Hi everyone,
I would like to characterize changes in the T cell receptor repertoire after ex vivo stimulation with different types of antigens (conventional peptide/MHC, unconventional antigens, superantigens, other unspecific proliferation inductors). Specifically, I would like to simultaneously quantify expansion of T cell types (gamma/delta vs alpha/beta), certain chains (like changes in Vbeta-profiles after SAg stimulation), T cells with (semi-)invariant TCR chains (like Vg9/d2 cells) and single clones (CDR3-sequences), in order to screen unconventional T cell antigens for their effects.
I am very unsure of how to approach this. I can find quite a lot of information on CDR3 sequencing, but most papers either focus only on specific T cell subsets or try to determine the overall TCR diversity in an individual. I still don't know, whether information from CDR3 is sufficient to determine which genes the respective cell used when creating the TCR. Can the CDR3 sequence be linked back to the corresponding alpha, beta, gamma or delta chains or do I need to analyze different regions in order to see, for instance, an expansion of Vbeta2?
Are there universal primers to capture the whole TCR-repertoire within my culture? Currently I intend to rewrite and expand the TCR-RNA in an RT-PCR. How can I ensure that the semi-quantitative nature of RT-PCR is sustained when primers bind to different chains with different affinity? Is there a bias?
I hope you can give me some pointers or "must-reads" for TCR sequencing.
I really appreciate any help you can provide.
Great questions Johannes. I will try to answer each in turn.
Can the CDR3 sequence be linked back to the corresponding alpha, beta, gamma or delta chains or do I need to analyze different regions in order to see, for instance, an expansion of Vbeta2?
--> This depends entirely on how much of the variable (V) and joining (J) regions you capture with your amplicon sequencing. The CDR3 region has the structure: V-n-D-n-J-C where n are potential non-templated nucleotides. If you prime at least 50 bases upstream of the invariant Cysteine in the V region, you should be able to identify the V gene usage - For many V genes, you need less than 20 bases there to uniquely identify them. If you prime in the constant region, you will capture all of the joining region as well.
Are there universal primers to capture the whole TCR-repertoire within my culture?
--> Sadly no. You will have to use multiplex priming to capture them all, unless you want to use RACE and sequence the RACE amplicons.
How can I ensure that the semi-quantitative nature of RT-PCR is sustained when primers bind to different chains with different affinity? Is there a bias?
--> Yes there is a bias, and you must adjust for it to have the most accurate results. At our institution we use the unique primer ID or "molecular tag" method (PMID 23995388, PMID: 25149069). There is a huge commercial player in this space (Adaptive Biotechnologies, www.adaptivebiotech.com) who uses a different method: PMID: 24157944.
I hope you can give me some pointers or "must-reads" for TCR sequencing.
--> Seminal paper: PMID: 19541912
--> This and the Adaptive Biotechnologies paper should get you started.
--> R package for analysis: PMID: 26017500
Best Wishes,
Ben
Good questions Johannes. I will try to answer each in turn.
Can the CDR3 sequence be linked back to the corresponding alpha, beta, gamma or delta chains or do I need to analyze different regions in order to see, for instance, an expansion of Vbeta2
--> This depends on how many bases into the relevant regions you capture with your priming strategy. The TCR CDR3 region has the structure V-n-D-n-J-C, where V is the variable region, n are potential non-templated nucleotides, D the diversity region, J the joining region, and C the constant region. If you prime 50 bases upstream of the invariant Cysteine residue contributed to the CDR3 by the V region, you should be able to identify the V families. If you prime from the Constant region (in the reverse sense) you will capture the entire J region. Bear in mind that if your amplification assay starts with DNA, you cannot reverse-prime from the C region as there are intronic bases between J and C that would make your amplicon prohibitively large. If you are priming from V and C, you would need to start with RNA as your template material.
Are there universal primers to capture the whole TCR-repertoire within my culture?
--> Sadly no. You will have to use multiplex priming, or RACE and sequence the RACE amplicons.
How can I ensure that the semi-quantitative nature of RT-PCR is sustained when primers bind to different chains with different affinity? Is there a bias?
--> There is definitely bias, and you will have to adjust for this to get the most accurate results. At our institution we use the unique primer ID or "molecular tag" method (PMID: 23995388, PMID: 25149069) . There is a huge commercial player in this space (Adaptive Biotechnologies, www.adaptivebiotech.com) that uses a DNA assay with a different method of bias control: PMID: 24157944.
I hope you can give me some pointers or "must-reads" for TCR sequencing.
--> Seminal technical papers: PMID: 19541912, PMID: 21349924, Adaptive Biotech paper cited above.
--> A perhaps helpful R package for analysis: PMID: 26017500
Best Regards,
Ben
Thank you so much for your comprehensive response! While diving into the literature you provided, I stumbled upon a relatively new publication (referencing several of the papers you mentioned), which on first glance appears to provide a perfect solution to my problems.
Even if it doesn't, I am positive that the pointers you provided will suffice to get me started into the right direction.
Article High-resolution analysis of the human T-cell receptor repertoire
Thanks Johannes. This is a nice method - Getting the RNA digestion right seems to be key. I haven't tried this - We are pursuing the RACE method with longer-read sequencing for now.
- Badges
- Science method