I cannot say for sure that these E-protein segments are presented on the outside, though my understanding is that parts of the E-protein may be. So spatially where is the C-terminus of the E-protein in the virion? I don’t know.
Re-reading the paper on the E-protein makes me think that it is often expressed outwardly to the cytoplasm but sometimes inwardly depending on conditions.
Fantastic. I bet the other 2 big companies in this space (as well as others) have defined their formulations but have not disclosed them yet. @DigitalEmbrace, if you can find more information on what formulations have been reported by these companies or other groups, particularly if the lipids are commercially available, the links and written out information would be extremely helpful to collect in this thread. We would order those reagents immediately for Eterna experiments. Thanks!
I have been reading about the SARS-CoV-2, vaccines, and how the Eterna community can help address the recent pandemic. I wanted to share my thoughts on choosing an mRNA molecule that can serve as a starting point for the Eterna community to design a vaccine against the virus. I strongly believe Eterna-based designs will not only have an impact on the current outbreak, but the lessons learned from this effort will change our approach to many other infectious disease vaccines in the future.
Before I reveal my antigen target, here is a brief summary of what I’ve learned from my research. I include links/sources that hopefully will allow for those looking to develop a quick understanding of this topic to get caught up. I want to give much credit to Ivan (Vanya) Zheludev, a graduate student in the Das Lab at Stanford for initial research and pointing me to key papers.
Moderna has a vaccine (named mRNA-1273) that is currently going into clinical trials. Of note, no mRNA vaccine has gained FDA approval, although many candidates are in the clinical trial pipeline. mRNA vaccines are faster to production than traditional protein-based vaccines, so their accelerated development timeline highlights one of the advantage of mRNA-based vaccines. As their antigen, they have chosen a prefusion stabilized form of the S (spike) protein like the one described here.
Antigens against viruses are often chunks of proteins from the virus that allow B cells and T cells to recognize and develop an immune response. A more specific portion of the antigen is an epitope, which is the part of the antigen that is bound by the antibody. Epitopes by definition are smaller than antigens, which will come into play later. For reference, epitopes are around 7-11 amino acids long, and the binding groove of MHC (major histocompatibility complex, responsible for binding antigens) accommodates ~9 amino acids.
An obvious choice for an antigen is the entire membrane (M), nucleocapsid (N), spike (S), or envelope (E) proteins, along with various other proteins from the viruses’s genomes (called Open Reading Frames, or ORFs). However, this can be tough because these proteins can be large (the spike protein is 1273 amino acids–so the mRNA encoding it is over 3600 nucleotides long), the immune responses may not be robust, or hard to express.
Thus, designing a vaccine that encodes for an epitope can be advantageous. Advancements in computational biology have enabled predictions of epitopes from viral genomes (like the example here, and explained here), and because epitopes individually are smaller than an antigen, you get more “shots on goal” when it comes to eliciting an immune response by stringing together multiple epitopes. Such an example of multiple epitope vaccines (MEVs) have shown promise for mRNA-based cancer vaccines. A recent paper applied computational techniques to predict MEV designs for SARS-CoV-2.
In the case of SARS-CoV-2, the virus which causes COVID-19 (coronavirus disease-2019), we can borrow from previous literature that have identified epitopes from SARS-Coronavirus which is a previously known/studied version of the virus. The hypothesis is that if the epitopes (or more accurately, protein sequences of those epitopes) are also encoded in the genome of SARS-CoV-2, those epitopes can be applied to fight SARS-CoV-2. A peer-reviewed article showcasing that approach can be found here. The epitopes identified in Table 2, Table 3, and Table 4 may be good starting targets. For the first Eterna challenge, I suggest the following amino acid sequences (epitopes) linked together by glycine-serine linkers:
FIAGLIAIV, GLIAIVMVTI, IITTDNTFV, ALNTLVKQL, LITGRLQSL, LLLQYGSFC, LQYGSFCT, NLNESLIDL, RLDKVEAEV, RLNEVAKNL, RLQSLQTYV, VLNDILSRL, VVFLHVTYV, ILLNKHID, FPRGQGVPI, LLLLDRLNQ, GMSRIGMEV, ILLNKHIDA, ALNTPKDHI, LALLLLDRL, LLLDRLNQL, LLLLDRLNQL, LQLPQGTTL, AQFAPSASA, TTLPKGFYA, VLQLPQGTTL
and if space allows, these additional epitopes linked to the one above:
GYQPYRVVVL, PYRVVVLSF, LSPRWYFYY
DSFKEELDKY, LIDLQELGKY, PYRVVVLSF, GTTLPKGFY, VTPSGTWLTY
A parallel target is:
GSFCTQLNR, GVVFLHVTY, AQALNTLVK, MTSCCSCLK, ASANLAATK, SLIDLQELGK, SVLNDILSR, TQNVLYENQK, CMTSCCSCLK, VQIDRLITGR, KTFPPTEPK, KTFPPTEPKK, LSPRWYFYY, ASAFFGMSR, ATEGALNTPK, QLPQGTTLPK, QQQGQTVTK, QQQQGQTVTK, SASAFFGMSR, SQASSRSSSR, TPSGTWLTY
Excited to discuss further and launch these puzzles!
Yes–and that one was shown to elicit a positive T-cell response! I think your approach of checking with SwissTargetPrediction is a really good one. If you see “HLA” pop up under the “Target” column, that is probably a really good hit!
or a salt or isomer thereof, wherein 1 is selected from 1, 2, 3, 4, and 5; m is selected from 5, 6, 7, 8, and 9; Mi is a bond or M’ ; R4 is unsubstituted Ci_3 alkyl, or -(CH2)nQ, in which Q is OH, -NHC(S)N®2, -NHC(0)N®2, -N®C(0)R, -N®S(0)2R, -N®R8, -NHC(=NR9)N®2, -NHC(=CHR9)N®2, -OC(0)N®2, -N®C(0)OR, heteroaryl or heterocycloalkyl; M and M’ are independently selected
from -C(0)0-, -OC(O)-, -C(0)N(R’)-, -P(0)(OR’)0-, -S-S-, an aryl group, and a heteroaryl group; and R2 and R3 are independently selected from the group consisting of H, CM4 alkyl, and C2-i4 alkenyl.