Downloaded by guest on September 27, 2021 iuaetepooto fltn n yi el enwithin seen cells lytic and latent of proportion the pitulate accessible existing, with targeted identi- be in therapies. can utility that has products cells gene infected preva- fying for cellular more essential the use are of is that limited examination KSHV processes comprehensive of a where Therefore, are resources but (1). Tradi- limited lent they patients, with cells. mean some settings tumor inaccessibility in in the and in effective toxicity is virus their are latency effec- the chemotherapeutics cellular been of because tional not state presumably has predominant treatment, this classically the KS but are for analogs, infections tive nucleoside Herpesvirus express- with cells (2). of treated transcripts percentage small lytic a ing just her- with sarcoma-associated Kaposi’s (KSHV), spindle with pesvirus of infected majority latently The are cell are endothelium. lesions cells a of cases cell, vascularized markers spindle all expresses the highly that is not complex, component proliferating AIDS, HIV- are primary with whose both tumors patients in KS many resolved. common therapy in antiretroviral is KS While (1). eliminates KS patients -negative where and positive Africa, sub-Saharan in K KSHV KSHV- for avenue respira- therapeutic mitochondrial potential cancers. of a associated leading provides inhibition biology, infec- which to latent mitochondrial tion, sensitivity During alters enhanced the mitochondria. significantly to of the products KSHV to gene tion, and localize multiple number, that locus copy found latent genome We size, levels. mitochondrial transcript in cells. increases there infected but are latently numbers, mitochondrial in Direct decreases death infection mitochon- lines. latent increased KSHV of to cell ablation leads PEL or genomes drial respiration patient-derived led mitochondrial in and of cells death inhibition endothelial cell infected increased that latently to of inhib- Antibiotics expansion specifically the translation an translation. including ited mitochondrial was mitochondria, mitochondrial and bacterial there the in inhibit identified, to counter- involved genes localizing uninfected genes are genes host their infected in many that not latently enrichment the but of factors Among culture, proliferation host parts. in or cells identified survival exploited endothelial the CRISPR- we be pooled for library, a could essential Using lentivirus cells vulnerabili- cells. endothelial tumor Cas9 direct host KSHV-associated infected no of inhibit is latently to identification there in KSHV, As percent- latent ties replication. a small viral cells for is a undergoing Spindle treatment only cells tumors cell. with of spindle KS infected age the latently of termed predominantly origin component are endothelial proliferating lymphoma of effusion main cell primary The and (KS) etiologic (PEL). 9, sarcoma the June Kaposi’s review is for of (received (KSHV) agent 2020 30, herpesvirus September approved sarcoma-associated and Kaposi’s CA, Emeryville, Inc., Research, Biomedical for Institutes Novartis 2020) Ganem, E. Donald by Edited a Holmes cells L. Daniel endothelial KSHV human latent of in infection process essential an as translation mitochondrial identifies screen CRISPR-Cas9 A www.pnas.org/cgi/doi/10.1073/pnas.2011645117 eateto irbooy nvriyo ahntn etl,W 98109 WA Seattle, Washington, of University Microbiology, of Department utrdhmnedteilclsifce ihKH reca- KSHV with infected cells endothelial human Cultured fimn upeso ntedvlpdwrdbtnot but world developed the context in the in suppression arises immune predominately of (KS) sarcoma aposi’s | HHV-8 | mitochondria a ailT Vogt T. Daniel , | Crispr/Cas9 | a aoissarcoma Kaposi’s n ihe Lagunoff Michael and , a,1 epromdagnm-ieCIP-a9sre nhuman in mock- cells. screen (TIME) with endothelial CRISPR-Cas9 cells, microvascular endothelial genome-wide cells tert-immortalized for infected a spindle required latently performed are of KS we that proliferation factors to and host relevant survival identify To in type controls. performed cell infected been date human to have for However, screens a important system. essentiality cell genes CRISPR-Cas9 rat identify no this to in mesenchy- (11) transformation rat cells KSHV with stem KSHV-infected EBV-positive for embryonic utilized control compare mal uninfected study true to Another a also iden- able lacked KSHV. are were it was cells, cases genes study PEL of interesting EBV-negative the subset many and a While All tified but EBV. infection. KSHV with KSHV with infected infected latent are maintain non-Hodgkin’s PELs which a are lym- primary lymphoma PELs of cell (10). B-cell survival Another cells B for (PEL) (9). lymphoma of critical effusion (EBV) genes survival identified virus the study Epstein-Barr to for important with used critical been infected the are to have phomas that essential approaches are genes Similar that cells. identify factors cancer These identify (8). of to cells survival used human so-called be of Recently can of population screens (4–7). a interrogation within large-scale vitro genes selec- “Achilles” enabled in screening to have cells used CRISPR-Cas9–based platforms be infected lentivirus-encoded can in latently developed which latency target iden- pathways, has KSHV tively laboratory cellular for our several systems, model culture tified cell culture Using a (3). tumors providing tumors, KS doi:10.1073/pnas.2011645117/-/DCSupplemental at online information supporting contains article This 1 the under Published Submission.y Direct PNAS a is article This paper.y interest.y the competing wrote no performed M.L. declare D.T.V. and authors D.L.H. and The and D.L.H. data; analyzed research; M.L. designed and M.L. D.L.H. research; and D.L.H. contributions: Author owo orsodnemyb drse.Eal [email protected]. Email: addressed. be may correspondence whom To fmtcodi uigltn SVinfection. KSHV function latent essential during the mitochondria into to of infected insights latently provide able findings KSHV Our are of cells. proliferation ribosomes endothelial the bacterial inhibit infected selectively targeting as latently translation Antibiotics of mitochondrial cells. proliferation iden- in the for latency involved KSHV necessary test factors during to host genes human tified screen all knockout of support requirement CRISPR-Cas9 KSHV. to with genome-wide Therefore, essential infected latently tumors. targets A cells in cell of treat- survival virus host and the direct identify proliferation of to no state sought is latent we the There the for Africa. is ment sub-Saharan common most in (KSHV) the of neoplasms one sarcoma, Kaposi’s herpesvirus of agent causative sarcoma-associated Kaposi’s Significance NSlicense.y PNAS . y https://www.pnas.org/lookup/suppl/ NSLts Articles Latest PNAS y | f9 of 1

MICROBIOLOGY Our screen produced 146 gene targets, which were depleted in latent infection. Interestingly, we found that multiple latent pro- the KSHV-infected population relative to mock controls. Many teins localize, at least partially, to mitochondria, suggesting that of the hits encode proteins that are known to localize to the more than one may be needed to sensitize cells to inhibition of mitochondria, and further analysis of this subset revealed mito- mitochondrial translation. chondrial translation to be the predominant function of these genes. Importantly, the mitochondrial ribosome has a shared Results ancestry with the bacterial ribosome. We found that antibiotics CRISPR-Cas9 Pooled Lentivirus Screen in TIME Cells during Latent targeting either the large or small subunit of bacterial ribosomes KSHV Infection. To determine the host factors required for cells and therefore, the mitochondrial ribosome can strongly sup- latently infected with KSHV to proliferate and survive, we used press cell proliferation in infected endothelial cells as compared half of the Human Activity-Optimized CRISPR-Cas9 lentivirus with treated mock-infected controls and also potently induce library to generate a mutant pool of cells containing single guide cell death in KSHV-infected B cell lymphomas. Suppression of RNA (sgRNA) and Cas9-expressing lentiviruses targeting 18,166 respiration directly, either chemically or by removal of mitochon- human genes with five sgRNAs each (Fig. 1A); ∼150 million drial genomes, produces a similar defect in cellular proliferation cells were transduced at a multiplicity of infection (MOI) of 0.6 specifically in latently infected cells. We found changes in the and 2 d later, were placed under selection with puromycin for mitochondrial network of cells during KSHV infection, where 3 d, the time by which a control flask of nontransduced TIME mitochondria become larger and longer but are fewer in number, cells was completely killed. The cells were grown out for an which is consistent with but not definitive evidence of increased additional 2 d and then split into two sets of flasks, each rep- mitochondrial fusion. Mitochondrial transcript levels as well resenting approximately 1,000-times coverage of the lentiviral as mitochondrial genome copy numbers were increased during library in 90 million cells. After the cells were settled, half were

Fig. 1. CRISPR-Cas9 whole-genome screen to identify essential host factors during KSHV infection of endothelial cells. (A) Schematic of TIME cell whole- genome screen of KSHV-infected cells (T0 and T48 refer to zero and 48 hours post initial KSHV infection, respectively. NGS stands for next generation sequencing). (B) Plot of the results of the live cell screen. The back line represents the false discovery rate cutoff of 0.25. The size of the circles represents the magnitude of the median log fold change for all sgRNAs for that particular gene. All red circles accompanied by red text are genes whose gene products localize to mitochondria.

2 of 9 | www.pnas.org/cgi/doi/10.1073/pnas.2011645117 Holmes et al. Downloaded by guest on September 27, 2021 Downloaded by guest on September 27, 2021 i.2. Fig. to relative KSHV with infected cells live the in sgR- represen- and depleted (12). MAGeCK infected in NAs using between determined changes genes were the samples different uninfected targeting and guides library, of the tation to mapping reads by cell calculated sequencing were dead abundances the sgRNA as DNA sequenced. well prepared Genomic and were as harvested. libraries samples sequencing were next-generation cell cells and live samples, live the the from pelleted. d, harvested were 8 was of samples supernatant end KSHV-infected the the and in At control cells dead the the both passage, representation of each maintain At to library. splitting the passaged, For of were latency. cells asso- KSHV the of latency d, marker for 8 a positive (LANA), the antigen stained After nuclear cells ciated experiment. of remained the were 91% of rates infection, cells remainder infection second the these that for later, ensure high cavity- d to sufficiently body KSHV 2 of with cells; induction reinfected (BCBL1) from lymphoma-1 obtained based KSHV with infected h eei oto ftasue IEclsadsgicnl euecl rwhrltv omc neto hncmae ihnnagtn control nontargeting with compared when infection mock to relative growth (KO) out protein cell knock reduce of to significantly able suppression are and showing MRPS34 cells blot targeting sgRNAs TIME Western separate transduced (D) The (paired of sgRNA. controls. cells (NTC) portion transduced when nontargeting control a cells nontargeting the uninfected in a with as in gene compared well counts the as to when MRPS34 relative controls targeting ( depleted sgRNAs uninfected rate. are with the MRPL40, discovery expression and false to MRPS39 FDR, relative ribosome, genes. cells mitochondrial these KSHV-infected the of (C of many controls. components for nontargeting two function with for common compared screen a cell is live translation the mitochondrial from that indicate mitochondria the to omse al. et Holmes ee novdi iohnra rnlto r seta uigKH neto.( infection. KSHV during essential are translation mitochondrial in involved Genes t test; B A n ) *P 4). = < gN onsfo h edcl cenfrtrecmoet ftemtcodilrbsm r nihdi the in enriched are ribosome mitochondrial the of components three for screen cell dead the from counts sgRNA ) D 0.05. eltdfo u iecl cenecd rtista localize that proteins encode screen cell live our from Infection. depleted KSHV Essen- Latent Are Machinery for Translation tial Mitochondrial the of the Components with cells sgR- infected across the change S2). comparing (Dataset fold when cells uninfected gene log each median for population 2.75-fold NAs cell than KSHV dead the greater during in by enriched growth were suppress genes 1,600 selectively infection; genes sug- these cells, KSHV-infected gesting in enriched and the 0.25 significantly KCTD10 genes, of were with two HSPA4, cutoff pop- for compared rate sgRNAs discovery Interestingly, cell as false S1). (Dataset a live cells using TIME the cells TIME infected from uninfected latently depleted of significantly ulation were the S1) iden- in Table were enriched uninfected to 1B sgRNAs (Fig. relative tified KSHV as with well infected cells as dead population uninfected the C eeotlg G)ercmn nlsso h ista localize that hits the of analysis enrichment (GO) ontology Gene A) and IAppendix, (SI genes 146 S1); Fig. Appendix, SI prxmtl 0 ftegenes the of 10% Approximately NSLts Articles Latest PNAS h gN counts sgRNA The B) | f9 of 3

MICROBIOLOGY to the mitochondria (shown in red in Fig. 1B). This represents an control sgRNA transduced cells from 48 to 72 h postinfec- enrichment from 6% of the total nuclear genes whose products tion (Fig. 2D). Knockout of MRPS34 led to the suppression localize to the mitochondria. The largest subset of mitochon- of the mitochondrial genome encoded COXII (SI Appendix, drial genes identified in the screen has functions in mitochon- Fig. S2). Overall, the screen implicated mitochondrial transla- drial translation and respiration (Fig. 2A). The mitochondrial tion as a potential vulnerability, which can be used to inhibit the ribosomal subunits MRPS39 and MRPL40 had mean counts proliferation and survival of latently infected TIME cells. roughly four times lower in the KSHV-infected population rel- ative the uninfected control in the live cell screen (Fig. 2B). Antibiotics Interfere with Proliferation and Survival of KSHV- The mitochondrial ribosomal subunits MPRS34, MRPL20, and Infected Cells. Since mitochondrial ribosomes are structurally MRPL41 were counted in our dead cell screen while having lit- more closely related to bacterial ribosomes than eukaryotic tle to no counts in our mock infected population (Fig. 2C). In an ribosomes, antibiotics that target bacterial ribosomes can be independent set of experiments, two sgRNAs targeting MRPS34 used to selectively inhibit mitochondrial translation (13). To were able to moderately reduce expression of MRPS34 when test if antibiotics that inhibit mitochondrial translation also visualized by western blot (Fig. 2D). These MRPS34 knockout inhibit cells latently infected with KSHV, chloramphenicol, cells showed reduced proliferation compared with nontargeting which interferes with the large mitochondrial ribosomal subunit,

Fig. 3. Antibiotics targeting bacterial translation inhibit the growth and survival of cells latently infected with KSHV. Treatment of KSHV-infected TIME cells suppresses cell growth in a dose-dependent manner with both chloramphenicol (A) and tigecycline (B; mean with SEM of three replicate experiments). KSHV-infected PEL, BCBL1 and AIDS body cavity-based lymphoma 1 (BC1), cells exhibit a dose-dependent induction of cell death for both chloramphenicol (C) and tigecycline (D), which is not seen in B cell lymphomas lacking KSHV (the EBV negative Burkitt’s lymphoma BJAB, and the EBV positive Burkitt’s lymphoma RAJI).

4 of 9 | www.pnas.org/cgi/doi/10.1073/pnas.2011645117 Holmes et al. Downloaded by guest on September 27, 2021 Downloaded by guest on September 27, 2021 omse al. et Holmes cells. required PEL is KSHV-infected respiration TIME in mitochondrial that of death suggest case cell results the of These in induction proliferation cells and cellular from cells of 4C). genomes inhibition mitochondrial (Fig. to respiration functional lead controls of of inhibitors infected elimination that show or mock results mitochon- these with functional together, Taken dou- compared lacking phase cells when log in dria in death infection increase induces KSHV Additionally, mito- selectively an S6B). during Fig. of and Appendix , proliferation (SI depletion 4B) time in bling that decrease (Fig. changes show a media infection with to curves KSHV d leads resulting 24-well 6 genomes The for to chondrial d. Incucyte plated 2 an were on every Cells imaged infection. and KSHV plates of absence presence the those and both in and genomes con- mitochondria mitochondrial intact functioning was with without cells (16), between stability compared was Appendix complex (SI blot decreased the western by to as firmed mitochon- due well expression nuclear-encoded genes as some of of drial COXII, expression loss encoded of induction, loss and genome expected of mitochondrial doxycycline, weeks the from 2 of removed After placed system. were is expression RNA, cells from identical uracil an remove in tet-inducible only uracil a can wild-type of enzyme which control wild-type glycosylase, target- the the control, mitochondrial under a a As placed promoter. is has and enzyme sequence This ing DNA. uracil (15). from mutant system thymine a mutagenesis uses inducible system genetic, This a from using eliminated of to cells were infection TIME way genomes differ- KSHV alternative mitochondrial during the respiration an cells, the of endothelial As if in essentiality to factors. differences unclear the cell-intrinsic added for on is test to dependent pyruvate due It is mM or types S5 ). 10 media cell Fig. or the 1 Appendix, between ence either (SI with media reduc- death a the observe cell not did in we however, tion pyruvate; to death with cell cells similar rotenone-induced PEL tigecycline in rescue 4A), and to chloramphenicol (Fig. attempted from We death death treatment. cell cell in found increase in and avail- the increase pyruvate no lacking dose-dependent media is to a in tested there cells we known Therefore, PEL and (14). is KSHV-infected media sensitivity, endothelial which rotenone pyruvate-free pyruvate, able from contain cells com- media rescue when S4 cell Fig. proliferation Appendix, endothelial (SI cellular controls rotenone in uninfected inhibitor with difference I pared no complex the showed with but treated the were discov- cells by KSHV-infected previously TIME (10). supported the dependencies oncogenic among proliferation is PEL-specific respira- genes ered decreased hypothesis mitochondrial of and This of death expression presence cells. cell of KSHV-infected increased should loss in to respiration the of lead inhibitors to also then components, due complex essential tory during is Process lation Infection. Essential an KSHV Is Latent Function Respiratory Mitochondrial S3). sensitive Fig. Appendix, more (SI were KSHV lacking cells lymphomas PEL than treat- KSHV-infected antibiotic however, to response ment; in proliferation 3 cellular (Fig. of inhibition posttreatment h 48 to at and sensitivity antibiotic dose-dependent either sensitiv- a with for show treatment tested cells and also B KSHV-negative KSHV-positive were cells, ity. lymphomas sensitiv- endothelial cell this infected if B to determine KSHV-positive To unique expansion. is cell ity of treatment leads inhibition Therefore, chloramphenicol an or treatment. tigecycline to of with tige- cells h TIME and 48 infected 3A) of after (Fig. mock 3B) chloramphenicol to (Fig. both cycline relative for confluence controls cell to infected in showed used cells decrease were TIME dose-dependent KSHV-infected subunit, a cells. small infected the latently inhibits treat which tigecycline, and .AlBcl ypoa etdehbtddose-dependent exhibited tested lymphomas cell B All D). fsprsino iohnra trans- mitochondrial of suppression If lcsls,wihremoves which N-glycosylase, .Cl proliferation Cell S6A). Fig. , .However, ). N- C hwa nuto ncl et pnifcinrltv ocnrl (paired controls to *P relative N-glycosylase. infection upon death cell t in lacking induction cells an infected show mock with compared (C when rate mitochondria. slower a at grow mitochon- ablation lacking dria After cells (B) KSHV-infected concentrations. and genomes, Rotenones tested BJAB mitochondrial the functional while of at manner, (A) sensitivity dose-dependent no a cells. show in cells KSHV-infected cells RAJI PEL of in death survival cell induces and proliferation presses 4. Fig. test; n ) tN,mtn uracil mutant mtUNG, 3). = niiino eprto rls fmtcodilgnmssup- genomes mitochondrial of loss or respiration of Inhibition diinly SVifce el akn mitochondria lacking cells KSHV-infected Additionally, ) < .5 s nonsignificant. ns, 0.05; lcsls;wUG idtp uracil wild-type wtUNG, N-glycosylase; NSLts Articles Latest PNAS | f9 of 5

MICROBIOLOGY for the proliferation and survival of cells latently infected This increase in expression is primarily from transfer RNAs (SI with KSHV. Appendix, Fig. S7B), which are the only transcripts stabilized by polyadenylation in mitochondria, and our complementary DNA Latent KSHV Infection Increases Mitochondrial Transcripts and library was poly-deoxythymidine primed. To confirm that tran- Genome Copy Number. Previous work from our laboratory estab- scription of the entire mitochondrial genome is increased during lished that mitochondrial function is decreased during KSHV KSHV infection, qRT-PCR primed with random primers showed infection (4). One possible explanation that is consistent with that transcripts of both strands of the mitochondrial genome are both the decreased mitochondrial function and sensitivity to increased during KSHV infection (Fig. 5D). To determine if this antibiotic treatment is that mitochondria are physically defective increase in transcript levels was related to genome copy num- in cells latently infected with KSHV. Defects in mitochondrial ber, qPCR of mitochondrial genome copy number showed that function can lead to a fragmented mitochondrial network, which KSHV-infected TIME cells possess approximately twice as many is a previously observed characteristic of certain cancers exhibit- mitochondrial genomes as mock infected cells (Fig. 5E). Overall, ing Warburg metabolism (17). To better characterize the state KSHV infection of TIME cells leads to an increase in the size, of the mitochondrial network during KSHV infection, KSHV- length, transcript levels, and genome copies per cell of mitochon- infected and mock infected endothelial cells were fixed to slides dria while decreasing the overall number of mitochondria. These and probed for COXIV as a marker for mitochondria. Images results are consistent with increased mitochondrial fusion during were analyzed using an ImageJ plug-in for examining mitochon- KSHV infection, rather than defective mitochondria. The reduc- drial networks (18). There was a decrease in the average number tion in oxygen consumption observed during KSHV infection is of mitochondria per cell (effect size of 47.9) and increases in independent of mitochondrial network structure, mitochondrial the average size and length of mitochondria (effect sizes of transcription, and mitochondrial genome copy number. 38.6 and 25.1% in KSHV-infected cells relative to mock cells, respectively) (Fig. 5 A–C). KSHV Latency Gene Products Localize to Mitochondrial Fractions. Mitochondrial function can be altered at the level of mito- To determine if any KSHV latent genes could play a direct chondrial gene transcription. Our previously published RNA- role in the alteration of mitochondrial biology, the localization sequencing dataset (7) was analyzed to look at average gene of latent proteins to the mitochondria was examined. Plasmids expression across each chromosome, including the mitochondrial for transient overexpression of each of the latent proteins with chromosome. While the mean change in transcript level for all 3xFLAG tag on the N-terminal end were transfected into 293T genes on a given chromosome was near zero for every nuclear cells; 48 h later, the cytoplasmic and mitochondrial fractions of chromosome, the mitochondrial genome showed a 1.29-fold the cells were separated using differential centrifugation. West- average increase during KSHV infection (SI Appendix, Fig. S7A). ern blot analysis of whole-cell lysates and the mitochondrial and

AB C

DE

Fig. 5. KSHV infection induces changes in the mitochondrial network and increases mitochondrial transcription and genome copy number. KSHV-infected cells were analyzed by immunofluorescence for changes in the mitochondrial network in single cells. The average mitochondrial number decreased per cell (A), while the average area (B) and length (C) of mitochondria increased (unpaired t test; n = 50 per sample). *P < 0.05; **P < 0.01; ***P < 0.001. (D) qRT-PCR shows that transcript levels from the mitochondrial (MT) genome increase during KSHV infection (paired t test; n = 3). *P < 0.05. (E) qPCR for mitochondrial genome copies relative to host genome copies shows an increase in mitochondrial genome copy number during KSHV infection of TIME cells (paired t test; n = 6). ***P < 0.001.

6 of 9 | www.pnas.org/cgi/doi/10.1073/pnas.2011645117 Holmes et al. Downloaded by guest on September 27, 2021 Downloaded by guest on September 27, 2021 omse al. et Holmes ntecntuta ela T5 n AD.vy,vLP n aoi B Kaposin Kaposin and and vFLIP, LANA absent. vCyc, while are mitochondria, GAPDH. (KapC) the C and to mito- ATP5A localized and as partially are well (Cyt) (KapB) as depending cytosolic tag construct indicated reduced the the for the on probed were with Blots along fractions. (MT) Whole-cell run chondrial used). was were transduction (WCL) lentivirus (in where lysates vFLIP, transfection but after all h of 48 case cells separate the 293T of to fractions used mitochondrial was and cytosolic centrifugation differential by fractionation chondrial 6. Fig. fraction. mitochondrial the to localizes vFLIP indicated of and lentiviral- detected proportion a was tag a However, V5 C-terminal detected. a be with vector not based could with tag construct N-terminal (vFLIP) frac- an protein mitochondrial FLICE-inhibitory the viral in The present also of tion. portion was small (vCyc) mitochondria. a cyclin and viral the fraction, mitochondrial the cytoplasmic the for the in as marker levels well low as a at fraction frac- present as cytoplasmic was (KapB) used the B for was Kaposin marker ATP5A a dehydroge- and as 3-phosphate tion, used Glyceraldehyde was (GAPDH) proteins. nase viral localization identify the to used of was cells the of fractions cytoplasmic SVltn rtislclz otemtcodilfato.Mito- fraction. mitochondrial the to localize proteins latent KSHV N oynme n rncitlvl,teewsadecrease endothelial a infected was latently in there number levels, mitochondrial transcript mitochondrial overall and in in increase lev- number the copy RNA despite DNA lower Interestingly, 30). and (29, genomes, els fewer networks, mitochondrial fractured produces consistent usually not which dysfunction, were mitochondrial in with genomes observed and we transcripts, changes structure, the network However, in cells. struc- levels endothelial protein network and infected mitochondrial number, mitochondrial copy of in genome levels, presence changes transcript the ture, for confirm looking to by him- sought defects Warburg Otto We by (28). made defects observation self an mitochondrial is of which presence cells, the within phe- with of consistent collection is This notypes (27). consumption oxygen lactate, decreased of and production includes increased This consumption, 24–26). glucose (4, increased metabolism Warburg of characteristics unclear. currently is types was cell difference there the between for responses cells, reason in PEL mechanistic the The death. in cell while widespread death, reach not cell was to there large-scale but unable obvious controls, infected were mock as cells quickly different as slightly endothelial confluence were infected there latently The cells, cells PEL outcomes. endothelial and KSHV both with inhibited infected antibiotics the Interest- while and respectively. ingly, subunit, machinery, ribosomal import respiratory mitochondrial protein of mitochondrial small part form the and I, mitochondria complex the and to PTCD3, NDUFB7, localize genes, lym- TAZ, three cell These B (10). KSHV-positive lines cell in phoma dependencies done three oncogenic screen contains PEL-specific CRISPR-Cas9 a of dataset from list screen the to cell common live hits Our mito- of function. levels basal chondrial lower having despite function some require mitochondrial cells KSHV-infected that suggest findings respiratory our the chain, in pri- of components is death synthesizing ribosome for cell responsible mitochondrial marily induce the Since and cells. pro- cells B suppress KSHV-infected endothelial to infected able of that are showed liferation tigecycline We Health and (23). World chloramphenicol Medicines the both Essential on of and of List available many Organization readily Importantly, are has (19–22). antibiotics cancers biology these other mitochondrial for of exploited aspect been This ribosomes. chondrial mito- eukaryotic and the ribosomes to bacterial due between ancestry antibiotics shared using or therapeutically targeted ribo- proliferation be Mitochondrial can KSHV. for somes with infected essential latently cells is of hypothesis survival the translation to mitochondrial leading mitochondria, that the to localize be trials. that clinical teins can in which are targets that of potential drugs number or 146 drugs a approved found screen, with human targeted whole-genome We of initial cells. our survival spindle most in type the KS cell the for for KSHV, relevant with cellular essential infected latently identify used is cells endothelial We to function work. whose screening preclinical genes library source pathogen in powerful lentiviral as a targets well CRISPR-Cas9 be therapeutic as to potential available proven cancers of have different readily screens of for These number infections. need cancer a common a been to already most applied is have third screens essentiality there the Whole-genome therapies. Africa, is KS sub-Saharan devel- As in in (1). the utility their in countries limit KS availability oping for and outcomes cost world, clinical developed improved have med- approaches personalized newer ical and chemotherapies traditional While mitochondrial Discussion with interfering idea directly the infection. latent be with during consistent might function is virus fraction the latent mitochondrial that multiple the of presence in the The genes to 6). localization (Fig. display not fraction did mitochondrial LANA and (KapC) C Kaposin eadohr aesonta SVifce el display cells KSHV-infected that shown have others and We pro- encode screen the by identified genes of number large A NSLts Articles Latest PNAS | f9 of 7

MICROBIOLOGY cells. At the same time, there was an increase in mitochondrial DNA Mini Kit (Invitrogen) followed by the LookOut Mycoplasma Detection size. These results are consistent with, but not definitive proof Kit (Sigma-Aldrich); 293 and 293T cells were grown in Dulbecco’s modi- of, increased mitochondrial fusion. Further experiments will be fied Eagle media (DMEM) (+L-glutamine, +penicillin-streptomycin, +4.5g/L needed to examine changes in mitochondrial fusion dynamics dur- glucose, +sodium pyruvate, + 10% FBS; Fisher). B cell lymphomas were ing KSHV infection and determine the role latent genes might be grown in Roswell Park Memorial Institute media (RPMI) 1640 (+L-glutamine, +penicillin-streptomycin, + 2-mercaptoethanol, + 10% FBS; Fisher). iSLK cells playing to facilitate or counteract mitochondrial changes during for BAC16 virus production were grown in DMEM (+L-glutamine, +penicillin- latent infection of human endothelial cells. streptomycin, +4.5g/L glucose, +sodium pyruvate, + 10% FBS). All 293T cells, Many viruses encode proteins, which modify mitochon- B cell lymphomas, iSLK cells, and the mitochondria lacking TIME cells in Fig. 4 drial function for a variety of different reasons. Among the B and C were negative for mycoplasma by both the Invivogen and Invitrogen observed functions are preventing apoptosis, inducing changes kits listed above. in metabolism, and avoiding innate immune activation. Some viruses encode proteins, which have been found to interact with Transfection. The 293T cells were seeded at 10,000 cells per centimeter2 the mitochondrial ribosomes in large-scale protein interaction stud- night before transfection. Transit 293T (Mirus Bio) was used to transfect plas- ies. These include the capsid proteins of flaviviruses (31) and mids as indicated by the manufacturer. After 24 h, the media on the cells Nsp8 of the severe acute respiratory syndrome coronavirus-2 were replaced with fresh serum containing media. In the case of lentivirus production, the masses of each plasmid used were 8 µg of psPAX2, 4 µg of (32). A previously published KSHV protein interaction network pMD2.G, and 8 µg of the lentiviral vector. Culture supernatants were col- identified potential interactions between KSHV latent genes and lected at 48 and 72 h posttransfection and filtered through 0.45-µm filters mitochondrial proteins (33). Ninety-two mitochondrial genes before aliquoting and freezing. were within vFLIP’s interaction network, representing 19.1% of all proteins in the interaction network; 5% of vFLIP’s interac- Proliferation Assay. Time cells were seeded at 1 × 104 cells per centimeter2 tion network is made up of proteins involved in mitochondrial and then treated once settled. After the specified length of time, when translation. Consistent with these findings, we found that vFLIP assessing cell death, the time cells were treated with trypsin and counted physically localizes to the mitochondria. Further studies exam- after trypan blue staining for cell viability. When assessing cell prolifera- ining the direct effects of viral proteins in the mitochondria will tion, TIME cells are washed twice with phosphate buffered saline (PBS) and fixed with 100% methanol. After fixing, the cells were stained with crystal be necessary to determine the mechanism of KSHV alteration of violet for 10 min and then rinsed with distilled water until dye no longer mitochondrial biology. washed off. The plates are allowed to dry overnight before scanning with a Whole-genome screening for factors that are essential to Typhoon imager (GE) using a 532-nm laser and a 670BP30 filter. The result- latent KSHV infection led us to unexplored therapeutic targets ing images were analyzed in ImageJ to measure cell confluence in each well. and pointed us toward an additional point of viral interaction The MRPS34 data in Fig. 2 were obtained by resuspending the crystal violet with the host cell. While KS is a multifactorial disease, the fact with 500 µL 10% glacial acetic acid after staining and measuring 200 µL of that the proliferating component is a latently infected endothe- the resulting solution in a 96-well plate reader at 570 nm. All transformed lial cell has prevented direct targeting of the viral infection. B cell proliferation assays began with viable cell concentrations of 2 × 105. Identifying host genes whose inhibition can selectively eliminate Forty-eight hours after treatment, cells were stained with trypan blue and KSHV-infected cells has potential for application in the clinic. counted using a TC-20 Cell Counter (BioRad). While previous work has demonstrated that specific metabolic qRT-PCR. Cell lysates were harvested using Nucleospin RNA II Kits (Macherey processes can be used to kill latently infected cells, the cen- Nagel). RNA was quantified by Nanodrop, and the quality of each sample tral nature of many of those metabolic pathways limits their was assessed on a 1% agarose gel in Tris-acetate-ethylenediaminetetraacetic utility. A narrowing of the target range to even more specific acid (EDTA) buffer to ensure that no degradation of samples had occurred; cellular pathways can make a therapeutic more attractive than 500 ng of RNA was used for each 20-µL reverse transcription reaction with broad metabolic inhibitors. While mitochondrial translation is the iScript Select cDNA Synthesis Kit (BioRad) using random primers. qPCR an intriguing target, the concentrations of antibiotics used in this was subsequently carried out using SsoAdvanced Universal SYBR Green study to inhibit KSHV-infected cell proliferation were relatively Supermix (BioRad). Primers targeting hypoxanthine-guanine phosphoribo- high. Future work will determine if antibiotic combinations or syltransferase (HPRT) were used as a reference, and primers targeting other antibiotics known to target the mitochondria better than mitochondrial genome sequences were used to assess transcript levels. chloramphenicol and tigecycline could identify effective treat- Primer sequences are available in SI Appendix, Table S5. ments at more reasonable concentrations of antibiotic. While targeting mitochondrial translation is likely still problematic, this qPCR for Mitochondrial Genome Quantitation. Cell lysates were harvested using PureLink Genomic DNA Mini Kits (Invitrogen). The resulting genomic work encompasses a step toward gaining a better understanding DNA is used as template for qPCR with SsoAdvanced Universal SYBR Green of viral alteration of a cellular metabolic hub, which can then Supermix (BioRad). Primers targeting Prox1 were used as a reference for cel- enable targeting more precise pathways. lular genomic DNA, and primers targeting the mitochondrial genome were used to quantify mitochondrial genome copy number. Primer sequences are Datasets. Gene scores and sgRNA counts from the live and dead available in SI Appendix, Table S5. cell screens can be found in Datasets S1 (live cell screen) and S2 (dead cell screen). SI Appendix, Figs. S1–S5 as well as materials CRISPR-Cas9 Whole-Genome Library. One-half of the Human Activity- tables for antibodies, primers, and plasmids can be found in SI Optimized CRISPR Knockout Library was transformed into Endura elec- Appendix. Raw sequencing data are available at Gene Expression trocompetent cells (Lucigen). Sufficient colony-forming units to achieve Omnibus (GEO) accession number GSE152284 (34). 1,000-times coverage of the library were harvested using the Plasmid Plus Maxi Kit (Qiagen). The resulting plasmid prep was transfected into 293T cells as mentioned above for production of lentivirus. The resulting lysate Materials and Methods was titered onto TIME cells using cell viability after selection as a proxy for Cell Lines. TIME cells were maintained in endothelial cell basal medium- infection. Thirty-six T225s containing 4.5 × 106 TIME cells per flask seeding 2 (EBM-2) media (Lonza), which was supplemented with an EGM-2 MV the evening before transduction were treated with the lentiviral library. This SingeQuot Microvascular Endothelial Cell Growth Medium Bullet Kit (Lonza) is roughly 150 million cells transduced at an MOI of 0.6 to achieve at least containing 5% fetal bovine serum (FBS), hydrocortisone, hFGF-B, VEGF, R3- 500-times coverage of the library. Transduced time cells were selected for 3 d IGF-1, ascorbic acid, and hEGF, as well as gentamycin and amphotericin-B. and grown for an additional 4 d until two sets of 34 T225s could be seeded. After the screen was performed, we discovered the presence of Mycoplasma The day after seeding, one set of the TIME cells was infected with KSHV arginini in TIME cells. The Plasmotest mycoplasma detection kit (Invivogen) purified from BCBL1 cells as previously described (35). Two days after infec- was used to find that TIME cells test positive for mycoplasma after a min- tion, the TIME cells were split and infected again. The infection rates for imum of 72 h of growth in the absence of antibiotic. This was confirmed the cells were counted by immunofluorescence for LANA and open reading by performing a genomic DNA extraction using the PureLink Genomic frame 59. For the next 8 d, the cells were split every 2 d and reseeded to 4.5

8 of 9 | www.pnas.org/cgi/doi/10.1073/pnas.2011645117 Holmes et al. Downloaded by guest on September 27, 2021 Downloaded by guest on September 27, 2021 8 .A erl,K .Fip,S tak Mauigmtcodilsaewt mgJ in ImageJ” with shape mitochondrial “Measuring Strack, S. Flippo, H. K. Merrill, A. R. 18. Giang A.-H. 17. 6 .Rsca .Klsesa .Brnk .To K. Bartnik, mitochon- E. induces Kaliszewska, damage M. Rusecka, Persistent J. Alexeyev, F. 16. M. exogenous Wilson, with L. G. Repopulation Shokolenko, mtDNA: N. lacking I. cells 15. Human Attardi, G. King, P. pharmacopeia. M. the in 14. inhibitors translational Mitochondrial Saneto, P. R. Cohen, H. B. 13. Li W. 12. 1 .Gruffaz M. 11. Manzano M. 10. omse al. et Holmes elsie ih4 aaomlyye fe lcigadpoigwt an with probing and blocking After paraformaldyhyde. 4% with slides well or Immunofluorescence. Mitochondrial CLx Odyssey LI-COR IRdye a LI-COR either with on treated system. imaging Fc were Odyssey to Blots prior above. antibodies antibody poly- mentioned polyvinylidene secondary primary 20% a dilutions appropriate to the to the 4 using transferred at blotted a was and to protein membrane loaded The difluoride were (BioRad). protein gel of (ThermoFisher acrylamide assay masses BCA equal Peirce and the using Scientific), 0.1% fluo- quantified was deoxycholate, sodium lysate mM 0.5% Cell 1 Roche). P-40, orthovanadate, Nonidet mM sodium 40 mM 1% ride, 1 EDTA, sulfate, mM dodecyl 1 sodium 7.6, NaCl, pH Tris-HCl, mM mM lysed (50 were 150 buffer pellets Cell (RIPA) PBS. assay with radioimmunoprecipitation once washed with and pelleted then and cells ent Analysis. Blot Western parameters other All MAGeCK. counts. for read cell settings total default dead the on sgRNA The used the based distribution. except normalized manner, null identical were an the counts nontar- in and generate analyzed of were control to distribution results the used sequencing the were and was counts as counts, sgRNAs sgRNA read uninfected geting The median group. using on treatment after based analyzed the normalized counted as was guide. cells per were screen TIME base KSHV-infected uncalled sgRNAs cell one to live (12). up The for 5.6 allowing samples, version of deconvolution MAGeCK using analyzed Screening. Library CRISPR-Cas9 of Analysis (8). for al. template et Wang as in used described was as DNA cell amplicons genomic and sequencing The blood Illumina (Qiagen). infected the kit using and maxi samples uninfected DNA all from for culture harvested pooled was were were DNA cells Genomic samples live cells. cell all experiment, dead the and of each end collected, were the At pellets At cell sample. centrifugation. dead each after and for frozen collected, flasks were supernatants 34 culture maintaining the flask, passage, per cells TIME million .T ag .J e,D .Sbtn,E .Lne,Gntcsresi ua el using cells human in screens Genetic Lander, Ma S. Y. E. Sabatini, 9. M. D. Wei, J. J. Wang, T. 8. infected KSHV Latent Lagunoff, Sychev E. M. Z. Thalhofer, 7. B. A. Carroll, A. P. Sanchez, L. E. 6. infec- of profiling metabolic Global Lagunoff, M. Roman, C. Sanchez, L. E. Delgado, T. 5. Delgado T. 4. .M Lagunoff M. 3. sarcoma- Kaposi of expression Restricted Ganem, D. Herndier, B. Wang, H. Zhong, W. 2. Cesarman E. 1. d.(ermtos uaaPes e ok Y 07,vl 2,p.31–48. pp. 123, Usachev, vol. Y. 2017), Strack, NY, York, S. New Neurons, Press, Humana in (Neuromethods, Function Eds. Mitochondrial Investigate to Techniques effect. Warburg the showing cells coma DNA. mitochondrial of instability (2018). by caused diseases drial ra N degradation. DNA drial complementation. by mitochondria Acta Biophys. Biochim. screens. knockout cells. CRISPR/Cas9 cancer scale of target vulnerable a as (2019). XPO1 e00866 identifies cells transformed isi epsia rmr fuinlymphoma. effusion primary herpesviral in cies factors. dependency system. CRISPR-Cas9 the metabolism. lipid (2017). mediated e1006256 peroxisome 13, on reliance and induction viral survival. for glutaminolysis require of Pathog. and addicted survival glutamine are for cells lipogenesis endothelial requires and induces KSHV virus: infection. oncogenic latent an by tion cells. endothelial infected U.S.A. latently Sci. of maintenance the for required soitdhrevrsi utrdedteilcells. endothelial cultured in herpesvirus associated sarcoma. Kaposi in genes U.S.A. 8) Sci. Acad. herpesvirus (human herpesvirus associated AeKealsrbs dnicto fesnilgnsfo genome- from genes essential of identification robust enables MAGeCK al., et RSRCs cen eelEsenBr iu-rnfre elhost cell B virus-transformed Epstein-Barr reveal screens CRISPR/Cas9 al., et 1002(2015). e1005052 11, 09–00 (2010). 10696–10701 107, nuto fteWrugefc yKps’ acm epsiu is herpesvirus sarcoma Kaposi’s by effect Warburg the of Induction al., et RSRCs cenn fKps’ acm-soitdherpesvirus- sarcoma-associated Kaposi’s of screening CRISPR-Cas9 al., et gyeohsht,Cmlt iipoes niio tablet; inhibitor protease Mini Complete β-glycerophosphate, nertdssesbooyaayi fKH aetifcinreveals infection latent KSHV of analysis biology systems Integrated al., et iohnra yfnto n emaiiytasto nosteosar- in transition permeability and dysfunction Mitochondrial al., et aoisarcoma. Kaposi al., et eeesnilt adcp n rgal noei dependen- oncogenic druggable and landscape essentiality Gene al., et env neto n eiltasiso fKps’ sarcoma- Kaposi’s of transmission serial and infection novo De al., et 6164 (1996). 6641–6646 93, LSPathog. PLoS elHs Microbe Host Cell el eehretduigtysnt eoeadher- remove to trypsin using harvested were Cells 0717 (2012). 1067–1074 1819, N Repair DNA Science 1086(2012). e1002866 8, 08 (2014). 80–84 343, a.Rv i.Primers Dis. Rev. Nat. Science eoeBiol. Genome 8–9 (2013). 488–499 12, IEclswr xdt ls chamber glass to fixed were cells TIME 8–9.7(2017). 580–591.e7 21, .Bo.Chem. Biol. J. sa ula ee novdi mitochon- in involved genes Nuclear nska, ´ 0–0 (1989). 500–503 246, luiasqecn eut were results sequencing Illumina a.Commun. Nat. 5 (2014). 554 15, .Virol. J. 30–31 (2013). 33303–33311 288, (2019). 9 5, pl Genet Appl. J 4024 (2002). 2440–2448 76, 23(2018). 3263 9, rc al Acad. Natl. Proc. LSPathog. PLoS . rc Natl. Proc. 43–57 59, mBio PLoS 10, 6 .Johannes S. 36. Punjabi, S. A. 35. health. in and sickness In Mitochondria: Suomalainen, A. Nunnari, J. 30. body. cells. the cancer Trifunovic in of A. origin tumors the 29. of On Warburg, metabolism O. The 28. Negelein, E. Wind, F. Warburg, O. 27. microRNAs herpesvirus sarcoma Kaposi’s Ma Boshoff, T. C. 26. Enver, T. Lagos, D. Yogev, O. 25. Bhatt P. A. 24. essential of use and selection The summary: “Executive activ- Organization, cisplatin Health World enhances 23. tigecycline Antibiotic Hu, Y. Zhou, M. Song, M. Tan, J. 22. M. 21. Lamb R. 20. Kalghatgi S. 19. 4 .Egr .Dmahv .E ah eeEpeso miu:NB eeexpression gene NCBI Omnibus: Expression Gene Lash, E. A. Domrachev, M. Edgar, R. 34. Davis H. Z. 33. Gordon E. D. 32. Shah S. P. 31. a orl nsuydsg,dt olcinadaayi,dcso opublish, to funders decision manuscript. analysis, The the and manuscript. collection of data preparation the design, or of study in for preparation role DiMaio the Terri no thank had during also We editing advice project. for and the M.L. of advice of course laboratory the the over the the discussion thank of performing and members for We and core M.L.). sequencing genomics deep (to Center Illumina Research R21CA240479 (to Cancer and RO1CA189986 Hutchinson M.L.), Grants Fred Institute (to Cancer RO1CA217788 This National T32AI083203. M.L.), by Grant Training supported NIH was the work of Diseases Infectious and Allergy ACKNOWLEDGMENTS. (GSE152284). GEO the in deposited been have data in results found sequencing be can plasmids and primers, screen). elsrescnb on in found be can screens cell Availability. Data of contamination were to cytoplasmic proteins blots fraction. for mitochondrial check mitochondrial purification, the to five GAPDH the and targeting of enrichment antibodies confirm validity pre- primary the as with processed check lysate probed were To whole-cell fraction each and mentioned. of fraction, viously masses equal cytoplasmic and paired quantified, lysate, were concentrations Protein resulting buffer. RIPA centrifugation, the with of differential lysed was through purification, after fraction which mitochondrial achieve to high-purity followed a were (ThermoFisher recommendations Cells manufacturer’s Cultured The for Scientific). Kit Isolation Mitochondria the using cessed Fractionation. in Mitochondrial plotted and cell per quantified were Software). number, (Graphpad volume mitochondrial PRISM8 total for and outputs length, Resulting using 18. average (36) FIJI ref. Retiga in from analyzed adapted a were macros using Images Photometrics). captured (Teledyne were camera R6 Images 4 with Laboratories). medium (Vector mounting phenylindole Vectashield with cells mounted the antibodies, were secondary fluorescent and antibody primary anti-COXIV xdtv aaei amla cells. mammalian in damage oxidative Methods cells. endothelial of infection herpesvirus polymerase. Physiol. sarcoma. Kaposi’s in 2 (2015). 477–488 kinase pyruvate of upregulation dependent cells. infected of transformation metabolic induce lymphoma. selection the on Health committee-on-the-selection-and-use-of-essential-medicines). https://www.who.int/publications/i/item/executive-summary- Committee World (Rep.WHO/MVP/EMP/IAU/2019.05, Expert the-selection-and-use-of-essential-medicines-2019-report-of-the-22nd-who-expert- 2019; medicines” WHO essential 22nd Organization, of the use of and Report 2019: medicines dys- damage. mitochondrial inducing oxidative (2017). through and carcinoma hepatocellular function human against ity leukemia. disease. myeloid infectious acute human an like cancer Treating types: tumor Oncotarget multiple across cells, stem n yrdzto ra aarepository. data array hybridization and genes. late for strategy transcription repurposing. pathogenesis. virus zika and dengue of mechanisms (2012). 1159 Skrti ˇ is S1–S5 Figs. Appendix, SI SVidcsarbcgyoyi n nignsstruhHIF-1- through angiogenesis and glycolysis aerobic induces KSHV al., et c ´ 1–3 (1927). 519–530 8, 7–8 (2012). 676–682 9, niiino iohnra rnlto sateaetcsrtg for strategy therapeutic a as translation mitochondrial of Inhibition al., et niitc httre iohnraefcieyeaiaecancer eradicate effectively mitochondria target that Antibiotics al., et oprtv aiiu-otpoenitrcinmpigreveals mapping interaction protein flavivirus-host Comparative al. , et yrglto fftyai ytei n lclssi non-Hodgkin in glycolysis and synthesis acid fatty of Dysregulation al., et lblmpigo epsiu-otpoencmlxsrvasa reveals complexes protein herpesvirus-host of mapping Global al., et rc al cd c.U.S.A. Sci. Acad. Natl. Proc. 5948 (2015). 4569–4584 6, ii noe-orepafr o ilgcliaeanalysis. biological-image for platform open-source An Fiji: al., et Nature rmtr gigi ieepesn eetv iohnra DNA mitochondrial defective expressing mice in ageing Premature al., et essetatvto fSA3b aetKps’ sarcoma-associated Kaposi’s latent by STAT3 of activation Persistent al., et Nature atrcdlatboisidc iohnra yfnto and dysfunction mitochondrial induce antibiotics Bactericidal al., et ASCV2poenitrcinmprvastresfrdrug for targets reveals map interaction protein SARS-CoV-2 A al., et eesoe n gN onsfo h ieaddead and live the from counts sgRNA and scores Gene 1–2 (2004). 417–423 429, 5–6 (2020). 459–468 583, ...wsspotdi atb ainlIsiueof Institute National by part in supported was D.L.H. elpleswr eupne nPSadpro- and PBS in resuspended were pellets Cell aaesS1 Datasets acrCell. Cancer swl smtrastbe o antibodies, for tables materials as well as ice.Bohs e.Commun. Res. Biophys. Biochem. o.Cell Mol. c.Tas.Med. Transl. Sci. 11–12 (2012). 11818–11823 109, uli cd Res. Acids Nucleic .Virol. J. IAppendix SI Science lv elsre)and screen) cell (live 7–8 (2011). 674–688 20, 4–6 (2015). 349–360 57, 4925 (2007). 2449–2458 81, 0–1 (1956). 309–314 123, LSPathog. PLoS NSLts Articles Latest PNAS Cell 9r8 (2013). 192ra85 5, cennx generation next Screen . , 9114.1 (2018). 1931–1945.e18 175, 0–1 (2002). 207–210 30, 1040(2014). e1004400 10, 0 ,6-diamidino-2- Angiogenesis Cell S2 17–23 483, da cell (dead 1145– 148, | .Gen. J. f9 of 9 Nat. 18,

MICROBIOLOGY