Watch this on-demand webinar presented by Professor Kevin Hiom
Listen in for an introduction to post-translational modifications.
About the Presenter
Professor Kevin Hiom, is the Pat McPherson Chair of Cancer Biology at the Biomedical Research Institute and Dundee Cancer Centre at the Ninewells Hospital and Medical School.
Kevin received his BSc in Genetics from the University of Liverpool and carried out his PhD work at the MRC National Institute for Medical Research London, where he developed an interest in how cells are able to repair damage to their DNA. Postdoctoral fellowships followed at the Imperial Cancer Research Fund Laboratories (now known as Cancer Research UK) and the National Institutes of Health in the United States.
In 1998, Kevin started his own research group at the MRC Laboratory of Molecular Biology in Cambridge where his attention turned towards understanding the relationship between defects in DNA repair and cancer.
In 2009, Professor Hiom relocated his group to the Medical Research Institute in Dundee where he continues to study the role of ubiquitin in the maintenance of genome stability.
A very clear presentation, you can also use ELM and HPRD to predict your phospho sites"
- Detecting ubiquitylation
- Troubleshooting ubiquitylation detection
- Historic overview of the ubiquitin system discovery and ubiquitin biology
- Ubiquitin and its importance in human diseases
- Detecting S/T/Y phosphorylation; identifying phosphorylation sites and elucidating the function of the modification
- Troubleshooting phosphorylation
Hello. Welcome to Abcam's webinar on an Introduction to Post-Translational Modifications. Today's guest speaker is Professor Kevin Hiom, the Pat McPherson Chair of Cancer Biology at the Biomedical Research Institute and Dundee Cancer Centre at the Ninewells Hospital and Medical School. Kevin received his BSc in Genetics from the University of Liverpool and carried out his PhD work at the MRC National Institute for Medical Research London, where he developed an interest in how cells are able to repair damage to their DNA. Postdoctoral fellowships followed at the Imperial Cancer Research Fund Laboratories (now known as Cancer Research UK) and the National Institutes of Health in the United States. In 1998, Kevin started his own research group at the MRC Laboratory of Molecular Biology in Cambridge where his attention turned towards understanding the relationship between defects in DNA repair and cancer. In 2009, Kevin relocated his group to the Medical Research Institute in Dundee where he continues to study the role of ubiquitin in the maintenance of genome stability.
Joining Kevin today will be Miriam Ferrer and Judith Langenick, members of Abcam's new products team. Miriam completed her Biology Degree at the University of Barcelona and a PhD from the Vrije University in Amsterdam. Judith completed her Molecular Biology Degree and PhD at the University of Dundee. Both Miriam and Judith have experience in DNA repair.
Just before we start, a quick reminder that questions for the question and answer session at the end of the webinar can be submitted at any time during the webinar via the Q&A panel at the bottom right hand side of the screen. Also, when you log-off from the webinar you will be directed to a webpage where a copy of the presentation can be downloaded. I will now handover to Miriam who will start this webinar.
MF: Thank you, Lucy. Welcome to Abcam’s Post-Translational Modifications webinar. I will start with a brief introduction on post-translational modifications before moving into ubiquitylation. After discussing ubiquitylation, our guest speaker, Kevin, will give you a historic overview on the field, and discuss how ubiquitylation is increasing its medical relevance. Finally, Judith will take you through phosphorylation and she will end this webinar by sharing with you several resources and related products available from Abcam.
So what are post-translational modifications? Post-translational modifications, or PTMs, are in vivo chemical modifications of a protein after its translation, which can occur at any step in the life of a protein. PTMs have been mainly studied in nucleated cells, and they are conserved from yeast to humans which demonstrates how essential they are for cellular function ability. PTMs play a key role in functional proteomics regulating activity, localization and interaction of proteins with other proteins, nucleic acid or co-factors. In fact, it is estimated that about 5% of the human disease-associated mutations may affect PTM status in cells.
Post-translational modifications are generally classified by the type of change they create. The most common one is the addition of a functional group or proteins to the amino acids. Examples are acetylation, phosphorylation, ubiquitylation and SUMOylation. Modifications can also be caused by a change on the chemical nature of the residues, such as carbamylation or deamination. Lastly, modifications can be caused by structural changes in the protein, like proteolytic cleavage or formation of disulfide bridges. In this webinar we will focus on ubiquitylation and phosphorylation.
So first let's talk about ubiquitylation. What is ubiquitylation? Ubiquitylation, also known as ubiquitination, is the addition of one or multiple ubiquitin molecules to lysine residues in a particular protein. Ubiquitin is a fairly small protein, only 76 residues, but it's found in all eukaryotic cells, and it is highly conserved among species. The main key features of ubiquitin are a C-terminal tail and 7 lysine residues. Ubiquitin can be linked to proteins either as a mono-ubiquitin, a single molecule attached in which case can be involved in regulation or receptor endocytosis, or as a poly-ubiquitin chain, which then can be involved in proteasomal degradation.
So how does ubiquitin get attached to a protein? The ubiquitylation process consists of three main steps. Initially, ubiquitin is activated by an ubiquitin-activating enzyme or E1, in an ATP-dependent process. The C-terminal carboxyl tail of ubiquitin will bind to the E1 enzyme. From the E1 the ubiquitin is then transferred to the cysteine on the active site of the ubiquitin conjugating enzymes, also known as E2 via a trans(thio)esterification reaction. The final step of the cascade creates an isopeptide bond between a lysine residue on the substrate and the C-terminal end of the ubiquitin molecule. These steps require the activity of an E3 ubiquitin ligase enzyme. E3 ligases function at substrate recognition models, and are capable of interacting with both the E2 and the substrate.
Ubiquitylation itself is a reversible reaction and the enzymes responsible for clearing ubiquitin protein bonds are known as the ubiquitinating enzymes or DUBs. The ubiquitylation enzymes are structured in a phenomenal way. The E1 enzyme family is quite small with only nine different E1s described in humans. One E1 enzyme can associate with several E2s. There is a great diversity of E2s when compared to E1s, and each type of E2 enzyme can generally associate with several E3s. There is even a greater diversity of E3s compared to E2s, and that's because E3 enzymes are responsible for setting substrate specificity. There are commonly more than 600 putative E3 ligases described in humans. It is important not to forget the ubiquitinating enzymes, although to its fairly recent discovery, there are still a lot of things we don't know about them. What we know is that they are more promiscuous than E3s. Over 80 putative DUBs have been described so far in humans, which means that one DUB can interact with several E3 ligases.
We are going to look briefly now at the different types of ubiquitin E3 ligases. I will focus mainly on the two main important type of E3 ligases. The main family of E3 ligases are the RING ligases, which contain a zinc finger ring domain. Known members of this family are BRCA1, Mdm2 and XIAP. As you can see from this diagram, the RING ligases mediate the direct transfer of the ubiquitin molecules from the E2 to the lysine in the substrate. The other main family, the HECT ligases share a common HECT catalytic domain. The best known member of this family is E6AP, in which the family takes the name, as well as other proteins such as Nedd4. Contrary to RING ligases, HECT ligases mediate ubiquitylation by forming an intermediate between the E2 protein and the ubiquitin. From this intermediate, the ubiquitin molecules will then be transferred to the lysine residue on the substrate. I will now focus on the different types of ubiquitylation, and how they affect the ubiquitylated substrate.
Mono-ubiquitylation is the addition of one ubiquitin molecule to one lysine residue in the substrate. This type of modification is generally involved in signaling and regulation, as well as receptor tyrosine kinase endocytosis. Multi-ubiquitylation is the addition of one ubiquitin to several lysine residues in the substrate. This type of modification is generally involved in receptor tyrosine kinase endocytosis, and endosome sorting. Poly-ubiquitylation is the addition of several ubiquitin molecules chained to each other and linked to one lysine residue in the substrate.
As previously mentioned, ubiquitin contains several lysine residues that all can be used as linking points for poly-ubiquitin chain formation. The ubiquitin residue to which all the ubiquitin molecules RING bind to form the chain, defines the space of the poly-ubiquitylated protein. For example, lysine 48 poly chain is the most well-known ubiquitin modification. Formation of a chain of poly-ubiquitins all link through lysine 48 will lead to proteasome degradation of the ubiquitylated protein. Other well-known chains are lysine 63 involved in DNA repair, and lysine 11 which is involved in NF-kB activation.
Let's look now at the steps to follow to determine whether our protein of interest is ubiquitylated. There are generally three main ways to investigate any post-translational modifications: in silico, to a computational tool; in vivo and in vitro. I will briefly touch on in silico detections. Briefly, because, unfortunately, there are no clear consensus motifs for ubiquitylation, and the current ubiquitylation prediction sites are only available for yeast proteins at the moment. If you would like more information about this site, just contact us through the Q&A panel on the right of your screen so we can give you more information later on about these sites.
The main way to study whether a protein is ubiquitylated is through in vivo studies, and a lysine ubiquitylation pattern. You start with cells or tissue where the protein to study is expressed. If your protein is naturally not abundant in those particular cells, you can use overexpressed systems. For ubiquitylation analysis, you will perform a western blot. Needless to say, if you have used another expressed protein, you should always confirm your resource with the endogenous protein. To enrich the potentially ubiquitinated proteins as well to get a better detection, you should perform an immunoprecipitation once you have seen evidence for ubiquitylation.
So let's review these steps one-by-one. You have prepared your samples, you have run your gels and blotted your membrane, and the detected protein you want using an antibody against that protein. If your protein is ubiquitylated you will notice that it migrates slower than the expected molecular weight, and you probably will see a smear of a shift on top of the band. In this first example, cells were treated with a proteasome inhibitor, which led to the appearance of a ladder smear when detecting p53. This is a typical signal of proteasome-dependent K48 poly-ubiquitination.
In the other example on the right you can see FANCD2, which is a DNA repair protein involved in homologous combination. FANCD2 usually runs at 155 kDa, which is known as the FANCD2-S form, but upon treatment with cross-linking reagents such as cisplatin, FANCD2 becomes mono-ubiquitylated and can be detected as this lower band running at 162 kDa, which is then known as FANCD2-L form.
To confirm that your shift is due to ubiquitylation, you should always perform an immunoprecipitation. Immunoprecipitation not only helps to enrich your protein fractions, but also clears your lysates from extra protein that might interfere with your detection. The protein of interest is generally immunoprecipitated with a specific antibody against the protein itself, or against the tag if you are using an overexpressed protein, and the blot is incubated with an ubiquitin-specific antibody. On the blot appearing on this slide you can see how an overexpressed protein was pulled down using a FLAG antibody, and then blot using a generic ubiquitin antibody. The smear you can see on the right lane is typical of a poly-ubiquitylated protein via the proteasome pathway. In addition to general ubiquitin antibodies, you can also use antibody-specific for ubiquitin chain detection. The most common antibodies used are FK1 and FK2, and these antibodies don't bind to free ubiquitin, and whereas FK2 will bind to mono- and poly-ubiquitin, FK1 will only bind to poly-ubiquitin. The use of both of these antibodies will always help you to understand which type of chain it is.
As you can see on the set of blots shown on this slide, cells were treated with proteasome inhibitors and separated in different cellular fractions. Then p53, which was the target protein investigated in this case, was pulled down with a p53-specific antibody. The membranes were then proofed with FK2, which are section A and section B; and FK1, which is shown at the bottom. You can see that there is a clear p53 smear present in all three blots in the nucleolus fraction, which is where the golden arrow is shown. From this experiment we can conclude that p53 is fully ubiquitylated as is picked by both FK1 and FK2 antibodies.
In vitro analysis will also help us to define which specific lysine residue is ubiquitylated, as well as identifying which type of ubiquitin chain is being formed. This is usually done through a combination of truncation analysis, and in vitro ubiquitylation assay. In a truncation analysis different sections of the ubiquitylated protein are expressed. When detecting the ubiquitylation reaction on a western blot using an ubiquitin antibody, only the section containing the modified lysine will show on the blot. This allows to pinpoint which specific area the lysine of interest is located. Once the ubiquitylated region has been defined, the next step is to mutate lysines to arginines to define which is the actual residue that will be ubiquitylated. Arginines are hydrophobic and structurally very similar to lysines, but cannot be ubiquitylated which makes them the perfect substituting mutations. There is a caveat though: when modifying a lysine on ubiquitylation sequences, be aware that you are not disturbing a SUMOylation site. For instance, on PCNA the same lysine is both ubiquitylated and SUMOylated, and although this will not affect the ubiquitylation reaction, it will probably have a great effect on the in vivo action that you see on the protein. Of course, once you have identified the lysine you think is ubiquitylated, always confirm through mass spectrometry.
An in vitro ubiquitylation reaction is the ideal system to identify the specifics of the reaction in which factors are essential. To set up an ubiquitylation reaction you will need a substrate, which is your protein of interest. The substrate can be recombinant or purified from cells.
You will also need the whole set of ubiquitin conjugated enzymes as well as ubiquitin, so make sure that you are using the appropriate set of enzymes, especially the E3 ligase, so check the available literature, or check with your colleagues if they have done similar experiments previously. If your protein is fully ubiquitylated you might also want to use ubiquitin mutants as well as the wild-type, because this will help you to identify which type of poly chain is being formed. As with any reaction, you will also need a negative and a positive control. Ubiquitylation is an ATP-dependent reaction, so you will need ATP in the mixture. Also, check the literature to see if there are any co-factors that are necessary for the ubiquitylation of your protein.
The experiment shown in this slide you can see the in vitro ubiquitylation reaction of a histone H2A by BRCA1/BARD1 dimer using radioactively labeled ubiquitin. In lane 1, you can see how the substrate histone H2A is mono-ubiquitylated when all the factors are present, and that's when it's labeled complete. If you take out any of the conjugated enzymes or the ATP, the ubiquitylation doesn't take place, as you can see from lanes 2, 3, 4 and 6. Interestingly, this experiment also shows that BRCA1/BARD1 auto-ubiquitylates itself, as you can see at the top of lane 1 and lane 5. You can see in the image there is a big band appearing even in the absence of the substrate. This is a good example of mono-ubiquitylation reaction, because there's a fairly clear, distinct band both in the BRCA1 locations and in the ubiquitylated histone H2A.
In this experiment, the researchers used a ubiquitin antibody instead of radioactive ubiquitin. This blot shows how Pellino 1 acts as an E3 ligase catalyzing the formation of a poly-ubiquitin chain in IRAK1, which is a serine threonine kinase involved in NFkB upregulation.
To further identify which type of chain was formed, the researchers used different mutated ubiquitin variants. As you can see from the blot, when wild-type or K48 ubiquitin mutant was used, a poly-ubiquitin chain was still formed. However, when a K63R ubiquitin mutant was used, no chain was seen. This confirms then that Pellino 1 is an E3 ligase for IRAK and catalyzes the formation of K63 poly chain ubiquitins.
I would like to share now with you some tips and troubleshooting to keep in mind when determining whether the protein of interest that you are studying is ubiquitylated or not. So what can you do if you're not sure about a shift/smear that you see in your western blot? When you're preparing your cells or your lysates add MG-132, which is a very well-known proteasome inhibitor. As you probably remember, this has been used in some of the examples shown on the previous slides. But the proteins that are signaling for proteasome degradation via a K48 poly chain, additional MG-132 will increase the concentration. You can also add NEM, N-ethylmaleimide, which is a general deubiquitinating enzyme inhibitor. Similarly to MG-132, inhibiting the degradation of the ubiquitin signal will increase the chances of detection. Also, make sure when you are preparing your cells that you are using the optimal conditions. If, for example, your protein is ubiquitylated at a particular time during the cell cycle, try to synchronize your cells and harvest them at the right time. If your ubiquitylated protein accumulates in a particular cell compartment, make sure that you're harvesting the right fraction as this will give you the strongest signal.
Make sure that your running gel conditions are optimal. Always ensure that your protein is fully denatured by adding beta-mercaptoethanol or DTT to your sample buffer. Although this might sound quite obvious, make sure that you are using a gel percentage that is optimal for your protein and will allow you to see the shift. Always run your gels low to avoid smiling. When running a gel for a long time it is always recommended to run it at 4°C to avoid overheating. Either run your gels in a cold room overnight or in a bucket of ice.
If your signal is very weak, try then to use BSA instead of milk for blocking your membrane. Again, make sure that the antibody you're using is specific for your protein, because otherwise you could pick up ubiquitylation for similar proteins, or several isoforms of the protein you are studying.
If your in vitro reaction hasn't worked, make sure that you have all the co-factors needed for the reaction. A good tip is to avoid mixing E1 and E2 enzymes from different sources, either use them either from a commercial source, or from your own purification. Again, this might sound obvious, but make sure that you are using the appropriate E3 ligase for the reaction. If you don't know just check the literature, or maybe just get some people that have done this reaction previously.
I would like to conclude this section with a quick summary on the main points discussed on ubiquitylation. Ubiquitylation is a post-translational modification that controls many processes within cells. The addition of ubiquitin to proteins is a tightly regulated mechanism. The E3 ubiquitin ligases are the enzymes responsible for the substrate-specificity. The substrate can be either mono-ubiquitylated, when one ubiquitin is attached to one lysine; multi-ubiquitylated, when several lysines receive one ubiquitin each; or poly-ubiquitylated, when one lysine receives a chain of several ubiquitins. The type of chain form will regulate the space of the ubiquitylated protein. Determining the site and function of a ubiquitylation event should be done through a combination of in vivo and in vitro studies.
If you have any questions about the topics we have discussed and any other one that we couldn't discuss through time constraints, please submit your questions through the Q&A panel on the right of your screen. Without further delay, I will pass the microphone to Kevin who will talk to you about historical review of ubiquitylation and the future of the field.
KH: Thanks very much, Miriam. So as you’ve seen from Miriam ubiquitylation is one of the most versatile and powerful forms of post-translational modification in cells, controlling many important cellular functions and involves more than 1,000 proteins. In this part of the presentation I'm going to give a bit of historic perspective to the discovery of ubiquitylation. I'm then going to talk about the versatility of the system and its control of cellular function, and then give some views on how this system impacts on medicine and drug discovery, and perhaps where the future lies for research.
So the process of protein ubiquitylation was discovered in the late-70’s and early-80‘s, and was awarded the Nobel Prize for Chemistry in 2004. The recipients of this award were Avram Hershko and Aaron Ciechanover, both at the Technion Israel Institute for Technology in Haifa; and Irwin Rose, at the time of the citation, at the University of California at Irvine, but did his seminal work at the Fox Chase Cancer Centre in Philadelphia. The award was actually given for the discovery of ubiquitin-mediated protein degradation, a very important finding at the time, but back then it was probably not quite clear how important this process was going to be generally in addition to its role in protein degradation. Now, like most major discoveries, the science often starts off with a very simple question, and this is no exception.
The question at the time was how are proteins degraded in cells, and why is energy required in this process? So the prevaling view in the 70’s was that the majority of misfolded and aggregated proteins were degraded through an organelle called the lysosome, into which proteins were subsumed and proteolytically degraded. However, this clearly wasn't the complete story. Firstly, it didn't explain how different proteins had turnover rates that were different. Secondly, the inhibitors of lysosomes could be applied to cells which were still able to turnover different proteins in an ATP-dependent manner; and, therefore, clearly was not responsible for the complete degradation of proteins in the cell. Finally, it was not clear why energy would be required in a process which is essentially exergonic and, therefore, produces energy, so what was this energy being required for?
So this is where Avram Hershko and his PhD student, Aaron Ciechanover, started out and they took the classical biochemical approach to try and identify ATP-dependent proteolytic degradation in cell extracts. In this classical piece of biochemistry they were very successful in doing this. They started off with a reticulocyte lysate which they fractionated into several fractions, which they could mix together and get ATP-dependent proteolysis. In particular, they had two fractions: fraction 1 of which contained a heat-stable component of unknown identity, which they called ATP-dependent proteolytic factor 1. Now, this fraction could be added back to the second fraction to reconstitute the whole ATP-dependent proteolytic reaction.
In later work with Irwin Rose, Ciechanover and Hershko subdivided fraction 2 into two more fractions; one of which contained an ATP stabilized protease, and another contained a number of factors required for ATP-dependent proteolytic degradation. The important thing here being then, you could mix these 3 fractions and get back the reconstituted reaction. Analysis of this very powerful system continued and a number of important findings were made.
Firstly, ATP-dependent proteolytic factor 1 was identified as a protein ubiquitin. Secondly, ubiquitin was shown to become conjugated to substrates which then underwent degradation, and this covalent attachment of the ubiquitin occurred through an isopeptide bond. They also isolated various other components to the reaction, outlined by Miriam, the E1, E2 and E3 ubiquitin-activating-conjugating and ligase proteins. The importance of their findings was firstly that they identified an ATP-dependent proteolytic system that was different from that of the lysosome. They also showed that it contained multiple components, and it wasn't an individual protease. Thirdly, it explained how you could have a tagging system which enables the protease and the substrate to be in the same cellular compartment, but in a regulated way so that protease activity does not go rampant on the substrates. This also explained the versatility of the degradation system in that it was tightly regulated, and different proteins could be regulated at different rates by specific tagging of the ubiquitin.
So, basically, they outlined a system of ubiquitin tagging involving these E1, E2 and E3 components that Miriam talked about, which identified a substrate for proteolytic degradation by the proteasome. But this was a biochemical demonstration, and what was really lacking at this point was a biological analysis of the importance of this protein system in cells.
This is where we introduce another major figure in ubiquitin biology, Alexander Varshavsky. He was a Russian scientist working at MIT and he was aware of Hershko, Ciechanover and Rose's work, but wanted to know what the impact of this system was in a cell. Now, genetically manipulating mammalian cells was not so easy back then, but Varshavsky identified a cell line which had a very interesting property. It was the conditional lethal cell line which when elevated in temperature, lost a protein in the cell which looked like ubiquitylated Histone H2A, suggesting that the cell line may be defective in one or more components of the ubiquitylation system.
He characterized this further to show that this cell line ts85 has a temperature-sensitive mutation in the ubiquitin-activating E1 enzyme. Because it has an inactive E1 at higher temperatures, histone H2A was not becoming ubiquitylated. At the same time, because the ubiquitin mediated degradation process wasn't going on, the bulk proteins which are normally degraded by the ubiquitin system accumulated at elevated temperature. But most importantly, because these cells died, he showed that the ubiquitin proteasome system was actually essential for cell existence.
To follow his ideas about generating a genetic system with which to study ubiquitylation, Varshavsky then turned his attention to the ubiquitin system in yeast where he made many seminal discoveries, along with his colleagues, such as Dan Finley. For example, it was the Varshavsky lab which first showed that the ubiquitin existed as a poly-ubiquitin precursor, which needed to be degraded before ubiquitin could then be added to substrates. He also identified a degron sequence which targeted specific proteins for degradation. Finally, he also identified a role for this ubiquitylation system in the cellular response to DNA damage.
So let's turn now to the versatility of the system. Miriam went through the formation of ubiquitin chains earlier, but let's just go over this again. Initially, ubiquitin is added to an internal lysine in a substrate, but iterative rounds of ubiquitylation often no longer target the substrate directly, but become conjugated to the first ubiquitin that has been added. Now, this ubiquitin, as Miriam pointed out, has seven lysines and so the conjugation to ubiquitin can occur at one of seven positions. In the case of K48 chains, which are known for tagging for proteolytic degradation, this chain is fairly long, and it is recognized by the proteasome.
However, you can also form chains on other ubiquitin residues such as lysine 63, and the point here is that by forming this different form of chains you'll end up with a different product. The structure of a lysine 63 link chain is likely to be very different from a lysine 48 link chain, it may be stiffer, it may be longer, it presents the individual different ubiquitin subunits in a different fashion for recognition by ubiquitin binding proteins of which there are many. So this just gives you an example of the kind of different flavors of product that ubiquitylation can give rise to.
So what about the function of these different chains, and different ubiquitin modifications? Well, we've talked before about K48 chains and how they're involved primarily in protein turnover. For example, the K48 chains which regulate the availability of cyclins, which control the cell division cycle. Lysine 11 chains are also involved in targeting proteins for degradation, but, interestingly, they seem to arise through the activity of a particular complex: the anaphase-promoting complex and, therefore, are almost exclusively associated with mitosis. K63 chains, rather than being involved in protein turnover, act as platforms in signaling pathways, for example, in NF-kB and in the DNA damage response. For example, K63 linked ubiquitin chains added to histone H2A calls the downstream recruitment of several proteins, including another ubiquitin ligase, BRCA1, which has been shown to capitalize the formation of lysine 6 chains. The function of lysine 6 chains is currently unknown. Not to be forgotten, is one of the earliest ubiquitin modifications identified, that which occurs on histones. This is mono-ubiquitylation, it's not involved in protein degradation. But the modification of histone H2A and H2B has been shown to act as an important marker for regulating gene expression profiles. This is achieved, it is thought largely, through its ability to modulate chromatin compaction and, therefore, the ubiquitylation of H2A and H2B is likely to be very important in epigenetics and the marking of cellular identity.
So ubiquitylation has been linked with a number of diseases, some of which are listed here. I'll just go over a few of them, for example, Fanconi's anaemia, a haematological disease in which mutation in one or more of the factors which impair the ubiquitylation of the Fanconi D2 protein, cause the patients to have attrition of their bone marrow stem cells and aplastic anaemia; and later on cancer predisposition. Muscle atrophy, which can occur following long-term immobilization or as a consequence of other diseases such as cancer, occurs largely as an imbalance between protein synthesis and protein breakdown, and we can see how loss of the ubiquitin proteasome system and protein breakdown might destabilize this. In cancer, a number of different cancers have been linked with the ubiquitin proteasome system, in particular p53 which we know is mutated in 50% of cancers, but is probably somehow disregulated in almost 100% of cancers. Many defects occur in the proteolytic system for regulating p53 function, and that is associated with ubiquitylation. Other diseases such as inherited blindness, inherited Parkinsonism and Angelman syndrome have also been linked to defects in ubiquitylation, and more recently the ability of viruses to infect cells has been shown to be dependent on ubiquitylation.
So how can we interfere with these ubiquitin pathways as a potential to disease therapy? Well, we could interfere with the activity of the E1 protein, but that's a rather blunt tool as it should disrupt the whole of the ubiquitin pathway. Slightly more specific is the inhibition of E2 proteins, which may be an individual E2 protein, may be limited to a particular cellular function but it's still going to affect a lot of individual proteins. More specific would be in interfering with the activity on the individual E3 ligase. Now, - inhibition of any of these three proteins - would result in a decrease in ubiquitylation, and an increase in the protein substrate. However, as Miriam said, deubiquitylating enzymes are very important and if we inhibit those we are likely to increase ubiquitylation of the substrate, targeting it for degradation.
Finally, there's a possibility of inhibiting the proteasome. Now, this, again, seems to be a blunt tool, but actually turns out to be rather successful. Inhibition of the proteasome using velcade or bortezamib has been found to be a very good treatment for multiple myeloma. This proteasome inhibitor binds to the catalytic site of the 26S proteasome and inhibits clearly not all protein degradation, but sufficient to help in remission of multiple myeloma. Other compounds listed here interfere with the p53 pathway by inhibiting Mdm2 or inhibiting the interaction between Mdm2 and p53 to effect p53 turnover. Again, we talked about muscle atrophy earlier, P013222 is a compound which inhibits the MuRF1 E3 ligase which is overexpressed in muscle atrophy, with potential therapeutic responses.
So what's the future? Well, there's a number of things to point to. Firstly, there's likely to be an increase in the number of ubiquitin-related diseases identified. Secondly, we can use the information we have to therapeutically target the ubiquitin pathway for therapy in these diseases. Finally, I think there's a lot of subtleties left in understanding the ubiquitin system, the different forms of ubiquitin chains that are formed, the different proteins that are involved in this process in order to identify potential new targets of therapy. At this point, I'm going to hand you over to Judith who's going to talk to you about another very important point of post-translational modification, and that is phosphorylation.
JL: Hello everybody, my name is Judith and I'm going to take you through the phosphorylation part of the webinar. Before I start I would quickly like to remind you that you can submit questions for Miriam, our guest speaker, and the phosphorylation part via the Q&A panel on the right hand side of your screen.
The basic mechanism of phosphorylation is kind of comparable to ubiquitylation. Indeed, ubiquitylation has been named the new phosphorylation. A kinase is a type of enzyme that transfers phosphate groups from high energy donor molecules such as ATP, and covalently attaches them to a substrate. This process is known as phosphorylation and reversed by enzymes known as phosphatases. Before I continue, I would quickly like to summarize the key facts about phosphorylation. Phosphorylation is the addition of phosphate; it is the reversible process that is conserved in prokaryotes and eukaryotes. In most eukaryotes serine, threonine and tyrosine residues become phosphorylated; I’m saying most as some eukaryotes such as plants and slime molds have been reported to assist within phosphorylation. In eukaryotes, serine phosphorylation is the most common, and tyrosine phosphorylation is the most rare. In prokaryotes, phosphorylation does not only occur in serine, threonine and tyrosine, but also in histidine, arginine and lysine residues.
After having given you a brief overview, I would like to look in a bit more detail at the key enzymes involved in the phosphorylation process. This diagram shows the human kinome, which encompasses 518 kinases that fall into eight families. These families have been often named after the substrates they phosphorylate. Most kinases act on serine and threonine residues, others on tyrosine and some are dual specificity kinases, which can phosphorylate serine, threonine and tyrosine residues. Specific inhibitors are available for kinases such as glycogen synthase kinase 3. These originally started off as broad spectrum kinase inhibitors such as lithium, and that's been developed into more sophisticated inhibitors that specifically inhibit the activity of a single kinase.
How do I know that the protein I'm working on is a kinase? You can use websites such as InterProScan from the European Bioinformatics Institute, which predict motifs. As you can see, the protein sequence I submitted is a kinase and, more specifically, a dual specificity kinase. If you have any preferred modes of prediction sites, please feel free to share them with the audience by entering them into the panel in the right hand corner of your screen. We will make them available to attendees after the webinar.
Because kinases have profound effects on the cell their activity is highly-regulated. Kinases themselves are turned on or off by phosphorylation. This could be the autophosphorylation or by the activation of an upstream kinase. Their activity also gets regulated by binding activator and inhibitor proteins. On this slide, you can see the sequence alignment of three human kinases. The kinase domain starts in the orange box and contains a p-loop, a catalytic loop and an activation loop.
Phosphatases fall into similar families to kinases, they are serine/threonine phosphatases, tyrosine phosphatases, also called PTPs, and dual specificity phosphatases. In total, there are hundreds of serine/threonine phosphatases, and around 90 dual specificity and tyrosine phosphatases in the human genome. As for kinases, you can use prediction programs to identify whether your protein of interest is a phosphatase. In general, phosphatases are a lot less well understood due to the lack of specific inhibitors.
Next, I would like to look at the different types of phosphorylation. There is mono-phosphorylation: one phosphate group is added to one residue. There is multi-phosphorylation: one phosphate group is added to multiple residues. Poly-phosphorylation: the equivalent of poly-ubiquitylation does not exist.
In the next part of the webinar I would like to help you identify whether the protein you are working on is phosphorylated, and, if so, identify the type and actual site of phosphorylation. To get a first idea of whether the protein you are working on could be phosphorylated, it is good to use phosphorylation prediction sites. I would start off with identifying whether your protein is conserved in other species, and, if so, check the literature to find out the phosphorylation site. This can be quickly done by using UniProt, as is shown here for beta-catenin. I would then follow this up with a sequence alignment to check whether the residue is conserved in the species you are working on.
Several websites are available to predict phosphorylation sites, for example, NetPhos 2.0. The example shown here is for tyrosine phosphorylation predictions. As you can see, all tyrosine residues within the protein are listed according to where they are located. The program awards a score to each residue. One is the highest possible score and zero the lowest, therefore, the tyrosine residue shown in blue with the respective scores of 0.92 and 0.884 are the ones most likely to be phosphorylated.
Another example is the Scansite Motif Scanner, which allows you to look at whether your protein of interest could be the substrate of a specific kinase. An example is shown here for Abl and Akt kinase substrates. If you're using different websites to predict phosphorylation sites, please feel free to share them with the audience by entering them into the right hand panel corner of your screen. We will make them available to all delegates after the webinar.
Should you have been lucky enough to find that only one residue as predicted to be phosphorylated, you might be able to perform western blot analysis for commercially available antibodies. The western blot shown here is detecting serine 10 phosphorylation of histone H3. The histone 3 variants are predominantly found in non-dividing cells. As we can see here, colcemid treatment results in the metaphase arrest and, therefore, an increased serine 10 phosphorylation on histone H3. Abcam offers a variety of phospho-specific antibodies covering various research areas ranging from epigenetics to microbiology.
This western blot looks at the difference in phosphorylation of a STAT protein in response to different treatment periods. STAT proteins are transcription factors that become activated and move to the nucleus in response to phosphorylation. This blot compares STAT activation in the knockout versus a random integrant strain. A random integrant is the cell line where the gene disruption cassette is randomly integrated into the genome. It looks like there is a clear difference between these two cell lines. However, to confirm and quantify the difference of phosphorylation, you have to first ensure that equal amounts of protein have been loaded. We can determine this by using a general loading control such as tubulin.
However, it is better to probe for the non-phosphorylated form of the protein to ensure that the protein levels of the protein you're interested in were equal to start off with. This can either be done by stripping and re-probing the membrane, or by starting of running duplicate gels. For normalization and quantifications, you can use free software such as ImageJ. One thing to keep in mind is that you should not overload the gel, do not overexpose the film, and as film is never linear we recommend using a fluorescent scanner as these allow more accurate quantifications. Also, repeat the experiment multiple times to determine the average and the standard deviations.
What if the prediction software programs or commercial antibodies have not shed any light as to whether your protein of interest is phosphorylated? Then the best way forward is to run a western blot of an immunoprecipitated protein. Phosphorylation will normally reside in a reduction of a protein's mobility on SDS-PAGE. This can be observed in the below blot where treatment results and increased mobility, also referred to as a mobility shift. Alternatively, two bands can be observed. Of the two bands, the upper one is the phosphorylated variant and the lower one is the non-phosphorylated form.
How can I ensure to get a clear mobility shift? Use IP samples as this results in a concentration of your protein of interest. Find the best percentage gel and for smaller proteins, you should increase the percentage, and for larger proteins you should reduce the percentage. Find the best type of gel and buffer. It is better to run a single percentage gel than a gradient gel, and consider different gel types such as Tris-Glycine and Tris-Acetate gel. You can also consider different running buffers such as MOPS and MES to increase resolution. Run the gel slowly to avoid smiling and also run the gel in an ice-bucket or in the cold room to avoid overheating.
Another indication of whether a protein is phosphorylated or not is whether you can see a shift on a 2D gel. Indeed, phosphorylation replaces neutral hydroxide groups and negatively charged phosphate groups. Below pH 5.5, phosphates add a single negative charge; around pH6.5, they add 1.5, and around pH7.5 two negative charges. In the example shown here, you can see that the protein migrates as a series of irregular spots. Upon treatment, the protein shifts towards the acidic range. IgG was used as a loading control here to ensure that equal amounts of proteins were loaded to start off with. As not only phosphorylation can result in such a shift, but also as a modification such as acetylation, it is best to perform phosphatase treatment. In this case, a serine/threonine/tyrosine phosphatase such as lambda or alkaline phosphatase was used. As we can see, the dots disappear and the protein shifts towards the basic range. Phosphatase-treated samples can also be run on a 1D gel. As we can see here, treatment with the serine/threonine phosphatase such as PP2A abolishes the mobility shift. As a control, PP2A treatment together with a phosphatase inhibitor was performed. In that sample, again, the mobility shift can be observed. Phosphatase inhibitor treatment should always be performed as a negative control.
As we have seen on the previous slides, treatment with serine/threonine versus serine/threonine/tyrosine phosphatases can be performed to distinguish between tyrosine and serine/threonine phosphorylation. Alternatively, you can perform western blot analysis with phospho-serine-specific antibodies. An example of this is shown here. For p53 serine phosphorylation after your retreatment. You can also perform western blotting with phospho-threonine, and phospho-tyrosine-specific antibodies, in addition to phospho-serine and threonine antibodies.
After having determined the type of phosphorylation, it is key to identify the site. To identify the actual phosphorylation site, it is best to perform immunoprecipitation of the protein of interest from your cell line of choice, or a cell line overexpressing your tag protein of interest. The IP sample is run on SDS-PAGE, Coomassie staining is performed, and the protein band is excised with a clean scalpel. This is followed by enzyme digestion. Commonly used enzymes are trypsin or chymotrypsin. However, you have to choose the proper digestion enzyme and the method according to your protein sequence. The last step in identifying the phosphorylation site is mass spectrometry analysis.
What to keep in mind when performing phosphorylation site analysis? Avoid degradation by having the samples on ice at all times, and by adding sufficient amounts of protease inhibitors. Use sterile equipment to avoid keratin contamination, and do multiple repeats to ensure that the phosphopeptides you identify are real.
How to identify the role of the phosphorylation? Once the peptide has been identified by mass spec it is important to determine the actual residue. For example, in the peptide sequence shown here, either the serine or the threonine, or both residues could be phosphorylated. Should you not have overlapping peptide sequence information available that helps you to identify either the serine or threonine is phosphorylated, it is easier to perform point mutations and knock these into your cell line of interest. Mutating a serine or threonine to an alanine will make the peptide unable to be phosphorylated and create a phospho-defective mutant. If you're analyzing tyrosine phosphorylations, the tyrosine residue would have to be mutated to phenylalanine to create phospho-defective mutants. Mutating the serine/threonine to glutamic acid or aspartic acid will mimic phosphorylation; you are creating pseudo-phosphorylation as you're replacing the residue with an amino acid that has an endogenous negative charge.
In general, it is best to do both single and multiple point mutations, and to analyze whether mobility shift can still be observed in these. It is also important to establish whether the mutants are viable, and to determine whether there are morphological phenotypes. For example, phosphorylation could alter a protein's activity, cellular localizations or interaction partners. To further investigate, it is best to generate phospho-specific antibodies and to identify the kinase responsible for the phosphorylation by performing in vitro kinase assays.
Next, we will look at how to troubleshoot the detection of phosphorylation. What to keep in mind when wanting to detect phosphorylation? Keep the samples on ice at all times and add freshly prepared phosphatase inhibitors, including general inhibitors such as phosphatase inhibitor cocktails and specific inhibitors like sodium fluoride, which is a protein serine phosphatase inhibitor. Or sodium orthovanadate which specifically inhibits protein tyrosine phosphatases. If you establish the optimal phosphorylation conditions, are you using the correct stimulus and the right amount of stimulation at the correct developmental stage? For SDS-PAGE conditions, is your sample properly denatured? Also make sure that you're using the optimal gel percentage buffer and gel type, and that you're running the gel slowly and at 4°C.
In order to avoid a high background signal, use BSA as a blocking agent. Do not use milk, as milk contains the phospho-protein casein, which will be picked up by the phospho-specific antibodies. Use TBST and not PBST as your buffer, as PBS contains phosphate. Should you like more information about how to perform western blot detection of phospho-proteins, have a look at our phospho-western blot protocols. For more general western blot and IP information, check out our previous webinars on western blotting and immunoprecipitation, and have a look at our western blot troubleshooting tips.
To summarize the phosphorylation part of the webinar, phosphorylation is a post-translational modification that controls many processes within cells. Indeed, it is estimated that 30% of the proteome is phosphorylated. It is the addition of phosphate to proteins, and this mechanism is tightly regulated. Kinases are responsible for the specificity of the phosphorylation reaction. Proteins can either be mono- or multi-phosphorylated, and serine, threonine and tyrosine residues. Phosphorylation can alter a protein's activity, cellular localization and interaction partners. Determining the site and function of a phosphorylation event should be a combination of experiments.
I would now like to introduce some of our products that will enable you to detect and identify post-translational modifications. Abcam offers a range of active ubiquitin and ubiquitin-related proteins, including single lysine ubiquitin proteins, lysine to arginine ubiquitin mutants, E1s, E2s, E3, DUBs and ubiquitin-like proteins such as SUMO. For more information please have a look at our ubiquitin resource page.
For superior western blot results, Abcam offers our Optiblot western blot product range. This new range contains pre-cast gels, running buffers, prism protein ladders and gel-staining reagents. In addition to that, we have Optiblot electrophoresis kits providing denatured/unnatured proteins. If you are intrigued, please have a look at our Optiblot resource pages.
For the detection step of your western blotting experiment, we recommend AbExcel secondary antibodies. These 19 AP- and HRP-conjugated secondaries have been extensively tested in the Abcam labs. The optimal dilution has been determined, and we are pleased to tell you that you can do over 1,500 blots with a vial of AP-conjugated, and over 600 blots with a vial of HRP-conjugated AbExcel secondary antibodies. If you would like to learn more, please view our AbExcel pages.
An alternative to determining the amount of phosphorylated proteins by western blotting, is to perform ELISAs. In general, ELISAs tend to be more accurate than western blotting and allow the detection of lower banding proteins. Our fast and sensitive Phospho-Tracer ELISA kits allow the accurate detection of a single phosphorylation site, such as Akt serine 473 phosphorylations. We also have Phospho-Tracer ELISA kits to measure the modified and total form of a protein, and to detect a combination of modifications of several members of a protein family at once. For more information, please have a look at our Phospho-Tracer resource pages.
AbcamBiochemicals offer a large selection of high purity and exceptional quality products. To elucidate phosphorylation processes, various kinase activity modulators and phosphatase inhibitors are available. Should this slide not contain what you're looking for, why not check out the AbcamBiochemicals website.
Abcam's Scientific Support Team is there to answer any questions you may have. The key members are multilingual and offer support in a range of languages including French, Spanish, German, Chinese and Japanese. You can contact them in the US, UK, Hong Kong and Japan.
Also, I would like to highlight our future webinars. We would be delighted if you could join us on June 21st for a webinar on In-Cell ELISAs. This webinar will be presented by our colleagues at MitoSciences.
As a special thank you for attending this webinar, we would like to offer you a special 20% discount on all AbExcel secondary antibodies, Optiblot products, ubiquitin proteins, kinase and phosphatase activity modulators and Phospho-Tracer ELISA kits. All you have to do in order to take advantage of this offer is to quote promo code PTM-E71W4 when placing your order. I would like to thank Miriam and Kevin, and all of you for attending and would like to let you know that the three of us are happy to now answer your questions.
MF: Thank you, Judith. We have a question from Sophie and she's asking: In a poly-ubiquitin chain can you have different linkages, for instance, K83 then K67? Why or why not, and, if so, which ubiquitin determines the fate of the substrate? I think Kevin can answer this question.
KH: Okay Sophie, well, there is evidence that you can get mixed chains, mixed ubiquitin chains both in vitro and in vivo, but I could say at this point, the function of those mixed chains is unlikely to be understood, probably even partially. So in vitro it's very difficult, because ubiquitin ligases are very promiscuous and so you don't know if what you're looking at is really biologically relevant. In vivo looking at chains is very difficult without reagents that specifically target those ubiquitin chains and, again, those are proving very elusive. So, yes, it can occur, but its function was not at all clear.
MF: I have another question for Judy from Jung Li, and she's asking: Does phosphorylation occur in a particular subcellular site, for instance, the Golgi, or can this modification occur anywhere in the cell?
JL: It is known that phosphorylation can occur anywhere within the cells; it's not limited to a specific site.
MF: We think we have time for one more question. Now, John is asking: What would you recommend as the best negative control for ubiquitylation assays? We would recommend a mix of controls, so, for instance, inactive E1, E2 or E3 enzymes, and ATP is always a very good control. Also, it's possible a non-specific substrate could be used. Ubiquitin mutants, as you can see, can also be used as a negative control. Also, for instance, a negative substrate which is unrelated to the ubiquitylation, so you’re sure this will not be ubiquitylated.
JL: We've also got a question from David who asked us whether there is no easier way to determine that his protein of interest is tyrosine phosphorylated? To answer your question, David, you could do treatment with T-cell protein tyrosine phosphatase, or you could perform an IP with commercially-available phospho-tyrosine sepharose or agarose beads. That's a lot quicker than going via the 2G gel experiments I showed you earlier.
Thank you Kevin, Judith and Miriam. Unfortunately, we have now run out of time, but if we have not answered your questions we will get back to you very shortly with answers. On behalf of Abcam, I would like to thank you all for attending and we hope you have found this webinar useful and informative. The webinar presentation is available for download. When you log-off from the webinar you will be automatically redirected to a webpage where a PDF can be found and downloaded. Also on this webpage you'll find information about this special webinar promotion we're running. If you have any questions about post-translational modifications or any scientific enquiries, please do not hesitate to contact our scientific support team who are more than happy to help you. They can be contacted at [email protected] That's the end of the webinar; we look forward to welcome you to another webinar in the future. Thank you again for attending and good luck with your research!