We are searching data for your request:
Upon completion, a link will appear to access the found materials.
The cell still has mutations, but those mutations only occur in the noncoding sequences, such as promoters, which drives over expression of proto-oncogenes and downregulation of tumor suppressor genes. So the cell will proliferate uncontrollably, but won't express any neoantigens. Will such cell be invisible to our immune system?
Immunosurveillance of cancer cells is predicated upon multiple "signals" derived from the cancerous cells and their microenvironment.
I should point out tumor mutational load (Figure 1). A sort of take-home point here is that certain indications have less mutations and thus less neoantigens than others. You also have to take into consideration the stage of the tumor, though. Stage IV tumors have amassed more mutations than stage I/II.
Figure 1. Mutational prevalence by indication. A higher prevalence predicts a higher neoantigen frequency or mutational load for that cancer histology. (1, 2)
It's interesting to point out the bottom part of figure 1 where the authors of a separate review note for us where it's believed the disconnect between traditional checkpoint blockade immunotherapy and CAR-T immunotherapy exists. Just to educate a bit on the subject, checkpoint blockade targets T cells, the drivers of antigen-dependent cell killing. T cells naturally infiltrate the tumor as it develops and attempt to mount an attack. One of the methods tumors use to evade T cell killing is through persistent antigen, expression of inhibitory receptors and secretion of inhibitory molecules. Checkpoint blockade immunotherapy uses antibodies against the inhibitory receptors such as PD-1 (Opdivo, Keytruda), it's ligand PD-L1, CTLA-4 (Yervoy), Tim3, Lag3, and so forth. This attempts to lift the dampening effect these receptors have on T cells so that they may resume killing. The T-antigen-dependent paradigm is also supported by such publications which look at checkpoint blockade in the setting of clonal T cell diversity and confirm that (1) a diverse repertoire of T cell clones result in successful checkpoint blockade, and (2) enrichment of certain clonotypes within the repertoire, presumably the antigen-specific clonotypes, further indicates a response (3).
So below that are indications circled in red, notably the histologies with a low neoantigen frequency. The idea behind CAR T cells is that I can take advantage of T cell biology, notably that T cells recognize antigen with a receptor and kill the target cell. So just backing up to checkpoint blockade immunotherapy real quick, say a patient has melanoma which can consist of 1000s of mutations and 100s of neoantigens. There is a greater chance that an infiltrated T cell has specificity for any particular antigen. So how about a glioblastoma patient? They may have 10-20 neoantigens, or as you said they may have no neoantigens, at least none that their T cells are specific against. So checkpoint blockade may not be so effective. CAR stands for chimeric antigen receptor. It's essentially a monoclonal antibody against one target connected to a linker connected to the intracellular domain of a T cell receptor. Hematologic malignancies such as NHL like diffuse large B-cell lymphoma are commonly targeted for a B cell receptor known as CD19. So all you have to do is make your CAR CD19 specific and transfect the T cells with it in place of their own T cell receptor or TCR. So the basic process behind CAR is I isolate your T cells as with leukapheresis, lentivirally transfect with my anti-CD19 CAR, expand, and infuse. There are two CD19 CAR T products with great response rates on the market (Kymriah, YESCARTA). The whole reason these are good is because we can direct them to whatever antigen we want (of course be careful about self-antigens, see autoimmunity).
Figure 2. Schematic of CAR T cell therapy. (4)
The CAR T approach is in fact because the neoantigen load is low or non-existent, which is problematic for un-modified T cells, and thus problematic for standard-of-care treatments like checkpoint blockade.
As such, there are other ways, and other cells that can handle some of the abnormalities presented by tumors as a whole. This includes the presence of danger/damage-associated molecular patterns (DAMPs) and the regulation of surface receptors, for example, NK receptor ligands such as MIC.
The way DAMPs work is that a milieu of molecules are secreted by cells that are damaged or transformed, signaling cells of the innate immune system. One of the issues is that the DAMPs promote an inflammatory environment. This means that yes, you get immune cells checking out the situation, but that some of these are good immune cells and some are "bad" (myeloid-derived suppressor cells (MDSC), tumor-associated M2 macrophages (TAM), or T-regulatory cells (Treg)). These dampen the immune response both through inhibitory cytokine release like IL-10, and expression of inhibitory receptors like PD-L1.
Figure 5. The molecules secreted by the inflammatory tumor microenvironment both help and hurt the immune response. (5).
And then the idea behind NKR ligands is pretty straightforward: Natural killer (NK) cells patrol your body routinely for abnormalities and are licensed to kill cells that are abnormal. For example, all nucleated cells express class I major histocompatibility complex (MHC-I) on their surface. Some receptors like MIC are also present on the cell surface that engage receptors on the NK cells like NKG2D in distress. So what do cancers do? They sometimes express truncated and thus inactive MIC, express no MHC but also express inhibitory receptors to NK cells, and so forth.
Everything is summed nicely in the following figure on cancer immunoediting (Figure 7). Note that while things are working normally, the immune system can handle the tumor load. As the scales tip, and this could be for example, your T cell response edits out an immunoantigen but doesn't take care of the whole tumor, resulting in growth that those T cells no longer respond to. As this happens, the cancer becomes increasingly invasive.
Figure 7. Cancer immunoediting. (7).
So just to clarify at the end, there probably isn't a case where the immune system isn't responding at all unless you had chemotherapy that destroyed your immune system. Inflammatory environments alone are sufficient for your immune cells to at least check out the tumor. In fact, and I cant find the paper now but I'll try, investigators even found naïve T cells in the tumor microenvironment simply due to the inflammation. We could also get into "hot" and "cold" tumors, but that's out of scope here (take-home point, cold tumors aren't inflamed and are known to contain less immune infiltrate). I used immunotherapy as a vehicle to get my point across, note that this answer isn't about immunotherapy but rather about immune escape and mutational load theory.
No, Really, mRNA Vaccines Are Not Going To Affect Your DNA
The short version: There is no plausible way that mRNA vaccines are going to alter your DNA. It would violate basically everything we know about cell biology.
Lodish H, Berk A, Kaiser C, Krieger M, Bretscher A, Ploegh H, Amon A, Martin K. Molecular cell biology. 8th ed. New York: W.H. Freeman 2016. Figure 5-1
This figure is useful because you can clearly see the two compartments we care about: the nucleus, which houses almost all of the DNA (exception discussed), and the cytosol, which is where translation happens.
Recently, I’ve gotten an influx of questions about how we can really be sure that mRNA vaccines will not affect our DNA. In my prior post on the subject I wrote:
Lodish H, Berk A, Kaiser C, Krieger M, Bretscher A, Ploegh H, Amon A, Martin K. Molecular cell biology. 8th ed. New York: W.H. Freeman 2016. Figure 13-37B which demonstrates the export of mRNA from the nucleus.
Another concern raised has been the idea that mRNA can somehow alter the host’s genome. That would actually be super cool and be huge for gene therapy (and I could finally give myself the giant bat wings I’ve always wanted) but this is not so. This is ordinarily impossible except if there is also a reverse transcriptase enzyme present that produces DNA from the RNA template, which is how retroviruses work. There is no such risk with any mRNA vaccine candidate. mRNA vaccines act entirely within the cytosol of the cell- they do not go near the nucleus where all the DNA is. That’s actually a major advantage of RNA-based vaccines over DNA ones.
Flint S, Racaniello V, Rall G, Skalka A, Enquist L. Principles of virology. Washington, DC: ASM Press 2015. Figure 5.23C which shows the nuclear import cycle with influenza ribonucleoproteins as an example. Nucleear localization signals are recognized by importin-α which then recruits importin-β which recruits a small GTPase called Ran. When binding GDP, Ran is able to transport the RNP across the nuclear pore complex. The complex of importins and Ran will then dissociate, and a guanine nucleotide exchange factor will exchange GDP for GTP, that allows Ran-GTP to be exported out of the nucleus. RanGAP-1 or RanBPI, 2 can then catalyze hydrolysis of GTP to GDP which allows Ran to bind another importin-β to initiate the import cycle once more.
I gave this response in large part because I felt that the detailed discussion of reverse transcription, nuclear trafficking, the endocytic pathway, and the other 11 or so advanced cell biology topics that I would have to invoke to give this a rigorous answer was too complex to be of benefit to the average person wanting to know simply whether or not this is possible. However, I had a flurry of questions about “what ifs” relating to retroviruses or hepadnaviruses (hepatitis B), and I can grant that this response doesn’t address that, so here I will attempt to answer that as explicitly and with minimal complexity as I am capable.
To simplify the discussion so as to avoid having to explain the phases of phospholipid bilayers and the molecular composition of the lipid nanoparticle as it relates to stability (discussed in 1, 2, 3, 4), I will ask readers take for granted that mRNA vaccines are endocytosed and liberated (and this) into the cytoplasm of the cell.
Firstly, for mRNA to affect your DNA, at a minimum we need to establish that it would need to gain access to the DNA in question. There are two subcellular compartments where this can be accomplished. The first is the nucleus, so let’s start with a discussion of the trafficking of cargo in the nucleus. The nucleus of the cell is an isolated compartment with pore complexes (NPCs) that impose limits on the size of the particles that can freely enter. RNA is readily transported out as transcription occurs within the nucleus but the ribosomes required to produce proteins are in the cytosol or on the rough endoplasmic reticulum. This process is mediated by several accessory proteins which you can see to your right. Note however that there isn’t any physiological circumstance in which one might need RNA from the cytosol to be transported back to the nucleus. RNA is synthesized within the nucleus. Viruses which have a nuclear phase in their replication cycle have to have various tricks to be able to allow their RNA payload to enter. Though RNA is not readily transported into cells, proteins can be. This occurs via a network of proteins called importins (see figure 5-23C on the right). Proteins containing an amino acid sequence called the nuclear localization sequence (NLS there are 2 common ones) are able to bind the importins, which can then transport them across the nuclear pore complex as shown on the right. RNA viruses often have replication cycles that do not require access to the nucleus, but there are some exceptions. Influenza viruses for example are RNA viruses that have their genomes associated with ribonucleoproteins, and these ribonucleoproteins express nuclear localization signals that facilitates the entry of their genomic RNA into the nucleus. mRNA vaccines, on the other hand, are not associated with any proteins. Once inside the cytosol, the mRNA is naked and exposed to the harsh environment of ribosomes and exonucleases which destroy the mRNA in a matter of hours (at most). There is no conceivable mechanism by which mRNA can spontaneously be trafficked into the nucleus. Being made of nucleotides, it cannot contain a nuclear localization sequence.
The other relevant compartment would be the mitochondrion. Mitochondria are actually vestigial bacteria with their own genomes, and it’s thought that billions of years ago an ancient cell (probably an archaean- the cousins of bacteria) tried to consume the ancestor of the mitochondria but lacked the machinery to actually do the digesting and the two established a symbiotic relationship. Since that instance, the mitochondria have been an essential feature of our cell’s biologies. This allowed the mitochondria to develop an extremely reduced genome containing only 37 genes (most of the genes relevant to mitochondrial function are still in the nucleus). Mitochondria have their own ribosomes and even their own genetic code (sort of). There is also a specialized process for the clearance of diseased mitochondria called mitophagy, which is the subject of many excellent reviews e.g. this, this, and this.
The collective conclusion from our understanding of these biological process is that a naked mRNA in the cytosol has no potential to end up in a cellular compartment that contains our own DNA means that, irrespective of the presence or absence of other factors, there is no chance of harm to the DNA from the mRNA vaccine. But still people wanted to ask me about reverse transcriptases so let’s discuss those.
The process of going from RNA to DNA (the exact opposite of what the central dogma of molecular biology dictates) is known as reverse transcription, and it is carried out with an enzyme called a reverse transcriptase (which are a really interesting group of enzymes). In general, reverse transcription is performed by a few different genetic entities: retroviruses, hepadnaviruses, telomeres, and retrotransposons. These are worth defining.
Retroviruses are viruses who have an RNA genome, from which they create a DNA copy through reverse transcription that then integrates into the cell of the host (by which I mean, literally inserts itself into the host cell’s genome and becomes a permanent part of it, in the form of a sequence called a provirus). The proviral sequence itself can then be transcribed in the host cell to produce viral proteins and particles that can go on to spread to the next cell. The most famous retrovirus is HIV-1.
Hepadnaviruses are DNA viruses which have gapped genomes (there is one complete DNA strand and another partial DNA strand which is linked to a pregenomic RNA), and unlike retroviruses, do not integrate into the genome of the host cell they infect. The most famous example is Hepatitis B virus, for which multiple effective vaccines exist.
Telomeres are structures present at the ends of human chromosomes which are maintained by a protein complex called telomerase that uses a reverse transcriptase called TERT to maintain them. The reasons this is necessary are discussed below. They are about 5-15 kilobases long normally, and shortening results in arrest of cell growth and replication (senescence), or can even trigger cell death by apoptosis.
Get full journal access for 1 year
All prices are NET prices.
VAT will be added later in the checkout.
Tax calculation will be finalised during checkout.
Get time limited or full article access on ReadCube.
All prices are NET prices.
Rebecca&rsquos family tree, as illustrated in Figure (PageIndex<1>), shows a high incidence of cancer among close relatives. But are genes the cause of cancer in this family? Only genetic testing, which is the sequencing of specific genes in an individual, can reveal whether a cancer-causing gene is being inherited in this family.
Figure (PageIndex<1>): Pedigree for Rebecca's family, as described at the beginning of this chapter, showing individuals with cancer (red) and those that do not have cancer (blue). Circles represent women, squares represent men.
Fortunately for Rebecca, the results of her genetic testing show that she does not have the mutations in the BRCA1 and BRCA2 genes that most commonly increase a person&rsquos risk of getting cancer. However, it does not mean that she doesn&rsquot have other mutations in these genes that could increase her risk of getting cancer. There are many other mutations in BRCA genes whose effect on cancer risk is not known, and there may be many more yet to be discovered. It is important to continue to study the variations in genes such as BRCA in different people to better assess their possible contribution to the development of the disease. As you now know from this chapter, many mutations are harmless, while others can cause significant health effects, depending on the specific mutation and the gene involved.
Mutations in BRCA genes are particularly likely to cause cancer because these genes encode for tumor-suppressor proteins that normally repair damaged DNA and control cell division. If these genes are mutated in a way that causes the proteins to not function properly, other mutations can accumulate and cell division can run out of control, which can cause cancer.
BRCA1 and BRCA2 are on chromosomes 17 and 13, respectively, which are autosomes. As Rebecca&rsquos genetic counselor mentioned, mutations in these genes have a dominant inheritance pattern. Now that you know the pattern of inheritance of autosomal dominant genes if Rebecca&rsquos grandmother did have one copy of a mutated BRCA gene, what are the chances that Rebecca&rsquos mother also has this mutation? Because it is dominant, only one copy of the gene is needed to increase the risk of cancer, and because it is on autosomes instead of sex chromosomes, the sex of the parent or offspring does not matter in the inheritance pattern. In this situation, Rebecca&rsquos grandmother&rsquos eggs would have had a 50% chance of having a BRCA gene mutation, due to Mendel&rsquos law of segregation. Therefore, Rebecca&rsquos mother would have had a 50% chance of inheriting this gene. Even though Rebecca does not have the most common BRCA mutations that increase the risk of cancer, it does not mean that her also mother does not, because there would also only be a 50% chance that she would pass it on to Rebecca. Therefore, Rebecca&rsquos mother should consider getting tested for mutations in the BRCA genes as well. Ideally, the individuals with cancer in a family should be tested first when a genetic cause is suspected so that if there is a specific mutation being inherited, it can be identified and the other family members can be tested for that same mutation.
Mutations in both BRCA1 and BRCA2 are often found in Ashkenazi Jewish families. However, these genes are not linked in the chromosomal sense, because they are on different chromosomes and are therefore inherited independently, in accordance with Mendel&rsquos law of independent assortment. Why would certain gene mutations be prevalent in particular ethnic groups? If people within an ethnic group tend to produce offspring with each other, their genes will remain prevalent within the group. These may be genes for harmless variations such as skin, hair, or eye color, or harmful variations such as the mutations in the BRCA genes. Other genetically based diseases and disorders are sometimes more commonly found in particular ethnic groups, such as cystic fibrosis in people of European descent and sickle-cell anemia in people of African descent. You will learn more about the prevalence of certain genes and traits in particular ethnic groups and populations in the chapter on Human Variation.
As you learned in this chapter, genetics is not the sole determinant of phenotype. The environment can also influence many traits, such as adult height and skin color. The environment also plays a major role in the development of cancer. 90 to 95% of all cancers do not have an identified genetic cause and are often caused by mutagens in the environment such as UV radiation from the sun or toxic chemicals in cigarette smoke. But for families like Rebecca&rsquos, knowing their family health history and genetic makeup may help them better prevent or treat diseases that are caused by their genetic inheritance. If a person knows they have a gene that can increase their risk of cancer, they can make lifestyle changes, have early and more frequent cancer screenings, and may even choose to have preventative surgeries that may help reduce their risk of getting cancer and increase their odds of long-term survival if cancer does occur. The next time you go to the doctor and they ask whether any members of your family have had cancer, you will have a deeper understanding of why this information is so important to your health.
Lying with science
So let’s take a look at the press release and the scientific paper the press release was promoting. The press release is entitled “In a Twist, Scientists Find Cancer Drivers Hiding in RNA, Not DNA“, and the paper was published in Nature by Christine Mayr, a molecular and cellular biologist at MSKCC who studies mRNA. The study is entitled “Widespread intronic polyadenylation inactivates tumour suppressor genes in leukaemia“, which is, of course, not the sort of title that most people would understand. Don’t worry. I’ll explain.
But first, I need to explain some background. When discussing the mRNA-based COVID-19 vaccines and how they don’t “reprogram your DNA” and are not “gene therapy“, I discussed how DNA encodes RNA that ultimately encodes protein. Before I discuss what the paper actually shows and why Wells is misapplying its findings, I think it helpful to review the basics of this aspect of molecular biology again. Here is a very basic diagram of the process:
The “Central Dogma of Molecular Biology”. Information flows from DNA to RNA and then is used to make protein.
Basically, DNA replicates from a DNA template and results in a double-stranded molecule that is very stable, as it has complementary sequences that tightly bind to each other in a sequence-specific fashion. This DNA template is unwound by enzymes that use the template to make RNA strands, which are single-stranded, which is then used by a ribosome to make protein out of amino acids. Again, to put it simply, each nucleotide equals one letter of the code each three-nucleotide sequence (codon) equals one “word” that translates to an amino acid. Given that there are four nucleotides, there are 64 possible codons. Since there are only 20 amino acids, that means that most amino acids are encoded by more than one combination of nucleotides or more than one codon i.e., the genetic code is redundant. Of course, it’s more complicated than that, as this diagram shows:
After the genetic code was cracked 60 years ago, it soon became apparent that the RNAs encoding for proteins are often not fully formed right after they’re transcribed. Often RNA starts out as a longer precursor RNA (a pre-mRNA) that is spliced to the final mRNA sequence before being transported out of the nucleus into the cytoplasm to be used to drive the production of protein in the cytoplasm. In brief, the precursor RNA that is initially transcribed contains sequences known as “exons” and “introns”. In genes, exons contain the nucleotide sequences that encode actual protein, while introns contain nucleotide sequences that do not code for anything but can have important sequences that regulate gene production and activity. Here’s an illustration of the splicing process from Wikipedia:
This diagram is actually fairly simple, with two exons and one intron. Some genes have many exons and introns, requiring multiple splices, as in this diagram:
Did I forget to mention that mRNAs are also processed to have a “cap” at one end and a stretch of As (a poly-A tail) at the other end? The poly-A tail is very important in regulating mRNA stability and therefore its half-life in the cytoplasm. In any event, as with any biological process, things can go wrong with these splicing events. Splice site mutations, for instance, can result in mis-spliced mRNAs and proteins lacking exons:
I could go on and on and on. There are normal genes can produce more than one protein through alternative splicing:
Sometimes when splicing goes awry, it can result in a truncated protein lacking one end. For instance, if an intron is left attached to two exons, chances are that the ribosome (the enzyme complex that translates mRNA into protein) will hit a “stop” codon (three nucleotide code that tells transcription to stop) long before it reaches the other end of the intron, at which point transcription will just stop. All of this is not even counting the other molecular modifications that the RNA can undergo on its journey from transcription to pre-mRNA through splicing to the final “mature” mRNA.
Unsurprisingly, if these sorts of errors occur in genes important to processes regulating cell growth and invasion, cancer can result, either from a mis-splicing removing a regulatory region in the protein that keeps it in check or by producing a protein that doesn’t function as it should. Examples of cancers that are caused or accelerated by a splice site mutation are accumulating. It is this latter possibility, a truncated protein that doesn’t function, that the paper being misapplied by Wells examines. The proteins compromised by the truncation are tumor suppressor proteins, whose function is to shut down growth or other processes that can result in cancer when they are overly active.
So what did the paper show? What was interesting about the paper is that it showed the existence of splicing errors in tumor suppressor genes in a specific cancer, chronic lymphocytic leukemia, without mutations in splice sites to explain how these proteins became truncated due to splicing errors, or, as the authors put it in the manuscript:
We discovered widespread upregulation of truncated mRNAs and proteins in primary CLL cells that were not generated by genetic alterations but instead occurred by intronic polyadenylation
The truncated proteins generated by intronic polyadenylation often lack the tumour-suppressive functions of the corresponding full-length proteins (such as DICER and FOXN3), and several even acted in an oncogenic manner (such as CARD11, MGA and CHST11). In CLL, the inactivation of tumour-suppressor genes by aberrant mRNA processing is substantially more prevalent than the functional loss of such genes through genetic events.
So what does this all mean? First, to reiterate and simplify, the authors detected truncated mRNAs and proteins for a number of tumor suppressor genes in CLL that could not be explained by DNA mutations in the genes themselves, such as splice site mutations. They did a lot of other controls, such as making sure that the explanation for the truncated proteins was not cleavage by proteases, enzymes that cut proteins at specific amino acid sequences. After ruling out other possibilities, the authors demonstrated that these mRNAs and proteins were truncated because of a process called intronic polyadenylation. But what is that?
Polyadenylation is the process of adding a bunch of adenosines (As) to the 3′ end of an RNA molecule. It’s how the poly-A tail is added to the end of an mRNA, but it turns out that it’s a common process for polyadenylation to occur in introns. This process is very widespread and was appreciated well over a decade ago. It’s involved in diversifying the products of immune cell mRNAs, the process explained thusly:
In the splicing literature, isoforms generated through recognition of an IpA signal are often described as ‘alternative last exon’ events. Genes that generate IpA isoforms are thought to harbor competing splicing and polyadenylation signals, producing a full-length messenger RNA (mRNA) when splicing outcompetes polyadenylation and otherwise producing a truncated mRNA. As the defining event is the recognition of an IpA signal, we call these transcripts IpA isoforms. It is now possible to recognize the widespread expression of IpA isoforms through the analysis of 3ʹ-end sequencing data.
Or, to put it more simply, whether there is a truncated protein or a full-length protein depends on the balance of splicing to poly-adenylation at the site in the intron. If there’s more splicing activity, you get much more of the whole protein. If there’s more polyadenylation, you get much more of the truncated protein. What Mayr’s lab found was that too much polyadenylation can result in truncated tumor suppressor proteins in CLL, contributing to the development of the cancer, which is why, according to the MSKCC press release which explains the findings pretty well:
These findings help explain a long-standing conundrum, which is that CLL cells have relatively few known DNA mutations. Some CLL cells lack even known mutations. In effect, the mRNA changes that Dr. Mayr’s team discovered could account for the missing DNA mutations.
Because CLL is such a slow-growing cancer and people with CLL often live for many years, it’s too early to say whether these mRNA changes are associated with a poorer prognosis.
There are some important differences between the mRNA changes and a bona fide DNA mutation. Most important, the inactivation of tumor suppressors through mRNA is usually only partial only about half of the relevant protein molecules in the tumor cells are truncated. But in many cases this is enough to completely override the function of the normal versions that are present. And because this truncation could apply to 100 different genes at once, the changes can add up.
So why does none of this have anything to do with mRNA-based vaccines causing cancer? I’m glad you asked and hope you don’t mind that I took this opportunity to geek out a bit, in a molecular biology sense. The biology being abused by Wells is actually quite complex and fascinating, and I don’t get to discuss pure molecular biology very often any more. I hope I didn’t lose too many readers with the explanation, but I also bet more than a few of you have already figured out why what Wells is peddling is utter nonsense. If not, here we go.
The complex role of NMD in cancer
Tumors have found ways to leverage their unconstrained growth and the progression of cancer by taking advantage of both functions of NMD. On one hand, cancer cells use NMD activity to selectively downregulate tumor-suppressive genes through PTC acquisition, but on the other hand, these cells fine-tune NMD magnitude to allow the upregulation of stress-corrective genes responsible for their adaptation to the tumor microenvironment [21, 26]. Although apparently straightforward, the role of NMD in cancer development and progression can, in fact, be quite complex [2, 26, 148, 149]. While in some contexts NMD may work as a tumor-suppressive pathway, in other scenarios its activity might aggravate the disease, depending on the genetic evolutionary history of the tumor [2, 151].
NMD activity against tumorigenesis
Disabling mutations in genes encoding NMD factors have been found in several types of cancers, which raised the possibility of NMD having some type of protective role against tumorigenesis. For instance, in pancreatic adenosquamous carcinoma (ASC) tumors and in lung inflammatory myofibroblastic tumors (IMTs), UPF1 gene exhibits splicing-altering mutations that compromise NMD activity [27, 28]. This may lead to the upregulation of genes typically controlled by NMD that, therefore, contribute for the disease phenotype. For example, the NMD target encoding the mitogen activated protein kinase kinase kinase 14 (MAP 3 K14 or NIK), a potent activator of the proinflammatory nuclear factor-kappa B (NF-κB) signaling pathway, is upregulated in the UPF1-mutated IMTs, promoting chemokine production and immune infiltrations that characterize this type of tumors . Similarly, lung adenocarcinomas (ADCs) and hepatocellular carcinomas (HCCs) display lower expression of UPF1 when compared to normal tissues due to promoter hypermethylation, a finding that correlates with poor prognosis in patients with HCC [147, 152]. The resultant impairment of the NMD pathway leads to the upregulation of factors from the transforming growth factor beta (TGF-β) signaling pathway, which drive epithelial-mesenchymal transition (EMT) and, consequently, tumorigenesis and neoplasm metastasis . In line with these clinical findings, there is experimental data showing that overexpression of UPF1 reduces the number and size of cultured tumor cell colonies when compared to control cells of several cancer cell lines . Also, UPF1-overexpressing prostate cancer (PC3) cells injected as tumor explants in nude mice present no significant tumor growth . Interestingly, an expression array analysis performed in human bone osteosarcoma (U2OS) cells subjected to NMD inhibition revealed that NMD controls the expression of a wide variety of transcripts that encode important factors involved in tumorigenesis-related processes, including cell growth, cell cycle, growth factor signaling, apoptosis and cell migration . Altogether, it seems that typically NMD works as a tumor suppressor pathway by regulating the expression of genes involved in cell proliferation, differentiation and survival, and that tumors with impaired NMD, like the ones with mutated UPF1, have favorable conditions for tumor proliferation (Fig. 3a). Further supporting this idea is a study reporting that UPF3A, a paralog of UPF3B that inhibits NMD by sequestering UPF2 , is highly expressed in metastatic tissues of colorectal cancer (CRC) when compared to primary tissues . This higher expression is associated with liver metastasis, recurrence and poor prognosis in CRC patients . This finding was also reported in a previous study in which the interaction of the oncogenic transcription factors, signal transducer and activator of transcription 3 (STAT3), glioma oncogene homolog 1 (GLI1), and truncated GLI1 (tGLI1), was seen to promote UPF3A upregulation . That higher UPF3A expression enhances the aggressiveness of triple-negative and human epidermal growth factor receptor 2 (HER2)-enriched breast cancers and worsens metastasis free survival, according to the gene expression profile of breast tumors retrieved from the GEO database . Thus, it is possible that some tumors with fully functional NMD factors may have found another way to control NMD activity by increasing the expression of modulators, like UPF3A, to leverage tumorigenesis and/or drive the disease (Fig. 3a).
Roles of nonsense-mediated mRNA decay (NMD) in cancer. a Disabling mutations or changes in the gene expression level of the key NMD factors (UPF1 as an example) occur in different cancer types [for example, lung inflammatory myofibroblastic tumor (IMT), pancreatic adenosquamous carcinoma (ASC), lung adenocarcinoma (ADC), and hepatocellular carcinoma (HCC)]. The case on the left represents mutated (Mut) UPF1, which leads to a decreased NMD activity resulting in the upregulation of an NMD target encoding the mitogen activated protein kinase kinase kinase 14 (MAP 3 K14) and stimulating NF-κB (nuclear factor kappa-light-chain-enhancer of activated B cells) activity, thus inducing chemokine production and immune infiltrations. The example in the middle illustrates lower NMD activity due to the downregulation of UPF1, which causes the upregulation of several factors of the transforming growth factor beta (TGF-β) pathway. This favors the epithelial-mesenchymal transition (EMT) and consequently, the number of metastatic events. The example on the right shows the interaction between three oncoproteins, STAT3, GLI1 and tGL1, to induce higher protein levels of UPF3A, which inhibits NMD activity, increasing malignant progression of the tumor. b Tumor suppressor genes can completely loss their function by PTC acquisition and subsequent NMD degradation, combined with either deletion of the wild-type allele, or haploinsufficiency of the remaining allele. On the other hand, a tumor suppressor gene can experience an NMD-resistant mutation, leading to a dominant-negative protein that hampers the wild-type function. c The tumor microenvironment modulates NMD in order to overcome different types of cellular stresses associated with the unconstrained growth of the tumor. Stresses such as hypoxia, production of reactive oxygen species (ROS), or nutrient deprivation promote, eIF2α phosphorylation, which inhibits NMD and therefore, several mRNAs encoding stress-responsive factors are stabilized, allowing tumor progression and adaptation. WT: wild type PTC: premature termination codon aa: amino acid
In addition to its role in gene expression regulation, NMD ability to degrade PTC-harboring transcripts may also be involved in the protection of cancer development. It was shown that tumor suppressor genes have a higher propensity to acquire nonsense mutations than oncogenes, which present mostly missense mutations . Moreover, many of these nonsense mutations are predicted to induce NMD . Indeed, several tumor suppressor genes were found to present PTC-introducing mutations in a plethora of cancers, like p53 in mantle-cell lymphoma  and breast cancers , E-cadherin (CDH1) gene in hereditary diffuse gastric cancers (HDGC) , retinoblastoma 1 (RB1) gene in mantle-cell lymphoma , breast cancer type 1 susceptibility protein (BRCA1) gene in breast and ovarian cancers , and breast cancer type 2 susceptibility protein (BRCA2) gene in breast cancers . Interestingly, the abnormal transcripts that result from these mutated tumor suppressor genes are stabilized in conditions of inhibited NMD, suggesting that this pathway is usually responsible for degrading them [20, 22, 23, 25, 155]. Therefore, in these contexts, the quality-control function of NMD may protect the cell from production of potential dominant-negative proteins that would, otherwise, lead to tumorigenesis, as has been described in different models, like BRCA1 in nude mice , p53 in human samples , or the Wilms’ tumor protein 1 (WT1) gene in in vitro studies .
NMD activity in cancer development and progression
Despite the clear notion that NMD can work against cancer, in some contexts its activity may have the opposite effect. For instance, it was recently reported that NMD inhibition in CRCs with microsatellite instability (MSI) leads to decreased tumor growth in xenograft models, suggesting that NMD typically plays a pro-tumorigenic role in these tumors . Indeed, CRCs with MSI present a higher expression of UPF1, UPF2, SMG1, SMG6, and SMG7, when compared to the counterparts CRCs with microsatellite stability. The overexpression of NMD factors is thought to potentiate degradation of the increased number of potentially toxic PTC-bearing transcripts that MSI CRCs characteristically produce, promoting their survival . In a different context, if a PTC-harboring tumor suppressor gene encodes a truncated protein that totally- or partially- preserves its original function, rather than a dominant-negative one like described above, the targeted degradation of its mRNA by NMD could promote cancer development. Accordingly, patients with NMD-sensitive mutations in CDH1 present a higher risk of developing HDGC than patients with NMD-resistant mutations, possibly because the latter produce truncated but still functional forms of E-cadherin . Interestingly, in a study matching exome and transcriptome data from human tumors it was reported that nonsense mutations are enriched in regions of tumor suppressor genes expected to induce NMD . As these mutations usually occur in heterozygosity , this raises the question of how NMD potentiates cancer development when at least one allele of a tumor suppressor is functional . Regarding this, Lindeboom et al. have revealed that cancer cells take advantage of NMD activity to accomplish complete tumor suppressor inactivation, which can occur by three mechanisms: i) tumor selection for NMD-inducing mutations in one allele combined with a deletion in the second allele, thus achieving biallelic inactivation by a “two-hit” process ii) selection for NMD-inducing mutations in haploinsufficient versions of the wild-type allele, to eliminate residual function and less frequently, iii) selection for NMD-resistant mutations in alleles that produce dominant-negative proteins  (Fig. 3b).
In addition to participating in the process of cancer development, NMD activity also seems to impact tumor evolution. During cancer progression, tumor cells acquire several somatic mutations that may, or not, favor the tumorigenic process. This is accompanied by a positive and negative selection that allows the proliferation of subclones with favorable mutations, such as PTCs in tumor suppressor genes or NMD-resistant/missense mutations in oncogenes, while eliminating the ones bearing detrimental mutations. By this means, cancer cells take advantage of NMD activity and of the rules that govern its induction to favor proliferation of transformed cells overproducing oncoproteins and other pro-tumorigenic proteins, thus driving further and aggravating the disease [19, 21].
A role for AS-NMD in cancer
As explained before, NMD activity is intimately linked to AS, together forming a key post-transcriptional mechanism of gene expression regulation. During the last decade, many dysregulated AS events have been observed in several cancer types, typically involving mutated splicing factors that lead to genome-wide altered patterns of gene expression. This can result in the production of NMD-sensitive isoforms of oncogenes and tumor suppressor genes that will, therefore, contribute for cancer development. A well-documented example is the SR protein, SRSF2, frequently mutated in patients with acute myeloid leukemia (AML) [159, 160]. A recent study reported that Pro95 hot spot mutation in SRSF2 enhances the stabilization of EJCs downstream from the PTC, thus favoring the association of key NMD factors to elicit NMD . A robust target of SRSF2Mut is EZH2, which encodes a protein that catalyzes histone methylation and acts as tumor suppressor in myeloid malignancies . SRSF2Mut drives the inclusion of a poison exon in the EZH2 pre-mRNA that triggers NMD and consequently shuts down its protein expression . In agreement with this finding, there are studies reporting EZH2 loss-of function mutations in the same spectrum of myeloid disorders displaying mutated SRSF2 [162, 163]. These data suggest that Pro95 mutation turns SRSF2 into an oncoprotein that uses AS-NMD to shut down the expression of a tumor-suppressor gene.
Epithelial cadherin (E-cadherin) is a crucial factor to maintain tissue integrity and polarization of epithelial cell layers and it is well-known that E-cadherin loss is a key event during cancer progression that contributes to the epithelial-mesenchymal transition . Interestingly, Matos et al. discovered that one of the causes leading to E-cadherin decrease is an mRNA variant produced by AS that is committed to NMD . That novel isoform arises from the usage of an alternative 3′ splice site that ends in the depletion of 34 nucleotides in the exon 14, introducing a PTC. Moreover, stable breast cancer MCF7 cells expressing this novel variant resulted in a concomitant decrease of the wild type E-cadherin mRNA levels and higher cell migration and invasiveness . However, it has to be elucidated the mechanism by which AS promotes this alternative isoform. Another example of an AS-NMD event impacting EMT is the one orchestrated by SRSF1. This splicing factor promotes a constitutively active isoform of the proto-oncogene MST1R ((Macrophage Stimulating 1 Receptor), by inducing skipping of exon 11 . Consequently, the active isoform of MST1R induces EMT, as well as increases resistance to apoptosis [167,168,169]. AS-NMD operates upstream in this pathway, inducing 3’UTR intron retention in SRSF1 under physiological conditions, which creates a stop codon premature context inducing NMD. However, in a tumorigenic context, another splicing factor, KHDRBS1 (KH RNA Binding Domain Containing, Signal Transduction Associated 1), stabilizes SRSF1 mRNA, which turns into a positive regulation of constitutive active MST1R.
Hypoxia is a major feature in solid tumors, given the high proliferating mass of cancer cells encountering an avascular environment that limits oxygen supply . Interestingly, hypoxia seems to impact AS-NMD, as observed for the cysteine-rich angiogenic inducer 61 (CYR61), a matricellular protein that promotes cell proliferation, migration and angiogenesis in numerous solid tumors [171,172,173]. Under normal conditions, CYR61 experiences retention of intron 3, which translates into a downstream PTC, leading to an NMD-sensitive isoform . Nevertheless, hypoxic conditions alter this AS-NMD gene regulation, inducing skipping of intron 3, which makes the transcript resistant to degradation by NMD.
N 6 -Methyladenosine (m 6 A) RNA methylation is a common mRNA modification dynamically regulated in mammalian cells  that controls several steps of the mRNA metabolism, from mRNA processing and transport to translation or decay [176,177,178,179]. METTL3, the catalytic subunit of the m 6 A methyltransferase complex, plays a critical role on tumorigenesis, promoting cell proliferation, survival, and invasion of cancer cells [180,181,182]. Interestingly, Li et al. recently discovered that METTL3 modulates alternative splicing of splicing factors affecting the pool of PTC-bearing isoforms in glioblastoma cells . Transcriptome studies showed that impaired METTL3 methylation results in the generation of PTCs in the mRNAs of several SRSFs, contrary to the existing scenario in malignant gliomas, which are associated with elevated expression of METTL3 and lower expression levels of SRSFs NMD spliced forms . Moreover, authors demonstrated that oncogenicity derived from METTL3 activity is due to changes in alternative splicing events of genes with relevant implications in cancer cell death and migration, like BCL-X and NCOR2. Altogether, this clearly indicates that dysregulated AS events in cancer cells contribute for the involvement of NMD in cancer development and/or progression.
NMD modulation in the tumor microenvironment
In striking contrast with the previous examples in which NMD promotes cancer is the finding that the tumor microenvironment induces NMD attenuation to potentiate cancer progression and adaptation, stressing out the idea that NMD activity has paradoxical outcomes. As mentioned above, during the unconstrained growth of the tumor, the blood supply to cancer cells becomes insufficient, creating a cellular environment of hypoxia, nutrient deprivation, production of ROS and ER stress, all stimuli that promote phosphorylation of eIF2α and inhibit NMD [26, 105, 107, 109, 148] (Fig. 3c). Accordingly, it has been reported that cancer cells grown as three-dimensional tumor explants present decreased NMD activity when compared to cancer cells cultured in monolayers, displaying significant levels of eIF2α-P . In fact, several studies showed increased levels of eIF2α-P in a plethora of cancers, including bronchioloalveolar, gastrointestinal and thyroid carcinomas, Hodgkin’s lymphoma, melanocytic and colonic epithelial neoplasms, and breast cancer [184,185,186,187,188,189]. The downside effect of this stress-mediated NMD inhibition in tumors is the stabilization and upregulation of several transcripts encoding stress-responsive factors, which will help cancer cells to proliferate in the adverse environment of the tumor  (Fig. 3c). For instance, it has been reported that NMD inhibition by stress stabilizes the mRNA of the cystine/glutamate antiporter xCT [Solute Carrier Family 7 Member 11 (SLC7A11)], a subunit of the xCT amino acid transporter system. This rate-limiting channel is responsible for the uptake of cystine for production of the cellular antioxidant, glutathione (GSH), a tripeptide that neutralizes free radicals and reactive oxygen compounds. Curiously, it was shown that UPF1-depleted cells can survive to higher doses of H2O2 in a SLC7A11-dependent manner, suggesting that NMD impairment during stress, and the consequent upregulation of SLC7A11, provides protection to cancer cells against the oxidative damage that may result from the overproduction of ROS . The ATF4 mRNA is another NMD target frequently found to be upregulated in several types of solid tumors, consistent with NMD inhibition in these conditions [190, 191]. Nutrient deprivation and oxidative stress promote ATF4 expression to transcriptionally induce genes involved in amino acid synthesis and transport, protein folding, cell differentiation, and autophagy, as a way to counterbalance the stress. Interestingly, it has been reported that the simultaneous molecular/pharmacological inhibition of autophagy and NMD leads to synergistic cell death in CRC cell lines, suggesting that autophagy is an adaptive response to NMD inhibition that allows cancer cells to overtake metabolic stress . Moreover, ATF4 activity has been implicated in the chemo-resistance of CRC cells , further highlighting that NMD shut down by the tumor microenvironment can favor signaling cascades that potentiate tumor proliferation and malignancy. Altogether, these findings indicate that NMD impairment is a consequence and part of the adaptative mechanisms cancer cells use to thrive in the harmful microenvironment of the tumor.
Cell Surface Discovery May Lead to Major Breakthrough in Cancer Treatment
University of Virginia School of Medicine researchers have discovered a new strategy for attacking cancer cells that could fundamentally alter the way doctors treat and prevent the deadly disease. By more selectively targeting cancer cells, this method offers a strategy to reduce the length of and physical toll associated with current treatments.
“We think we have a way not only to more specifically target cancer cells, but a way that could become a frontline treatment for women who have cancers of many types and want to preserve fertility,” said reproduction researcher John Herr of U.Va.’s Department of Cell Biology.
An Unexpected Cancer Connection
Herr and his research partner, Department of Obstetrics and Gynecology biologist Eusebio Pires, both specialize in germ cells – the reproductive cells that make up sperm and eggs. While researching new methods of contraception, Pires and Herr discovered a surprising link between developing egg cells and tumors. That link may allow doctors to use antibodies to deliver medication directly to tumors while sparing healthy tissue.
At the time, Herr and Pires were studying a protein called SAS1B that is typically only on the surface of developing and mature egg cells.
“Except for the small group of growing eggs in the ovary, the SAS1B protein is virtually absent in other tissues in the body,” Pires said. “So SAS1B has the promising features of a candidate contraceptive target.”
The restriction of SAS1B to growing eggs suggests strategies for developing improved female contraceptives that selectively target only the pool of growing eggs, potentially reducing unwanted side effects of current steroidal contraceptives.
Although the team originally focused on SAS1B because of its possible use in contraception, a single piece of data in the National Cancer Institutes’ GenBank database, showing expression of SAS1B in a uterine cancer, led their team to begin a search for it on the surface of various cancer types. So far, they have found SAS1B expressed in breast, melanoma, uterine, renal, ovarian, head and neck, and pancreatic cancers. There is also evidence to suggest that it appears in bladder cancers.
“The research opens a new field of enquiry, termed cancer-oocyte neoantigens, and reveals a previously little know fundamental aspect of cancer – that many types of cancer, when they dysregulate or go awry, revert back and take on features of the egg, the original cell from which all the tissues in the body derive,” Herr said.
He and Pires have found a way to exploit this fundamental insight by developing a method for delivering medication using the SAS1B protein as a target.
Tiny Cancer Trackers: Antibodies Armed with Drug Payloads
Since the SAS1B protein appears only on egg cells and cancer cells, the molecule can serve as a target for tiny tracking probes created using monoclonal antibodies. Monoclonal antibodies are highly pure antibodies designed to bind to one single target protein with a uniform affinity. The monoclonal antibodies attach to any cell marked with SAS1B proteins. Then the monoclonal antibody-SAS1B complexes can function as tiny injectors for targeted medication.
“You add a SAS1B-targeted antibody with a drug on it, and within 15 minutes of contacting the cancer cells, the antibody binds at the cell surface and the antibody-SAS1B complexes begin the internalization process,” Herr said.
After about an hour, the antibody-SAS1B complexes reach compartments inside the cell and release their toxic drug payload, triggering changes leading to cell death within a few days.
This kind of targeted drug delivery could mean a dramatic reduction in the difficult side-effects of traditional cancer treatment such as hair loss, nausea, anemia and neuropathy. Both women and men can use the treatment, which is predicted to dramatically limit unwanted side effects on healthy normal cells. For female cancer patients especially, a drug that doesn’t touch their body’s reserve of quiescent eggs could be a huge breakthrough.
While the monoclonal antibodies would have to attack the pool of growing egg cells in addition to the SAS1B positive cancer cells, the ovaries’ supply of dormant eggs would remain healthy and untouched by the treatment. This means normal ovulation could begin again once treatment is complete and oocytes are again recruited to develop and ovulate a process anticipated to take approximately 200 days.
In addition to treating cancer, these selective antibodies could also lead to a new method for early cancer detection and prevention. Pires explained that the research team is developing a way to find tiny amounts of free-circulating SAS1B proteins in the body.
“It could be a valuable discovery in terms of prerequisite testing to identify those patients who have tumors making SAS1B,” he said.
Pires and Herr hope that doctors will one day be able to use the monoclonal antibodies to measure patients’ SAS1B protein levels in blood. Those with elevated levels of the protein would be tested for the early stages of cancer and receive treatment sooner.
As they fine-tune this early detection method, Herr and Pires plan to begin testing their qualified antibody treatment in model organisms that carry human tumors at U.Va.’s labs this fall.
“If everything goes well, less than a year from now we will know if we’re ready to propose a study for testing within select human populations,” Pires said.
The researchers’ findings have been published in the scientific journal Oncotarget in an article written by Eusebio S. Pires, Ryan S. D’Souza, Marisa A. Needham, Austin K. Herr, Amir A. Jazaeri, Hui Li, Mark H. Stoler, Kiley L. Anderson-Knapp, Theodore Thomas, Arabinda Mandal, Alain Gougeon, Charles J. Flickinger, David E. Bruns, Brian A. Pollok, and John C. Herr. The full article is titled, “Membrane Associated Cancer-Oocyte Neoantigen SAS1B/Ovastacin is a Candidate Immunotherapeutic Target for Uterine Tumors.”
This work was supported by grants from the NIH Fogarty International Center, the Grand Challenges Exploration Program of the Bill and Melinda Gates Foundation, The Wallace H. Coulter Translational Research Partnership Endowment, The Paul Mellon Urologic Oncology Institute in the Cancer Center at the University of Virginia, The Virginia Center for Innovative Technology with matching support from Neoantigenics, Inc., and The Virginia Biosciences Health Research Corporation.
Is it possible to make a cancer cell that doesn't encode any neoantigens? - Biology
Synthetic biology will play an important role in advancing adoptive T cell therapy.
Engineered receptors and genetic circuits can make cell-based therapies safer and more powerful.
Cellular engineering and genome editing can further improve the T cell as a chassis for therapy.
The adoptive transfer of genetically engineered T cells with cancer-targeting receptors has shown tremendous promise for eradicating tumors in clinical trials. This form of cellular immunotherapy presents a unique opportunity to incorporate advanced systems and synthetic biology approaches to create cancer therapeutics with novel functions. We first review the development of synthetic receptors, switches, and circuits to control the location, duration, and strength of T cell activity against tumors. In addition, we discuss the cellular engineering and genome editing of host cells (or the chassis) to improve the efficacy of cell-based cancer therapeutics, and to reduce the time and cost of manufacturing.
Get full journal access for 1 year
All prices are NET prices.
VAT will be added later in the checkout.
Tax calculation will be finalised during checkout.
Get time limited or full article access on ReadCube.
All prices are NET prices.
Mutations accumulate in cells due to environmental insults such as UV light and cigarette smoke and from sporadic DNA replication errors that occur during normal cell proliferation. Mutations that confer the ability to proliferate unchecked by the body’s normal regulatory systems are often referred to as driver mutations. Cells with such driver mutations can become abundant in the tumor population. Every time these cells divide, there is a chance that additional mutations will occur due to errors copying the DNA. Thus, in addition to driver mutations, tumor cells often accumulate random damage to many other parts of the genome, including those that do not accelerate cancer’s growth these are called passenger mutations.
The mutational landscape of a tumor is composed of both driver and passenger mutations, which can be identified using high-throughput next-generation sequencing. Studying the number of each, their abundance in the population, and which mutations seem to have evolved together can reveal key information about the selective pressure the tumor is under (due to competing for limited resources like nutrients and oxygen, struggling to maintain essential cell processes despite rapid growth, or being attacked by the immune system) and can help us choose precise combinations of therapies to target the genetic and immunogenic weaknesses of the tumor.
Our current mutation-calling pipeline implements four somatic mutation callers (MuTect, Strelka, Somatic Sniper, and VarScan) to increase the confidence of calling. Low confidence mutations, such as low coverage variants, regions of low mapability (source: encodeproject.org/annotations/ENCSR636HFF/) and loci of DNA sequences with repeats and low complexity (source: www.repeatmasker.org) are filtered out automatically and selected mutations are reviewed manually for quality assurance.
We use whole-exome sequencing, whole-genome sequencing, and targeted gene sequencing to identify the genomic factors affecting antitumor immune activity. Briefly, our refined pipeline maps raw sequence reads to the human reference genome (GRCh37/hg19) the positions of insertions, deletions, and nucleotide variations are annotated and artifacts from library preparation are removed.
Personnel: Nadeem Riaz, Nils Weinhold, Rajarsi Mandal, Jonathan Havel, Luc Morris
We are interested in understanding the clonal composition of tumors. A clone is defined as a cluster of cells that share the same mutations, possibly due to a shared lineage. When a tumor contains many such lineages, it is called “subclonal,” and these distinctly arising subclones can accumulate new mutations that provide growth advantages, allowing them to out-grow less competitive subclones. Over time, the most competitive subclones make up a higher overall proportion of the tumor.
Not all subclones in a tumor necessarily respond to immunotherapy the same way. Some subclones may carry mutations that cause a stronger immune response than others. Therefore it is important to understand the clonal composition of tumors in order to design strategies that target enough of the tumor to perturb its growth at a clinically measurable level.
We estimate the relative frequency of cells within a tumor that carry a mutation based on genome sequencing data. For each mutation, we calculate the cancer cell fraction (CCF) based on variant allele frequency of the mutation, its copy number, as well as the sample’s purity. Analysis of CCF can help us identify subclones of cells that develop independently over the lifetime of a tumor, and deduce the relationship between the fitness of those subclones relative to others, as well as their susceptibility to immune targeting.
In a tumor cell population, particular mutations (colored circles) are often found in subsets of cells any given tumor cell contains some but not all of the mutations observed in the population as a whole. The efficacy of an immunotherapy that bolsters the T cell response to a particular mutated tumor protein may be strongly influenced by how much of the tumor cell population expresses that mutant protein. (Top) Immunotherapy #1 enhances the T cell response to the mutant protein A (pink), which occurs in 50% of tumor cells. Thus, when immunotherapy #1 allows these T cells to become activated and kill their targets, mutation A is eliminated from the tumor, but the other 50% of tumor cells remain. (Bottom) Immunotherapy #2 enhances the T cell response to the mutant protein B (purple), which occurs in 75% of tumor cells, so treatment results in only 25% of tumor cells (those without mutation B) persisting. By measuring the abundance of mutations in a tumor cell population over time, including during therapy, we can learn about how mutations are linked. For example, when one mutation disappears or becomes more common, which other mutations go with it? We can also determine how immunotherapies are acting on the immune response. Do some therapies only bolster T cell responses to a small number of antigens, while others support more broad T cell stimulation? Together this information about both the tumor target and the nature of the stimulated immune response can be used to more precisely design therapy for each patient.
A major obstacle to the development of a strong, effective immune response to a growing tumor is the fact that tumor cells are very similar to healthy tissue. Antigens that arise in tumor cells due to mutations (neo-antigens) allow the immune system to recognize those tumor cells as non-self and can thereby trigger a tumor-specific immune response. It is thought that the number of neo-antigens present in a tumor is a crucial factor determining whether an immunotherapy will be successful at marshaling an effective antitumor immune response.
We are actively developing novel computational approaches to identify neo-antigens in human cancers. Our current method utilizes the same somatic mutation-calling pipeline as described above (see Genomic Sequencing), followed by neo-epitope analysis.
Our algorithm for predicting neo-epitopes translates all missense mutations identified by the genomic mutation pipeline, generates all 9-amino acid peptides (the most frequently presented peptide length) that would contain the mutation, and uses the netMHC tool to compare the predicted rate of presentation of the mutated peptide to that of the 9-amino acid peptide with the unmutated (wild type) amino acid at the same position by that patient’s particular HLA alleles. Peptides for which the mutated version is more strongly presented than the wild type are considered potential neo-antigens.
Personnel: Nadeem Riaz, Vladimir Makarov, Diego Chowell
To better understand the specific interplay between a patient’s mutations and the immune system, mutant peptides are systematically tested for immunogenicity — the ability to activate T cells taken from the same patient. Results of this type of antigen screening can help in the creation of more personalized immunotherapies, such as tumor-specific vaccines or adoptive T cell therapies. Furthermore, IPOP seeks to understand the relative contributions of different types of mutations and antigens to effective immune responses with the goal of making patient-specific therapies more precise.
In order to maximize screening efficiency, plasmids encoding multiple tandem minigenes (TMGs) are generated. A single minigene consists of the DNA encoding a somatic mutation flanked on both sides by twelve amino acids from the wild type source protein. Up to ten minigenes are strung together to generate the TMGs used in screening. In vitro transcribed mRNA is then introduced into autologous dendritic cells (DCs) via electroporation to enable processing and HLA-presentation of the somatic mutation-containing peptides. Patient-derived T cells are co-cultured with TMG-transfected DCs. Neo-antigen peptide-induced T cell activation is quantified via detection of cytokine (e.g., interferon gamma) production using the highly sensitive ELISpot assay. Results are deconvolved by back-mutating (to wild type) each of the ten mutations contained in a reactive TMG and testing each for cytokine production in the co-culture assay described above. Intracellular cytokine staining is used for orthogonal validation of any positive hits from a minigene antigen screen.
Personnel: Raghu Srivastava, Jonathan Havel, Wei Wu
Adaptive immune cells — T cells and B cells — help us to recognize specific threats, such as microbial pathogens (e.g., bacteria, viruses, fungi) and tumors. Each T cell or B cell expresses a receptor on its surface — the T cell receptor (TCR) or B cell receptor (BCR), respectively — that can bind to a particular molecular target, and differs from one immune cell to the next. When a TCR or BCR finds its target molecule, called an antigen, the T or B cell is signaled to divide and multiply. Each receptor is unique, generated by random recombination and alteration of DNA during development into a mature T or B cell, and the number of different TCRs that can be generated by one person is huge: between 10 12 -10 20 over the course of a lifetime, with
10 9 present in the repertoire at any given time. It is the vast diversity of these receptors that enables any one person to respond to antigens his or her immune system has never encountered before, and to raise an “army” against a particular antigen if it represents a threat.
Expansion of tumor-specific T cells. A. When the TCR of a T cell binds a target antigen strongly, the T cell becomes activated and proliferates. Thus, that TCR is represented at an expanded frequency in the population. B. In the context of a tumor, a T cell whose TCR recognizes a tumor-specific antigen may be represented in higher abundance by the same mechanism. With immunogenomics, we can learn about tumors and the T cells that recognize them in parallel: How many TCRs are there? What do they have in common? Do patients with the same tumor share the same expanded TCRs? How do the proportions of these TCRs reflect and predict the changes in the abundance of tumor antigen targets?
Many of these immune cells are not circulating freely in the blood, but infiltrate and provide surveillance in tissues. This population of tissue-infiltrating lymphocytes (TILs) differs from the circulating population in that the former represents only a small sample of the total repertoire T cells surveilling any particular tissue may be selected — on the basis of their receptors as well as growth factors and other signaling molecules — to reside in that particular organ or tissue.
TCRseq libraries represent a sample the peripheral blood and tumor-infiltrating lymphocyte (TIL) cell repertoires. While some TCRs occur at the same rate in both populations (pink), some TCRs are relatively more abundant in the TILs than in the peripheral blood (green, purple), while some are at much lower abundance, to the point that they aren’t detected (cyan). The TCRs that are most enriched in the tumor tissue may reside there disproportionally because their TCRs bind antigens that are present only in the tumor tissue, making these sequences of interest for further study, as possible biomarkers of tumor progression or as therapeutic templates or targets (see below).
Recent advances in high-throughput next-generation sequencing let us capture the TCRs from a whole sample (TCRseq) — circulating blood cells or T cell-infiltrated tissue — and describe the population in terms of the distribution of those TCRs. Using statistics, we analyze the diversity of these populations, compare them to one another, and look for patterns across groups of patients being treated for cancer. How does the TCR repertoire inside a tumor differ from that in the circulating blood?
We are currently defining properties that indicate tumor-specific reactivity: What does the antitumor T cell response look like when it’s working? When it’s failing? When it has been restored through immunotherapy? These properties may be useful as multi-dimensional biomarkers to monitor tumor progression and therapeutic response. We are also using TCR repertoire sequencing to identify receptors that could be adapted for use as antitumor therapeutics.
Particular TCR sequences associated with either the progression or regression of a tumor could be used directly to develop therapeutics. (Top) The TCRs of cytolytic T cells (CTL) found to expand concomitantly with the regression of cancer could be tested as templates for engineered chimeric antigen receptor (CAR) T cells, which would be able to recognize and attack the tumor using a receptor based on that TCR. (Bottom) The regulatory T cells (Tregs) that inhibit the activity of active antitumor CTLs (thus protecting the tumor) could be blocked by immunotherapies targeting their TCRs.
TCRseq @ MSK
We perform TCR sequencing of clinical samples on-site in collaboration with the Integrated Genomics Operation, provide analysis of the raw sequencing data (where applicable), as well as supported end-user analysis (under development). IPOP currently supports the following commercial TCR library generation platforms:
- iR Profile (TCRa and TCRb) — iRepertoire
- SMARTer Human TCR a/b Profiling (TCRa and TCRb) — Clontech
- ImmunoSEQ (TCRb only) — Adaptive Biotechnologies
We are actively testing and integrating new products and platforms, and developing immunotherapy-related analytical tools in collaboration with cBioPortal.
Personnel: Jennifer Sims, Jonathan Havel
One of IPOP’s goals is to extract the immunogenomic information that will allow doctors to anticipate which patients are most likely to respond to immunotherapy. We study tumor phenotype, or cell behavior, which is largely determined by the levels at which each gene is expressed. In particular, we use high-throughput sequencing of RNA from tumor biopsies to study how expression of genes changes as cancer progresses, when therapy is given, and when therapy is effective. Comparing tumors from patients who respond with those from patients who do not allows us to identify any distinct sets of tumor features that can be translated into diagnostic, prognostic, and therapeutic biomarkers to be used for future patients.
The expression levels of genes also provide information about the environment in which the tumor evolves, particularly how the patient’s own immune system reacts to it. Using cutting edge computational techniques, we can integrate this information to understand what types of immune cells are successful in this process.
Tumors differ from one another, in part because each patient’s immune system reacts to a tumor using a unique set of cells to try to destroy it. Abnormal tumor cell behavior, specific antitumor immune activity, non-specific inflammatory immune activity, and tissue damage shape the gene expression profiles of both tumor and non-tumor cells in unique ways.
RNA from tumor tissue is subjected to high-throughput next-generation sequencing, which gives short nucleotide “reads” as output. These reads are ordered and aligned to the human genome, giving the amount of RNA from each gene. For each gene, we can then statistically test for differentially higher or lower expression between two groups of samples (for example, the tumors of therapy-responsive patients and non-responsive patients). We can then identify differentially expressed genes, or functionally related groups of genes, that reflect different programs of expressed genes or different cell compositions between tumors.
One application of differential gene expression analyses is to compare the pre-treatment and post-treatment profiles of tumors that responded to immunotherapy with those that did not. We can also identify marker genes or groups of functionally related genes that, if unusually high or low prior to treatment, correspond with better responses to particular therapies. Such predictive signatures could enable a simple pre-treatment biopsy to help tailor a patient’s treatment regimen.
In IPOP, our pipeline for automating and visualizing these analyses is constantly improving. High-dimensional data visualization tools such as oncoprints and Visne maps allow us to organize and render dozens of parameters (e.g., RNASeq gene expression data in parallel with clinical parameters) simultaneously, without sacrificing their complexity, to enrich our understanding of the cancer immune environment.
In one recent study, hierarchical clustering of the expression of genes across tumor biopsies from patients who were strongly responsive (+++), weakly responsive (+), or non-responsive (-) to anti-PD-1 therapy identified two subsets of genes, one of which was highly expressed among the clinically responsive patients, and one of which was highly expressed among the non-responsive patients. Enrichment of functionally related genes in these clusters can be used to infer how such a gene signature is related to these levels of immune response.
Cell composition (in silico deconvolution)
Many different immune cell types infiltrate tissues, where they perform different roles in surveilling for tumors, injuries, or infections. For example, certain types of T cells are capable of directly killing dysfunctional, tumorigenic, or infected cells, while monocytes and macrophages take up free-floating cell debris and present these potential antigens to T cells. This interaction, which requires both T cells and antigen-presenting cells, can help locally activate or suppress all the T cells that recognize the same antigens. Meanwhile, B cells produce antibodies that can rapidly spread throughout the body to neutralize a particular threat. Thus the relative abundance of different cell types can indicate which modes of tumor recognition are active, and which may be suppressed.
The type and degree of immune infiltration into tumors plays an important role in the efficacy of immunotherapy. The abundance of the messenger RNA (mRNA) of particular genes in a tissue biopsy not only allows us to identify differential expression gene between samples, but also enables us to calculate the relative abundance of different immune cell types in the local microenvironment. Briefly, from the mRNA of the bulk sample, we can detect high expression of signature genes or enrichment of a subset of genes that are specific to one cell type, and compare it to the expression of genes specific to other cell types. We use computational algorithms such as Supporter Vector Machines (SVMs) or Single-Sample Gene Set Enrichment (ssGSEA) to translate the expression of these signatures into relative abundances of the corresponding immune cell populations.
Because immunotherapies perform different functions — such as maintaining immune cell activation, rescuing immune cells that were activated then became exhausted, or stimulating antitumor reactivity among immune cells that were previously unexposed to tumor antigens — understanding which types of immune cells are present (or not) in the tumor microenvironment has implications for predicting response to these immunotherapies, and choosing the right one for each patient.
From a bulk tumor sample, mRNA molecules are extracted from the mix of immune cells (colored and gray) and non-immune cells (brown, such as tumor). Running high-throughput next generation sequencing on the mRNA mixture provides the gene expression profile of the biopsy sample (left). Using known gene expression signatures, the proportion of mRNA molecules representing each infiltrated immune populations can be deconvolved (bottom), and the composition (relative abundance) of the immune cell types in the population can be inferred (right).
Personnel: Fengshen Kuo, Alexis Desrichard
T cells recognize microbial threats and cancer by binding to degraded bits of foreign proteins (peptides) presented to them by the molecules of the major histocompatibility complex (MHC). These presentation molecules are expressed on the surface of most cell types, but especially strongly on certain immune cells that provide surveillance of tissues.
The genes that encode MHC class I proteins (called the HLA class I genes in humans) are located on chromosome 6, and there are three of them: HLA-A, HLA-B, and HLA-C. Every person has two copies (alleles) of each gene (one from each parent), and since these genes are the most polymorphic (variable in DNA sequence) in the entire human genome, the six alleles each person has are often all different, and rarely do they match those of genetically unrelated individuals. There are specific alleles (e.g. HLA-A*02:01) that are more prevalent worldwide. Moreover, the frequency of HLA alleles varies across geographic regions and populations.
Frequency of select HLA-A alleles across different geographic regions. Shown are the normalized frequencies of some HLA-A alleles in diverse geographic regions. For each HLA-A allele, each colored bar represents the frequency of the allele in a particular geographic region. The data were obtained from http://www.allelefrequencies.net/.
We are examining how the HLA alleles a patient uses affect responsiveness to immunotherapy. The presentation of peptides to T cells by the MHC proteins plays a critical role in the adaptive immune response, and strongly influences how T cells respond to that peptide. For example, some MHC molecules activate T cells strongly, which is desirable if that antigen represents a threat (such as a viral infection or a dysfunctional or mutated protein produced by a tumor) but can be dangerous if the antigen is normal and occurs on healthy cells. Because potentiating the correct recognition of self versus non-self peptides by T cells is a major function of MHCs, and this distinction becomes muddled in the case of cancer, it is important to use genomic sequencing data to identify which six HLA alleles any patient has when trying to determine how his or her immune system will react to their mutated tumor peptides.
Currently the gold standard for identifying which HLA alleles a patient has is PCR-based typing, in which the HLA locus is specifically amplified and then sequenced. As genomic sequencing has achieved higher and higher coverage, in silico HLA genotyping offers an efficient alternative that is economical when a patient’s genome is already being sequenced. Current software tools provide up to 99% accurate resolution for most clinical applications. For clinical applications that require higher accuracy, such as prediction of tumor antigen presentation by certain HLA alleles, which differ from their closest other alleles by only a few nucleotides, we are refining the computational pipelines for HLA identification using ensemble approaches, population-based weighting, and alternative assemblies of the human reference genomes.
Personnel: Diego Chowell, Vladimir Makarov, Fengshen Kuo
Understanding the cellular composition of tumor and immune cells on the level of phenotypic protein markers is a critical part of investigating tumor immunology. IPOP utilizes several experimental techniques to better quantify the expression of proteins of interest in individual tumor and immune cells. Antibody-based flow cytometry allows for the precise quantification of extracellular and intracellular proteins of interest. Using fluorescence-activated cell sorting (FACS), individual immune or tumor cell populations can be further subdivided for downstream analysis including DNA and RNA sequencing.
Occasionally, investigators may wish to quantify the expression of a large number of intracellular and extracellular proteins simultaneously from a single sample. Conventional flow cytometry limits the number of simultaneous parameters detectable due to fluorophore-generated spectral overlap. To overcome this barrier, IPOP utilizes mass cytometry by time-of-flight (CyTOF) technology. CyTOF identifies intracellular and extracellular proteins using antibodies conjugated to rare earth heavy metals. After antibody-based staining, the sample is ionized and the antibody composition of single cells are subsequently identified. The primary advantage of CyTOF is its ability to analyze a robust user-defined panel of cellular targets simultaneously from a single sample using an antibody-based approach. Multi-parametric data can subsequently be analyzed using conventional flow cytometry software or more sophisticated techniques including SPADE or ViSNE plots. IPOP CyTOF projects are currently done in collaboration with the Mount Sinai Human Immune Monitoring CORE (HIMC) (212-824-9354, [email protected]).
Tumor-infiltrating lymphocytes and peripheral blood mononuclear cells from a patient with head and neck squamous carcinoma were clustered by their staining for immunophenotypic markers (clusters represent cells with similar phenotypes, where closeness on the plot indicates similarity), using the t-distributed stochastic neighbor embedding (t-SNE) algorithm. Color scale indicates CD8 protein detected, normalized across cells, which distinguishes the cell types (clusters) expressing this marker of the cytolytic T cell lineage.