
  ERKVAQLAEENERLKQEIERLTREVEATRRALIDRMVNLHQA
  

  As we can see in the protein sequence represented above, there are several aminoacids in the middle of the sequence that seems that are not derived from any of the two original proteins. Because of that, we need to analyze the mRNA sequence of TLS/CHOP, extracted from the NCBI Database, which is showed below. There are two columns which shows the same sequence but each represents different things. On the one hand, the first column shows the total number of exons that the sequence has; we have marked them in blue and black colours to differenciate them. On the other hand, the second column shows the conservation of the sequence with the FUS mRNA sequence (marked in purple colour) and with the DDIT3 mRNA sequence (marked in orange colour). We also have marked in red bolds the starting-codon ATG and the stop-codon TGA used in the translation process in the sequence of both columns.
  
  
  

  TLS/CHOP mRNA Sequence (exons)	TLS/CHOP mRNA Sequence (from FUS and DDIT3 alignment)
  ATGCTCAGTCCTCCAGGCGTCGGTGCTCAGCGGTGTTGGAACTTCGTTGCTTGCTTGC CTGTGCGCGCGTGCGCGGACATGGCCTCAAACGATTATACCCAACAAGCAACCCAAAG CTATGGGGCCTACCCCACCCAGCCCGGGCAGGGCTATTCCCAGCAGAGCAGTCAGCCC TACGGACAGCAGAGTTACAGTGGTTATAGCCAGTCCACGGACACTTCAGGCTATGGCC AGAGCAGCTATTCTTCTTATGGCCAGAGCCAGAACACAGGCTATGGAACTCAGTCAAC TCCCCAGGGATATGGCTCGACTGGCGGCTATGGCAGTAGCCAGAGCTCCCAATCGTCT TACGGGCAGCAGTCCTCCTACCCTGGCTATGGCCAGCAGCCAGCTCCCAGCAGCACCT CGGGAAGTTACGGTAGCAGTTCTCAGAGCAGCAGCTATGGGCAGCCCCAGAGTGGGAG CTACAGCCAGCAGCCTAGCTATGGTGGACAGCAGCAAAGCTATGGACAGCAGCAAAGC TATAATCCCCCTCAGGGCTATGGACAGCAGAACCAGTACAACAGCAGCAGTGGTGGTG GAGGTGGAGGTGGAGGTGGAGGTAACTATGGCCAAGATCAATCCTCCATGAGTAGTGG TGGTGGCAGTGGTGGCGGTTATGGCAATCAAGACCAGAGTGGTGGAGGTGGCAGCGGT GGCTATGGACAGCAGGACCGTGGAGGCCGCGGCAGGGGTGGCAGTGGTGGCGGCGGCG GCGGCGGCGGTGGTGGTTACAACCGCAGCAGTGGTGGCTATGAACCCAGAGGTCGTGG AGGTGGCCGTGGAGGCAGAGGTGGCATGGGCGGAAGTGACCGTGGTGGCTTCAATAAA TTTGGTGTGTTCAAGAAGGAAGTGTATCTTCATACATCA CCACACCTGAAAGCAGATGTGCTTTTCCAGACTGATCCAACTGCAGAGATGGCAGCTG AGTCATTGCCTTTCTCCTTCGGGACACTGTCCAGCTGGGAGCTGGAAGCCTGGTATGA GGACCTGCAAGAGGTCCTGTCTTCAGATGAAAATGGGGGTACCTATGTTTCACCTCCT GGAAATGAAGAGGAAGAATCAAAAATCTTCACCACTCTTGACCCTGCTTCTCTGGCTT GGCTGACTGAGGAGGAGCCAGAACCAGCAGAGGTCACAAGCACCTCCCAGAGCCCTCA CTCTCCAGATTCCAGTCAGAGCTCCCTGGCTCAGGAGGAAGAGGAGGAAGACCAAGGG AGAACCAGGAAACGGAAACAGAGTGGTCATTCCCCAGCCCGGGCTGGAAAGCAGCGCA TGAAGGAGAAAGAACAGGAGAATGAAAGGAAAGTGGCACAGCTAGCTGAAGAGAATGA ACGGCTCAAGCAGGAAATCGAGCGCCTGACCAGGGAAGTAGAGGCGACTCGCCGAGCT CTGATTGACCGAATGGTGAATCTGCACCAAGCATGAACAATTGGGAGCATCAGTCCCC CACTTGGGCCACACTACCCACCTTTCCCAGAAGTGGCTACTGACTACCCTCTCACTAG TGCCAATGATGTGACCCTCAATCCCACATACGCAGGGGGAAGGCTTGGAGTAGACAAA AGGAAAGGTCTCAGCTTGTATATAGAGATTGTACATTTATTTATTACTGTCCCTATCT ATTAAAGTGACTTTCTATG	ATGCTCAGTCCTCCAGGCGTCGGTGCTCAGCGGTGTTGGAACTTCGTTGCTTGCTTGC CTGTGCGCGCGTGCGCGGACATGGCCTCAAACGATTATACCCAACAAGCAACCCA AAGCTATGGGGCCTACCCCACCCAGCCCGGGCAGGGCTATTCCCAGCAGAGCAGTCAG CCCTACGGACAGCAGAGTTACAGTGGTTATAGCCAGTCCACGGACACTTCAGGCTATG GCCAGAGCAGCTATTCTTCTTATGGCCAGAGCCAGAACACAGGCTATGGAACTCAGTC AACTCCCCAGGGATATGGCTCGACTGGCGGCTATGGCAGTAGCCAGAGCTCCCAATCG TCTTACGGGCAGCAGTCCTCCTACCCTGGCTATGGCCAGCAGCCAGCTCCCAGCAGCA CCTCGGGAAGTTACGGTAGCAGTTCTCAGAGCAGCAGCTATGGGCAGCCCCAGAGTGG GAGCTACAGCCAGCAGCCTAGCTATGGTGGACAGCAGCAAAGCTATGGACAGCAGCAA AGCTATAATCCCCCTCAGGGCTATGGACAGCAGAACCAGTACAACAGCAGCAGTGGTG GTGGAGGTGGAGGTGGAGGTGGAGGTAACTATGGCCAAGATCAATCCTCCATGAGTAG TGGTGGTGGCAGTGGTGGCGGTTATGGCAATCAAGACCAGAGTGGTGGAGGTGGCAGC GGTGGCTATGGACAGCAGGACCGTGGAGGCCGCGGCAGGGGTGGCAGTGGTGGCGGCG GCGGCGGCGGCGGTGGTGGTTACAACCGCAGCAGTGGTGGCTATGAACCCAGAGGTCG TGGAGGTGGCCGTGGAGGCAGAGGTGGCATGGGCGGAAGTGACCGTGGTGGCTTCAAT AAATTTGGTGTGTTCAAGAAGGAAGTGTATCTTCATACATCACCACACCTGAAAGCA GATGTGCTTTTCCAGACTGATCCAACTGCAGAGATGGCAGCTGAGTCATTGCCTTTCT CCTTCGGGACACTGTCCAGCTGGGAGCTGGAAGCCTGGTATGAGGACCTGCAAGAGGT CCTGTCTTCAGATGAAAATGGGGGTACCTATGTTTCACCTCCTGGAAATGAAGAGGAA GAATCAAAAATCTTCACCACTCTTGACCCTGCTTCTCTGGCTTGGCTGACTGAGGAGG AGCCAGAACCAGCAGAGGTCACAAGCACCTCCCAGAGCCCTCACTCTCCAGATTCCAG TCAGAGCTCCCTGGCTCAGGAGGAAGAGGAGGAAGACCAAGGGAGAACCAGGAAACGG AAACAGAGTGGTCATTCCCCAGCCCGGGCTGGAAAGCAGCGCATGAAGGAGAAAGAAC AGGAGAATGAAAGGAAAGTGGCACAGCTAGCTGAAGAGAATGAACGGCTCAAGCAGGA AATCGAGCGCCTGACCAGGGAAGTAGAGGCGACTCGCCGAGCTCTGATTGACCGAATG GTGAATCTGCACCAAGCATGAACAATTGGGAGCATCAGTCCCCCACTTGGGCCACA CTACCCACCTTTCCCAGAAGTGGCTACTGACTACCCTCTCACTAGTGCCAATGATGTG ACCCTCAATCCCACATACGCAGGGGGAAGGCTTGGAGTAGACAAAAGGAAAGGTCTCA GCTTGTATATAGAGATTGTACATTTATTTATTACTGTCCCTATCTATTAAAGTGACTT TCTATG

  
  

  When we analized the protein sequence (look before) we found that some residues weren't present on FUS protien, neither CHOP protein. But now when we compare the mRNA sequence of the hybrid gene with the mRNA of the original genes, the whole sequence alignes. It is due to the fact that these residues are encoded by the non-translated second exon of DDIT3 mRNA. Using the information provided by the right column of the table, we can see which region of TLS/CHOP mRNA sequence is derived from the FUS mRNA sequence and which from the DDIT3 mRNA sequence. Considering that the whole sequence contains 1682 nucleotides, the region compressed between the first nucleotide and the 877th is derived from the first part of the FUS mRNA sequence, whereas the other region (from 877th nucleotide to the last one) is derived from the DDIT3 mRNA sequence.
  

  On the left column we can see that TLS/CHOP mRNA sequence contains 10 exons in total which all encode for the final protein, because the starting-codon ATG it's present on the first exon and the stop-codon TGA it's on the last one. If we relacionate the number of exons with the information provided on the right column, we will be able to realize that the final of the seventh exon of the TLS/CHOP mRNA sequence coincides exactly with the final conserved part of the FUS mRNA sequence, and also happens with the beginning of the eighth exon of TLS/CHOP mRNA sequence, which is the same as the beginning of the conserved part of the DDIT3 mRNA sequence. So now we can extract two conclusions about this coincidences:
  

  • Exons of the TLS/CHOP gene compressed from the first to the seventh are derived from the exons of the FUS gene, whereas the other three located downstream are derived from the exons of the DDIT3 gene.
  • The translocation point between the FUS gene and the DDIT3 gene to form the hybrid TLS/CHOP gene is located on the 877th position of the FUS mRNA sequence which finalizes with the seventh exon, and continues with the downstream DDIT3 mRNA sequence from the 102th position, which coincides with the beginning of the second exon of this mRNA sequence.

  We can analyze in the next table the different exons of TLS/CHOP, from which original gene are derived, which are the positions that occupy in both genes and finally for which residues encode on the final protein.
  
  

  Exon number	Exon sequence	Exon position in TLS/CHOP gene	Exon length	Exon derivation	Codifying aminoacid residues
  1	ATGCTCAGTCCTCCAGGCGTCGGTGCTCAGCGGTGTTGGAACTTC GTTGCTTGCTTGCCTGTGCGCGCGTGCGCGGACATGGCCTCAAACG	1-91	91	FUS gene Exon 1	MASND
  2	ATTATACCCAACAAGCAACCCAAAG	92-116	25	FUS gene Exon 2	DYTQQATQS
  3	CTATGGGGCCTACCCCACCCAGCCCGGGCAGGGCTATTCCCAGCAG AGCAGTCAGCCCTACGGACAGCAGAGTTACAGTGGTTATAGCCAGTC CACGGACACTTCAGGCTATGGCCAGAGCAGCTATTCTTCTTATGGCC AGAGCCAGAACA	117-268	152	FUS gene Exon 3	SYGAYPTQPGQGYSQQSSQPYGQQSYSGY SQSTDTSGYGQSSYSSYGQSQNT
  4	CAGGCTATGGAACTCAGTCAACTCCCCAGGGATATGGCTCGACTGGC GGCTATGGCAGTAGCCAGAGCTCCCAATCGTCTTACGGGCAGCAGTC CTCCTACCCTGGCTATGGCCAGCAGCCAGCTCCCAGCAGCACCTCGG GAAG	269-412	145	FUS gene Exon 4	TGYGTQS TPQGYGSTGGYGSSQS SQSSYGQQSSYPGYGQQPAPSSTSGS
  5	TTACGGTAGCAGTTCTCAGAGCAGCAGCTATGGGCAGCCCCAGAGTGG GAGCTACAGCCAGCAGCCTAGCTATGGTGGACAGCAGCAAAGCTATGG ACAGCAGCAAAGCTATAATCCCCCTCAGGGCTATGGACAGCAGAACCA GTACAACAGCAGCAGTGGTGGTGGAGGTGGAGGTGGAGGTGGAG	413-601	188	FUS gene Exon 5	SYGSSSQSSSYGQPQSGSYSQQPSYGGQQ QSY GQQQSYNPPQGYGQQNQYNSSSGGGGGGGGGG
  6	GTAACTATGGCCAAGATCAATCCTCCATGAGTAGTGGTGGTGGCAGTGG TGGCGGTTATGGCAATCAAGACCAGAGTGGTGGAGGTGGCAGCGGTGGC TATGGACAGCAGGACCGTGGAGGCCGCGGCAGGGGTGGCAGTGGTGGCG GCGGCGGCGGCGGCGGTGGTGGTTACAACCGCAGCAGTGGTGGCTATGA ACCCAGAGGTCGTGGAGGTGGCCGTGGAGGCAGAGGTGGCATGGG	602-842	241	FUS gene Exon 6	GNYGQDQSSMSSGGGSGGGYGNQDQSGGGGSG GYGQQDRGGRGRGGSGGGGGGGGGGYNRSSG GYEPRGRGGGRGGRGGMG
  7	CGGAAGTGACCGTGGTGGCTTCAATAAATTTGGTG	843-877	35	FUS gene Exon 7	GGSDRGGFNKFGV
  8	TGTTCAAGAAGGAAGTGTATCTTCATACATCACCACACCTGAAAGCAG	878-925	48	DDIT3 gene Exon 2	VFKKEVYLHTSPHL KAD
  9	ATGTGCTTTTCCAGACTGATCCAACTGCAGAGATGGCAGCTGAGTCATTG CCTTTCTCCTTTGGGACACTGTCCAGCTGGGAGCTGGAAGCCTGGTATGA GGACCTGCAAGAGGTCCTGTCTTCAGATGAAAATGGGGGTACCTATGTTT CACCTCCTGGAAATGAAGAG	926-1095	170	DDIT3 gene Exon 3	DVLFQTDPTAEMAAESLPFSFGTLSSWELEAWYED LQEVLSSDENGGTYVSPPGNEE
  10	GAAGAATCAAAAATCTTCACCACTCTTGACCCTGCTTCTCTGGCTTGGCTG ACTGAGGAGGAGCCAGAACCAGCAGAGGTCACAAGCACCTCCCAGAGCCC TCACTCTCCAGATTCCAGTCAGAGCTCCCTGGCTCAGGAGGAAGAGGAGG AAGACCAAGGGAGAACCAGGAAACGGAAACAGAGTGGTCATTCCCCAGCC CGGGCTGGAAAGCAGCGCATGAAGGAGAAAGAACAGGAGAATGAAAGGAA AGTGGCACAGCTAGCTGAAGAGAATGAACGGCTCAAGCAGGAAATCGAGC GCCTGACCAGGGAAGTAGAGGCGACTCGCCGAGCTCTGATTGACCGAATG GTGAATCTGCACCAAGCATGAACAATTGGGAGCATCAGTCCCCCACTTGG GCCACACTACCCACCTTTCCCAGAAGTGGCTACTGACTACCCTCTCACTA GTGCCAATGATGTGACCCTCAATCCCACATACGCAGGGGGAAGGCTTGGA GTAGACAAAAGGAAAGGTCTCAGCTTGTATATAGAGATTGTACATTTATT TATTACTGTCCCTATCTATTAAAGTGACTTTCTATG	1096-1682	587	DDIT3 gene Exon 4	EESKIFTTLDPASLAWLTEEEPEPAEVTSTSQS PHSPDSSQSSLAQEEEEEDQGRTRKRKQSG HSPARAGKQRMKEKEQENERKVAQLAEENE RLKQEIERLTREVEATRRALIDRMVNLHQA

  In this table we have marked in blue bolds the starting-codon ATG and the stop-codon TGA in the first and last exon sequences, and also with red bolds the aminoacid residues which are shared between the different exons. The purple bolded ATG codon located on the nineth exon corresponds to the starting-codon of the DDIT3 gene.
  

  There are some things we can see with the information provided by the table:
  

  • The translocation point between the FUS exon 7 and the DDIT3 exon 2 derives to the formation of a new codon GTG (marked in orange bolds) which codifies for a valine residue. This valine is absent in the FUS protein and also in the translated-DDIT3 Exon 2 (which is normally untranslated).
  • The DDIT3 exon 2 translocation into the codifying region of the FUS mRNA causes that this exon becomes codifying on the hybrid TLS/CHOP gene, which derives to a new aminoacid sequence located between the FUS protein region and the DDIT3 protein conserved in the hybrid protein.

  We have made below those lines a similar version of the TLS/CHOP protein sequence we have presented before where we have marked in blue this new codified aminoacid sequence (by the original untranslated exons of the DDIT3 gene) and in red the valine which is codified by the new codon CTG generated by the translocation point:
  

  TLS/CHOP Protein Sequence
  

  MASNDYTQQATQSYGAYPTQPGQGYSQQSSQPYGQQSYSGYSQSTDTSGYGQSSYSSYGQSQNTGYGTQS
  TPQGYGSTGGYGSSQSSQSSYGQQSSYPGYGQQPAPSSTSGSYGSSSQSSSYGQPQSGSYSQQPSYGGQQ
  QSYGQQQSYNPPQGYGQQNQYNSSSGGGGGGGGGGNYGQDQSSMSSGGGSGGGYGNQDQSGGGGSGGYGQ
  QDRGGRGRGGSGGGGGGGGGGYNRSSGGYEPRGRGGGRGGRGGMGGSDRGGFNKFGVFKKEVYLHTSPHL
  KADVLFQTDPTAEMAAESLPFSFGTLSSWELEAWYEDLQEVLSSDENGGTYVSPPGNEEEESKIFTTLDP
  ASLAWLTEEEPEPAEVTSTSQSPHSPDSSQSSLAQEEEEEDQGRTRKRKQSGHSPARAGKQRMKEKEQEN
  ERKVAQLAEENERLKQEIERLTREVEATRRALIDRMVNLHQA
  

  
  
  Go back
  
  
  

• CONSERVATION OF GENE AMONG SPECIES

  In this section it's shown the information about the ortholog genes of FUS and DDIT3. Here are mentioned the species that have an ortholog gene of FUS or DDIT3, the Ensembl identification of the ortholog gene and the percentage of homology.
  
  

  ORTHOLOG GENES OF FUS

  Picture	Homo sapiens vs Specie	Ensembl ID	% homology peptide
  	Anopheles gambiae	-	43
  	Caenorhabditis elegans	-	37
  	Drosophila melanogaster	-	43
  	Mus musculus	-	95
  	Pan troglodytes	-	96
  	Rattus norvegicus	-	94
  	Saccharomyces cerevisiae	-	34

  
  
  
  
  

  ORTHOLOG GENES OF DDIT3

  Picture	Homo sapiens vs Specie	Ensembl ID	% homology peptide
  	Bos taurus	ENSBTAG00000031544	92
  	Canis familiaris	ENSCAFG00000000232	92
  	Cavia porcellus	ENSCPOG00000012871	89
  	Caenorhabditis elegans	-	32
  	Danio rerio	ENSDARG00000059836	41
  	Dasypus novemcinctus	ENSDNOG00000017504	90
  	Drosophila melanogaster	-	30
  	Echinops telfairi	ENSETEG00000002521	57
  	Erinaceus europaeus	ENSEEUG00000003634	91
  	Felis catus	ENSFCAG00000002084	51
  	Gasterosteus aculeatus	ENSGACG00000006480	23
  	Macaca mulatta	ENSMMUG00000011286	98
  	Mus musculus	ENSMUSG00000025408	88
  	Oryzias latipes	ENSORLG00000005361	24
  	Pan troglodytes	ENSPTRG00000022781	99
  	Rattus norvegicus	ENSRNOG00000006789	89
  	Takifugu rubripes	SINFRUG00000133607	24
  	Tetraodon nigroviridis	GSTENG00013575001	24
  	Tupaia belangeri	ENSTBEG00000000521	94
  	Xenopus tropicalis	ENSXETG00000006245	47

  
  
  Go back
  
  
  

• EXPRESSION CARACTERITZATION

  As we have said before FUS is ligated downstream with DDIT3 to form the hybrid gene TLS-CHOP. Due to this we can confirm that the hybrid gene has the same promoter as FUS and the control of the expression of both genes will be similar. So to study the expression of TLS-CHOP we focus on the expression of FUS. Here are some pictures that show the expression of FUS in some experiments using different microarrays and tissues. If the gene has a high expression in the tissue it is shown in red, a basal expression is represented in black and a low expression in showed in green. We have made the same for the DDIT3 gene, in order to know how is it's basal expression and in which tissues it takes place.
  
  

  FUS GENE

  • GNF Expression Atlas 2 Data from U133A and GNF1H Chips

    
    

  • Normal Human Tissue cDNA Microarrays

    
    

  • GNF Expression Atlas 1 Human Data on Affy U95 Chips

    
    

  • Gladstone Microarray Data Including Stem Cell Tissue

    
  
  
  
  

  DDIT3 GENE

  • GNF Expression Atlas 2 Data from U133A and GNF1H Chips

    
    

  • Normal Human Tissue cDNA Microarrays

    
    

  • GNF Expression Atlas 1 Human Data on Affy U95 Chips

    
    

  • Gladstone Microarray Data Including Stem Cell Tissue

    
  
  
  Go back
  
  
  

• PROMOTOR REGION CARACTERITZATION

As we have said on the characterization of the genomic structure section, the hybrid gene TLS/CHOP contains at the beginning the genomic sequence of the FUS gene, so in order to study the promoter region of the TLS/CHOP gene we have to analyze which transcription factors are susceptible to bind to the promotor region of this gene.

The promoter sequence is represented below (which is extracted from the UCSC Genome Browser Database), which has marked in purple bolds the TSS (Transcription Starting Site) nucleotide, in blue the first exon and in red the first intron. The promoter region is written in small letters:

FUS promotor region

tttgcagttacaagacctggattcgaatcacgactcctcttagctgccctgtaatcaggcacaattacttgggtctctgagtctcactttccttatctag
aaaacggaggtatctttacttccttcgtaagactgatgacaaggaaattatctgtgcattttgaaaccacttaagccttgtacacgttttatttctggga
tcgccctggtagggcttcagaaaaataaaaaggaggtccctgagaaaaggctgggtaccgtacatctgaggtcaaccctctctggtcccaaggatggcct
gggctgttccgccccgtggctccccaggggcaaagccatgaggatccgggtgagagcccagtgctggacgagcccggggcccaggggtcccggccgaaat
ccctgctgtctttcaggtcaaacgtcataatccccgaaccccagaaaggccgaaaggcaaggcaaccctgaaagacgacgaagtcaacctcagggcgcag
gagagggagggccagtgtgctgccgacgagggaggctggagccgcggggacgaggcgccccatacagcggcaagagggtggagggcaggagctcgccatc
ctgggtgaaagcggggcccagcgaaggggcccggccacaggaatctcggttccaccccgctactcccggctgtgactccagtttcgtccccagccgccgg
gaccgccccctcgccccgcccccagcgggcactcaggccgtaccactgtgccttcatgggggtggagatagatcgtgggctagtcctgccgaggagagag
gggttcttcctcaaaaaatatgattatgtatagtattcgcatgattctagttaacttgtttcccttctgcctgctcggaccctctacctgccctacgaag
ggggcggagtgcgttcctgcctccccctgctcttccgcgtttggtgcgcgcctgcgcggtgcgtaggcggcggagcgtacttaagcttcgacgcaggagg
CGGGGCTGCTCAGTCCTCCAGGCGTCGGTACTCAGCGGTGTTGGAACTTCGTTG
CTTGCTTGCCTGTGCGCGCGTGCGCGGACATGGCCTCAAACGgtag

Once we had the promoter sequence isolated we had to introduce it on a Perl program made following the instructions on the web in order to know which transcription factors were able to bind into the sequence. This program is described on the Methods section. Below these lines we have presented the results of the program on a table format. We have for each transcription factor the region which it probably binds, it's punctuation and it's p value, which we have to consider as the probability to bind this region by a random situation. If a transcription factor has a low p value (it means that we have a low probability to bind by random that position with that TF) and a positive punctuation we can consider it as a good candidate to bind the promoter sequence of the FUS gene and obviously for the promoter sequence of TLS/CHOP gene.

Transcription factor	Start binding region	End binding region	Binding region sequence	Punctuation	p value
AP-1 [T00029]	480	486	gaagtca	2,651	0,24
AR [T00040]	1044	1050	GAACTTC	2,747	0,43
c-Myc [T00140]	183	188	cacgtt	-996,022	0,66
NF-AT1 [T00550]	143	149	ggaaatt	3,201	0,21
NF-kappaB [T00590]	856	864	ttgtttccc	-996.421	0,96
SRF [T00764]	751	759	ccttcatgg	-996.355	0,43
YY1 [T00915]	294	299	atggcc	2,678	0,58
RXR-alpha [T01345]	162	167	tgaaac	2,496	0,64
HIF-1 [T01609]	1067	1075	TGCGCGCGT	-996.58	0,99
AhR [T01795]	1072	1078	GCGTGCG	3,011	0,29
PU.1 [T02068]	638	644	caggaat	2,664	0,32
HNF-4 [T02758]	1042	1049	TGGAACTT	-995.559	0,42
NRSF [T06124]	307	315	ttccgcccc	-1994.202	0,91

As we have said before, the transcription factors that have positive punctuation and lower p value are good candidates to bind our sequence. So in this case we could consider the transcription factors AP-1, NF-AT1 or AhR as good options. However, we used another method to find these transcription factors, using the PROMO Database. Below these lines we have the results in a table format. Again, we are going to analyze the binding region of the same transcription factors used in our Perl Program and their p value, in this case represented by the RE query column. This term shows the probability to obtain the same results on another sequence with the same length and the same nucleotide proportions as ours.

Transcription factor	Start binding region	End binding region	Binding region sequence	RE query
AP-1 [T00029]	672	680	tgactccag	0,07
AR [T00040]	1079	1087	GGACATGGC	0,15
NF-AT1 [T00550]	142	150	ggaaattat	0,03
NF-kappaB [T00590]	314	324	cgtggctcccc	0,03
YY1 [T00915]	293	296	atgg	4,011
RXR-alpha [T01345]	271	277	tcaaccc	0,07
AhR [T01795]	1067	1076	GCGCGCGTGC	0,03

First of all there are two things that we have to consider on the PROMO results:

• c-Myc, SRF, HIF-1, PU.1, HNF-4 and NRSF transcription factors are not found to bind the FUS promotor region because they don't appear on the PROMO results.
• There is a high numeric differency between the p values given by the Perl program and the RE query given by the PROMO results, but this is because both programs use a different type of matrix to analyze each transcription factor.

Now we have to compare both results to determinate which transcription factors are good candidates to bind the FUS promotor sequence. To do this comparation we have to ensure that the binding regions in the both results match approximately and the p value and the RE query should be similar. For example, the NF-AT1 transcription factor which has the lowest p value (0,21) also has a low RE query value (0,03), so it means that we have a low probability of random binding. In addition, the binding region of the promotor sequence is approximately the same in the Perl program and PROMO results. With all these data we can affirmate that the NF-AT1 transcription factor will probably bind the TLS/CHOP and FUS promotor sequences.

We also have another two transcription factors that could be considered good to bind the promotor sequence, such as AhR and YY1. The AhR transcription factor has low p and RE query values, and also would bind in a similar region in both results, the problem is that this binding region is located inside the first exon of the transcribed mRNA, so it really means that it's not a good option. On the other hand, the YY1 transcription factor matches the binding region in both results but has a very high RE query value, so it means that it has a high probability to bind the FUS or TLS/CHOP sequences by random.

To finalize this section we present again the TLS/CHOP (FUS) promotor sequence with the binding region of the NF-AT1 transcription factor marked in purple. We have also marked in orange the binding region of the other two transcription factors AhR and YY1:

FUS promotor region

tttgcagttacaagacctggattcgaatcacgactcctcttagctgccctgtaatcaggcacaattacttgggtctctgagtctcactttccttatctag
aaaacggaggtatctttacttccttcgtaagactgatgacaaggaaattatctgtgcattttgaaaccacttaagccttgtacacgttttatttctggga
tcgccctggtagggcttcagaaaaataaaaaggaggtccctgagaaaaggctgggtaccgtacatctgaggtcaaccctctctggtcccaaggatggcct
gggctgttccgccccgtggctccccaggggcaaagccatgaggatccgggtgagagcccagtgctggacgagcccggggcccaggggtcccggccgaaat
ccctgctgtctttcaggtcaaacgtcataatccccgaaccccagaaaggccgaaaggcaaggcaaccctgaaagacgacgaagtcaacctcagggcgcag
gagagggagggccagtgtgctgccgacgagggaggctggagccgcggggacgaggcgccccatacagcggcaagagggtggagggcaggagctcgccatc
ctgggtgaaagcggggcccagcgaaggggcccggccacaggaatctcggttccaccccgctactcccggctgtgactccagtttcgtccccagccgccgg
gaccgccccctcgccccgcccccagcgggcactcaggccgtaccactgtgccttcatgggggtggagatagatcgtgggctagtcctgccgaggagagag
gggttcttcctcaaaaaatatgattatgtatagtattcgcatgattctagttaacttgtttcccttctgcctgctcggaccctctacctgccctacgaag
ggggcggagtgcgttcctgcctccccctgctcttccgcgtttggtgcgcgcctgcgcggtgcgtaggcggcggagcgtacttaagcttcgacgcaggagg
CGGGGCTGCTCAGTCCTCCAGGCGTCGGTACTCAGCGGTGTTGGAACTTCGTTG
CTTGCTTGCCTGTGCGCGCGTGCGCGGACATGGCCTCAAACGgtag

Go back

Methods

As we said in the summary, we have studied the hybrid protein using the information provided by each original gene separately, because of the fact that we found less information of TLS/CHOP than those ones.

The function information of the three proteins (hybrid and the two originals) has been consulted in diverse literary sources that are mentioned in the References section.

All the information obtained to characterise the genomic structure of the hybrid gene involved in the myoxid liposarcoma (TLS-CHOP) was provided by UCSC genome browser, NCBI and Ensembl. We also used some blast tools to analyze the genomic comparations between the hybrid gene and the FUS and DDIT3 genes, such as Blastn and Blastp, and also the ClustalW software to make more specific study about the aminoacid conservation between the proteins. It is important to say that in order to make the ClustalW between the TLS/CHOP protein and the FUS protein we used the protein sequence of the first transcript of FUS, because it conserves the original exons and is not affected by the alternative splicing mechanism.

In order to study the gene conservation and know if FUS and CHOP are highly conserved among the species we have search for information in Ensembl, NCBI and UCSC. At the first genome browser mentioned, Ensembl, it is possible to find a lot of information of ortholog genes for DDIT3 but not for FUS. It's because of that we have looked for more information in other genome browsers. So the genes that don't have an ensembl ID are the homoleg genes that we have found in NCBI and UCSC.

To analyze the expression in diferent tissues of TLS-CHOP we have consulted the expression of FUS in UCSC. We also looked for the expression of DDIT3 just to compare it with the expression of FUS. In that genome browser it's possible to access to the Gene Sorter where some tables that show the expression are found. This expression were represented by a gradiation of colours. In order to make it more clear we gave an upper brigthness to the colours.

On the study of the promoter characterization section, we first had to found the 5' upstream region from the TSS (Transcription Starting Site) of the FUS gene using the information provided by the USCS Genome Browser Database. We looked for the DNA sequence of the first 1000 nucleotides upstream region from the TSS of the FUS gene and the 100 nucleotides downstream region from it.

After that we programmed an algorism based in Perl language to be able to read a sequence (we use it with our promoter sequence) with the help of different TF matrix and find which of those transcription factors bind to the sequence. This program is based in four parts:

• The first part consists in reading the matrix for each TF and save them as a hash of vectors.
• The second part consists in transform this hash into a weight matrix, considering the nucleotide proportions in the sequence we have introduced.
• The third part consists in calculate the best punctuation of the matrix in every possible binding position of the transcription factor along the sequence.
• The final part consists in calculate the p value for each transcription factor, in where we consider the probability we have to find that TF in that position by random.

We have presented the program algorism we used clicking here.

And the results of this program are showed by clicking here.

In addition, we have used the PROMO software. In the SelectSpecies section we only chase human factors and sites in order to reduce the number of TF obtained. However, we obtained so many TF so we decided to decrease the dissimilarity rate into 10 and give special attention only to those transcription factors that were analyzed with the Perl algorism.

Go back

Discusion

To begin this section we are going to explain what we think it could happen with the hybrid protein TLS/CHOP. On the one hand, the FUS part doesn't conserve the RNA-binding region from the original protein so this is not able to stabilize the cellular mRNA when it's transcribed. This happens, as we said before, because this domain it's replaced by the CHOP DNA-binding and dimerization domains. As a result the FUS part of the hybrid protein would acquire a new function consisting in binding to another gene promoters using the CHOP DNA domain, causing the expression of oncogenic proteins which can help to develop the myoxid liposarcoma. On the other hand, in spite of the whole CHOP protein is conserved in TLS/CHOP, the fact that it is bound to the FUS protein and has translated it's second untranslated-exon would cause that the protein doesn't fold correctly and that would be the reason why it becomes to an afunctional transcription factor, incapable to bind it's normal target genes, which include the adipocyte growth and diferenttiation genes. This second fact also contributes to develope the disease.

We think that FUS has been discovered recently because as we said in the first section, it was named TLS when they found that it was involved in myoxid liposarcoma. Furthermore, there are other evidences like the fact that there are more ortholog genes for DDIT3 than for FUS. Actually in Ensembl database you will find lots of ortholog genes for CHOP but zero ortholog genes for FUS. We would affirmate that both genes are highly conservated in the mammal family (there is an exception on the conservation of the DDIT3 in the cat, only 51%), so it means that both play an important role in the organism. We can say that DDIT3 has an important function because not only controls the differentiation of an specific tissue, but also acts as a tumour supressor gene.

As we can see in the tables from expression caracteritzation, FUS is highly expressed in thymus, thyroid, CD4+ T cells and testis. It has a higher expression than the normal but not so intense in bood and lung. We can also apreciate that it has a low expression in lots of difernet organs like kidney, liver, heart, pancreas, skin, stomach and brain. The results of the expression of DDIT3 show that it also has a wide variety of expression on different tissues. DDIT3 is highly expressed in testis, thymus, lymph node,stomach, lung, trachea and bone marrow. It is shown that is has a low expression in ovary, heart, liver, kidney, blood and pancreas. We think that it could be related to the fact that these organs have a low composition of fat tissue. As CHOP is involved in cell division control and FUS stabilizes the mRNA it is natural that they have such an extensive expression.

We have used different strategies to find the TF that bind to the TLS-CHOP promoter. This results aren't quite reliable because the two methods used don't match exactly in all the FT. There is only one transcription factor that matches more or less in both methods and is suceptible to bind the promoter: NF-AT1. In addition, looking on the literature we found that there are some studies that demonstrated that there are three TFs that can bind the TLS-CHOP promoter. They are AP-2, GFC and Sp-1, so no one is NF-AT1. In the Perl program we have analyzed a variant of AP-2, the AP-1, which has a p value of 0,24, so we could consider this family of TF also good candidates.

Go back

References

1. Crozac,A.,Aman,P.,Mandahi,N.,Ron,D. Fusion of CHOP to a novel RNA-binding protein in human myoxid liposarcoma. Nature363(6430):640-4.17 Jun 1993


2. Thelin-Jarnum,S.,Lassen,C.,Panagopoulos,I.,Mandahl,N.,Aman,P. Identification of genes differentially expressed in TLS-CHOP carrying myoxid liposarcomas. Int J Cancer83(1):30-3.24 Sept 1999


3. Rabbitts,TH.,Foster,A.,Larsson,R.,Nathan,P. Fusion of the dominant negative transcription regulator CHOP with a novel gene FUS by translocation t(12;16) in malignant liposarcoma. Nat Genet.4(2):175-80.Jun 1993


4. Atlas of Genetics and Cytogenetics in Oncology and Hematology


5. Online Book: Cancer Medicine. 6th ed. Kufe, Donald W.; Pollock, Raphael E.; Weichselbaum, Ralph R.; Bast, Robert C., Jr.; Gansler, Ted S.; Holland, James F.; Frei III, Emil, editors. Hamilton (Canada): BC Decker Inc; c2003


6. OMIM data base: *137070 FUSION, DERIVES FROM 12-16 TRANSLOCATION, MALIGNANT LIPOSARCOMA;FUS


7. Ensembl Genome Browser Database


8. UCSC Genome Browser Database


9. NCBI Database


10. ClustalW


11. PROMO Software

Go back