Project Description

MMRF » 2018 Vaccine Development Conference #05: Spliced Peptides — Novel Candidates for effective influenza immunity? [2018-06-27. Patricia Therese Illing. HVP/USC]

2018 Vaccine Development Conference Session #05: Spliced Peptides — Novel Candidates for effective influenza immunity? [2018-06-27. Patricia Therese Illing, Marie-Paule Kieny. HVP/USC]

2018 Vaccine Development Conference – Session #05: Spliced Peptides — Novel Candidates for effective influenza immunity? — Patricia Therese Illing, Marie-Paule Kieny.


Organized in conjunction with the Human Vaccines Project, the 1st Annual Conference on the Future of Vaccine Development was a one day event which took place at the USC Michelson for Convergent Bioscience on June 27, 2018.

By bringing together some of the world’s leading scientists in the fields of immunology, genomics, bioinformatics, and bioengineering, the Future of Vaccine Development annual conference aims to explore how the convergence of new technologies across disciplines is impacting the future of vaccine development. The conference will also honor the three inaugural winners of the Michelson Prizes for Human Immunology and Vaccine Research, both via their respective presentations and the remittance of their prizes during the Awards dinner ceremony following the conference itself.

About the Presenter: A research fellow at Monash University, Dr. Patricia Therese Illing was the first to identify spliced peptides during a viral infection. This work involves an innovative new approach for identifying influenza specific peptide antigens with implications for the development of vaccines against both seasonal and pandemic influenza strains. The Prize money will provide greater resources to expand understanding of how a viral antigen is recognized by the human immune system.


  • Patricia Therese Illing, PhD, Research Fellow, Immunoproteomics Laboratory, Biochemistry & Molecular Biology, Biomedicine Discovery Institute, Monash University, Melbourne, Australia.
  • Marie-Paule Kieny, PhD, Director of Research @ Inserm (Institut national de la santé et de la recherche médicale), former Assistant Director-General for Health Systems and Innovation at the World Health Organization (WHO).

Marie-Paule Kieny: It’s my pleasure to introduce the first of awardees of Michelson’s Prize, so Patricia Illing, who’s a research fellow at Monash University and the fact that she was selected has nothing to do with the fact that the chair of the board of HVP is from Australia, absolutely promise. So Patricia, don’t feel, how would I say, embarrassed talking about the previous speakers who have so many awards and prizes. You are still young. The future is yours, and I’m sure by that time, you’ll be also honored and loaded up with prizes.

Patricia received her Ph.D. actually quite recently in 2014 from the University of Melbourne and where you have been studying human leukocyte antigen associated adverse drug reaction abacavir sensitivity syndrome, so this has provided a new paradigm for presentation of drugs by HLA molecules that you published in Nature and was recognized also already by a Victorian Premier Award in 2013. So we are very happy to have you here and to present a transformative project in the area of immunology and I also give you then 15 minutes and then we’ll have a few questions for you. Go ahead.

Patricia Therese Illing: All right, thank you for the introduction. Can I firstly say that I’m very honored to be here as one of the recipients of the Michelson Prize and to have a chance to talk to you about some of the work that we’re doing to try and understand the contribution of a novel subset of peptides to the immune response in particular, in particular to immune responses to influenza virus.

So today, what I’d like to do is basically give a little bit of background, firstly starting with something that’s quite familiar, which is the classical HLA Class I antigen processing presentation pathway for looking at the generation of spliced peptides and how they enter into this pathway, how we and others are trying to identify these peptides in a high-throughput manner using mass spectrometry, and then where we’re going with this with regards to mapping the contribution of these peptides to the immunopeptidome in influenza infection and looking to see whether these can actually be harnessed in an immune response.

So the classical HLA Class 1 antigen processing presentation pathway is probably fairly familiar to most people in this room, so the HLA molecules are basically expressed by most nucleated cells in the body and their role is to present peptides at the cell’s surface that generate a sort of summary of the protein production ongoing within the cell. So just turn here, we have our proteins within the cell. These are broken down by the proteasome through a hydrolysis reaction and then become available to be transported into the endoplasmic reticulum by the transporter associated with antigen processing where they’re available for binding to the actual HLA molecules, and this stabilizes the HLA molecule allowing it to go back to the cell’s surface and we then have this array of peptides being presented by the HLA, that’s known as immunopeptidome.

So when we have a virus, the virus also is producing proteins within the cell. These get broken down and enter into the same pathway, and hopefully, some of these peptides when they’re presented at the cell’s surface, are actually recognized as noble and stimulate T-cell responses, so a CD8 T-cell response targeted against these specific peptides and then helping eliminate the virus-infected cells.

So the HLA molecules are incredibly diverse across the human population. We have three loci encoding the classical HLA Class 1 and there are actually thousands of different variants. The variations within these molecules predominantly map to this region here which are these antigen binding cleft where we have peptides of 9 to 12 amino acids usually binding within this groove, and what these polymorphisms basically mean is that we have slight differences in the chemistry within this group that bias presentation of peptides towards peptides that have anchor residues that will help them anchor within these anchor pockets. So you can imagine that different HLA molecules, if you sequence large enough peptides isolated from these molecules, you’ll see biases towards particular amino acids at a couple of these anchor locations and this will be different between different variants of the HLA.

And so when it comes to basically defining what peptides are stimulating our CD8 T-cell responses, there’s a few different workflows that you can go about to isolate these peptides, and one quite longstanding workflow is to screen overlapping peptide libraries, so if you have a CD8 T-cell that’s responding to a particular antigen looking for basically taking the sequence of that antigen and generating overlapping peptide libraries, adding them to the antigen-presenting cell, and looking for response, and so this is based on knowledge of the protein sequence encoded by the antigen.

A second approach and one that is basically the bread and butter within our laboratory is the sequencing of peptides isolated from HLA molecules using mass spectrometry. So in this case, we isolate these molecules from cells. We then elute off the peptides and use tandem mass spectrometry to fragment these peptides and determine their sequence by aligning their masses and their fragmentation spectra against the source proteome, and this is what’s known as database searching.

So both of these approaches, however, are based on one really key assumption, and that is that the peptides that are produced by a breakdown of proteins by the proteasome match directly to the protein sequences or the initial protein sequence, however over the last 10 to 15 years, we’re becoming increasingly aware that this isn’t always the case, so the proteasome, as well as performing these hydrolysis reactions, can also perform reactions known as transpeptidation, where it takes two segments, sometimes within the same protein, and it actually stitches these together, so these can be separated by numerous amino acids, or alternatively, it can also do this between peptides from separate proteins, and this is what’s known as trans-splicing. So we have cis-spliced peptides from the same protein or trans-spliced from two different proteins, and this generates, as you can imagine, a much greater diversity within the pool of peptides that are able to be presented by the HLA molecules.

So the first such epitope to be described, this was back in 2004, and this was from FGF5, and it was an epitope that was being targeted by T-cells responding to renal cell carcinoma and presented by the A3 molecule, and it was quite an elegant study using antigen-presenting cells expressing mini-gene constructs of the FGF5 protein, and they showed by truncating and deleting various regions of this protein that the epitope was matched to these two distinct regions here, so these five amino acids here and these four amino acids here stitched together, and in the years since, there’s been a handful of other such spliced epitopes that have been defined in cancer models, and more recently as well, a few such spliced epitopes have been found in listeria infection in mice being presented by mass MHC molecules.

So whilst this does show that these spliced epitopes are involved in immune responses, it doesn’t really give us much idea of how common they are, and this is where mass spectrometry can really come to the fore. So as I mentioned earlier, what we do with regards to trying to identify peptides being presented by the HLA molecules is we take these HLA molecules, we elute off the peptides, and then we use tandem mass spectrometry to basically fragment the peptides and look at their spectra.

So in these data sets, we have all of these peptides available, however when we search against the human proteome or against the viral proteome, we’re only asking it to look for linear peptides. These spliced peptides don’t exist within this database and so when we’re doing this database searching approach, they disappear. They become invisible. We can’t see them.

And so in 2016, Liepe et. al published a beautiful paper in Science where they tried to get around this problem by generating a theoretical spliced peptide database, so what they did was they basically looked at the proteins of the human proteome and generated theoretical spliced fragments basically by taking segments of up to 25 amino acid separation and generating all 9 to 12 amino acid peptides that could be generated by these splicing events, and when they did this, they were able to bring the cis-spliced peptides back into focus, and quite staggeringly, what they found was that around about 30 percent of the immunopeptidome of these different cell lines could be explained by cis splicing, and so this was remarkable. We had no idea that there was that amount of contribution from these amounts.

However, there are some limitations to this workflow. Firstly, it’s computationally quite expensive, so the human proteome is about a 30 megabyte FASTA file when you’re wanting to search this. These databases end up being 200, 300 gigabytes, so your search space is massively expanded, increasing the computational time. Secondly, it only considers cis-splicing and it also is biased against distant recombination partners, so we’re only looking at 25 amino acid separation, and that very first FGF5 epitope was actually separated by 40 amino acids, so it would have been missed using this workflow.

So we’re taking a slightly different approach to try and identify cis and trans-spliced peptides within the immunopeptidome, and this is using a de novo-assisted database searching algorithm which is incorporated within Peaks 8.5 software, and so basically what this does is it actually starts out with the peptide spectra and it comes up with hypotheses with regards to the sequences that could explain that spectra, and it searches for these within the database, so within the source proteome that you’ve given it.

When these are found within the proteome, these are all linear peptides, however high confidence de novo-sequence assignments that fail to be matched to the proteome we then put through our in-house software known as Hybrid Finder looking for explanations of two segments either within a single protein that could give rise to these peptides, so cis-splicing, as shown here, or two separate proteins, so trans-splicing, and we’ve put through various data sets from the immunopeptodome data sets from within our laboratory and also publicly available across various different alleles and what we can find is that we can actually attribute various sequences to either linear, cis, or trans-spliced peptides, and there’s different balances of these in the immunopeptidomes of different alleles.

But what we really want to know with this technique is can we actually find spliced peptides that are produced from viral antigens during infection? So as part of a collaboration with Katherine Kedzierska’s laboratory at the Peter Doherty Institute, we’ve been performing various epitope discovery experiments, and one of the early experiments that we did was looking for influenza B derived epitopes being presented by the common HLA module HLA AO201, and to do this, we took a b lymphoblastic cell line, which was transfected with the A2 molecule.

We then infected this with B Malaysia virus for 12 hours before confirming infection using flow cytometry for the MP protein, and then lysing these infected cells, immunoaffinity purifying out the HLA Class 1 molecules, in this case using an antibody specific for AO201, dilute these acetic acid, separate the proteins from the peptides using reverse phase HPLC, and then sequencing our peptides using tandem mass spectrometry.

Initially, we were just going through a very standard workflow using database searching looking for peptides within the human or the B Malaysia proteome, and so as we expected with this, we found that most of the peptides that we were isolating were 9 to 11 amino acids in length consistent with Class 1 ligands. They were biased towards leucine at Position 2 and valine and leucine at Position 9, consistent with ligands at the A2 molecule, and in amongst these we were able to map some of these peptides or a small number of these peptides to the B Malaysia proteome and these traverse across various proteins within the B Malaysia proteome, and Marios Koutsakos, a Ph.D. student within Katherine’s laboratory, then went on to test these for immunogenicity.

So we separated these into six different pools and took PBMCs from A2 positive individuals, cultured them with the peptide pools, and then after nine to ten days of culture performed a restimulation with peptide pools looking for interferon gamma production by the CD8 T-cells, and in this particular individual, we saw quite robust responses to this Pool 2. This was seen across all of the donors that we tested and we also saw some responses to other pools as well, and these could further be dissected down to the specific peptides within the pool. So this is the peptides within Pool 2 and we see that the response is predominantly to this hemagglutinin and peptide here, and overall, we were able to map responses of varying degrees across different donors to five different peptides from the influenza proteome.

So this looks like an ideal data set for us then to look for cis and trans-spliced peptides derived from the influenza virus, and so we put these data sets through this de novo searching algorithm or de novo searching workflow, and for simplicity’s sake, I’m just going to focus on one of our infection data sets. So using this workflow, firstly, you can see we’re seeing peptides that tend to be 9 to 11 amino acids in length and we see both cis and trans-spliced peptides coming up within the database using this workflow. These map to both the human and the flu proteome, and when we look at those that are mapping to the flu proteome, we see an array of the linear peptides across the proteome as we’ve seen previously, so consistent with our previous workflow, but with our cis-spliced peptides, we saw these quite striking enrichment within the hemagglutinin molecule.

What was even more exciting when we turned down and looked at these actual sequences that were providing these hits is that most of these sequences are actually sharing a C terminal splice fragment which is sitting here within the hemagglutinin protein, and these could either be explained by cis-splicing with portions of the hemagglutinin protein N terminal to it or C terminal to it, and we also saw another peptide here that actually had this segment slightly extended at the N terminus of the spliced peptide as well as a couple of linear epitopes that were overlapping this section, whilst our remaining linear epitopes were spread throughout the protein.

And so to confirm these sequences, we’ve ordered in synthetic peptides for some of these and what you can see is when we fragment the synthetic peptide, we get an exact match between this peptide spectra of this synthetic and the eluted peptides, giving us high confidence in our sequence assignment.

So this is where we stand at the moment and I would have to say that we actually have more questions than we have answers. Really what we want to know right now is do these peptides stimulate immune responses during infection? If so, can they be harnessed to prime an immune response? Can we also predict what antigens and regions of proteins are prone to splicing? We saw this massive enrichment of spliced peptides derived from a single region within the hemagglutinin molecule. Are similar regions being presented by different HLA or will we see enrichment of other parts of the protein or different antigens? And we really don’t know the answers to these questions at the moment.

So what we’re moving to do now is to basically expand our analysis across influenza A and B viruses looking at a couple of common HLA molecules, so the A2 molecule as well as the B8 molecule, and to try and define the rules for peptide splicing in the presentation of spliced peptides from a virus, and then also moving to test the epitopes or potential epitopes that we detect alongside known and linear HLA ligands in hopes that we can identify novel contributors to the anti-influenza T-cell response that might be novel candidates for epitope-based vaccines.

So it just remains to thank all of those that have been involved in this work and the work going forward, and thank you.


Marie-Paule Kieny: Thanks a lot, Patricia. Congratulations, again.

Patricia Therese Illing: Thank you.

Marie-Paule Kieny: Yes, go ahead.

Audience: Do you also see evidence of modified peptides that might have a phosphorylation moity or something like that?

Patricia Therese Illing: So in this workflow, we’ve actually tailored it so that we’re not really considering the modifications ’cause it expands the search space quite astronomically. We have done some assessments where we actually do go through and look with various modifications to make sure we’re not misassigning spliced peptides to modified peptides, but at this stage, no, we’re not doing that. That might be something to consider going forward but it does massively expand the search space.

Marie-Paule Kieny: Let’s go ahead.

Audience: That was beautiful. I have a question about where the peptides are coming from. What percent of the peptides come from non-protein coding sequences, so non-splicing RNAs or things like that?

Patricia Therese Illing: So from the viral proteins or from –

Audience: From _____ _____.

Patricia Therese Illing: So at the moment, we have been searching predominantly against what’s recognized as the express proteome. We have gone in and done some searches also against basically different reading frames and going back and forth to make sure we’re misassigning to things that are coming from alternate reading frames. We don’t believe we are. It doesn’t look like it. We don’t get many hits in that regard, but it’s all an open-ended question at the moment.

Marie-Paule Kieny: Okay. I see no more questions so thank you very much and –


Transcript curation: Alison Deshong

Dr. Patricia Therese Illing, PhD, Monash University; 2018 Michelson Prize Winner for Human Immunology and Vaccine ResearchDr. Patricia Therese Illing – 2018 Michelson Prize Winner for Human Immunology and Vaccine Research.

Dr. Patricia Therese Illing; Spliced Peptides — Novel Candidates for effective influenza immunity? 1st Annual Conference on the Future of Vaccine Development. [2018-06-27] {#151} (Credit: Marv Steindler / Steve Cohn Photography)Dr. Patricia Therese Illing checks on a slide of her presentation ‘Spliced Peptides — Novel Candidates for effective influenza immunity?’ during the 1st Annual Conference on the Future of Vaccine Development. [2018-06-27] {#0151} (Credit: Marv Steindler / Steve Cohn Photography)

Dr. Patricia Therese Illing lectures on spliced peptides at the 1st Annual Conference on the Future of Vaccine Development.Professor Ian Gust and Dr. Wayne Koff, founder of the Human Vaccines Project, attend Dr. Patricia Therese Illing’s presentation ‘Spliced Peptides — Novel Candidates for effective influenza immunity?’ during the 1st Annual Conference on the Future of Vaccine Development. [2018-06-27] {#0161} (Credit: Marv Steindler / Steve Cohn Photography)

Dr. Patricia Therese Illing, Ian Gust AO; The Future of Vaccine Development Symposium Dinner. [2018-06-27] {#0667} (Credit: Marv Steindler / Steve Cohn Photography)Dr. Patricia Therese Illing receives the Michelson Prize Award from the hands of distinguished Professor Ian Gust AO, founder director of the Burnet Institute and recipient of the 2016 Peter Wills Medal. [2018-06-27] {#0667} (Credit: Marv Steindler / Steve Cohn Photography)

Dr. Patricia Therese Illing, Dr. Ansuman Satpathy, Dr. Laura Kate Mackay, Michelson Prizes 2018, Dr. Gary K. Michelson, Wayne Koff, PhD, Ian Gust AO, Dr. Steve A. Kay; The Future of Vaccine Development Conference. {#0005} [2018-06-27. Marv Steindler / Steve Cohn Photography](Left to Right) The 3 winners of the 2018 Michelson Prize for Human Immunology and Vaccine Research: Dr. Patricia Therese Illing, Dr. Ansuman Satpathy and Dr. Laura Kate Mackay pose with Dr. Gary K. Michelson, Founder of the Michelson Medical Research Foundation, Wayne Koff PhD, Ian Gust AO of the Human Vaccines Project and Steve A. Kay, Director of the USC Michelson Center for Convergent Bioscience, for a group picture ahead of the 1st Annual Conference on the Future of Vaccine Development held at the USC Michelson Center for Convergent Bioscience. [2018-06-27] {#0005} (Credit: Marv Steindler / Steve Cohn Photography)

Ian Gust AO, Dr. Ansuman Satpathy, Dr. Laura Kate Mackay, Dr. Gary K. Michelson, Wayne Koff, PhD, Steve A. Kay, PhD, Dr. Patricia Therese Illing, Michelson Prizes 2018; The Future of Vaccine Development Conference. {#0011} [2018-06-27. Marv Steindler / Steve Cohn Photography]Ian Gust, AO (Human Vaccines Project), Dr. Ansuman Satpathy, Dr. Laura Kate Mackay (2018 Michelson Prize winners of the Human Immunology and Vaccine Research Prize), Dr. Gary K. Michelson (Michelson Medical Research Foundation), Wayne Koff, PhD (Human Vaccines Project), Steve A. Kay, PhD (USC Michelson Center for Convergent Bioscience) and Dr. Patricia Therese Illing (2018 Michelson Prize winner of the Human Immunology and Vaccine Research Prize) pose outside the USC Michelson Center for Convergent Bioscience ahead of the 1st Annual Conference on the ‘Future of Vaccine Development’. [2018-06-27] {#0011} (Credit: Marv Steindler / Steve Cohn Photography)

Back to Top