Chromatine Isolatie door RNA-zuivering (Chirp)

Biology
JoVE Journal
Biology
AccessviaTrial
 

Summary

Chirp is een nieuwe en snelle techniek om genomische bindingsplaatsen van niet-coderende RNA's lang (lncRNAs) in kaart te brengen. De methode maakt gebruik van de specificiteit van anti-sense oligonucleotiden tegels aan de lijst van lncRNA-gebonden genomische locaties mogelijk te maken.

Cite this Article

Copy Citation | Download Citations | Reprints and Permissions

Chu, C., Quinn, J., Chang, H. Y. Chromatin Isolation by RNA Purification (ChIRP). J. Vis. Exp. (61), e3912, doi:10.3791/3912 (2012).

Please note that all translations are automatically generated.

Click here for the english version. For other languages click here.

Abstract

Lang niet-coderende RNA's zijn belangrijke regulatoren van chromatine staten voor belangrijke biologische processen, zoals de dosering compensatie, imprinting, en ontwikkelingsstoornissen genexpressie 1,2,3,4,5,6,7. De recente ontdekking van duizenden lncRNAs in samenwerking met specifieke chromatine modificatie complexen, zoals Polycomb Repressieve Complex 2 (PRC2) die bemiddelt histon H3 lysine 27 trimethylering (H3K27me3), brede rollen voor tal van lncRNAs in het beheren van chromatine staten in een gen-specifieke suggereert mode 8,9. Terwijl sommige lncRNAs worden verondersteld om te werken in cis op naburige genen, andere lncRNAs werken in trans naar de verte gelegen genen reguleren. Bijvoorbeeld, Drosophila lncRNAs roX1 en roX2 binden tal van regio's op het X-chromosoom van mannelijke cellen, en zijn van cruciaal belang voor de dosering compensatie 10,11. Echter, de exacte locaties van hun bindingsplaatsen niet bekend met een hoge resolutie. Ook de menselijke lncRNA Hotair kan PRC2 bezetting van invloed zijn op hundreds van genen genoom-brede 3,12,13, maar hoe specificiteit wordt bereikt, is onduidelijk. LncRNAs kan ook dienen als modulaire steigers voor de samenstelling van verschillende eiwitcomplexen te werven. De klassieke trans-werkende RNA-steiger is de TERC RNA dat dient als de sjabloon en steiger voor het telomerase complex 14; Hotair kan ook dienen als een platform voor PRC2 en een H3K4 demethylase complex 13.

Eerdere studies mapping RNA bezettingsgraad van chromatine gebleken aanzienlijke inzichten 15,16, maar alleen op een locus met genen voor een. De bezettingsgraad sites van de meeste lncRNAs zijn niet bekend, en de rollen van lncRNAs in chromatine verordening zijn vooral afgeleid uit de indirecte effecten van de lncRNA verstoring. Net zoals chromatine immunoprecipitatie gevolgd door middel van microarray of diep-sequencing (chip-chip of chip-seq, respectievelijk) is sterk ons ​​begrip van eiwit-DNA interacties op een genomische schaal verbeterd, hier hebben we illustreren een recentgevoerd gepubliceerde strategie op de lange RNA genoom-brede bezetting in kaart te brengen met een hoge resolutie 17. Deze methode, chromatine Isolatie van RNA-zuivering (Chirp) (figuur 1), is gebaseerd op affiniteit vangst van de beoogde lncRNA: chromatine complex door tegels antisense-oligo's, die vervolgens een kaart van genomische bindingsplaatsen genereert met een resolutie van enkele honderden bases met hoge gevoeligheid en weinig achtergrond. Chirp is voor vele lncRNAs omdat het ontwerp van affiniteit-probes is eenvoudig aangezien de RNA sequentie en vereist geen kennis van de structuur van de RNA of functionele domeinen.

Protocol

1. Probe-ontwerp

Ontwerp anti-sense DNA-tegels probes voor selectieve ophalen van RNA-doelwit door Chirp.

  1. Ontwerp anti-sense oligo-probes met behulp van de online sonde ontwerper bij singlemoleculefish.com 18.
  2. Gebruik deze parameters: aantal sondes = 1 probe / 100 bp van RNA lengte, 2) Target GC% = 45, 3) Oligonucleotide lengte = 20; 4) Afstand lengte = 60-80. Breek RNA in segmenten als er te lang voor de ontwerper. Laat gebieden herhaalt of uitgebreide homologie.
  3. Bestel anti-sense DNA-probes met BiotinTEG op 3-prime end.
  4. Label probes volgens de posities langs het RNA. Scheid ze in twee zwembaden, zodat de "even" pool bevat alle sondes nummering 2, 4, 6, etc. en de "oneven" pool bevat sondes nummering 1, 3, 5, enz. verdunnen pool van de sondes op 100 uM concentratie en Bewaar bij -20 ° C.
  5. Alle experimenten worden uitgevoerd metbeide groepen, die als interne controle elkaar. Real RNA-afhankelijke signaal aanwezig zou zijn van beide zwembaden, terwijl de probe-specifieke geluiden zou uniek zijn voor elk zwembad. Dit geldt zowel voor Chirp-qPCR en Chirp-seq.

2. Harvest Cellen

Verzamelen cellen die worden gebruikt voor Chirp experiment.

  1. Groei cellen in weefselkweek platen of kolven tot confluentie. Spoel met fosfaat gebufferde zoutoplossing (PBS) voor eens en trypsinize. Quench trypsine met> 2x volume van de media, pipet op en neer naar de cellen los te maken en in enkele celsuspensie mengen. Breng alle media en geresuspendeerde cellen in 50 ml Falcon buizen. 20 miljoen cellen zijn gewoonlijk voldoende voor een Chirp monster.
  2. Spin cellen 800RCF 4 min. Zuig media en hersuspenderen 40 miljoen cellen in 40 ml PBS, combineren buizen indien nodig. Spin cellen 800RCF 4 min. Giet PBS, zorgvuldig te zuigen op een hoek de resterende vloeistof.
  3. 3. Cross-link Cellen en Collect celpellet

    Crosslink verzamelde cellen met glutaaraldehyde op RNA-chromatine interacties te bewaren en voor te bereiden celpellet.

    1. Voer alle stappen bij kamertemperatuur.
    2. Bereid 1% glutaaraldehyde bij kamertemperatuur PBS. Doe in 10 ml per 10 miljoen cellen (0,4 ml 25% glutaaraldehyde voorraad + 9,6 ml PBS). Glutaraldehyde moet worden gebruikt fris.
    3. Tik onderkant van Falcon buizen aan pellets te verwijderen. Resuspendeer celpellet in 1% glutaaraldehyde, te beginnen met een klein volume te brokken te vermijden, dan herladen tot vol volume. Omkeren om te mengen. Crosslink 10 min. bij kamertemperatuur op een eind-tot-eind schudapparaat of rotator.
    4. Doof het verknopingsreactie met 1/10e volume van 1,25 M glycine bij kamertemperatuur gedurende 5 minuten.
    5. Spin op 2000RCF gedurende 5 minuten. Zuig supernatant en was pellet met 20 ml koud PBS een keer, het spinnen op 2000RCF gedurende 5 minuten.
    6. Zuig en resuspendeer de wverast, cross-linked pellet met 1 ml gekoelde PBS per 20 miljoen cellen. Breng elke ml aan een Eppendorf buis en draaien op 2000RCF gedurende 3 minuten bij 4 C. Verwijder zoveel mogelijk met PBS pipetpunt nauwkeurig.
    7. De cel pellets in vloeibare stikstof en opslaan Flash-invriezen bij -80 ° C voor onbepaalde tijd.

    4. Cell Lysis

    Lyseren verknoopt cellen cellysaat bereiden.

    1. Ontdooi bevroren celpellets bij kamertemperatuur. Tik op moeilijk te verjagen en meng de cel pellet. Draai beneden de pellet bij 2000RCF gedurende 3 minuten bij 4 ° C. Gebruik een scherpe 10 pl pipetpunt om de resterende PBS te verwijderen.
    2. Op een elektronische weegschaal (nauwkeurig tot op 1 mg) tarra de massa van een lege eppendorfbuisje (onze buizen wegen 1,060 gram zeer consequent). Weeg elke pellet en opnemen zijn gewicht. Een volledige 15 cm schotel van verknoopt HeLa cellen weegt meestal 100 mg.
    3. Supplement Lysis Buffer (10x de massa van de pellet, bijvoorbeeld 1 ml voor 100 mg) met FRESh Proteaseremmer, PMSF en Superase-in (zie bijgevoegde buffer lijst). Meng goed.
    4. Voeg 10X volume aangevuld Lysis Buffer aan elke buis en resuspendeer de pellet. Voor kleine pellets <25 mg, opnieuw in suspensie in 250 ul aangevuld lysis buffer. Schorsing moet glad zijn. Zo niet, dan verdeelt de schorsing in 500 pi porties en gebruik een gemotoriseerde pellet mixer te breken bosjes. Ga onmiddellijk naar ultrasoonapparaat.

    5. Sonicatie

    Shear DNA door sonicatie verknoopt cellysaten.

    1. Bewerk met ultrasone trillingen cellysaat in Bioruptor in 15 ml Falcon buizen. Gebruik <1,5 ml lysaat in elke buis, en een snellere sonicatie niet meer dan twee buizen ultrasone trillingen per keer.
    2. Ultrasone trillingen in een 4 ° C waterbad hoogste stand van 30 seconden, 45 seconden OFF puls intervallen. Controleer lysaat elke 30 minuten. Verder sonicatie tot het cellysaat niet meer troebel. Dit kan slechts 30 minuten en liefst 4 uur. Het aantalvan buizen, zal het monster volume, het bad temperatuur, en de periode van ultrasoonapparaat tijd invloed hebben op hoe lang het proces duurt. Buizen zal waarschijnlijk ultrasone trillingen met verschillende snelheden, dus bundelen ze samen om de 30 min en herverdelen in de originele buizen om homogeniteit te garanderen. Let op: glutaaraldehyde-verknoopt cellen nemen aanzienlijk langer aan ultrasone trillingen dan de formaldehyde-equivalenten.
    3. Bij het lysaat duidelijk wordt, over te dragen 5 pi lysaat een nieuwe Eppendorf buis. Voeg 90 uL DNA Protease K (PK) Buffer (zie buffer lijst) en 5 ul PK. Vortex te mengen en even spin down. Incubeer 45 min bij 50 ° C.
    4. Pak DNA met PCR Qiagen zuivering kit. Elueer DNA in 30 pi Qiagen elutiebuffer (EB) en controleer DNA grootte van 1% agarose gel. Als grootste deel van de DNA-uitstrijkje is 100-500 bp, sonicatie is voltooid. Zo niet, dan blijven ultrasone trillingen.
    5. Centrifugeer gesoniceerd monsters 16100RCF gedurende 10 min bij 4 ° C. Combineer supernatanten, aliquot in 1 ml monsters en flash-vries in vloeibare nitrogen. Bewaren bij -80 ° C.

    6. Chirp

    Hybridiseren gebiotinyleerde DNA probes om RNA te isoleren gebonden chromatine.

    1. Ontdooien buizen van chromatine bij kamertemperatuur.
    2. Bereid hybridisatiebuffer (zie de lijst van buffers, voor te bereiden 2 ml per ml van het chromatine). Vortex om te mengen.
    3. Voor een typische Chirp monster door 1 ml van lysaat, verwijder dan 10 pi voor RNA-INPUT en 10 ul voor DNA-INPUT en plaats in Eppendorf buizen. Houd op het ijs tot verder gebruik.
    4. Breng 1 ml chromatine tot 15 ml Falcon buis. Voeg 2 mL hybridisatiebuffer aan elke buis. Voor het totale volume <1,5 ml, gebruik Eppendorf buizen.
    5. Ontdooi sondes bij kamertemperatuur. NanoDrop sondes te bedragen controleren of u niet gebruikt in een lange tijd (100 uM sondes moeten spec ~ 500-600 ng / ul met behulp van enkele streng DNA-instelling). Voeg het juiste volume van de sondes aan specifieke buizen (100 pmol sonde per 1 ml chromatine, 1 pl van 100 pmol / ul sonde per 1 ml chromatine).Meng goed. Incuberen bij 37 ° C gedurende 4 uur onder schudden.
    6. Met 20 min resterende hybridisatie bereiden C1 magnetische kralen (bewaard bij 4 ° C). Gebruik 100 ul per 100 pmol van de sondes. Wassen met 1 ml unsupplemented Lysis Buffer drie keer, met behulp van de DynaMag-2 magneetstrip te scheiden kralen uit buffer.
    7. Resuspendeer kralen in oorspronkelijke volume van Lysis Buffer, te vullen met verse PMSF, PI en Superase-in. Na 4 uur hybridisatie reactie voltooid worden 100 ul kralen aan elk buisje. Meng goed. Incuberen bij 37 ° C gedurende 30 min onder schudden.
    8. Bereid Was Buffer (5 ml per monster). Vortex om te mengen. Voorverwarmen tot 37 ° C. Voeg PMSF voor gebruik.
    9. Was kralen met 1 ml wasbuffer vijf keer. Op de eerste wasbeurt, gebruik DynaMag-15 magneetstrip om aparte kralen, giet, en resuspendeer in 1 ml wasbuffer. Breng volume 1,5 mL Eppendorf buis. Incuberen bij 37 ° C met schudden gedurende 5 minuten.
    10. Bij volgende wasbeurten, spin down elke buis op een minicentrifuge, Stel monster op DynaMag-2 magnetische strip gedurende 1 minuut. Giet monster, veegt alle druppels met een Kimwipe, opnieuw in suspensie in 1 ml wasbuffer. Incuberen bij 37 ° C met schudden gedurende 5 minuten. Herhaal dit voor in totaal vijf wasbeurten.
    11. Eindelijk wassen, goed mengen van de kralen. Verwijder 100 ul en zet ze opzij voor RNA isolatie. Reserve 900 ul DNA voor fractie. Plaats alle buizen op DynaMag-2 magnetische strip en verwijder wasbuffer. Draai alle buizen naar beneden kort, leg ze op de magneet strip. Volledig verwijderen van de laatste beetje wasbuffer met een scherp 10 pl pipetpunt.

    7. RNA isolatie

    Uittreksel RNA fractie van Chirp monsters te kwantificeren door qRT-PCR.

    1. Neem 100 ul kraal monsters en 10 pl RNA INPUT monster. Voeg 85 uL RNA PK Buffer pH 7,0 tot INPUT RNA. Resuspendeer kralen in 95 ul RNA PK Buffer pH 7,0. Voeg 5 ul Proteinease K en incuberen bij 50 ° C gedurende 45 min met end-to-end schudden.
    2. In het kort spin down alle buizen enkook monsters gedurende 10 min op verwarmingsblok bij 95 ° C.
    3. Chill monsters op ijs, krachtig toe te voegen 500 pL TRIzol, vortex gedurende 10 sec. Incuberen bij kamertemperatuur gedurende 10 minuten. Bewaren bij -80 ° C of ga naar stap 4.
    4. Voeg 100 ul chloroform TRIzol behandelde monsters. Vortex krachtig gedurende 10 sec. Spin op 16100RCF op een benchtop centrifuge gedurende 15 minuten bij 4 ° C.
    5. Verwijder ~ 400 pL waterige supernatans, het vermijden van organische en interface.
    6. Voeg 600 ul (1,5 volume) 100% ethanol en meng goed. Spin monster door MIRNeasy mini-kolommen. Was 1x met RWT (MIRNeasy mini-kit), 2x met RPE per protocol van de fabrikant. Elueren met 30 pi nuclease vrij H2O (NFH 2 O).
    7. Behandel de RNA eluaat met DNA-free per protocol van de fabrikant. Nadat de reactie is voltooid, verwarmen van het monster gedurende 15 minuten bij 65 ° C volledig te inactiveren resterende DNase.
    8. Gebruik 1 pl RNA isolaat per goed voor qRT-PCR analyse lncRNA te bevestigenophalen. GAPDH wordt vaak gebruikt als een negatieve controle.

    8. DNA isolatie

    Uittreksel DNA-fractie van Chirp monsters te herkennen aan sequencing of door qPCR kwantificeren.

    1. Bereid DNA elutiebuffer (zie buffer lijst), 150 ul per monster, met inbegrip van DNA INPUT.
    2. Voeg 1 0μL RNase A (10 mg / ml) en 10 pi RNase H (10 U / pl) per ml DNA elutiebuffer en vortex te mengen.
    3. Resuspenderen elk monster korrels in 150 ul DNA elutiebuffer met RNAses. (Hersuspenderen DNA INPUT in 140 pi) Incubeer bij 37 ° C gedurende 30 min onder schudden.
    4. Aparte kralen en supernatant op DynaMag-2 magnetische strip. Verwijder supernatant en gelabeld aan buizen.
    5. Bereid een tweede hoeveelheid van DNA elutiebuffer met 10 pi RNase A (10 mg / ml) en RNaseH (10 U / ul) precies zoals gedaan 8.2). Voeg 150 ul aan elk monster (met inbegrip van DNA INPUT), incubeer en verwijder supernatant. Verzamel alle supernatant (schouderd te zijn ~ 300 pi).
    6. Voeg 15 ul PK aan elk monster. Incubeer bij 50 ° C gedurende 45 min met schudden.
    7. Pre-spin down geel phase-lock gel buizen (5PRIME). Overdracht DNA-monsters aan phase-lock gel buizen, en voeg 300 ul PhOH: Chloroform: Isoamyl per monster. Schudden gedurende 10 minuten en spin neer op een tafelmodel centrifuge bij 16100RCF gedurende 5 min bij 4 ° C. Neem waterige vanaf de top (~ 300 pi). Voeg 3 ul GlycoBlue, 30 pi NaOAc en 900 ul 100% EtOH. Meng goed en bewaar bij -20 ° C gedurende de nacht.
    8. Spin monsters 16100RCF gedurende 30 minuten bij 4 ° C.
    9. Giet supernatant voorzichtig. Voeg 1 ml 70% EtOH en vortex te mengen. Spin op 16100RCF gedurende 5 minuten. Verwijder de bovenstaande vloeistof met een pipet. Air drogen 1min. Resuspenderen in 30 pi EB.
    10. DNA-monsters zijn klaar voor analyse door qPCR of het bereiden van high-throughput sequencing bibliotheken per Illumina-protocol.

    10. Representatieve resultaten

    Figuur 1 Figuur 2 verrijking van de menselijke telomerase RNA (TERC) van HeLa cellen over GAPDH een overvloed van cellulaire RNA dat als een negatieve controle. De meerderheid van TERC RNA's (~ 88%) in de cel aanwezig zijn naar beneden getrokken door het uitvoeren van tjilpen, terwijl slechts 0,46% van de GAPDH RNA werd teruggehaald, aantonen dat er een verrijking factor van ~ 200 maal. Aspecifieke probes zoals probes gericht LacZ RNA, die niet wordt uitgedrukt in zoogdiercellen (figuur 2) kan worden gebruikt als additionele negatieve controles.

    DNA-regio's verwacht dat de doelgroep lncRNA te binden zijn meestal verrijkt over negatieve regio's, gemeten door qPCR. Figuur 3 toont qPCR validatie van vier Hotair-gebonden sites in primaire humane voorhuid fibroblasten dat we bepaald door het uitvoeren van Chirp-seq in dezelfde cel lijn, terwijl TERC en GAPDH DNA websites seRVE als negatieve controle regio's. Zowel de "even" en "oneven" probe sets leverde vergelijkbare verrijking van de verwachte Hotair-gebonden sites over negatieve regio's, een kenmerk van echte lncRNA-bindende sites.

    High-throughput sequencing van Chirp verrijkte DNA levert een globale kaart van lncRNA-bindingsplaatsen. De Drosophila lncRNA roX2 bekend is de X-chromosoom interactie op een wijze die nodig is voor dosering vergoeding. Figuur 4 toont roX2 bindingsprofiel over een gedeelte van het X-chromosoom. Zowel de "even" en "oneven" monsters zijn gesequenced en hun unieke geluiden zijn geëlimineerd om een ​​track van overlappende signalen te produceren. Elke "piek" geeft hier een sterke site van roX2 bindend. De volledige baan en de lijst van roX2 genen zijn beschreven in Chu et al.. 2011 17.

    Figuur 1.
    Figuur 1. Stroomdiagram van de Chirp proceduredure voorschrijft. Chromatine wordt verknoopt aan lncRNA: eiwitadducten in vivo Biotinylated tegelwerk probes gehybridiseerd aan lncRNA richten, en chromatine complexen worden gezuiverd met behulp van magnetische streptavidine kralen, gevolgd door strenge wasbeurten.. We elueren lncRNA gebonden DNA of eiwitten met een cocktail van RNase A en H. Een vermoedelijke lncRNA bindende sequentie wordt schematisch voorgesteld in de kleur oranje. Eerder verschenen in Chu et al.. 2011. 17

    Figuur 2.
    Figuur 2. Chirp verrijkt voor menselijke TERC RNA. TERC-asDNA sondes op te halen ~ 88% van de cellulaire TERC RNA en niet op te sporen GAPDH. LacZ-asDNA probes worden gebruikt als negatieve controles en ophalen niet RNAs. De gemiddelde + sd worden weergegeven. Eerder verschenen in Chu et al.. 2011. 17

    Figuur 3.
    Figuur 3. Hotair Chirp-qPCR in het primair mens voorEskin fibroblasten. NFKBIA, HOXD3-4, SERINC5 en ABCA2 zijn regio's die samenwerken met Hotair. TERC en GAPDH dienden als negatieve controles. De gemiddelde + sd worden weergegeven. Eerder verschenen in Chu et al.. 2011. 17

    Figuur 4.
    Figuur 4. Chirp volgende gegevens roX2 RNA in Sl2 Drosophila cellen. "Even" en "oneven" werden afzonderlijk gesequenced, hun gegevens samen te voegen tot enige gemeenschappelijke pieken weerspiegelen in beide. De samengevoegde traject wordt weergegeven. Eerder verschenen in Chu et al.. 2011. 17

Discussion

Hier hebben we beschreven Chirp-seq, een methode in kaart brengen van in-vivo lncRNA bindingsplaatsen genoom-breed. De belangrijkste parameters voor succes zijn de splitsing poelen van tiling oligonucleotideprobes en glutaaraldehyde verknoping. Het ontwerp van affiniteit-probes is eenvoudig, gezien de RNA-sequentie en vereist geen voorkennis van de structuur van het RNA's of functionele domeinen. Ons succes met roX2, TERC en Hotair - drie in plaats van verschillende RNA's in twee soorten - suggereert dat Chirp-seq waarschijnlijk generaliseerbaar veel lncRNAs. Zoals met alle experimenten, zijn zorg en de juiste controles nodig zijn om de resultaten te interpreteren. Verschillende lncRNA kan titratie van omstandigheden en oordeelkundige verandering van omstandigheden, zoals selectie van verschillende affiniteit probes of verknopingsmiddelen vereisen, kunnen belichten verschillende aspecten van RNA-interacties chromatine. Net als ChIP-seq, niet alle bindende gebeurtenissen zijn noodzakelijkerwijs functioneel, en bijkomende studies nodig zijn om de biologische gevolgen van RNA-o vast te stellenccupancy op chromatine. Toch voorzien we vele interessante toepassingen van deze technologie voor onderzoekers van andere chromatine-geassocieerde lncRNAs, die nu in de duizenden 8,9. Net zoals ChIP-seq heeft de deur geopend voor genoom-brede verkenningen van DNA-eiwit interacties, kan Chirp-seq studies van de "RNA interactoom" onthullen vele nieuwe mogelijkheden van de biologie.

Disclosures

C. Chu en HY Chang worden genoemd als uitvinders van een octrooiaanvraag op basis van deze methode.

Acknowledgments

Wij danken T. Hung, MC. Tsai, O. Manor, E. Segal, M. Kuroda, T. Swigut, en ik Shestopalov voor discussies. Ondersteund door het Agentschap van Wetenschap, Technologie en Onderzoek van Singapore (CC), NIH R01-CA118750 en R01-HG004361 (HYC), en Californië Instituut voor Regeneratieve Geneeskunde (HYC). HYC is een Early Career wetenschapper van het Howard Hughes Medical Institute.

Materials

Name Company Catalog Number Comments
Buffer List:
Dissolve a pellet of complete protease inhibitor in 1 ml water as 50x stock. Make 100 mM PMSF in isopropanol (100x stock). Superase-in is used as 200x stock. Store all at -20 °C.
Lysis Buffer:
50 mM Tris-Cl pH 7.0
10 mM EDTA
1% SDS
Always add PMSF, P.I. and Superase-in fresh before use except when washing beads
Proteinase K Buffer (for DNA)
100 mM NaCl
10 mM TrisCl pH 8.0 (For RNA use pH 7.0)
1 mM EDTA
0.5% SDS
Add 5% by volume Proteainse K (Ambion AM2546 20 mg/ml) fresh before use
Hybridization Buffer
750 mM NaCl
1% SDS
50 mM Tris-Cl pH 7.0
1 mM EDTA
15% formamide (store in the dark at 4 °C)
Always add PMSF, P.I. and Superase-in fresh before use
Wash Buffer
2x NaCl and Sodium citrate (SSC) (diluted from 20x SSC Invitrogen stock)
0.5% SDS
Always add PMSF fresh before use
DNA elution Buffer
50 mM NaHCO3
1% SDS
Table of specific reagents and equipment:
Glutaraldehyde (EM grade) Sigma-Aldrich G5882-10x10ml
Motorized pellet mixer VWR international V8185-904
Protease inhibitor Roche Group 11873580001
PMSF Sigma-Aldrich 78830
Superase-in Ambion AM2696
Bioruptor Diagenode UCD-200
Falcon tubes (for sonication) Corning 430790
Proteinase K Ambion AM2546
PCR purification kit Qiagen 28106
C-1 magnetic beads Invitrogen 65002
PMSF Sigma-Aldrich P7626-25G
DynaMag-15 magnet Invitrogen 123-01D
DynaMag-2 magnet Invitrogen 123-21D
MIRNeasy mini kit Qiagen 217004
Rnase H Epicentre Biotechnologies R0601K
Rnase A Sigma-Aldrich R4875-100MG
Phase Lock Gel Heavy 5 PRIME 2302810
Trizol Invitrogen 15596-018
Phenol:chloroform:Isoamyl Invitrogen 15593-031
Chloroform Ricca RSOC0020-1C
GlycoBlue Ambion AM9515
Glycine JT Baker 4057-06
PBS, pH 7.4 Invitrogen 10010-049
Elution Buffer (EB) Qiagen 19086
20x SSC Invitrogen 15557-036
10% SDS Invitrogen 15553-027
DNA-free Ambion AM1906
Buffer kit Ambion AM9010
Formamide Invitrogen 15515-026

DOWNLOAD MATERIALS LIST

References

  1. Koziol, M. J., Rinn, J. L. RNA traffic control of chromatin complexes. Curr. Opin. Genet. Dev. 20, 142-148 (2010).
  2. Mercer, T. R., Dinger, M. E., Mattick, J. S. Long non-coding RNAs: insights into functions. Nat. Rev. Genet. 10, 155-159 (2009).
  3. Rinn, J. L. Functional demarcation of active and silent chromatin domains in human HOX loci by noncoding RNAs. Cell. 129, 1311-1323 (2007).
  4. Zhao, J., Sun, B. K., Erwin, J. A., Song, J. J., Lee, J. T. Polycomb proteins targeted by a short repeat RNA to the mouse X chromosome. Science. 322, 750-756 (2008).
  5. Kelley, R. L. Epigenetic spreading of the Drosophila dosage compensation complex from roX RNA genes into flanking chromatin. Cell. 98, 513-522 (1999).
  6. Pandey, R. R. Kcnq1ot1 antisense noncoding RNA mediates lineage-specific transcriptional silencing through chromatin-level regulation. Mol. Cell. 32, 232-246 (2008).
  7. Wang, K. C. A long noncoding RNA maintains active chromatin to coordinate homeotic gene expression. Nature. 472, 120-124 (2011).
  8. Khalil, A. M. Many human large intergenic noncoding RNAs associate with chromatin-modifying complexes and affect gene expression. Proc. Natl. Acad. Sci. U.S.A. 106, 11667-11672 (2009).
  9. Zhao, J. Genome-wide identification of polycomb-associated RNAs by RIP-seq. Mol. Cell. 40, 939-953 (2010).
  10. Meller, V. H., Wu, K. H., Roman, G., Kuroda, M. I., Davis, R. L. roX1 RNA paints the X chromosome of male Drosophila and is regulated by the dosage compensation system. Cell. 88, 445-457 (1997).
  11. Franke, A., Baker, B. S. The rox1 and rox2 RNAs are essential components of the compensasome, which mediates dosage compensation in Drosophila. Mol. Cell. 4, 117-122 (1999).
  12. Gupta, R. A. Long non-coding RNA HOTAIR reprograms chromatin state to promote cancer metastasis. Nature. 464, 1071-1076 (2010).
  13. Tsai, M. C. Long noncoding RNA as modular scaffold of histone modification complexes. Science. 329, 689-693 (2010).
  14. Zappulla, D. C., Cech, T. R. RNA as a flexible scaffold for proteins: yeast telomerase and beyond. Cold Spring Harb. Symp. Quant. Biol.. 71, 217-224 (2006).
  15. Nagano, T. The Air noncoding RNA epigenetically silences transcription by targeting G9a to chromatin. Science. 322, 1717-1720 (2008).
  16. Carter, D., Chakalova, L., Osborne, C. S., Dai, Y. F., Fraser, P. Long-range chromatin regulatory interactions in vivo. Nature. 32, 623-626 (2002).
  17. Chu, C., Qu, K., Zhong, F. L., Artandi, S. E., Chang, H. Y. Genomic Maps of Long Noncoding RNA Occupancy Reveal Principles of RNA-Chromatin Interactions. Mol. Cell. (2011).
  18. Raj, A., van den Bogaard, P., Rifkin, S. A., van Oudenaarden, A., Tyagi, S. Imaging individual mRNA molecules using multiple singly labeled probes. Nat. Methods. 5, 877-879 (2008).

Comments

84 Comments

  1. hi,i has a great interest on CHIRP and your works are highly appreciated,but i want to know that where can i find the buffer list?thank you

    Reply
    Posted by: tang n.
    May 5, 2012 - 4:16 AM
  2. Hi there, thanks for the interest in our paper. As you pointed out the original article lacked the buffer list, which we have uploaded a while ago. Hope it helps!

    Reply
    Posted by: Ci C.
    December 18, 2012 - 5:20 PM
  3. Hello, I'm trying to perform ChIRP experiments and your protocol is very helpful to me.

    However, I have a question about the final concentration of biotinylated probes and the volume of the beads.
    My probes are 100µM so 1µL corresponds to 100pmol. I have 50 probes and my question is : should I use 100µL of magnetic C1 beads per 100pmol of probe, so in my case, 50 probes x 100µL = 5mL of beads ?

    Thanks a lot for the answer ?

    Reply
    Posted by: Sylvain F.
    October 31, 2012 - 4:47 AM
  4. I also have this question!

    Reply
    Posted by: Guihai f.
    November 28, 2012 - 7:51 AM
  5. Hi guys, thanks for the interest in our paper. That's indeed a common question I get, so apologize for not explaining that more clearly in the paper. The 100uM concentration refers to all probes, so if you have ² probes in the pool the individual concentration is 50uM per probe, and if you have 50 probes it's ²uM per probe. It follows that as you scale up number of probes in the pool you do not have to scale up the amount of beads to add. As a caution though, we have noticed that when I use too many probes the concentration of each probe drops too much and it actually adversely affects yield. Try not to use more than 50 probes if possible.

    Reply
    Posted by: Ci C.
    December 18, 2012 - 5:23 PM
  6. what is the fucntion of the Buffer kit ? I can not find it in the protocol.

    Thanks for your time!

    Reply
    Posted by: Guihai f.
    November 1, 2012 - 12:50 AM
  7. The buffer kit is simply a convenient collection of buffers provided by Ambion that includes 5M NaCl, 1M Tris, 0.5M EDTA, etc..

    Reply
    Posted by: Ci C.
    December 18, 2012 - 5:25 PM
  8. Hi, I refer to your paper in HOTAIR ChIRP. You used 48 oligo probes for this ncRNA. I am trying to do a HOTAIR ChIRP. I am wondering if the number of probes can be smaller to reduce cost without a huge compromise on specificity. Furthermore, I am not doing sequencing but a specific check on a genomic region of interest, so I supposed the background noise may be lower? Thank you!

    Reply
    Posted by: Yi Fang L.
    November 8, 2012 - 5:11 AM
  9. We share your desire to reduce number of probes used. In preliminary testing I've found that you can use as few as 1 probe per ²00 nucleotide of RNA target RNA. But that number has to be doubled considering if you need both "even" and "odd" set, so effectively tiling density is 1 probe / 100nt.

    Reply
    Posted by: Ci C.
    December 18, 2012 - 5:28 PM
  10. Hello,
    I enjoyed your paper very much.

    At the end of the video, and in the figure, you mention that in this process you can isolate RNA binding proteins that were associated with the RNA. My questions are: have you had success in doing this? What is the protocol for isolating the proteins? Do you just isolate the protein fraction from the Trizol, or is there a more specialized method? Is there a paper that may lay out use of this method?

    Thank you for your help.

    Reply
    Posted by: Gavin J.
    January 18, 2013 - 12:09 PM
  11. Hi,
    Great protocol. Just one question. You mention that all ChIRP steps have to be performed at 37C. I can imagine this is important for the hybridisation steps, but I was wondering why this was needed for the washing steps as well?

    Many thanks

    Reply
    Posted by: Sebastian V.
    March 15, 2013 - 10:21 AM
  12. The washing should be performed at 37C just like you would wash a northern blot at elevated temperatures. If this is logistically difficult for you, just warming up the buffer to 37C and washing on a room temp. shaker works well too.

    Reply
    Posted by: Ci C.
    March 15, 2013 - 1:47 PM
  13. Hello,
    Thanks for the detailed and very helpful protocol. Can you tell me what kind of DNA yields to expect from the pull down per input cell amount? I know it will depend on the lncRNA targeted, but I need an idea of what kind of amounts to expect. Could you tell me what you got for your HOTAIR hybridizations? My own hybs only result in a few nanograms of DNA, and I don't know how to determine whether that is nonspecific DNA binding to my beads or whether it is actual lncRNA-bound DNA? How do you control for this? Thank you for your help in advance.

    Reply
    Posted by: blake e.
    March 20, 2013 - 5:44 AM
  14. Hi,

    Congrats for setting up such a protocol. It might have been discussed already but I would like to have an idea about the typical amounts of DNA and RNA that one might expect from 20 x 10 ^6 cells. This of course is transcript specific, but should we expect nanograms, picograms for either DNA and RNA?

    Also have you tried FA instead of glutaraldehyde for x-linking? Many thanks for your time

    Reply
    Posted by: Antonis A.
    November 26, 2013 - 12:15 PM
  15. Hi,

    Thanks for the detailed protocol. I am about to embark on it over the next week or so and had a question regarding the sonication. You advise a Bioruptor for this step however we only have a Covaris available. Would that provide acceptable sonication and do you have any advice regarding appropriate settings?

    Thank you

    Joshua

    Reply
    Posted by: joshua b.
    March 2, 2014 - 12:24 AM
  16. Yes Covaris works, we've found that it generally reduces sonication time by 30-50% compared with Diagenode. Actual time and intensity needs to be optimized for each machine and cell line.

    Reply
    Posted by: Ci C.
    March 3, 2014 - 5:51 PM
  17. Hi,

    Thanks for the detailed protocol. I am about to embark on it over the next week or so and had a question regarding the sonication. You advise a Bioruptor for this step however we only have a Covaris available. Would that provide acceptable sonication and do you have any advice regarding appropriate settings?

    Thank you

    Joshua

    Reply
    Posted by: joshua b.
    March 2, 2014 - 6:32 AM
  18. To all interested ChIRPers, hybridization can be performed overnight instead of 4hours. Doing so greatly reduces hands-on time. Happy ChIRPing!

    Reply
    Posted by: Ci C.
    March 3, 2014 - 8:14 PM
  19. Hi Ci,

    I have one more issue to resolve regarding ChIRPseq that I would appreciate your input on please (or other ChIRPers that have had similar problems).

    We have been unable to produce lysates using glutaraldehyde despite following the protocol to the letter and getting the Sigma brand that is recommended. Even before sonication, the DNA is degraded to a thick band below 200bp. When we try exactly the same process but substitute 1% formaldehyde (which we normally use for ChIP and 3C) or even plain PBS, the DNA is fine. We have even tried doing everything on ice and reducing the centrifuge steps to only 500g but to no avail.

    My questions therefore are:

    1/ Has your team had similar issues with glutaraldehyde in the past or found it to behave quite differently to formaldehyde (apart from needed longer sonication and harder centrifuging)?
    2/ Though you have shown glutaraldehyde to give a better signal to noise ratio, could we still expect to get usable data using 1% formaldehyde or is it not worth the time and effort?

    Thank you for your time

    Joshua Betts

    Reply
    Posted by: joshua b.
    April 16, 2014 - 8:07 AM
  20. Dear Joshua,

    Thanks for your interest in our protocol. It's unfortunate to hear that you're experiencing DNA degradation. We've never experienced this issue, nor have we heard similar problems from our collaborators who successfully did ChIRP. Usually we're more concerned with RNA integrity, naturally, and if you're careful that's not a problem either.

    Could you tell me how you're extracting DNA, especially prior to sonication?

    Regarding alternative crosslinking methods, for certain lncRNAs we've found that strong formaldehyde crosslinking also works (3% for 30min is a good starting point). 1% formaldehyde usually does not work.

    Best,
    Chu

    Reply
    Posted by: Ci C.
    April 16, 2014 - 1:55 PM
  21. Dear Chu,

    Thank you for your advice - I will try the protocol again but substitute 3% formaldehyde for the glutaraldehyde. Does that require any other adjustments apart from slower centrifuge spins and shorter sonication times?

    In regards to my DNA isolation, I lysed the cells as per the protocol and then treated with proteinase K for 45mins (at 50deg or 65deg - either seemed ok). I then RNAse treated the sample and either phenol chloroform extracted the DNA or instead used the Qiagen PCR columns before running the product on a 1% agarose gel. The columns gave more of a smear on the gel, whilst the phenol chloroform extraction just had a broad band below 100bp.

    When I performed exactly the same procedures on formaldehyde treated cells, I obtained a bright band of gDNA at the top of the gel as well as a more faint diffuse band below 100bp. This is what we normally see when working with formaldehyde. Am I right to expect a strong band of genomic DNA on the gel when using glutaraldehyde before sonication, or does the glutaraldehyde break the gDNA down during the fixation process? Does your glutaraldehyde fixed sample normally start as a broad smear even prior to sonication?

    Our group has a lot of experience with ChIPseq, 3C and 4Cseq library preparation but we have never dealt with glutaraldehyde before so are unsure what to expect.

    Any further suggestions would be greatly appreciated.

    Thanks

    Joshua

    Reply
    Posted by: joshua b.
    April 24, 2014 - 9:26 AM
  22. Before preparing your ChIRP library for sequencing, do you check your DNA size distribution (for example using Agilent Bioanalyzer). If so, do you see a similar distribution to that seen for a typical ChIP sample (200-800 bp range)?

    Reply
    Posted by: Eric S.
    August 29, 2014 - 2:43 PM
  23. Dear Chu
    I have two questions:
    1.In designing your chirp oligos have you tried to reduce the spacer between biotin-oligonucleotides. Is there any steric reason for choosing 1 probe / 100nt?
    2. Did you tried to crosslink adherent cells directly on plate? we have problem in properly resuspend our differentiated murine myoblasts in PBS+glutaraldehyde.

    thanks in advance
    regards

    Mariangela

    Reply
    Posted by: Mariangela M.
    October 16, 2014 - 9:56 AM
  24. Dear Mariangela,

    1. It's both a cost consideration (fewer probes are cheaper) and also a biochemical noise consideration (more probes introduce higher chances for probe:DNA direct hybridization).

    2. Yes we've found no difference fixing on plate vs. in suspension. Feel free to pick and choose what works best for you.

    Best,
    Chu

    Reply
    Posted by: Ci C.
    October 17, 2014 - 10:43 PM
  25. Dear ChIRP community,

    We are having problems getting ChiRP peaks with our transcripts. We have tried ChiRPing against 4 different transcripts and for all of them we were able to get very good fold enrichments in the genomic locus of each transcript (on average 50 to 200 x over background) capturing native transcription but we cannot find other binding sites in the genome for any of them. We have good reasons to believe that some or all of our transcripts are chromatin localized and we also know that the signal in the genomic loci is RNA dependent because it is dramatically reduced when we ChiRP after knocking down our transcripts of interest.

    Our transcripts are low in expression, on average 3 to 10 copies per cell. Any ideas as to how can we improve/solve the issue? Is it abundance that is limiting the system? Has anyone experienced/solved something similar?
    We tried the normal protocol, and now we are trying with 3 % FA in parallel to glut. Any suggestions would be very welcomed!

    Antonis

    Reply
    Posted by: Antonis A.
    November 13, 2014 - 5:18 AM
  26. Hi Antonis, I'm afraid the copy number is simply too low. In all of the ChIRP paper published so far I don't think anyone went that low.

    Your enrichment of the genomic loci could be complicated by ChIRP probes pulling down the DNA directly. Do an RNase-control to see if the signal is still there.

    Reply
    Posted by: Ci C.
    November 13, 2014 - 8:52 AM
  27. Dear Chu,

    Thank you for the prompt reply-yes I know that with such expression levels, ChIRP is challenging. I can try the RNase control but as mentioned I am confident about the RNA-dependence of the signal, because it falls back to LacZ levels when I knock down the transcripts. I guess we will have to try increasing the sensitivity of the experiment. Could an increase in x-linking time help here, have you guys tried 20 min of glut x-linking?

    Reply
    Posted by: Antonis A.
    November 18, 2014 - 3:28 AM
  28. Hello,
    I enjoyed your paper very much.

    At the end of the video, and in the figure, you mention that in this process you can isolate RNA binding proteins that were associated with the RNA. My questions are: have you had success in doing this? What is the protocol for isolating the proteins? Do you just isolate the protein fraction from the Trizol, or is there a more specialized method? Is there a paper that may lay out use of this method?

    Thank you for your help.

    Reply
    Posted by: jason h.
    November 26, 2014 - 10:15 PM
  29. Dear ChIRPers, we're happy to announce that you can now officially design and order ChIRP from Biosearch Technologies (Single molecule fish). Scientists developing ChIRP probes using the Stellaris probe designer and following the guidelines from our publication can now email Sheila Semaan at Biosearch Technologies, Inc. for help with ordering the probes in the proper format. Sheila’s contact info is as follows:

    Sheila J. Semaan, Ph.D., Associate Product Manager
    Biosearch Technologies, Inc.
    Email: ssemaan@biosearchtech.com

    Reply
    Posted by: Ci C.
    December 17, 2014 - 9:33 AM
  30. Dear ChIRP community,

    You can soon study RNA-protein interactions in vivo using ChIRP-mass spec. The story is coming out in Cell on April 2nd. Stay tuned!

    Reply
    Posted by: Ci C.
    March 25, 2015 - 5:42 AM
  31. Hello, thank you for the protocol! I also want to perform a RNA pulldown using your protocol and i am very excited for your RNA-protein interaction protocol! I have one question regarding the biotin oligos. Do they need to be 3' biotinylated or can they also be 5' biotinylated. Have you tested this? And you say nothing about (salt, HPLC) purification of the oligos. Thanks for answering my questions.

    Reply
    Posted by: Christine C.
    March 30, 2015 - 10:01 AM
  32. and is it necessary to order Biotin-TEG oligos or is "standard" biotin sufficient?

    Reply
    Posted by: Christine C.
    March 30, 2015 - 11:33 AM
  33. Hi Christine, thanks for your interest in our technique. Regarding your questions:

    1) 3' biotinylation is preferred, because oligos are synthesized 3' to 5'. This way, we ensure all oligos start with a biotin tag. 5' biotin tags are usually added to a synthesized oligo, and that increases the chance of untagged probes, which is less than ideal. However if you can ensure high tagging efficiency on 5' end, and the cost is advantageous, I don't see why it won't work!

    2) We use a disposable reverse phase column called "Glen-Pak DNA purification cartridge," which is a very efficient and cost-effective system. These are really short DNA probes and HPLC or PAGE would be necessary (also probably can't tell tag from untag probes).

    3) TEG ensures no steric hindrance, I don't have hard evidence on this but I think it's necessary.

    Reply
    Posted by: Ci C.
    March 30, 2015 - 10:03 PM
  34. Thank you for your answers! As I am performing your protocol with the aim of an RNA-protein pulldown, i have some other questions regarding the protocol, if you don't mind.

    - in the protocol before washed your magnetic beads in unsupplemented lysis buffer and you blocked the magnetic beads using BSA and yeast RNA- in the latest protocol you don't block. Why is that?

    - ChIRP-MS requires 100 Mio cells, ChIRP-seq 20 Mio cells. If i want to perform ChIRP-MS (or western blot) do i need 100 pmol of probe for each 20 Mio cells --> 500 pmol of probe for ChIRP-MS? --> 500 µl beads(??)

    Reply
    Posted by: Christine C.
    April 8, 2015 - 8:12 AM
  35. Hi Christine,

    - We no longer block as there's no difference in results, and we prefer a more streamlined protocol. BSA would also potentially contaminate ms results.

    - yes scale up beads accordingly. We do 1ml beads per experiment routinely. It gets expensive, but hopefully you'd only have to do it once or twice, and validate with western in a smaller scale later. For western of more abundant chirp hits, I do from 1/5 of cell pellets, but for less abundant ones I still use full amount.

    Reply
    Posted by: Ci C.
    April 8, 2015 - 11:36 PM
  36. Dear ChIRPers,

    ChIRP mass spec has just come online today at Cell. We're very pleased to introduce a very robust and straight-forward method that's optimized to study lncRNA:protein interactions. Check out the story at:

    http://www.cell.com/cell/abstract/S0092-8674(15)00312-8

    Happy ChIRPing!

    Best,
    Chu

    Reply
    Posted by: Ci C.
    April 2, 2015 - 9:00 PM
  37. Dear ChIRPers,

    Now you can easily design and order ChIRP probes with the latest optimization from Biosearch Technologies, the company that invented Single Molecule FISH.

    https://www.biosearchtech.com/chirpdesigner/

    Best,
    Chu

    Reply
    Posted by: Ci C.
    April 2, 2015 - 10:24 PM
  38. Hi, i'm trying to perform ChIRP-western and i have some questions concerning the methods.

    1) for a control, i also want to see if my RNA is present: how can i get a small aliquot of post-ChIRP beads to perform RNA elution with this aliquot
    2) for the protein biotin elution buffer: resuspend beads in what volume?
    3) for ChIRP-MS lysates are incubated with 30µl beads - can i reuse them?

    thank you in advance!

    Reply
    Posted by: Carolin C.
    April 14, 2015 - 7:18 AM
  39. Hi Carolin,

    thank you for your interested in our method.

    1) yes that's what we we as well. Simply take a 1% or 10% aliquot of your post-wash beads, and extract RNA.

    2) that's dependent on your starting beads volume. But there's a practical constraint, you don't want to have so much volume that after addition of TCA will exceed the capacity of an eppendorf tube.

    3) I'm not sure if I understand your question. Reuse lysate or beads?

    Reply
    Posted by: Ci C.
    April 15, 2015 - 12:20 AM
  40. thank you for the answers!
    to 2) again: if my beads are 100µl starting volume - so i resuspend 2x in biotin elution buffer and pool. Is 2x100µl then enough? and then add 25% of 100%TCA (50µl)?
    to 3) my question is, if i reuse the beads in the hybridization step. and just add another 70 µl beads (=100µl beads)

    Reply
    Posted by: Carolin C.
    April 24, 2015 - 11:23 AM
  41. so what do you recommend for my previous questions 2) and 3)

    another question: how do you calculate the amount of RNA retrieved? I retrieve only 10% of my RNA of interest calculated as % of Input. Furthermore,when i took a sample of what hasn't bound (after hybridization and capture of the beads) i get only 15% (fraction of washing doesn't include much RNA). so where has my RNA gone? i repeated several times… degradation?

    Reply
    Posted by: Carolin C.
    May 20, 2015 - 5:13 AM
  42. Hi, Carolin
    I am also, using ChiRP. Can you tell me the equation you used to calculate the % of your retrieval

    thanks

    Reply
    Posted by: u.
    June 28, 2016 - 1:22 AM
  43. Hi Carolin, May I ask in the end with your 10-20% retrieval rate of RNA, can you identify new proteins by MS via UV crosslink? Thanks.

    Reply
    Posted by: Zhenqiu H.
    May 17, 2018 - 6:20 PM
  44. Dear ChIRPers, we're glad to introduce the Magna ChIRP kit from Millipore. It's a simple pre-made reagent set for scientists who prefer the reliability and user-friendliness of a reputable brand.

    https://www.emdmillipore.com/US/en/product/EZ--Magna-ChIRP-RNA-Interactome-Kit---Isolation-and-characterization-of-non-coding-RNA%3Achromatin-complexes,MM_NF-17-10495

    Happy ChIRPing!

    Best,
    Chu

    Reply
    Posted by: Ci C.
    April 15, 2015 - 12:35 AM
  45. Hi Chu

    Thanks for sharing this information.
    Did anyone try this kit already and how good is it? Kits are easy to use and save time for reagents.

    Reply
    Posted by: Liu Z.
    April 16, 2015 - 5:17 PM
  46. Dear Chu,

    Many thanks for your excellent protocol. I'm designing a probe set and have been looking at the probes you designed for HOTAIR and XIST for inspiration. I noticed that some of the probes designed for XIST target the sequence at multiple loci (e.g. Probe 10), albeit ocassionally with a single base mismatch. Either way, I assume this may reduce the number of probes needed. Could I ask whether this was a deliberate design strategy and, if so, does the Biosearchtech ChIRP designer incorporate this strategy?

    Thanks

    Reply
    Posted by: Sebastian V.
    April 16, 2015 - 11:31 AM
  47. Dear Chu

    I feel exciting when I am reading your chirp paper in molecular cell. I have 2 questions that need your input.

    1) Have you or others you know tried in vivo tissue samples and if so how it works? I realized that your molecular cell and the latest cell paper used cell lines primarily.

    2) If I have a non-coding RNA which is 400 bp alone, and cannot generate more than 8 probes based on the parameters provided by https://www.biosearchtech.com/chirpdesigner/ ? What should I do ? Could I just reduce the spacer length between odd and even probes ? do you think it will increase background due to the increased probe density?

    Thanks

    Zhiyong

    Reply
    Posted by: Liu Z.
    April 16, 2015 - 4:49 PM
  48. Hi Zhiyong,

    Thank you for your kind words about our paper. Regarding your questions:

    1) Yes we have ChIRPed in drosophila tissues with great success. The key is probably to have small enough chunks of tissue that can be crosslinked effectively and thoroughly.

    2) Email the Biosearchtech team and they can help you manually reduce the number of probes.

    Good luck!

    Best,
    Chu

    Reply
    Posted by: Ci C.
    May 22, 2015 - 11:15 PM
  49. hey there, i was able to retrieve RNA with your method, but i'm not retrieving RNA bound protein (even though i did see a milky pellet after TCA and acetone treatment). i cross link with UV but the other steps i perform are just the same you are doing.
    i'm not so sure about this competetive elution method by using free d-biotin - and do you no longer add protease inhibitors in your buffer? the beads are simply eluted by using biotin buffer, rotate sample for 20 min,RT and 10 min mixing at 65°C (x2). This is it? then everything should be in my buffer and i continue with TCA precipitation ON.
    sorry if i this sounds stupid to you. i'm just looking for mistakes i could have made...

    Reply
    Posted by: Carolin C.
    June 10, 2015 - 10:32 AM
  50. Hi Carolin, I'm glad you're able to get RNA yield, that's a very good sign. The reason that you're not getting proteins could be: 1) mass spec instrument not sensitive enough. The quality of your results varies A LOT with who runs your mass spec. We've had amazing experience with Bill Lane at Harvard. 2) UV cross linking is known to be super low yield. Unless your lncRNA is very abundant or you're prepared to start with a huge amount of material, I would suggest formaldehyde as your first step. 3) TCA and cold acetone with give you a tiny pellet no matter what (the detergent acts as a carrier that will precipitate by itself even in the absence of proteins). So unless you see a sizable pellet, the presence of a pellet itself doesn't mean anything. d-biotin elution is complete, assuming you're using the C1 beads and biotin from invitrogen. Protease inhibitors are absent at the elution step, because your solution should be quite pure and free of contaminating proteases at that point assuming good lab practice. Hope this helps!

    Reply
    Posted by: Ci C.
    June 10, 2015 - 5:45 PM
  51. hey ci! thank you for your answers!
    1) i'm doing western blot as a start, later on i will do mass spec - so thank you for your tip&hint! as i already know a 100% candidate to get retrieved with my (cytoplasmic) mRNA, i stained for this candidate and got no result.
    2) you also mentioned uv-crosslinking in your first protocol, do you have a comment or any experience on the time of cross linking or amount of joules? so far i cross linked 150mj/cm^2 but I'm thinking to increase that. also i need to scale up my cells/beads/probe as you're saying. i do not know if formaldehyde cross linking works in my case, as i have cytosolic mrna.
    3) yes I'm using c1 dynabeads and d-biotin from invitrogen! thanks for the explanation, that helps!

    thank you very much for your quick answers, tips and hints!

    Reply
    Posted by: Carolin C.
    June 11, 2015 - 8:43 AM
  52. hey there, repeated the experiment - this time i used 200 Mio MEF cells - 10 ml lysate and 1 ml beads, 10µl probe. i also increased cross linking to 300 mj/cm^2 - still no proteins (known RNA binding protein) detectable via western blot. i confirmed RNA pulldown via qPCR after cross linking from a small fraction of beads.
    is sonication necessary?
    any troubleshooting ideas?

    Reply
    Posted by: Carolin C.
    June 17, 2015 - 11:38 AM
  53. I would start with formaldehyde first. UV is known to cross link very inefficiently, and certain residues of contact may not be amenable to photocrosslinking at all. Yes sonication is necessary to soluble the cell lysate.

    Reply
    Posted by: Ci C.
    June 19, 2015 - 2:37 PM
  54. i'm using your 25%TCA / Aceton precipitation method to precipitate my proteins. Have you also experienced a pH change (when adding laemmli it turns green/yellow) due to rest TCA? And i think i don't get my pellet resuspended - even after mixing and resuspending. Isn't boiling for 30 min, 95°C destroying the proteins?

    Posted by: Carolin C.
    July 10, 2015 - 7:30 AM
  55. Yes I have had similar experiences before: pH change usually can be prevented by more careful and thorough acetone washes (although need to care not to disturb pellet), and over drying will cause difficulty in resuspension. I never air dry for over a minute. Boiling is completely fine. Try using the LDS samples buffer / Bis-tris gel system from invitrogen, these work great on chirp samples. If you are still having issues with pH, you can manually adjust a little with a moderate alkaline buffer.

    Posted by: Ci C.
    July 10, 2015 - 10:59 PM
  56. Hi Carolin,

    I've the same problem as you had. I could only retrieve 10-15% of RNA when compare to input RNA. Could you please share how you sort this problem?

    Thank you in advance!

    Reply
    Posted by: Ganeshkumar A.
    July 23, 2015 - 4:25 AM
  57. Hi Ganeshkumar, unfortunately i didn't solve the problem… but i found more RNA of interest in my beads when i directly resuspended them in trizol (leaving out proteinase K treatment) - since i'm interested in my proteins bound to the RNA the proteinase K treatment step is of no importance to me. But so far i still couldn't retrieve my RNA bound proteins.

    Reply
    Posted by: Carolin C.
    July 23, 2015 - 5:19 AM
  58. Hi Carolin,

    Thanks for your immediate reply. I'll also avoid Proteinase K step and try again.
    I've one more question. We always do two step RT-qPCR. So, the amount of RNA that I retrieve isn't enough to perform a RT reaction. How do you do your RT-qPCR? This question maybe a bit stupid, but i really wanna sort out all the issues as early as possible.

    Thanks in advance.

    Reply
    Posted by: Ganeshkumar A.
    July 23, 2015 - 5:27 AM
  59. Hello,
    my RNA amount is also quite low, measured by Nanodrop (Once i measured with qubit and could't get anything). I'm using max. volume possible for RT reaction and adapt for input, not bound… Then i perform qPCR with my cDNA

    But i suppose you cannot skip proteinase K step when your lysate is formaldehyde treated, or you need to boil it first to reverse the crosslink. i diluted my beads in trizol when i had non-crosslinked samples to check for retrieval percentage

    Posted by: Carolin C.
    July 23, 2015 - 7:12 AM
  60. We always do a one-step qRT-PCR (i.e. each qPCR well runs it's own RT). This in my hands is much more sensitive and convenient as well (you don't have to measure RNA or cDNA, just load equal portions of each sample). We use Strategene Brilliant II SYBR reagents and Roche 480 machine. Let me know if this helps with yield detected.

    Posted by: Ci C.
    August 4, 2015 - 8:32 AM
  61. Hi Ci,

    I did gene specific cDNA synthesis with 10ng of RNA from all conditions and treated with RNase H and A (in excess) and then proceeded to qPCR. This worked good for me and I'm getting around 14-25 fold enrichment in my odd and even probe samples when compared to input.

    Thank you and Carolin for your suggestions.

    Posted by: Ganeshkumar A.
    August 4, 2015 - 10:43 AM
  62. Hello,
    i have not yet retrieved any RNA bound protein, so i'm thinking about what issues to sort out. i have to say i can retrieve my RNA of interest (but only max. 20% of input)

    i'm wondering about the protein amount in your formaldehyde cross linked, sonicated and hypotone lysed cell lysats. I experienced the protein amount (measured with Bradford) to be 5-8 fold lower compared to uncross linked or Uv cross linked cells treated with a NP40 lysis buffer.
    Why is that, and will it affect my experiments? Any ideas why the protein amount is so low?
    Thank you in advance!

    Reply
    Posted by: Christine C.
    July 23, 2015 - 5:47 AM
  63. Hi Christine, off the top of my head the two main reason for less protein detected upon cross linking is 1) insufficient sonication, and 2) insufficient reverse cross linking (i.e. boiling). Regarding 1), you have to titrate sonication time of your lysate, and pick the minimum sonication required to release all proteins and RNA of interest. To test this, precipitate the lysate post sonication for 10 min at max speed, and run qRTPCR against your RNA of interest and western against the protein on both the supernatant and pellet, the latter resuspended in appropriate buffer. This way you'll know if there's anything left to be solubilized. 3% cross linked lysates can take significantly longer to solublized than your typical ChIP cross linking.

    Regarding 2), do no less than 30min boiling in sample buffer. I've titrate that also, and 10min or 20min recovers much less protein.

    Let me know if these help and let's start from here. Good luck!

    Reply
    Posted by: Ci C.
    August 4, 2015 - 8:29 AM
  64. Hello,
    Many thanks again for this superb protocol. I'm currently trying to optimise this for mRNA pull-down from fixed cells. I've tried various fixing conditions in parallel and dot-blotted for my RBP of interest vs a non-binding protein (GAPDH) after performing ChIRP. I found that 1% glutaraldehyde (10 min) samples blotted strongly for my RBP, but produced an equally strong signal for GAPDH. The latter was also the case for 3% formaldehyde (10 min), but blots of both GAPDH and my RBP weren't as strong. The only sample that showed a slightly stronger signal of the RBP vs GAPDH was 1% formaldehyde (10 min), but both signals were quite weak. All samples were sonicated for several 30 sec ON and 45 OFF cycles at maximum intensity before foaming to produce sheared RNA of <500 bp in optically clear samples.

    I tested the expression of mRNA of interest in samples pulled down with my tiling oligo set compared with a run using non-targeting scrabled oligos. A capture on unfixed cells produced a very high target yield in the tiling oligo sample and no mRNA in the scrambled control sample. However, performing the same pull down in the 1% formaldehyde-fixed+sonicated samples dramatically increased background (GAPDH mRNA) compared to the scrambled control. I'm curious if you found anything similar during the development of ChIRP and ChIRP-MS and if you might have any solutions? Would extra sonication perhaps help?

    Many thanks

    Sebastian Vencken PhD
    Post-Doctoral Researcher

    Reply
    Posted by: Sebastian V.
    September 13, 2015 - 8:06 PM
  65. Hi Sebastian,

    Have you had any luck adapting this technique to mRNA capture?

    Best,
    Mike

    Reply
    Posted by: Michael K.
    December 7, 2016 - 4:48 PM
  66. Mixed results to be honest. We went forward with 1% formaldehyde and tried different forms of sonication (incl. covaris which didn't give us effective fragmentation). Biased background was still an issue and the copurification of our RBP (Ago+miRNA) was much lower than expected. This has apparently successfully been performed before (https://www.ncbi.nlm.nih.gov/pubmed/23325846) but we didn't manage to reproduce these results for our target.

    Reply
    Posted by: Sebastian V.
    December 8, 2016 - 4:15 AM
  67. Thanks for the feedback. Does your miR-CATCH work well/better? Similar idea, no sonication, and probes attached to beads first. Have you ever tried to use this to look for bound proteins?

    I also ran across this mRNA pulldown from yeast (https://www.ncbi.nlm.nih.gov/pubmed/27641505). They also attached the oligos to the beads first, but the buffers look more like ChIRP.

    Reply
    Posted by: Michael K.
    December 8, 2016 - 11:44 AM
  68. We've had some pretty good results (and publications) from miR-CATCH, but have also discovered that the performance can be very dependent on both the target and oligo design.
    With thorough optimisation we managed to reduce background in miR-CATCH, but each target was a new challenge in this respect.

    We hoped that ChIRP would significantly reduce the issue of background, but we have found balancing fragmentation and signal to be difficult. Dot blots for Ago2 gave us very high background signal in scrambled controls when using gluteraldehyde, while 1% formaldehyde gave some better signal differentiation, but at reduced strength. If I'd take ChIRP further for mRNA:RBP(:miRNA) capture, I'd look at the sonication and fixing steps again as there may be further room for improvement. Unfortunately, I've moved to other things for now.

    Reply
    Posted by: Sebastian V.
    December 9, 2016 - 10:42 AM
  69. Hello!

    Many thanks for the detailed protocol, however i am having trouble at crosslinking and sonication steps! I am sorry if it sounds silly but I just want to identifiy what I am missing/doing wrong.

    In the near future, I want to scale up for ChIRP MS, however first want to get these steps right as CHIRP MS would consume 10 times the beads.

    Crosslinking - When I add Glycine to quench formaldehyde, it doesnt change color?

    Sonication - As used by your team, I am using Bioruptor to sonicate the Hela S3 lysate (20 Mi. cells/1ml supplemeted lysis buffer) in 15 ml falcon (Corning) with a probe attached to the cap. I cant seem to get a clear lysate as shown in your video. I have tried sonicating starting from 30 cycles to 300 cycles and it just froths towards the end. After centrifugation to remove cell debris, the lysate is still considerably cloudy.

    I see that a lot of people are working on optimising this method and would appreciate any tips / tricks that might help me get going.

    Looking forward to some suggestions.

    Best regards,
    Minakshi
    PhD Student

    Reply
    Posted by: Minakshi G.
    September 30, 2015 - 5:06 AM
  70. Hi Minakshi, thanks for your interest in our protocol. Re your questions:

    - glycine doesn't change the color of formaldehyde, just glutaraldehyde.
    - check our bioruptor to make sure power is normal and probes are centered correctly. It shouldn't take more than a couple hours even for hardy cell lines. Frosting is strange and indicates to me that you may not be centering the probe correctly and thus losing efficiency.

    Reply
    Posted by: Ci C.
    November 2, 2015 - 2:46 PM
  71. Hi,
    I would have a question regarding the usage of formamide. My concern is that once the bottle is opened and the
    formamide is exposed to oxygen, it will begin to oxidize to formic acid. Do you take any precautions to prevent oxidation? How do you store formamide for further use? Sigma suggests to purge formamide with nitrogen and store it frozen to prevent oxidation.

    Thanks a lot for your help.

    Best wishes,
    matjaz

    Reply
    Posted by: Matjaz B.
    November 2, 2015 - 4:53 AM
  72. Hi Matjaz, thank you for your interest in our technique. formamide is indeed subject to oxidation and ionization. Re-deionization is laborious, so we tried our best to protect the reagents by parafilming around the bottle cap after each use and store it at 4C. I usually run through a 1 liter bottle in a year with repeated opening and closing, and have not observed any effect on results over time. Hope this helps.

    Reply
    Posted by: Ci C.
    November 2, 2015 - 2:44 PM
  73. Hello,

    I have another technical question. Even though our probes are very specific (we detect target RNA using different probes with high specificity), we have problem with high DNA background signal. Have you also had a similar problem? One way to resolve the issue would be to pre-clear lysates with beads before the hybridisation step. Have you ever implemented this step? If yes, have you done the pre-clearing in lysis buffer with or without hybridisation buffer?

    Many thanks,

    - matjaz

    Reply
    Posted by: Matjaz B.
    November 3, 2015 - 7:36 AM
  74. Hi Ci,

    Thanks for providing such an excellent protocol. I have a question about probe design. The 3' half of lncRNA I interested in (5kb in length) was fully cover by repeat elements. Because ChIRP probe Designer ignored the repeat sequence when designing probe, I found almost no probe target to the 3' half of this lncRNA. I'm afraid that I will miss all proteins bind to the 3' end of lncRNA if I only use the probe target to the 5' end of lncRNA. I wonder how do you handle lncRNA contains large number of repeat sequence?

    Thanks for any help.

    Sincerely,
    Jian-You

    Reply
    Posted by: Jian-You L.
    December 3, 2015 - 9:48 PM
  75. Hi all!

    Thank you for sharing this protocol!
    I have just analysed the RT-qPCR results coming from the RNA samples in order to analyse the lncRNA retrieval and I have retrieved a very small percentage of the lncRNA. From the input and the ChIRP samples I have used the same volume of sample (for the qPCR, around 36ng for the input and 21g for the ChIRP) and I have obtained a Ct of 30 for the input and 37 for the ChIRP measuring my lncRNA of interest. How do you normalise the results? Do you use the same volume or ng of the different RNA samples? Do you use a normalisation method similar to the ChIP one? https://www.thermofisher.com/uk/en/home/life-science/epigenetics-noncoding-rna-research/chromatin-remodeling/chromatin-immunoprecipitation-chip/chip-analysis.html

    Thank you in advance

    Isabel

    Reply
    Posted by: Isabel R.
    December 17, 2015 - 9:30 AM
  76. Hi, Isabel
    we experience along time to optimize the protocol in our lab around 9 month for lncRNA and we found the result unstable for different lncRNAs but for your normalization we usually use a normalized methods like Chip experiment for both GAPDH and our target,
    Hope that helps
    for more contact amr@mail.ustc.edu.cn

    Reply
    Posted by: u.
    December 8, 2016 - 7:17 PM
  77. Dear Chu:
    Recently I'm trying to perform ChIRP experiments and your detailed protocol is very helpful. However, I have some questions about cell lysis and sonication.
    1. I find that the supplemented lysis buffer cannot lyse the crosslinked cells. The crosslinked cells are only resuspended, after a few minutes they will sink to the bottom of the tube, is it normal?
    2. You emphasize that lysis buffer should be added fresh Protease Inhibitor, PMSF and Superase-in then proceed immediately to sonication after resuspended smooth, however, a sample will be added appropriate supplemented lysis buffer in proportion then devided into several parts to sonicate by limiting ultrasound equipment, and will it be affected?
    3. I only have a Covaris available for sonication in lab, I tried many times but could not find an appropriate setting. My DNA size distribution is normally a bright band of gDNA at the top, a bright band above 2000bp and a diffuse band from 2000bp to 100bp, if I continued to sonicate, the diffuse band would below 100bp, and there were considerably cloudy after centrifuged the sonicated samples. I want to know what and how much influence if DNA smear is less than 100bp or more than 500 bp? Do you have any advice regarding it?
    Looking forward to some suggestions, thank you in advance!
    Best regards
    mixue
    Ph.D. Student

    Reply
    Posted by: he j.
    July 17, 2016 - 3:37 AM
  78. Dear Dr CHU.
    First of all, sorry for my english, but I'm french.
    I work on lncRNA and I really appreciate your two articles on ChIRP and Chirp-MS, and since some months I try to set it up, but unfortunately without result until now.
    => I designed the probes as you know but my lncRNA is about 2.3kb so i was able to design only 26 probes. I also ordered your Positive and Negative Control.
    => I amplified the cells until 200 to 300 million and I cross linked them with 3% formaldehyde 30min and stop it with Glycine 0.125M 5 to 10 min, as you mention.
    => My sonication was tested in order to have a smear of DNA between 100 and 600bp and it took between 120 and 140min. I use a Bioruptore in cold room and I have to change water and ice every 15min in order to maintain the low temperature. Samples are centrifuged and debris are removed.
    => I use 2 to 3 aliquots of 1mL of lysed cells for each test, and the indicated quantity of probes. Of course before incubation, samples are pre-incubated with prewashed beads.
    => First incubation is done at 37 ° C in a shaking system, overnight. Second incubation (after addition of the prewashed beads) is done at 37 ° C for 1 hour.
    => Wash is done 5 times with wash buffer, then samples are eluted (using elution buffer freshly prepared with Sigma-Aldrich Biotine) and precipitated with TCA and Aceton. Pellet are directly dissolved in Laemmli, incubated 30min at 95°C and loaded with 12% acrylamide gel. After 1h at 200V the gel is stained with Coomassie Blue and destained 3 times 45min.
    I found NO BANDS even with Positive and negative controls, but after elution, when I checked the presence of my lncRNA in the sample selected with my probes, the result is positive.

    Could you tell me what are the possible mistakes that I could do ....?
    Thanks a lot.

    Reply
    Posted by: Sebastien C.
    January 3, 2017 - 5:02 AM
  79. Dear Dr. Chu,

    Good day! I am Anchilie, a PhD student, and I am really interested in applying your discovered technique in plants. I am currently doing the ChiRP-MS in Arabidopsis but I don't seem to get any RNA-bound proteins in my sample.

    I would just like to briefly tell you how I do the ChiRP-MS in plants. First, I harvest 3g of 14-day old seedlings and crosslinked it with 1% formaldehyde. From these, I then do nuclear extraction and do sonication. From the 1mL sonicated chromatin, I dilute it with 2mL of hybridization buffer, add 1ul probe and 100ul of beads and incubated then in 37C for O/N or 4 hours at 37C. From then I wash the beads 5x and precipitated the proteins using the biotin elution buffer and TCA method. The samples in 1x Laemmli buffer were submitted to LC/MS. I also check the enrichment in the qPCR and are enriched by 500-fold. However, upon analyzing my samples, the mass spec results only detected mitochondrial and chloroplastic proteins in the control and non-coding RNA

    I also tried many things like using 3% formaldehyde and UV crosslinking. I also did direct boiling of beads in Laemmli buffer. I also tried increasing the amount of material 10x accompanied with increasing the probe and the beads.
    But it doesn't really help.

    I would really appreciate to receive some feedback from you and your group with this matter.

    Best regards,
    Anchilie

    Reply
    Posted by: Anchilie M.
    February 27, 2017 - 4:49 AM
  80. Dear Anchilie,

    I am doing CHIRP in arabidopsis, you are the only only doing CHIRP as far as I know. I met some problems. I couldn't enrich my non coding RNA. So I want to know how did you do your CHIRP? Could you please share your protocol with me? Thanks!

    Reply
    Posted by: Zhi-Hong L.
    June 19, 2017 - 7:56 AM
  81. Hi guys, can anyone help me how the percentage calculation for lncRNA retrieval is done.

    Thank you

    Reply
    Posted by: Sunil B.
    June 22, 2017 - 3:58 PM
  82. Hi,

    Thank you for sharing this protocol, I have a question. If I want to study the potential interaction between a protein and a long non coding RNA in rDNA locus (or eventually with the rDNA directly), do you think that your kit will work ?
    Thank you for your answer

    Reply
    Posted by: Burette M.
    April 12, 2018 - 3:27 AM
  83. Dear professor,
    Thank you for sharing this protocol! It helps me a lot. I have a question to ask. Since ultrasound may break the hydrogen bond, I want to ask whether it is necessary to perform sonication when I study weak RNA-RNA interaction, for example: lncRNA-miRNA interaction?

    Reply
    Posted by: Zihao L.
    May 26, 2018 - 10:52 PM
  84. Dear professor,

    Any idea if using a probe-based sonication (like the Branson sonifier) would work? Anyone has tried it ?
    in the previous lab, we were using the bioruptor however we don't have it where I am now unfortunately and needed help to optimize it on the Branson or probe-based sonifiers?

    Thanks

    Reply
    Posted by: Anonymous
    March 1, 2020 - 4:26 AM

Post a Question / Comment / Request

You must be signed in to post a comment. Please or create an account.

Usage Statistics