– Good afternoon, everyone. I’m Lindsey Criswell, thevice chancellor of Research. I’d to welcome you tothe third of a seriesof COVID-19-related research town halls,sponsored by the Office of Research. Today’s event will focuson COVID-19 researchunderway within QuantitativeBiosciences Institute,or QBI, at UCSF,and as with the previousOffice of Research town halls,a recording of this event,along with the answers tothe questions submitted,will be available at theOffice of Research website. Please also note that another town hallis planned for tomorrow at four o’clock. That will be our thirdin a series of eventsco-hosted with the AcademicSenate that focus onresearch operations duringthe COVID-19 pandemic. I hope you will all be joining us tomorrowfor that town hall. And now I’m going to turnthings over to David Morgan,our vice dean for Researchfor the School of Medicine,who will moderate today’s town hall. Thanks, David. – Okay. Hello everyone and welcome. We have a very specialevent for you today. We’re gonna hear abouta lot of really excitingnew discoveries in the scienceof COVID-2 and COVID-19from the QuantitativeBiosciences Institute. I think many of you knowthat a lot of labs at UCSFhave pivoted recentlyto COVID-19 research,especially since the shutdown occurredin the last couple of months. But I don’t think any lab,not quite as early as startas the lab of Nevan Krogan,who started working on this problemprobably in a roundabout late January. And so they’ve been working on the virus,the infection processand potential treatmentsfor about three or four months now. And so we’re gonna hearabout a lot of their excitingrecent developments in that area. So, as the first slide showed,we’re going to hold all questionsuntil the end of all the talks. We have five talks, as you can see here. We’re gonna keep our eye onthe question and answer boxin the Zoom video. And then at the end,I will get together with Kevan Shokat,we’ll moderate thequestion period at the end. So, without further ado,here is Nevan Krogan,director of the QuantitativeBiosciences Instituteto start the show, Nevan. – Great. Thank you, Dave and Lindsayfor coordinating this. We are delighted to be tellingyou about some of the ongoingresearch at the QuantitativeBiosciences Institute,which is in the schoolof pharmacy here at UCSF. In particular the efforts of the QCRG,the QBI Coronavirus orCOVID-19 Research Group. And as Dave said, this is agroup that has come togethera couple of months ago. Initially it was 22 laboratoriesand since its inceptionat the end of February,it’s actually expanded nowto over 41 groups at UCSF,encompassing essentiallyhundreds of different scientists. And the group has gotten so large,we’ve actually had to now splitit into different subgroups. There’s now 10 differentsubgroups associated with QCRGfocused on different technologiesor approaches as well asdifferent biological areas. And you’re gonna hear talksfrom four different subgroups. The first hear from Davide Ruggero,it’s focused on translation. And then Brian Schoichetwho along with Kevan Schokatruns the drug discovery subgroup. And then Kliment Verbais gonna give an updateon the structural biology work. He runs this subgroup with Oren Rosenberg. And then Shaeri Mukherjeeat the end will talk aboutthe efforts with respectto approaching trafficking. And if anyone’s interested in joiningany of these subgroups,please reach out tothese different leaders. And if you’d like toform your own subgroups,please reach out to us. There’s actually a couple of new subgroupsthat are gonna be formingin the near future. And at QBI we’re reallyabout collaborationover the last three, four years. And in this example, it’s no different. The work you’re gonna hearabout today was done jointlyat QBI UCSF with Mount Sinaiand the Institut Pasteur inParis that it involves a numberof other groups really around the world. And for me it has been agreat honor and pleasureto be involved in thisreally collaborative effortthat’s going on overthe last three months. And I’m looking forwardfor more collaborationsultimately in the future. And I think as Dave alludedto this is one of the projectsthat came out of the QCRG. It was published a few days ago,and this is what I’m gonnabe briefly talking about. It’s the A SARS CoV-2 toprotein interaction mapthat we generated and howwe used it to identify drugsand compounds that we’retrying to repurpose to seeif they’re useful in termsof fighting off COVID-19. There’s over 120 authors onthis paper and over 35 groupsI think represented from around the world. And I know a lot of peopleare focusing their attentionon trying to get drugsor compounds to targetthe viral proteins themselves,which is a great strategy. In this paper we took alittle different approach,and ultimately we’retrying to target the host,target the human proteinsthat the virus needsin order to infect ourselves. And there’s advantages to that. You don’t have to worryso much about resistance’cause we don’t mutate asfast as the different viruses. And ultimately as we and others have shownis that it’s similarhuman genes and proteinsthat are being hijackedby different viruses. Not even in the samefamily as these viruses. So, ultimately the vision wouldbe if we’ve got a strategy,a whole structured strategy for COVID-19,it would work for COVID-22, COVID-24,and even other viruses. There’s obviously a disadvantage here. You have to worry about toxicity,But we initially focused ourattention on FDA approved drugsand compounds that are in clinical trialsthat are past the toxicity stage,so that the logic would bethat we wouldn’t have to worryso much about toxicity. So, there’s a close to 30 genes associatedwith the SARS CoV-2 virus comparedto over 20,000 in our cells. And so the virus itself obviouslycan’t live without our cells. It needs our cells, our genes,our proteins in order to liveand replicate and infect our our bodies. And the vision here is, well,let’s find all the humanproteins that the virusneeds in order to infect us. So, we carried on aproject where we identifiedover 300 proteins, which wethink are enriched four proteinsthat the virus does need. And then we use this set to identify69 different drugs and compounds. And just a little bit moredetail on the experiment itself. We used mass spectrometry afteraffinity take purificationaspect to generate this map,which you gonna hear alittle bit about today. And this map is now fuelingmore hypothesis driven researchin the world that biochemistry,structural biology,chemical biology and bioinformatics. And one of our goalsis to try to integrateall this information togetherto find the key nodesor pathways or proteins inour cells that the virus needsso that we can pharmacologicallyor genetically inhibit themand then carry out infection experiments. And then this data in a readerof weight would then helpre-inform this experimentaland computational pipeline. So, here’s a list ofthe genes that we thinkare associated with the virus. There’s some debate aboutsome of these genes. We tried to be all in acomprehensive, at least initially. So there’s 16 differentnonstructural proteins,four structural proteinsand then a number of veryinteresting accessory proteins. And what we did, and this iswhat we’ve done in the pastwith many other viruses,put affinity tags on them,express them at least initiallyhere in HEK293T cells,purify these proteins,analyze the material by mass spectrometry,and then use algorithms thatwe’ve developed to come upwith a high confidence SARS-CoV-2human protein protein interaction map. And I think the QCRG wasthe first group in the worldthat actually had clonedout each one of these genes. And when we initiallyreleased our paper in bioRxiv,we tweeted out to the world and we said,”All right, whoever wants these plasmids,”we’ll send them to you for free. “No MTAs, please feel free to distribute. “And Dave Gordon and his crew with few CRGsent out to 318 differentlabs in 38 countries,this set of a plasmid. I think that the plasmidwent around the worlda lot faster than the virus did. So, we’re really excited thatwe’re able to help expediteSARS-2 research around theworld in this particular way. So, here’s the map that was generatedand published a few days ago. So it’s 332 SARS-CoV-2 humanprotein protein interactions,including 69 differentdruggable host factors. The diamonds here are the SARS-2 proteins. The circles here are the human proteins. If it’s an orange, wethink it’s a drug target. And I’d like to say wecollaborated with the ZOIC Labswho generated a very interactiveway to peruse this data. So, I think it’s reallycool how you can lookat this information in anumber of different ways. So, I’d encourage youto look at this websitethat’s affiliated with thepaper if you’re interestedin looking deeper atsome of this information. So, then this map was lookedat by some really greatchemical biologists at UCSF Kevan Shokatand Brian Schoichet and and others. And they identified 69different drugs and compounds. And ultimately with collaborators,we tested about two thirds ofthese in virological assaysin a couple of differentplaces in the world. At the time unfortunately,we didn’t have the viruspropagating in UCSF. There’s been some great work being doneby a number of people now includingMelanie Ott Gladstone lab whonow has the virus propagating,and we’re starting to work with her. But there’s been twocollaborators that we’ve beenworking with over the lastfew months to test these drugsand compounds including theInstitute pasture in Parisin particular, Marco Vignuzzi. He was born in Italy,but raised in Canada. Therefore he’s a reallygreat collaborator. And then also Olivia Schwartzand Christophe D’Enfert. This has been a relationshipthat QBI had startedwith the Pasteur Institutseveral years ago,and it really bore fruit herein this particular pandemic. And then we’re alsocollaborating very closelywith a good friend of mineand Adolfo Garcia-Sastrereally one of the bestcolleges in the world at theDepartment of Microbiologyat Mount Sinai in New York. And we’ve actually got alot of support to carry outthis research from a numberof different entities. We’ve actually been workingwith the French consulatehere in San Francisco. QBI is having a couple events with them. And in order to make surethat these drugs and compoundsgot to the Pasteur Institut in Paris,the French embassy actually got involvedto make sure that FedExwas sending these compounds to Paris. And then we have Todd here. This guy’s one of the heroesof the story from FedEx. He was coming in and shippingout these drugs and compoundsto New York and Paris andshipping out all these plasmids. Actually we have his namein the acknowledgementsof the paper, that’s well deserved. And then we were able to sendthese drugs and compoundsto Pasteur and theywere able to test them. And just very briefly, justto go through the assays,you’re gonna hear aboutsome of the results fromBrian and Davide. So, we’re using Vero6 cells. These are African green monkey cells. There’s now some betterhuman cells that can be used,but initially at the timethey weren’t available. So we initially used vero6 cells,and we’re growing up the cells,were adding drugs twohours before we infect. There’s two different MOIs. One, the York MOIs is a little bit lowerthan what we see in Paris. The experiments it goes onnow for a couple of days. We kill everything with formaldehyde. And then in New York, whatthey’re doing is using anantibody against NP is thereadout in a microscopy set up,And then looking at viraltiters with the TCID 50 assay. In Paris, they’re notlooking at the protein,they’re looking at the RNAand they have an RT-PCR experiment. And they’re combining thatalso with a plaque assay. So, it’s similar assays, but different. And it was very reassuringthat we were seeing the sameresults in two different labs ultimatelyin two different continents. And in total, this is whatthey initially screened for,is about 40 each and the total. If you look at collectively47 of the drugs and compoundswere tested in total. And the first vignetteyou’re gonna hear aboutis from David Ruggero, theconnection to translation. And just to show you here on this map,we found a number of connectionswith some very interestingfactors linked to mRNAtranslation regulation. and Davide is gonna to take it over nowand talk about some of that work. – Okay, great. Thank you, Nevan. So, I wanna startactually with this slide. As many of you know, manyvirus, including the CoV-2 viruswhen entering the cellsseek massive productionof viral proteins. Actually they just formthe cells in a factoryfor viral production. This is achieved throughhijacking a specific componentof the translation ofmachinery in order to translatetheir own vital and mRNA at the expenseof the actually host and mRNA. And I actually didn’t make these slides,and I put the stock, butoften the use of this slideeven if I’m not a biologist. More to illustrateactually a similar paradigmthat occurred in cancer. Whether we realize thata key Oncogenic liaisons,also in cancer cell hijackedlike a specific componentof the translational machineryto change the translationor efficiency of the specificmRNAs that very importantfor tumor progressionor tumor translation. And actually we call now this Onco-mRNAs. And I’m gonna talk to the,It’s actually the two words met one day,but actually as you cansee this selection now,there’s an important lessonto learn because what we areactually acceptably learningin the cancer contextthis can be applicable for, of course,for the pandemic that tobe out of facing right now. And indeed the based onthe beautiful and the virusExosome interactingcharacterized in the Kroghan lab,you can see that the keyvital proteins, in red,again interacted withmany factor that belongedto the transmission machinery,including the factoringpart of ribosome biogenesisor the factor that are importantto initiate the translation. These are known as thetranslation incision factor. And I wanna highlight thatthe never know the dimensionof specifically this factorthat’s called the elF4H. That is a factor thatstimulate like the activityof an helicase importantfor capita dependent translation. So, perhaps the reason whyindeed like the CoV-2 virushijack mainly translationfactor because again,as I said that like liewhen entering the cells,this is an a plus friend of the virus,the first thing has to do is massivelytranslate their own mRNA. And all the mRNA of the virus are capped. So they have like, thisis like the cap of mRNA,and the translation of this mRNAaccording to the cappingin the translation. But what’s intriguing isactually the structureof the thatis here of this virus. As you can see, this is the. . . sorry, this is like a isa structure that has beenbasically characterized inthe dust lab of Stanford. As you can see that it’slike a lot of RPN loops,and usually this RPN loopsact as a translational barrierbecause the ribosome cannotscan before to reach the AUG. So, the virus must rely onnot in the helicaseto unwind this 5UTR. But the structure for UTRdon’t therefore derive soneefficiency to initiate a translationof the vitamin in mRNA. So, I already mentionedelF4H that the proteinthat actually interacted withthe specific viral protein. But what I wanna highlighthere is that the factthat elF4H is part of thiscomplex known also for elF4Fand this is a key. But I think complex that include the elF4Athat is the RNA helicase. elF4E that is the majorcap-binding proteinas well as to this protein. and this complex is reallyimportant to basicallyrecruit the ribosome onthe 5UTR of the same RNA. And then with the help of,again, the helicase activityelF4E, so this 5UTR is melting down,and now the ribosome reach ofthe UTR Scaffold translation. So, of course the ideahere is that perhaps elF4Acan be a point of vulnerabilityfor this CoV-2 virus,and indeed that for othercorona virus there’s beensome person in the literaturewhere the inhibitoryfor elF4A not necessarily verystrong has a good activity,I guess, like thiscoronavirus so like tighteras well as also other compoundsthat targets this elF4Fcomplex also they havean antiviral effect. But this is our two compounds. This is our two clinical compounds. So, based on this data,based on our own data,so, we of course societalrational to collaboratewith the effect of therapeutics. That is the first biotechdedicated to developfor the first clinical compounds to targettranslation control in cancer. And I want to highlightlike cannot specificallythese compounds known as zotafitin,this is the best in class,the first clinical compoundsthat target like this elF4A RNA Helicase. And this is zotatifin isalready in phase one and twofor solid cancer for the patients,exactly for cancer patients. So, as my colleague already mentioned,we send it to Mount Sinai or Paris,we send Zotatifin as wellas the other compound drugthat target compensation oftranslation such like this two. As well as we also basically Jack Taunton,also send this great compounds. This is is a new compound,very interest compoundsthat’s been generated in theJack Taunton Lab, in UCSF,is not necessarily Ternatin4. This compound that doesn’ttarget initiation targetin other factor is called the eEF1A. That is an elongationfactor that is importantfor the elongation of thelike of course the translationas well as is importantperhaps for the frame shiftthat the virus undergo inorder to change the frameand produce some different vital proteins. And also important tohighlight that there is anothercompound that the structure iscompletely different from Ternatin. This is like it’s a different compound,but targets also eEF1A,in medicine is a compoundthat is in clinical trial formultiple myeloma in Australiaas well as for COVID-19 patients in Spain. And I also wanna highlightanother important aspectof this research, andoften like people askthe question, “Oh, buttargeting translation,”this can be detrimentalfor host normal cells. “And actually what we learnedthroughout the years,that two master genetics thatoften this translation factorsare not a skippy factorthat actually are in accessin the cells because youcan reduce for example,elF4E or you can use elF4A. This mice are completely normal. You can go even below to 50%and this mice are still completely normal,but actually more protectedto actually develop for tumor. So, they’re saying of coursethis concept phase for usis to think about thisdrug report personnelto target to COVID, of course, 2-viral. And this is perhaps themost important slidewhere the like many compoundsthat Nevan have already mentioned. So, these are two compounds stood outso they’re not in four, again,the target the elongationstart and that targetedthe RNA elF4A helicase. And this is the two acid,and Nevan already mentionedin New York and Paris. So check of course, acidfor viral productioneither for assessing their viral proteinor for the virus RNA genome. As you can see from like themred the curve in both cases,in two independent labs showthat the nanomolar digitsthese two compounds isa very potent antiviralof course effect. For few compounds, this inNew York that has been alsotesting including Zotatifin,so it’s been also testedfor reducing the virustiter that it was producedby infected like cells, as you can see,that also in this case zotatifinhas very potent antiviral effect. Like reducing like viraltiter almost like to zero. So, what I want toactually a non molar digitsthat is even lower than this assay. And I wanna end off here,but of course again,highlighted the fact that the zotatifinis already in a human,is already in clinical trial for cancer. So we are working harderwith the QBI also to bring itso that it can also be in the clinicfor the COVID-19 patients. And I also end up to show exactlythe QCGR translation subgroup encapsulatedin the beautiful ribosome. They’re great. This is a great group, butthe majority uses UCSF. A few of them and alsofrom the Burke Institutethat actually we had a meetingroutinely to understandhow CoV-2 translationalwork and how we can target. And the thanks for your attention. – All right, thankyou Davide, fantastic. Next up is Brian Schoichet,who is gonna be talking abouta couple of receptorsQCRG have been working on. Very exciting receptors’cause they’re very druggable. So, Brian. – Okay. I hope you can see my screen. This is a figure you’ve seenbefore, at least partly. So there’s both the chemistry team here,the PIs and our collaboratorswho are more on the biology,and our structural biology. And of course none of thosepeople do any of the work. The work is really done by the fabulousstudents and postdocs in the labs. And I wish I could go throughtheir names more carefully. So, I think you’ve seen aversion of this slide as well. This is a subset of the targets. The 332 targets thatthe proteomics reveals. And it focuses on those of thetargets that can be drugged. And one of the startlingthings for me from this projectso far, well, one of them isjust how many human proteinsare subverted by the virus. And then the other thingis how big a subset it isof those human proteins thateither have drugs availablefor them or have really goodpreclinical molecules for them. And sometimes mechanism ofaction drugs, or drugs that hitthose targets as theirprimary mechanism of action. But sometimes there moleculesthat hit the targetsas a side effect or anunintended consequence. So overall we looked at 1600 FDA drugs. A much larger. . . There’s not that many FDA drugs. That’s something that’s sortof surprising to people. There’s only about 1800. We looked at about 20,000investigational new drugsmolecules that have been in people,but are not approved for use,and about 4,000 PROMOs,so preclinical molecules. And ultimately 29 drugs, 12molecules that are in the clinicand 28 preclinical moleculeswere tested against 63 targets. Not all of them have been tested yet,but they’re being tested. So, Davide mentioned the activity,the really fabulous activityof the protein biogenesisinhibitors like zotatifin andthat it was a target classthat had some precedents as he showed you. Last expected on on two countswere molecules targetingthe sigma-1 and sigma-2 receptors. They were, I think unexpected to usfor several reasons. One is sigma-1 and sigma-2 have not beenwidely implicated in antiviral biology. Once we knew about themwhen we went to lookthrough the literature,there is actually precedent,especially for sigma-1 being involvedin a viral lifestyle. And the other thing isthough, these are receptors,they’re sort of darkhorses in pharmacology,and have been for 40 years. The functions of these receptorsis still murky, especially the sigma-2. Sigma-1 is now widelythought to be involvedin cell stress response. But even there, the mechanismsare still really beingteased apart, and what itmight do as a drug targetis still kind of murky. But the great thingabout both receptors isthey find a lot of cationichydrophobic molecules. And cationic hydrophobicmolecule means a lot of drugs. And so we’ve been able tofind molecules that bindto those targets fairlytightly, and sure enough,are antiviral in the assaysrun in New York and Parisby Adolfo and Marco thatNevan took you through. And so these are a fewof them that were foundand published in the paper. The IC50s range from about 215nanomolar to 20 micromolar. These are here on the rightare these viral titer assaysthat Davide also took you through. The top one is PB-28, whichis a preclinical moleculethat’s potent on actuallyboth sigma receptors. With it got an IC 50 of about. . . sorry, in IC 90 about a 280 nanomolar. And then the the bottomone is hydroxy chloroquine,which sort of a notoriousmolecule now for Coronavirus,and it’s substantiallyworse in the TCID assays. This is I think my last,no, almost my last slide. So, it’s interesting tocounterpoint these observationswith the activity ofRemdesivir in the same assay. This is being published on bioRxiv,and then was redone actually. So the numbers agreed pretty well. It’s definitely antiviral,but it’s not super strong,and you can compare itto some of the othersigma active ligans, thebest of which are bindingin the 250 nanomolar range. Sorry, now binding areactive against the targetsin the 250 nanomolar range. And then finally there’s dextromethorphan. Dextromethorphan is ubiquitousin cough medications. It’s an antitussive. You basically can’t buy a coughmedication that doesn’t havedextromethorphan in it in the U. Sunless you’re getting one with coding. And dextromethorphan is unlikemany of these other moleculesis actually a sigma-1 agonist. It’s lots to activatethe sigma-1 receptor,and sure enough itshows pro viral effects,and that the red lineis the viral activity. So you can see us actually thePF use, the log PF are goingup without affecting the cell viability. So, that’s actually a little bit alarming,but maybe consistent with the mechanism. And and then maybe the lastthing to say on this topicis that there’s a reallybroad range of moleculesthat we found in terms oftheir primary mechanismsand targets that areantiviral in these assays. And the thing that unifiesthem is they all hitthe sigma receptor. So, I think I’m out of time,so I won’t dwell on this. The other approach we’retaking in collaborationwith a lot of people on thiscall is looking for target,starting with the target andlooking for new novel ligand. So this is a docking approach. The downside of this isthat it takes a lot longer. The first campaign westarted was in early Februaryfor the major protease and collaborationwith Charlie Craig’s lab and that’s take,we’re just now getting the compounds in,so it takes about threemonths from start to finish,but the advantage of this is can reallyget a novel chemical matter. So we’re looking forward totesting those things too. So, I’ll stop here. – Okay, fantastic. Brian, thank you so much. So next up is Kliment Verba, QBI fellow. He, along with OrenRosenberg have been in chargeof the structural biology QCRG subgroup,and he’s gonna tell you aboutsome very exciting pipelinesthat they have been setting up,along with some veryexciting antibody studiesas well that are beingdone in his subgroup. So go ahead, Klim. I think Klim, you’re on mute. – One second. Can everyone see my screen and hear me?- Yeah, go ahead. – Yeah. Right, so our group as partof QCRG has been taskedreally with bringing the molecular detailsto this picture. So, this is a part of the PMS dataset. And the idea is that ifyou want to really designa small molecule inhibitors or biologics,we really want to understand,how do the active sites ofthe viral proteins look like?How do the proteins cometogether so that we can seizethose interfaces andpotentially design therapeuticsto break them apart. With experts from a structural biologist,scientists from around theworld, we now have structuresfor some of the viral proteinsfeature indicated here. But the key point isthat most of them by farhas not been visualized at all. And so we’re really ends the ones,many of them are membrane proteins,which are involved withthree modelings and membranesin the host cells and human cells. And we really have no ideahow those proteins look like,how they work. And furthermore, evensort of a darker spaceor darker space for us ishow do this protein complexesbetween virus protein and host proteinslook like once they come together. And without, again,seeing those interfaces,it’s very hard to come up with waysof breaking those interactions apart. And so Nevan tasked Oren Rosenberg,a professor here at UCSFand me was trying to havea cohesive structural biology effortsreally bringing someclarity to this picture. And early on during our discussionsit became clear is that the canonical wayof doing structural biology where each labgrab a one or two of histargets and then go after themwas not really going to cut it,if you really want to goafter this in a major way. And so we thought, why notset this up as a consortiumwhere we make a call to thestructural biology communityat UCSF and say, “Hey, ifyou wanna be part of this,”if you wanna contribute, please come join”and then we’ll work together on this. “And we’ll just prioritize target”and we’ll be one big effortrather than splitting it apart”between different many different labs. “And so the amazing thingis when we made the calls,the community respondedand it has been with amazing enthusiasm. And so within days we hadover 60 volunteers, which are,graduate students, postdocs scientistsfrom 18 labs at UCSF who cameforward and say, basically,”Hey, I wanna be part of this. “How can I help?”And so, and I wanna knowthat without the buy in,not just from the community,but advisor faculty at UCSF,this would not have been possible. So, I think there’ssomething unique too hereis that we can reallycome together like this. And just put somepictures, faces the nameshere is a screenshot of ourFriday meeting from last week. And this is just somepeople in the consortium. So, we formed this QCRGstructural biology consortium. Once we had this many people,the question was how dowe organize really tohave an efficient wayof going forward. And first thing we did was we formedsort of leadershipcommunity, faculty at UCSFwho are really experiencedand wise to really help usstrategize the best way of doing this. And what came out of thatat the end of it is breakingthe effort up into lab groups,which are really broken up byfunctional groups, nodes to some extent. And this is very akin to what you would doin an industry setting. And so we have groups focusedon expressing proteinsin the mammalian of bacterial systems. Then we pass those proteinsto purification groupand then the other groups whichdo crystallography and data processing. And the key point is allthis groups are interlocked. It’s not that one lab doesone saying it was the other. It’s really the knowledgejust flows togetherand we figure out the bestway to do it as a community. And the way we set this up is each grouphas a faculty mentor. One to see faculty mentors for each group. And the next is actuallyteam leads who are eithergraduate students or postdocswho sort of have more controlof our everyday operationsof each of these groups. And the group size between 15 to 30 peopledepending on what they’re doing. And so now that we have theset up, what is it that weactually want to accomplish?What are our goals?And so we really havea three prong approach. The first part, depictedhere is really going afterthe viral proteins to predominantly,and really the focus isto enable a small molecule drug design. And so we already haveall the viral proteinsfor expression and Nikolaiand we’re going throughthem are currently,and we already have a number of crystalsfor some of the viral proteins. And so the idea hereis that by using thingslike fragment based, drugdesign, drugs screening,and then computational drug designand visualizing those compounds foundwe can help that effort. The second prong, the secondpart of their approachis really following upstrongly on the APMSand also actually on the functional datawhich will be coming in here. And so we narrowed downthis host proteins,which virus hijacks. We narrowed them down fromover 300 stories to about 60,which we think is the mostpromising one based onthe biochemistry and the biological data. And we set up this sophisticatedsystem of approaching tags,expression purification,who will be able to reallyvery rapidly screen for oneto five protein complexes,scales them up andeither do a purificationand go to prior electromicroscopy,if it’s amenable,go to crystallography or evenuses novel affinity gridsfrom the argo group and skipsimplification altogetherand just go straight into the prior, yeah. And then the last part of itis really is working togetherwith groups here at UCSFfor a focus of developingantibodies towards wild proteins,predominantly turns wefocused on spike proteinwith leveraging our consortiumstrengths as expression. Purification and structuralbiology to enablethem to just design better antibodies. And so, broadly altogether,just to recap again as a plans,the goal is a three prong approach. On the left side is the ideais to enable this rapid cycle. The turnover between thechemists, the bioinformaticiansand the structural biologistswhere we get poor hits for the target,we get a structure was it, wevisualize and we see how wecan improve it based on the structure. We synthesize a newcandidate, and we repeat. The second one is focused onsomething Nevan talked aboutis in the middle ishost directed therapies. And those are really excitingbecause we can uncovercompeting new modalities ofbeing able to target these onesand frankly, again, as Nevan pointed out,because a lot of these are conservativecould be cross virus too. And last is the antibody effortwe think that we can destruct. Biology can play a major rolein being able to design morerationally higher identityspecificity antibodies,which could actually endup being in the clinicand not too far future. And so we started outabout three weeks ago,and due to the incrediblededication and hard workof the consortium members,we already have crystals fora number of viral proteins. We already have a highestresolution structurefor one of the viral proteins. And that’s on the leftand the bottom left panel,and then on the right we have,you can see this gray box,the top quadrant is acry young roll micrographof a spike at the domainwith with antibodies foundand is being processed as we speak. And then on the bottomis actually somethingwe just got today is thefirst volume spike structurefrom the consortium atbetters in swing hamstrings. And again, thesestructures have been solvedby other people before,but this is an incrediblyimportant foundationon which we can build. And we are alreadybuilding right as we speak,so there’s lots more to come. And that’s about it. – All right, fantastic, Klim. Thank you very much. Very good, great stuff. So, our last speaker of fortoday is Shaeri Mukherjee. She’s leading up to proteintrafficking subgroup. And obviously protein traffickingis a process that’s beinghijacked and utilized bythe virus and other viruses. And she’s been doing a fantasticjob extracting mechanisticinsight from this biological process. So, go ahead Shaeri- Can you see my slides?- I can see you, but not your slides. – Oh. – I think you have to share your screen. Yeah, we can see your slides now. – Good. Thanks Nevan. So, our group is mostlyinterested in protein trafficking,and I would like tostart off by my thankingthe uprooting trafficking group. And here I’ve listed someof the PIs who participatedin our Zoom calls and also thetrainees who have exchangedideas and thoughtsabout how we can proceedwith understanding how the coronavirusco-ops the protein trafficking pathways. So, here I’m showing youan intracellular snapshotof the coronavirus life-cycle. And there are two nodes of intersectionwith the Coronavirus, withthe host trafficking pathwaysthat we are mostly interested in. The first is when the spikeprotein binds to the receptorof the host called ACE2,and this causesinternalization of the virus. And then the second is whenthe virus has to assembleand egress out of the host cell. And more and more data isstarting to emerge that suggeststhat this assembly of the virustakes place on the surfaceof a host cell organelles suchas the urgec and the golgi. Interestingly, when we lookedat the post translationalmodifications of several wild proteins,we notice that several ofthem showed or linked or inletglycosylation as well as liquidation,such as farmetilization. And this suggested to usthat the virus did enterthe secretory organellesbecause these enzymes are foundin the lumen of the secretory organelles. And very excitingly, a recentpaper in science that came outjust a few days agoshowed EM of SARS-CoV-2replication in organoids. And I was really excitedabout how the golgi lookedin these images because youcould see the viral particleson the rim of the golgi,and the golgi appeared really swollen. And this suggested to methat there might be an influxof protein and lipidtrafficking through the golgi. And it would be very interestingto study how the virusco-ops this traffickingpathways to maybe help itegress out of the host cell. Now, from the interactivestudies, when we analyze the hits,we found that roughly 40%of protein interactorswere localized or functionin the secretory pathway. And 20 out of the 26 viralproteins interact with at leastone secretory pathway protein. And this suggested to us thatthe virus is really co-optingthe whole secretory pathway. And to understand theseinteractions might give us betterunderstanding of the lifecycle of the coronavirusas well as figuring out how we can stop itfrom egressing out of the cell. Now how do we start from 132 proteins,and where do we even begin?So we thought instead of justlooking only for the hostproteins and seeing if theyhave interesting functions,we could also start bylooking at the viral proteinsand seeing if they havesomething interestingthat would point to a function. And the protein that we got very excitedabout is Orf8 in SARS-CoV-2. And this is a genome-widecomparison of the SARS-CoV-2genome shown here in blueversus several bad coronavirusgenomes as well as the SARS-CoVgenome from 2003 epidemic. And you can see here that thisis where the off Orf8 lies. And these spikes showa fast evolving regionsuggesting many changeshappening in this region. And this is the spikeregion also showing changes. This is expected becauseit’s a surface protein,but this sounded really curiouswhat’s happening to Orf8. And if we compare thetwo with the SARS-CoV-2with the SARS-CoV from 2003,it’s only 30% conserved. And interestingly in the 2003 epidemicOrf8 originally encodeda single polypeptide. But at the later stage of the epidemic,it underwent a 29 nucleotide deletion,which split Orf8 intotwo Orfs of 8A and 8B. And later on it was shownthat just this step aloneled to the attenuation of thevirus replication in cells. So what’s happening to Orf8in the current pandemic?So good news is that evenin this current pandemic,we have evidence that there isa propensity of communicatingthis pot and a complete deletion of Orf8was not noticed in agenome sequences of eighthospitalized patients in Singapore. So what’s known about Orf8is that it seems to localizeto the endoplasmic reticulumand induce ER stress. This is something that wenormally study in the lab,so we could use our expertiseto understand Orf8 is doingbecause even now the functionof Orf8 still remains unknown. So digging deep into the literatureand finding out what Orf8 could be doing. We found a very interestingpaper and this showed,this paper showed that Orf8 actuallyhas an immunoglobulin domain,which is shared in anotherprotein called Orf7a, but thisis the immunoglobulin domainthat is in Orf7 andthis is what is in Orf8. And you can notice here thatthere is an extra bit of uniqueinsert that is seen in Orf8that is absent in Orf7. And this is modeled here,shown here in orange is the insert. And you’ll see here there’sa Leu84 that seems to bethe most variable positionacross 54 closely relatedSARS-CoV genomes. And these are bad Corona viruses. The top one is the SARS-CoV-2and the bottom is the SARS-CoV from 2003. So, the other interestingdifference was that Orf7had a transmembrane domainattached to it afterthe immunoglobulin domain,whereas the Orf8 does not. And interestingly, when theOrf8 split into two proteinsin the 2003 later part of the endemic,this two proteins werestill could be linkedby a die cell fight bridgesuggesting that it would stillbe active, maybe not as active as Orf8. So, what is known aboutOrf7 is that it bindsto the cell restriction factor BST-2and inhibits its glycosylation. And this impaired BST-2 withtethering the plasma membraneimpaired antiviral response. So we are interested inunderstanding how Orf8differs from Orf7 in thesense that because it doesn’thave a transmembranedomain can it be secretedout of the cell anddoes that modify the waythe immune response happens?We are also interested in studying the ERbecause from the interimstudies you can see severalendoplasmic reticulum proteinsincluding stress responseprotein and ER degradationpathway are hit. And interestingly, we also found hitfrom the ER membrane complex subunit EMC-1which has been known toplay a role in the spreadof flaming viruses suchas Zika and West Nile. So if this interaction is real,it would be very interestingto understand thefunction of Orf in the ER,of Orf8 in the ER. And finally, I would like toalso highlight another proteinin the virus NSP13. Why we are excited about NSP13is because not only did itbind to the golgiproteins and coiled-coiledstructural tethering proteins,including proteins thathave gripped domains,it also bound to the proteinKinase A, signaling complex. And by complex I mean that it boundto not only the catalytic andregulatory subunit of PKA,but also the scaffolding protein AKAP9as well as the protein that turns it off,which is the phosphodiesterase or PDE. So, we think that thisprotein might be co-opting PKAsignaling at the golgi topromote virus parting and spread. And this could change the waythe microtubule is nucleated. The other interesting aspect of NSB13is that it was found to bind to GCC185. This particular golgi bindsto 16 different Rab proteins. And Rab GTPases are known to play a rolein membrane trafficking. So this would be very exciting to studybecause the phenotypeof the golgi swellingand the fact that there couldbe increased traffickingcould be explained if thisprotein was somehow activatingalso a Rab GPS function. So, Brunno Goud who is a Curie Institutean expert of Rab reached out to us,and we are currently collaboratingto find that function. In addition, it’s goodnews that PKA and PDEare both druggable targets. In a recent study showedthat PDE is actuallycan inhibit Corona viruses in retro. So with that I’d like to thank Nevanand the rest of the crew at QCRGas well as my two brilliantpostdocs, Advait and Julia. Advait is working on NSP13and Julia is working on Orf8along with Albert in Melanie’s laband also our collaboratorsin U. K and Curie. Thank you. – Thanks Shaeri. Awesome. Great stuff. I just have just a couple,two wrap up slides here,just to make a point here thatQBI we’ve really been cryingover last several yearsto break down silos. And I know this hasbeen a terrible tragedy,this particular pandemic,but a silver lining for meis to see scientistsreally around the worldcoming together in an unprecedented way. These silos have really been broken downacross laboratories, acrossdifferent institutions,between academia andpharmaceutical companies,in a really unprecedented way. And my question is, whycan’t we do this normally?Look how fast science can move. And to me, the problem withscience is it’s too siloed. So the question is, can weset up an infrastructure,really a new paradigm on how to do scienceso that we’re better preparedfor COVID-24, COVID-26what other other virus we want to work onor just working on anydisease for that matter. So, I would encourage thecommunity to really thinkalong these lines going forward. so we talked to you about thediscovery research, a component. We’re really gonna bepushing hard now to connectto other entities at UCSF,including the immunological worldand the clinical world. Hopefully the next QCRG town hall,we’ll talk about some of ourconnections to these worlds. We’re really excited to pushor translate some of ourfindings into the clinicalworld and we’re having numberof discussions withdifferent clinicians at UCSF. So stay tuned to that. And this is my last slide. We have a lot moreinformation on our websiteif you’d like to go and lookabout the COVID-19 research. So, I’ll stop there and then I’ll letKayvon and Dave moderate questions. – Okay. I’ll take care of a couple of these,and Kayvon you’re welcome tojoin in when you feel like it. The first one was came from Dave Egadand it’s directed mostly toDavide who was talking aboutthese various anti translationalcompounds and David Egadis curious to know howtoxic those compounds are. ‘Cause they looked a littlerough on the cell viabilityassays that you showed. So what do we know about their toxicityin those a cell-based assays?- Yes, thanks for the question. I didn’t go through inthe detail at that time. It’s not cytostatic, soit doesn’t kill the cells. There is a cytostatic effect on the cellsand the reason it’s because thecells are immortalized cellsand obviously they have non normal cellsand actually the viralcells they have a mutationin a negative regulator soul cycle. So, they are really addictedto proliferate more. And both of the compoundsare actually a target. Also the transmission of a keymRNA porter for cell cycle. In the primary cells actually the cellspecific to these compoundsthey’re not toxic. Ternatin as well aszotatifin has been alsoused already then in mouse models. Zotatifin is already in the human. So, I think it’s the one already answeredthe question of toxicity. And also ternatin reject totally. We actually recently addedlike very interesting datain the mouse model that the block tumorin a specific dose thereis no so much toxicity. – Okay. Thank you very much, sir. Next, there’s a couple ofquestions here that are bothprobably for Brian Schoichet. And they’re all about thesigma R-1 and R-2 receptors. Anita Seal is curiousto know which cell typesexpress these particular receptors,which obviously has an implicationfor various other issues. And Sarah Kang is askinga related question,which is what is the roleof the sigma-1 receptorand has it been implicatedin the infectionby any other viruses?- Okay, great question. So, that the sigma receptorsare most well knownbest characterizedneuronly, neuronal cells. That’s where most of thework has been done on them,but they’re expressed, Imean, tissue wise for sure,they’re expressed pretty ubiquitously,certainly in the lungs, for instance. As to the role of the sigma-1 receptorand whether it’s implicatedin other virus infections. It has a role in stillcell stress response. It’s thought to be active on activation. It forms a tremor in the inactive state. I only have two hands. This is a new discovery. It forms a tremor in the inactive state,and then it’s thought to monitorize. There’s good evidence for that. And when it gets activated andthen it acts as a chaperonewhen the cells are under stress. There’s actually good humangenetics that it’s importantfor neuron development. And then as to whetherit’s being implicatedin other viral infections?Yes, it has. Is not a large literature onit, but both knock-down studieshave shown that it is involved,you knock down the sigma-1receptor and viral infectivity will drop. Not, it’s not being done for SARS-2,but I think for HCV and oneother virus that I’m forgetting,but these are resultsover the last decade. So there is some actually mechanisticand very logical reasonto believe that of coursesis pharmacology can be involved in. . . can be subverted by viruses generally. – Okay. I have anotherquestion on the topic of drugs. Maybe Brian can do this ormaybe Nevan could do this. And that’s from David Earlcurious to know if you’ve begunto look for synergy between drugs. We’ve seen a lot of singledrugs in actions so are youstarting to combine them to seewhat effect that might have?- I think everybody saw me texting outthat I was gonna answer David’s question,so maybe I can take a shot at it. So, I mean, it’s a great question. There’s actually ongoing testsboth in Paris and New Yorkadding one of the sigma-1 or2 are the most promising one,the modulator with some ofthese translational inhibitors,putting them together andseeing what effects they have. Obviously, you got toworry about toxicity,but then also remdesivircombined with these. So just like with HIV, the cocktailwas the big breakthrough. And I personally thinka common trial strategyholds a lot of promise. Maybe combining this remdesivirwith one of our drugsor another drug that hits a host factorthat somebody else discovers. But then the big question is the resultsthat we see in the laboratory,do they translate into humans?I don’t know, going forward. I just wanted to answer,add something to Brian’s last comment. It was influenza, actuallyBrian that was connectedto sigma-1. There was a genome wide RNE ice screenthat connected Sigma-1 to influenza. So there’s a connection there as well. So, that’s my response to David. I don’t know, Brian, if youhave anything to add to that. – No, I think that’s a. . . Yeah, so it’s influence and HCV, right?HCVL is also?- Yeah, Kayvon found that connection. – Nevan also Mary Selleck isasking when we do the combos,do we think that orthogonaltargets will be betteror sort of in pathway might be better?That idea of where synergymight pick us found?- Yeah, that’s a good question. I guess it’s related togenetic interactions, right?I mean, it depends onhow much of one pathwayis ultimately perturbed. My thinking would be if youhad two different pathways,that would be a better approach. But you may have to worryabout toxicity there. So, it’s a great question,and more data needs to be collected. – I’ll try and take this next one. There’s a fairly complexquestion here from Cynthia Chu. So, are there microscopydata for cells treatedwith the Sigma ligand hits?From Nevan’s HPMS data set,the Sigma receptor interactswith NSP-6 which is partof a complex that formsa double membrane vesiclesseen in infected cells. So the question is whetherany of those compoundsthat affect the Sigmareceptor might actually affectthe membrane structuresof the secretory pathway. I don’t imagine thatyou’ve gotten that far yet,but is there, is there a plan to do so?- Yeah, definitely. Actually, this morning wejust got some microscopy datafrom a collaborator in Germanylooking at certain kinasesthat are being mis-localizedin the context of infection. So I mean, yeah, that’sa fantastic question is,especially in the contextof some of these drugs seenwhere some of theseviral proteins are goingor just see what’s happeningwith the biology of the cell. There’s a lot of work to do going forward. – One last question about drugs again,they’re always about drugsthat from an anonymous attendeewho is curious to know, I suspectI know the answer to this,but the question is whetheryou can actually use plasmafrom a drug treatedhuman to test the effectof that on another patient. This sounds like a unusual idea. Is there any thoughts along those lines?Maybe Nevan, you should try that one. – I think. . . Oh, Davide, go ahead. – No, I think this already. . . So, the question, if aplasma from already infectedpatients, if they can be used. So these are the questionor in combinationwith the drugs?’cause this has been— I think with the combination with drugs. Yeah, so it looks likethe idea of being. . . yeah, if that anonymousattendee wants to clarifythe question, go aheadand put it in the Q&A. but it looks like they’re referringto plasma from drug treated people. – Right, because that onehas been already startedto be used in the UK, in Italy. So, using the plasma frominfected people actuallyof course as enrichedin principle of courseof an antibiotic against the virus. But of course it canbe an interesting ideato also combine also withsome of the compoundsthat we described today. So, it can be interesting to think about. – Okay. We’ve got one more questionthat just popped infrom Lisa for Klim. If they have plans tolook at the mutation,the recently discoveredmutation in spike protein,the D-614 gene mutation thatI guess was just associatedwith increased violence. We were actually talking aboutthis before this webinar. Is there any local lead,that structure worth looking at?- Yeah absolutely, actually. I was just looking at itright before this meetingand it seems like it mightbe at the interface betweenthe spike as a tremor,so you might be at theinterface between the monomers,but parts of the structureI was looking at,there was a piece of itwhich was disordered. So we don’t really see thatas part of acid packingor forming like a soulbridge or something. But absolutely we should really look at itand see maybe, especiallywith the antibody efforts,maybe the difference is there. It’s in the plan, it’s in the pipeline. – Okay. Oh no, a provocativequestion has just shown up. It might be related to somethingNevan said in the newspaperactually whichis can Brian Schoichetcomment on the dextromethorphan issuein common flu and cold medication?Should you be taking thatmedication at this point?- Yep Brian, go ahead. – I was gonnadefer to Nevan on this. – The only thing you could be sued for. – Yeah, Well I’m sure theinstitution will protect me, Dave. So it’s a little bit worrying, right?Dextromethorphan looks proviral, reliably pro viral. That experiment was done more than once. What it means in terms of human health,I’d be very cautious to say,but I think it does merita careful controlled trial. Whether that’s ethicalor not, I’m not sure. But I think the signal is. . . So, there’s a mechanistic reason,and there’s empirical evidence. And you put those two togetherand it’s enough to pay attention to. Yeah, that’s my answer. – Perfect, thank you. I think we’ve reachedthe end of the questions,and we’ve reached 4:03. So, perhaps we could wrap this up. Obviously everyone out therein the audience is welcometo email all these folksanytime with further questionsand I’m sure they’d love to answer them. So, please do that. And thanks again for allthe wonderful presentations. Nevan, did you want to say one last wordor are you good to go?- Yup, sure. No, that’s fantastic. And we look forward toongoing QCRG town hallsand with other entities at UCSF,including the clinical worldsand the immunological worlds. We’re very excited about that. – Great. Okay, thanks everyone for attending,we’ll see you next time. And as Lindsey mentioned,there is another town halltomorrow as well on operational issuesabout how research is going at UCSF. So, we’ll see you there. Take care of everyone.