It's a sugar rush. Not the kind that accompanies consuming candy. But labs working on sugars are feeling quite a sugar rush about their field. One aspect that makes research into sugars important is that sugars are everywhere, for example they coat all cells in the human body.
By modifying sugars, labs are finding ways to convert one blood type to another.
Separately, projects are ramping up to tally and characterize the sugars in the human body. One project is called the Human Glycome Project. That’s an international initiative to explore and map the human body’s sugars. The second meeting of the Human Glycome Project is just getting underway in Split, Croatia as part of the International Society for Applied Biological Sciences meeting.
New resources in glycomics, the scientific field devoted to studying sugars, are being set up to invite more labs to add glycan analysis to their question of interest. It’s taken a while and a lot of methods development is still underway. This podcast is about some recent developments and about what is emerging.
I did a blog post a while ago that also includes some glycoscience resources. They are also pasted below.
(Music credit: Split Phase--Three Month Funk; Photo: PhotoDisc/Getty Images)
You will find a transcript of the podcast below. If you are able to do so, please listen to the audio, because spoken word is distinct. Please note that the transcript might not be identical to the spoken word in all instances. If you would like to quote from the podcast, please credit the source and please refer to the audio.
Hi, I’m Vivien Marx. And this is Sugar Rush, a podcast about sugar. These days, some labs are feeling a bit of a sugar rush.
Stephen Withers, Peter Rahfeld and colleagues at the University of British Columbia have recently developed a way to convert blood type A to O-type universal donor blood type. That conversion is all about sugar. And being able to do this conversion is potentially crucial help for blood banks and for people needing blood. For this conversion, the researchers found two enzymes and they found them in an anaerobic bacterium when they were doing metagenomic screening of the human gut microbiome.
Sugars play an important role with the human blood groups A, B, and O. Blood groups differ in the sugars on the surface of red blood cells. The sugars there are modified in different ways. And it’s these differences that make for different blood types. With blood type A, the conversion of A to O requires removal of one type of sugar and to convert blood type-B to O another type of sugar.
Labs have looked into ways to do such blood group conversions since the 1980s. And the biochemistry of it has worked. But one issue has been, as the UBC researchers point out, that gram quantities of the converting enzyme were needed. That makes the approach impractical.
The blood-type conversion from A to O from the UBC team is not yet an application. But it’s a step closer to enable conversion of A to O. Another example that underscores the importance of sugars, they’re also called glycans, is that therapeutic monoclonal antibodies are glycosylated, they are covered in sugars.
Pamela Marino All of the biotherapeutics are glycosylated and glycosylation affects how long they stay in the system circulating and their effector function especially monoclonal antibodies: you tune effector functions by the glycans that you attach
Vivien Marx That’s Pamela Marino, a biochemist at the National Institutes of Health (NIH) where she directs the biochemistry and bio-related chemistry branch in the division of pharmacology, physiology and biological chemistry at the National Institute of General Medical Sciences (NIGMS). And she is a program coordinator in the NIH Common Fund’s glycoscience working group. She’s been overseeing the funding and research in many labs through the NIH Common Fund and NIGMS.
Pamela Marino Glycobiology was coined in early 1980s, the term, people coalesced over common interests, they ran a meeting annually out of somebody’s back pocket somebody ran it out of their lab.
It had the potential of an ‘omics field. It was the third information-rich polymer in the body. You had nucleic acids, proteins, you had carbohydrates.
It’s all coming together now. You need the tools, the libraries, you need to be able to synthesize the right compounds, sequence these, you need to have things in databases. At this point, we’ve stood up this field, it’s taken the better part of 25 years to do it. But we managed to stand up this field.
Vivien Marx Labs come across glycans in different contexts. They might be studying cell biology, neurobiology, development, immunology. Their questions arise in any number of other fields. Marino sees many papers ‘with glyco’ in them even if they do not uniquely focus on glycans.
Pamela Marino I think everybody sort of backs into glycoscience. They’re working in different fields and as they progress from DNA to RNA, to proteins to post-translationally modified proteins to function, they realize the biggest post-translational modification is glycosylation. Oh, by the way even those glycans can be post-translationally modified themselves, that puts a plethora of regulation on how these things function.
Vivien Marx The Human Glycome Project is an effort that is setting out to tally and map all the sugars in the human body. There’s a lot to do: all cells are covered in sugars. The Human Glycome Project is organized by Rick Cummings at Harvard Medical School, who has put in place a high-throughput screening center for glycans. And by Gordan Lauc at the University of Zagreb. Lauc founded a company called Genos, where his lab is located and which has a number of EU-funded research projects.
In the Cummings and Lauc labs, large-scale characterization of human glycans is getting underway. And Lauc sees glycobiology as a rich field that is just waiting for discoveries.
Gordan Lauc One of the observations we can make is that practically every protein that has appeared after the appearance of multicellular life is glycosylated. That every single organism is covered with glycans on their surface. There has to be a reason why these glycans are there. In many cases, we still do not understand. But for example, we know that all interactions between humans and pathogens or commensal bacteria, go through glycans. All the viruses attach to glycans. Many proteins are regulated by glycosylation.
Depending on which glycans you put, they will have a different mode of action depending on which glycans are put. Now there are over 30 glyo-engineered monoclonal drugs in clinical trials. Definitely there are many glycans around, they do many important things. The problem, it’s not written in the genes, we can’t just read the genome and know anything about glycans.
If you produce recombinant protein it will either have no glycans or the wrong glycans so we cannot study glycosylation on model proteins. We have to look into the living organism take a sample and analyze the glycans.
Vivien Marx Lauc has just set up a collaboration with two companies, New England Biolabs and Waters. For three years--so that’s until December 2021--Lauc and his team will analyze 10,000 samples a year. To do glycan analysis, for that sum of 30,000 samples, he and his team will get reagents and consumables for free. Free is certainly enabling. Methods have evolved and glycan characterization has become faster. Gordan Lauc talks about the pros and cons of different methods.
Gordan Lauc Mass spec is one of the leading methods to determine exact structure of glycans, so fragmenting, analyzing fragments, you can really determine structure of a specific glycan. Mass spec is also becoming more quantitative. There are some high-throughput quantitative mass spec methods.
For IgG, you can analyze all glycoforms and all subclasses, using LC-MS. But MS is still relatively difficult to use, not so reliable machines, people need a lot of experience. There are maybe two labs in the world that do high throughput glycomics on mass specs.
Two that’s not enough, we definitely need more laboratories, which would do high-throughput glycomics. Alternative to mass spec.
Chromatographic approaches are an alternative to mass spec, fluorescent labeling then separation by high–performance HPLCs then you can have the profile and you can reliably quantify individual chromatographic peaks.
The problem there is you do not always know what is an exact peak. In principle, you have to couple this to mass spec, determine individual structures, and then when you do multiple measurement.
Some labs can do 20-30,000 samples a year using the approach. This is the bread and butter of the Human Glycome project. This is the bread and butter of the Human Glycome Project, this where we can do thousands of samples.
The problem with HPLCs, you still need a half an hour per machine per sample, so if you want to do a million people, it will be difficult.
Third method, which is also developing is capillary electrophoresis.
Practically, what people have done is that they have repurposed DNA sequencers.
The old Sanger DNA sequencers are kind of retiring, not completely, as they switch to second generation sequencer and there are thousands of CE machines not so used so much. Several companies have kind of repurposed them for glycan analysis.
In theory, with a 96-capillary machine, you can analyze a huge amount of glycans. The problem here you see the peaks from CE and you don’t know what is in a peak. And you cannot easily couple CE and mass spec.
You can, but not if you have a gel in the capillary and for glycan separation you have gels. It’s difficult to determine what is in a peak. So you definitely you have to do enzyme arrays, sequence the glycans.
Once you have determined them--for IgG, it’s well established--what is the peaks, you can do hundreds of thousands of samples with this technology.
I think this is one of the ways glycomics can move a step forward and do the big biobanks, like for example the ‘All of US’ biobank or the UK Biobank with half a million people. I think this will happen relatively soon, technology is here, interest is here.
The problem is somebody will have to pay for that. People paying for high-throughput screening don’t have a problem investing several hundred million dollars in DNA sequencing. But nobody is willing to invest a few million in glycan sequencing. Which will change with time. But for now, there is no interest to do it, among the funding bodies.
Vivien Marx Lauc’s collaboration with Waters and NEB is likely going to help move the Human Glycome Project ahead and help his and other labs explore ways of doing high-throughput screening of glycans. That type of analysis is done differently from one lab next to the next.
Chris Taron You go to different camps, different leaders. Everybody is doing glycan analysis but with a little different flavor of it.
Vivien Marx That’s Chris Taron. He’s a glycobiologist at New England Biolabs (NEB) who directs the protein expression and modification division at the company.
Chris Taron You have some just going onto UPLC and profiling getting a fingerprint of the glycans that are present, you have others only do mass spec and fragmentation, they just want to let the mass spec speak. You have some people doing capillary electrophoresis, they’re all good and they all have drawbacks. There are lots of different flavors of it.
Vivien Marx Taron and his team work on methods to make analyzing glycans easier. Around six years ago the company was exploring the bottlenecks that labs were experiencing in glycan analysis. One important enzyme that they use is PNGase F.
This enzyme helps scientists remove N-linked glycans from glycoproteins. N-glycans attach to aspargine. NEB sells PNGase F and so do other companies. The enzyme deglycosylates the glycoproteins and then the sugar and the protein can be analyzed further. NEB developed an enzyme that works faster than the classic PNGase F.
Chris Taron Ours is special. It’s called Rapid PNGaseF. I will not get into the technicalities of what makes it different. There is trade secret involved there.
Vivien Marx NEB and Waters have partnered to build a sample prep kit for glycan analysis. It’s called GlycoWorks RapiFluor-MS N-glycan kit. It includes Rapi-Fluor, that’s a proprietary fluorescent tag from Waters; the NEB enzyme Rapid PNGase F; there’s a customized buffer; and a surfactant. The kit lets a lab take a glycoprotein, deglycosylate it with the enzyme, tag it, clean it. And it’s ready for further analysis. Chis Taron talks about the company involvement with the Human Glycome Project.
Chris Taron Our story that led to our involvement with the Human Glycome Project initially began an internal evaluation on glycobio products where we were looking at anything else, glycobiology products, some of the bottlenecks. This was maybe 5-6 years ago.
One of the things we kept coming back to, with PNGase F at the front end of glycan analysis workflows, most people were running digests that would need to be run oftentimes 12 hours, sometimes longer.
The sample prep alone for just one sample would sometimes be an overnight process, most of the time be an overnight process. If you superimpose that on what Gordan and Rick are trying to do in terms of throughput for the Human Glycome Project you can see how quickly sample prep becomes cumbersome.
One of the projects we were working at the time was to try and innovate around PNGase F. Could we find conditions or make formulations, or new ersions of the enzyme that allow us to more rapidly deglycosylate, we were focusing largely on IgG and monoclonal antibodies.
Ultimately we were able to get that program down to 10 minutes for complete deglycosylation of a monoclonal antibody. That was a huge step in the right direction.
Concurrent with what we were doing, Waters was working on glycan labeling. So after the release step, when you have a freed N-glycan, you want to label it, you want to derivatize it with a fluorophore. They were working on new fluorophores and new labeling chemistry with super-fast chemistry. So you could label the glycan very quickly.
We teamed up with Waters for a project what would ultimately end up being their be Glycoworks kit that they currently sell, where we were able to get glycan analysis, the sample prep under one hour. Which was a huge. And the addition of kits helps to streamline conditions so glycan analysis could be more streamlined.
This was at the same time that Gordan was starting to organize academics and starting to think about how glycomics how to be expanded upon to go after large population cohorts. There the same concepts were true there. You couldn’t do it if sample prep was going to take 12 hours or more. You really needed things to be quick You could see the synergy and why we started speaking with Gordan at that time.
In 2016, Gordan brought a bunch people to Dubrovnik, Croatia, for an invitation conference to think about where glycomics could go: epidemiologists, clinicians, the leaders in glycomics at the time, whether you were using mass spec or CE.
Everybody was there and we were trying to think about ways that we could streamline glycan analysis, make it more uniform. With the ultimate aim of applying deciphering the glyco-code for humans.
Through that it was that born the concept what would become Human Glycome project. Over the ensuing couple years, we worked with Gordan and the folks Waters, to come up with a framework where Genos, a Croatian company that does high throughput glycomics, New England Biolabs that makes rapid deglycosylation with Rapid PNGase F, Waters, that makes and sells the Glycoworks kit for glycan analysis would donate service and materials to pool of reactions, so to speak, that would allow clinicians to petition for glycan analysis to be done alongside of their analysis of DNA or protein in a cohort. And we would be able to fund that from this pool of reactions that we’ve created.
This innovation, the workflow and various components is built with target audience of pharmaceutical analytics in mind. The same workflow can be absolutely applied academic samples and in the case of the Human Glycome Project to high-throughput sample analysis.
The problem, when you start to get to the numbers of samples that Gordan and Rick are interested, in analyzing: any component in a reaction can become a rather significant cost barrier. If the number of reactions goes high enough.
So the effort by Genos, Waters and NEB is really to take cost out of the equation on these first 30,000 reactions so we can start to get glycans alongside proteins and nucleic acids in some of these really interesting cohorts. So we can get a picture how the glycans are changing in these cohorts.
Vivien Marx As the project gets under way, labs will have Big Data challenges to solve. And they need to share data. With colleagues at the Swiss Institute of Bioinformatics, Lauc and his team are building a software tool that will let people query the data interactively.
As of right now, though, it’s not yet clear how much data will end up in the public domain, since use of the samples and data derived from the samples is restricted. The samples are from biobanks and in each case, the samples are governed by the ethics guidelines of the individual biobanks.
The team plans to work with the biobanks to put as much data as possible in the public domain. They want to expand the field, and to do so, they need to make data accessible and usable by researchers outside the field of glycoscience. Eventually there might be an atlas of the human glycome that many will want to access.
As NIH’s Pamela Marino explains, the research community is generally working to build homes for glycoscience data and working on ways to query these data. For example, there’s a project underway to better incorporate glycoproteins in the Protein Databank, the PDB. That’s a project tended to by the Complex Carbohydrate Research Center, the CCRC, at the University of Georgia. Within the CCRC it’s run by the lab of Rob Woods. NIH’s Marino offers a bit of context.
Pamela Marino The PDB continues to be supported. Using this Common Fund, we’re funding Rob Woods, who is cleaning up the PDB.
There are proteins in there and some of them have carbohydrates or carbohydrate attachment sites, or they know carbohydrates are there but they don’t have them in the proper linkage, they don’t the proper sugars, they don’t have them in the right places.
So our Common Fund effort is essentially paying Rob though a small grant to work with the PDB and to expand the PDB appropriately and clean up the glycan structures on these proteins, they’re glycoproteins, and to make sure that they’re correct that the linages and the sugars are correct.
Vivien Marx Databases have come in gone in glycoscience. There are issues. For example, databases have different naming conventions. There is, for example, a database called UniCarb DB. But in the mid-1990s funding ceased for that resource. It contained around 5,000 glycan structures.
Pamela Marino The value of that database was huge and Germany, Australia, the US, I think 4 or 5 major sites France, took that database, salvaged parts of that, sort of, tore it apart and rebuilt their own databases. So there were multiple glycan structural databases across the world. I think there were 5 or 6 major ones where people were trying to find ways to put for structures and to build these up. They didn’t talk to each other, because people use different nomenclature.
Vivien Marx In 2001, NIGMS launched an initiative with a number of research institutions called the Consortium for Functional Glycomics, the CFG. The grant was renewed in 2006 and funding ended in 2011. Its steering committee was directed by James Paulson at Scripps Research Institute. The consortium launched a Functional Glycomics Gateway where researchers can request resources and explore databases. That is now run by Rick Cummings at Harvard. Pam Marino talks about some projects now underway that build on all of these developments.
Pamela Marino Through the CFG brought these players to the table. Asked them to find a common ontology and find a way to get all these structures into a single database. The CFG built its own database, again leveraging data from Carb the databases. They decided that the best way to do things was to essentially continue doing things the way they were and develop a universal translator so that all the databases throughout the different countries could be interrogated from a web-based site.
They developed a language called GlydeCT and used that to interrogate all the databases through this universal translator.
There’s lots of different databases but we’re hoping to make this one home. And that would be this workbench where things can be plugged in and accessed by a web based interface, which would use middleware to interrogate the databases.
You have all the information on the backend from genome to glycome and the databases would be interrogated through a middleware. And that middleware would be accessed through a front-end that would be web-based.
Vivien Marx Beyond databases, there are plenty of issues to tackle related to glycans. The enzyme PNGaseF and its cousin Rapid PNGase F work with N-glycans but not with O-glycans, which link to threonine or serine in glycoproteins.
Chris Taron I would say emphatically that PNGase F is the most important enzyme in that space, there are other endo glycosidase that act on N-glycans. PNGaseF seems to be the enzyme that most people use at the front end of most analytical workflows. It’s been that way since the early 1990s when it was discovered. And it’s an incredibly important enzyme.
We have heard over the last 20 or 30-years how the field would love to have a broad specificity O-glycosidase that can remove O-glycans from glycoconjugates the way PNGaseF remove N-glycans from proteins. 9.31 It’s not for lack of trying to find such an enzyme on our part. We have really put a lot into that in the past. Never found as broad a specificity as people would like
Vivien Marx It’s not easy to study sugars. And there has been some prejudice against the field. Some even think sugars are just too hard to study. It’s something that Gordan Lauc has experienced time and again in his three decades in glycobiology.
Gordan Lauc As a field we are struggling with ignorance, because people just ignore glycosylation. Most of the scientific find this to be difficult, complicated and they just ignore. They live in denial.
Everybody knows sugars are there. Up until around 10 years ago it was nearly impossible to study them quantitatively and qualitatively. But this has changed. In the last decade, the methods have developed which can analyze glycans both in structural detail and quantitatively, the ratios of different structures.
Now, people got used to ignoring them and so they keep ignoring them. But there are more and more examples where adding glycans to your study give additional information and helps understand in biology.
Vivien Marx That was Sugar Rush, a podcast about sugars. I’m Vivien Marx. Thanks for listening.
Here are some glyco-science resources :
|General resources and funding information|
|Transforming Glycoscience: A Roadmap for the Future||Report by the National Research Council of the National Academies of Science|
|NIH Common Fund program in glycoscience||Funding opportunities from the NIH Common Fund program in glycoscience|
|A roadmap for Glycoscience In Europe by BBSRC, EGSF, European Science Foundation||Glycoscience roadmap for Europe|
|GlycoNet||Resources related to glycoscience research in Canada, based at the University of Alberta where the Alberta Glycomics Centre is located|
|National Center for Functional Glycomics||A Glycomics-related Biomedical Technology Resource Center based at Beth Israel Deaconess Medical Center, Harvard Medical School with resources on, for example, microarrays and microarray services, protocols, training and databases|
|Databases and portals|
|CAZy||Carbohydrate-Active Enzymes, a database of enzyme families that degrade, modify or create glycosidic bonds|
|Consortium for Functional Glycomics||Resources and glycoscience data. Part of the National Center for Functional Glycomics.|
|ExPASy||Software tools and databases to simulate, predict and visualize glycans, glycoproteins and glycan-binding proteins|
|Glycam||A resource for predicting the three-dimensional structures of carbohydrates and macromolecular structures involving carbohydrates.|
|Glycan Library||A list of lipid-linked sequence-defined glycan probes|
|Glyco3D||A portal for structural glycoscience|
|GlycoBase 3.2||A database of N– and O-linked glycan structures with HPLC, UPLC, exoglycosidase sequencing and mass spectrometry data|
|GlycoPattern||Portal for glycan array experimental results from the Consortium for Functional Glycomics|
|Glycosciences.de||Collection of databases and tools in glycoscience|
|GlyToucan||Repository for glycan structures based in Japan|
|MatrixDB||A database of experimental data of interactions by proteoglycans, polysaccharides and extracellular matrix proteins|
|Repository of Glyco-enzyme expression constructs||University of Georgia Complex Carbohydrate Research Center repository for glyco-enzyme constructs|
|SugarBind||A database of carbohydrate sequences to which bacteria, toxins and viruses adhere|
|UniCarbKB||A resource curated by scientists in in five countries. It includes GlycoSuiteDB, a database of glycan structures; EUROCarbDB, an experimental and structural database and UniCarb-DB, a mass spec database of glycan structures|
|CASPER||Web-based tool to calculate NMR chemical shifts of oligo- and polysaccharides|
|Glycan Builder||An online tool at ExPASy for predicting possible oligosaccharide structures on proteins|
|GlycoMiner/GlycoPattern||Software tools to automatically identify mass spec spectra of N-glycopeptides|
|GlyMAP||An online resource for mapping glyco-active enzymes|
|NetOGlyc||Software tool for predicting O--glycosylation sites on proteins|
|SweetUnityMol||Molecular visualization software|
Sources: NIH, R. Mazumder, George Washington University; New England Biolabs, Thermo Fisher Scientific, Nature Research