review article adapted lethality: what we can learn...

Review ArticleAdapted Lethality What We Can Learn from GuineaPig-Adapted Ebola Virus Infection Model

S V Cheresiz12 E A Semenova12 and A A Chepurnov3

1Department of Medicine Novosibirsk State University Pirogova Street 2 630090 Novosibirsk-90 Russia2Institute of Internal and Preventive Medicine Bogatkova Street 1751 630089 Novosibirsk-89 Russia3Institute of Clinical Immunology Yadrincevskaya Street 14 630047 Novosibirsk-47 Russia

Correspondence should be addressed to A A Chepurnov alexachepurnovgmailcom

Received 11 June 2015 Revised 14 October 2015 Accepted 24 November 2015

Academic Editor Jih-Hui Lin

Copyright copy 2016 S V Cheresiz et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Establishment of small animal models of Ebola virus (EBOV) infection is important both for the study of genetic determinantsinvolved in the complex pathology of EBOV disease and for the preliminary screening of antivirals production of therapeuticheterologic immunoglobulins and experimental vaccine development Since the wild-type EBOV is avirulent in rodents theadaptation series of passages in these animals are required for the virulencelethality to emerge in these models Here we providean overview of our several adaptation series in guinea pigs which resulted in the establishment of guinea pig-adapted EBOV (GPA-EBOV) variants different in their characteristics while uniformly lethal for the infected animals and compare the virologic geneticpathomorphologic and immunologic findings with those obtained in the adaptation experiments of the other research groups

1 Introduction

Several animal models for Ebola virus infection have beenestablished in rodents and nonhuman primates (NHPs) TheNHPs including rhesus and cynomolgus macaques are bestsuited for pathogenesis treatment and vaccine studies sinceonly they can be lethally infected by nonadapted EBOVstrains with the resulting pathology closely resembling thehuman EBOV disease [1] However due to ethical practicaland expense reasons small animalmodels of EBOV infectionwere developed including guinea pig mouse and recentlySyrian hamster models [2] Those are established by a serialpassage required for virus adaptation since the wild-typeEBOV is avirulent or causes a nonlethal disease in rodentsAlthough even the lethal adapted EBOV infection in rodentsis different in many aspects from the disease in primates theimportant similarities in the courses of both infections makesmall animal models useful especially in the study of geneticdeterminants of EBOV disease and in antiviral screening [1]

In primates the pathogenesis of EBOV infection isassociated with the viral replication in several majorcell targets accompanied with immune dysregulation and

coagulopathies Viral reproduction in primary targets themononuclear phagocytes of spleen and lymph nodes isfollowed by a massive replication in the liver mostly inmacrophages and hepatocytes and the virus spread to theother organs and tissues (adrenals kidneys reproductiveorgans and lungs) A bystander lymphocyte apoptosis by anunknown mechanism is proposed to be the cause of severelymphopenia occurring in EBOV infection Inhibition ofIFN-mediated response mediated by viral proteins VP24 andVP35 blocks the innate antiviral defense Vascular damageeither occurring directly due to lytic virus reproduction inthe endothelial cells or induced indirectly by the effects ofproinflammatory cytokines on the vascular wall is an impor-tant factor of pathogenesis The mechanisms of coagulationdysfunctions such as disseminated intravascular coagulation(DIC) and hemorrhages as well as thrombopenia occurringin primate EBOV infection are still to be investigated inmoredetail [3]

In guinea pigs the lethal EBOV variants are establishedthrough the sequential passages (4ndash8 times) of an originallywild-type virus in which first incomplete and furthercomplete lethality in the groups of inoculated animals are

Hindawi Publishing CorporationAdvances in VirologyVolume 2016 Article ID 8059607 14 pageshttpdxdoiorg10115520168059607

2 Advances in Virology

acquired [4ndash7] The guinea pig-adapted EBOV is causing alethal infection with minor manifestations in the first 4-5days and a subsequent rapid development of a highly febrilecondition resulting in the animal death on days 8ndash11 Firstdetected in lymph node macrophages as early as 24 h pithe virus spreads to the spleen and liver on day 2 and to theother organs and tissues further on The virus spread canbe accompanied with a progressive rise of tissue virus titers(from 17 to 48ndash64 log

10PFUg in different tissues including

spleen liver adrenals lungs kidneys and pancreas) ondays 1ndash9 of the infection and the peak viremia in blood isreached on day 7 with sim105 PFUmL [7] However in twoof our adaptation experiments an only modest [8] or evenzero increase in virus titer [9] between the nonlethal andlethal adapted EBOV was occurring A prolongation of theprothrombin time (PT) and the partial thromboplastin time(aPTT) is observed in the infected animals [1]

While resembling the course of EBOV infection inprimates in many aspects the EBOV disease in rodents hassome important differences Fever and maculopapular rashwhich are the typical signs of infection in primates are bothnot present in mice infected with mouse-adapted Ebola virus(MA-EBOV) [10] In guinea pigs infected with guinea pig-adapted Ebola virus (GPA-EBOV) only fever but not therash is present [5 7] Unlike in mice and similarly to Syrianhamster lethal EBOV infection in guinea pigs induces seriouscoagulation abnormalities including the drop of platelets andan increase in coagulation time however fibrin depositionsand disseminated intravascular coagulation (DIC) are notreadily observed in these animals [2 7 11] Occurrence ofhemorrhages in EBOV disease in guinea pigs is still dis-putable some researchers report that death of animals is notaccompanied by the visible signs of hemorrhage [6] howeverwe regularly observed typical hemorrhagic foci in the liverof at least part of the animals infected with GPA-EBOVin our experiments [12] Despite the severe lymphopenia inthe course of GPA-EBOV infection lymphocyte bystanderapoptosis which is an important feature of infection inprimates and mice is not generally observed in guinea pigs[7]

Here we present a comprehensive overview of the viro-logic pathomorphologic and hematologic data obtained inthe EBOV adaptation experiments in guinea pigs (includingfour passage series performed in our laboratory) and discussthe value of those findings obtained in the rodent models forthe general understanding of pathology of EBOV infection

2 Establishment of Guinea Pig-AdaptedLethal Strains of Ebola Virus and theStudy of Their Virulence and VirologicCharacteristics

Serial passage of different viruses in animals or cell culturescan result in the selection of mutants with different repro-duction characteristics and either higher or lower virulenceas compared to the original wild-type virus [13 14] Adultguinea pigs inoculated with the wild-type EBOV isolatedfrom primates typically develop a nonlethal mild febrile or

even asymptomatic disease However it has been previouslyshown that young guinea pigs may occasionally develop alethal EBOV infection and the serial passage of EBOV causesa substantial increase in lethality in these animals [4 15]In this review paper we summarize our own data on theestablishment of four independent adaptation series in guineapigs using the original stocks of EBOV subtype Zaire strainMayinga differing in their passage history [5 8 9 11] andcompare them with those from other labs where the EBOVadaptation series were successfully performed [6 7 10]

The original EBOV strain Mayinga stock of the firstadaptation series (8mc) was passaged 24 times in Vero E6cell line prior to the guinea pig adaptation experiments [5]while the stock of the second series (K-5) was passagedtwice in NHPs (green monkeys) followed by 15 passagesin human lung L68 cells [9] In the third series (GLA)the wild-type EBOV strain Mayinga was passaged twicein NHPs and further multiplied in Vero cell culture [11]The fourth series (GPA-P7) was started with an individualclone of strain Mayinga EBOV which was obtained by triple-cloningpassaging in Vero cells and was confirmed to causeno lethal disease prior to its adaptation to guinea pigs Thisvirus clone further used for the inoculation of animals wasdesignated as the precursor of guinea pig-adapted EBOV(pre-GPA clone) [8]

At first passage of 8mc adaptation series [5] a groupof animals (we used the groups of 6 animals in all ouradaptation experiments) was inoculated with virus derivedfrom a liver homogenate of a rhesusmacaque lethally infectedby Mayinga strain of EBOV and passaged 24 times in VeroE6 cells In some animals a nonlethal febrile conditiondeveloped with the body temperature rising up to 40∘CThe liver homogenates of the animals which demonstratedhigh-grade fevers and higher levels of liver virus titer wereused for the next passage inoculation With each passagethe febrile temperature was increasing (from 398ndash400∘Cat earlier passages to 415ndash420∘C at later passages) aswas the liver virus titer which showed the overall growthfrom 29 plusmn 04 log

10PFUmL to 55 plusmn 07 log

10PFUmL (as

determined by plaque forming assay in Vero cells) The firstlethal outcomes were detected at passage 3 and the lethalityand infectivity of the adapting virus isolate were graduallyincreasing until passage 8 In guinea pigs inoculated with thevirus isolates of late passage 8 the fever occurred at days 4ndash7 and represented a double-peaked pattern with the secondpeak at days 11ndash13 Fever episodes were associated with themaximum levels of tissue virus titersThe adapted virus strainwas designated EBOV-8mc and several clones were derivedfrom it and proved to have similar features as the nonclonedEBOV-8mc virus itself [5]

As mentioned above the original EBOV stock of K-5series was passaged twice in NHPs and further 15 times inL68 cell culture prior to its adaptation to guinea pig and itsinfectivity and virulence were evolving in the passage seriesThe biological titer of the virus harvest in PFU decreasedfrom the first to third passage and then began to increasereaching its plateau by passage 12 and further remainingstable However the virus infectivity in guinea pigs wasunchanged throughout the L68 passages Following the cell

Advances in Virology 3

culture passages the virus cloning was performed and thebiological titer (in PFU) of the intermediate virus stock(after passage 15 stock designation Ch-15) and the threeobtained clones were shown to be equal When adaptationto guinea pigs was started clone 2 turned out to cause nofebrile reaction and virus isolation from blood samples ofinoculated animals failed However clone 1 caused a rise inthe rectal temperature in guinea pigs up to 405ndash418∘C afterthe first passage affecting up to 100 of animals Beginningfrom the fourth passage lethal outcomeswere recorded Afterthe fifth passage the virus became highly lethal for guineapigsThe virus presence in the blood was recorded at a titer of35 log

10PFUmL at passage 1 and did not change throughout

the passage series The resulting virus strain was designatedK-5 [9]

In the third passage series GLA [11] the animals inocu-lated at passage 1 developed a low-grade (below 40∘C) feverwhich showed a double-peak pattern with the peaks at days7 and 20 A low amount of virus (162 log

10PFUmL) could

only be detected in the serum after the virus concentrationby centrifugation while no virus could be isolated fromthe whole blood The tissue virus titers were detectable at16 log

10PFUmL (in liver) and 322 log

10PFUmL (in peri-

toneal exudates) The first lethal outcomes were detected atpassage 2 (on day 10) concomitant with the increase of serumand tissue virus titers to 42 log

10PFUmL Interestingly the

second peak of fever was not observed in the survivinganimals At passage 3 the adapting EBOV became fully lethalto the guinea pigs with all animals dying between days 5 and7 and the serum and tissue virus titers stabilized at the levelsof 30ndash40 log

10PFUmL The GPA-EBOV strain obtained in

this series was used for a comprehensive analysis of path-omorphologic hematologic immunologic and coagulationpatterns of EBOV disease in guinea pigs

In the last series GPA-P7 [8] the precursor pre-GPAclone was obtained by a triple passage of original MayingaEBOV in Vero cells and subsequent cloning In the course ofadaptation series at each passage one animal that displayedthe most pronounced temperature response by day 7 pi wasused as a donor of the virus for the next passage At allpassages the fever onset in the infected animals was recordedon day 4 pi The duration of fever episodes increased frompassage to passage (from 6 days at passages 1-2 to 10ndash12days in the surviving animals at passages 5-6) By day 7 pithe animals of all passages displayed a low level of viremiawhich was not increasing throughout the passage seriesThe ability of virus to reproduce in the liver of infectedanimals was not changing significantly as well In the first twopassages EBOV induced a mild disease in guinea pigs withan increase in temperature to 400ndash404∘C on days 4ndash9 piand eventual recovery The first lethal outcome was recordedat passage 3 at day 15 pi No lethal outcomes were recordedat passage 4 while two lethal outcomes (40 lethality) wererecorded at passage 4 Further on the lethality increasedwhile the time of survival decreased At passage 6 80lethality was observed and finally at passage 7 full lethalityof infection was acquired The survival time in this fullylethal EBOV infection averaged 116 days In lethal infectionsthe animals rapidly lost weight displayed anorexia and

developed impaired motor coordination especially in theterminal stage Two-three days before death guinea pigshad diarrhea and the signs of intestinal hemorrhage wereobserved This highly lethal EBOV variant obtained by the 7-passage adaptation of pre-GPA clone was designated GPA-P7[8]

Our several attempts were made to adapt two originalMayinga strain EBOV stocks (of adaptation series 8mc andK-5) to adult ICR mice using the same approach as theadaptation series to guinea pigs All attempts to adapt theK-5 virus stock failed however using high doses of the firstEBOV stock we obtained a paradoxical variant of EBO Thispassage series in adult ICR mice caused no lethal cases butresulted in the increasing virus titer in the liver reaching 555times 1011 PFUmL at later passages Of note the EBOV titers inthe liver even at the first passage reached 35 times 109 PFUmLon day 11 This mouse-adapted virus variant was named D-5[9]

All adaptation series and the designations of all originaland derivative viral stocks are provided in Figure 1

The guinea pig-adapted EBOV strains 8mc K-5 GLAandGPA-P7 as well as two other adapted variants obtained bythe other research groups [6 7] thus represent the virus vari-ants which appear to be very similar in their virulence andlethality (despite the number of passages needed to acquirethe full-scale pathogenicity) while strikingly different in theserum and tissue virus titers in the course of adaptation Twoof our adaptation series 8mc and GLA and the two seriesreported by others [6 7] represent a 2-3-order-of-magnitudegrowth of tissue virus titers (in PFU) accompanying thedevelopment of strain virulence for guinea pigs while in theother two K-5 and GPA-P7 no or slight virus titer increasehas been observed despite the acquisition of full lethalityto the animals [5 8 9] The adaptation series [5ndash7 11] inwhich an increase in the virus titers in the liver or spleenhas been observed (Table 1) prompted the hypothesis thatthe development of high virulencelethality of the adaptedvirus variants could be due to the increasing virus replicationin the infected tissues However the establishment of K-5and GPA-P7 virus stocks which do not show any significantincrease in virus titers while demonstrating high lethality tothe animals does not support this hypothesis Similarly aproposed link between the increasing rate of virus replicationand the development of lethality in the adaptation ofMayingaEBOV strain to mice [10] is contradicted by our data on themouse-adapted D-5 variant demonstrating the extraordinaryhigh tissue virus titers while lacking any pathogenicity inICR mice [9] It seems therefore that an increase in virusreproduction may not be the central determinant of EBOVvirulence evolution in the course of virus adaptation todifferent animal models

3 Study of the Genetic Determinants ofVirulenceLethality in Guinea Pig-AdaptedEBOV Variants

The information on the genetic markers of EBOV viru-lence is important for the understanding of EBOV infection


Table 1 Comparison of lethality virus reproduction and the time to adaptation ( of passages) in different adaptation series

Adaptation series 8mc[5]

K-5[9]

GLA[11]

GPA-P7[8] [6] [7] D-5

[9] [10]

Original virus stock MayingaZEBOV

MayingaZEBOV Mayinga ZEBOV Mayinga

ZEBOVMayingaZEBOV

MayingaZEBOV

MayingaZEBOV

MayingaZEBOV

Preadaptation historyVero E6cell linepassages

NHPs +L68 cellline

passages

NHPs + Vero cellline passages

Individualclone obtainedfrom infectedVero cells

Vero E6cell linepassages

mdash mdash mdash

Adaptation to Guinea pig Guinea pig Guinea pig Guinea pig Guinea pig Guinea pig Mouse MouseAdapted viruslethality 100 100 100 100 80 100 0 100

Number of passagesto full lethality 8 5 3 7 7 4 5 9

Increase in virus titer(nonadaptednonlethal versusadapted lethallog10PFUmL)

2955Liver

3535Plasma

0042Plasma16242Liver

32242Peritoneal exudate

6370 2950

2152Plasma

17(48ndash64)In differenttissues

95117 4278

Note all virus titers in our adaptation series 8mc K-5 GLA GPA-P7 and D-5 were determined on day 6 pi

stock stock

Variant D-5(causes high virus titers in ICR mice

without death)

(adapted to guinea pigs)

Variant K-5(adapted to guinea pigs)

GPA-P7(adapted to guinea pigs)

stockstock

GLA(adapted to guinea pigs)

Clone 5Clone 4

Clone 3Clone 2Clone 1

Clone 2 Clone 3

Clone pre-GPA

5 passages inadult ICR mice

8 passages inguinea pigs

15 passagesin L68 cell line

Variant Ch-15

Clone 1



Triple-cloningpassaging in


Several passages in Vero cell line

2 passages in NHP

The 1st EBO The 2nd EBO The 3rd EBO The 4th EBO

Vero cell line

Strain 8mc

Figure 1 Adaptation series of wild-type Mayinga strain EBOV to guinea pigs and mice resulting in the establishment of 8mc K-5 GLA andGPA-P7 lethal guinea pig-adapted virus variants and a paradoxical high-titer avirulent mouse-adapted virus variant D-5


pathogenesis as well as for the development of vaccines andantivirals Since some new mutations (as compared to thewild-type virus stocks) can occur de novo and some raremutations can be amplified and fixed in the adapted virusstocks in the process of adaptation comparison of geneticchanges in adapted lethal and preadaptation nonlethal EBOVgenomes and matching of those to the virulence of differentEBOV variants can yield invaluable information on thegenetic determinants of virulencelethality

In our first adaptation series 8mc a detailed analysis ofnucleotide sequences of the EBOV genomes of the originaland guinea pig-adapted virus stocks revealed several changesin genome regions encoding for the five viral proteins [9 16]The 8mc clones showed in total eight common sequencechanges compared to the original first EBO stock Two silentbase changes were found in the ORF encoding NP proteinwhich are obviously not affecting the NP function whileanother base change inNP gene leads to a conservative aminoacid substitution PherarrLeu(648) also unlikely to affect thevirulence No base changes were found in vp35 and vp40genes A single uridine insertion at the gene-editing site ofviral glycoprotein (GP) gene and an amino acid substitutionAsprarrGly(397) were found in three of five 8mc cloneshowever all five clones were similar in their pathogenicity forguinea pigs despite the significant change of GP expressionpattern caused by the former mutation A single base changein the 31015840 noncoding region of vp30 gene also seems irrelevantto the acquisition of virus pathogenicity in the course ofadaptation

Contrarily to the above mutations at least some ofthe three nucleotide changes in the coding region of vp24gene which led to amino acid substitutions MetrarrIle(71)LeurarrPro(147) and ThrrarrIle(187) may be linked to theadaptation process since the gel mobility of 8mc-EBOVVP24 is lower than its wild-type counterpart suggesting amajor structural rearrangement of this protein and probablythe crucial role for VP24 in the adaptation process

A conservative amino acid substitutionThrrarrAla(820) inL gene similarly to PherarrLeu(648) substitution in NP genedoes not seem to be responsible for virus virulence Howeversince both are found in all five EBOV-8mc clones they mayhave a supportive role in the increase in virulence [16]

A comparative nucleotide sequence analysis of the virusvariants of the second adaptation series (K-5) reveals two sin-gle nucleotide substitutions in vp24 ORF in the adapted virusstrain K-5 (as compared to the original second EBOV stock)which lead to the amino acid substitutions AsprarrGly(165)and HisrarrTyr(186) [17] Since the former is present in theinitial virus variant Ch-15 used for adaptation in guinea pigsit has obviously been introduced during passaging in NHPsor L68 cells and seems irrelevant to the development ofpathogenicity of an adapted virus The other vp24 mutationis found at nearly the same location in EBOV variant K-5as in another guinea pig-adapted variant 8mc obtained inthe first passage series thus supporting the hypothesis of apossible involvement of this protein in the host adaptationand increase in pathogenicity [16]

To further evaluate the contribution of genetic differencespresent in guinea pig-adapted EBOV variants in the virus

virulence two different approaches were employed The firstone was another adaptation series started with a clonedEBOV stock (pre-GPA) in which the genetic events occur-ring were determined at every passage and compared to thepathogenic reactions in guinea pigs reflecting the emergingEBOV virulence [8] The other study was using the reversegenetic approach and was focused on the effects of individualmutations of EBOV-8mc variant and their combinations onEBOV virulence [18]

In the former study the GPA-P7 adapted strain genomewas found to differ from the genome of the pre-GPA cloneby 17 nucleotide changes No genetic differences were foundin the terminal untranslated regions of the GPA-P7 genome(31015840-leader and 51015840-trailer) The nucleotide sequence of theGPA-P7 NP gene contained 12 nucleotide substitutions8 of those being synonymous and 4 leading to aminoacid changes in the encoded NP protein (LeurarrPro(544)AsnrarrSer(566) SerrarrPro(598) and AsnrarrAsp(663)) Threesubstitutions leading to amino acid changes were detectedin the nucleotide sequence of the GPA-P7 VP24 gene(MetrarrIle(71) LeurarrPro(147) and ArgrarrLeu(154)) The Lgene (encoding the RNA-dependent RNA polymerase) ofGPA-P7 contained two nucleotide substitutions one synony-mous and one nonsynonymous the latter resulting in aminoacid change (ValrarrIle(236)) [8]

Correlation of the mutations in GPA-P7 with the timeof their occurrence (passage number) and the emergingvirus pathogenicity in guinea pigs allows one to evaluatetheir roles in the virulence Of 17 nucleotide substitutionsthat distinguish between GPA-P7 and the pre-GPA clone12 occurred as early as the first passage (GPA-P1 variant)These 12 substitutions comprise nine synonymous (eight inNP gene and one in L gene) and three nonsynonymoussubstitutions resulting in 3 amino acid changes in NP gene(LeurarrPro(544) SerrarrPro(598) and AsnrarrAsp(663))Thosemutationswere considered not related to the emerging EBOVvirulence since they occurred before the acquisition oflethality Two other mutations emerged at passage 3 leadingto the amino acid changes AsnrarrSer(566) in NP proteinand LeurarrPro(147) in VP24 protein Of note the first lethaloutcome in this series was recorded at passage 3 although nolethal outcomes among the infected animals were recordedat passage 4 Those two loci remained heterogenous untilthe substitutions were fixed in the genome of the GPA-P5variant (at passage 5) representing increased virulence forguinea pigs and causing the disease with 40 lethality (2lethal outcomes of five observed animals) At passage 6 amutation in the VP24 gene sequence G10557A occurredresulting in amino acid change MetrarrIle(71) This passagewas accompanied by a further increase in EBOV virulencefor guinea pigs appearing as an 80 lethal disease (fourlethal outcomes between days 10 and 12 out of 5 animals)At passage 7 a nucleotide substitution G10805U resultingin an amino acid change ArgrarrLeu(154) was fixed in theVP24 gene sequence and a heterogeneous site (G12286AValrarrIle(236)) appeared in the L gene sequence Emergenceof these mutations correlated with a further increase inEBOV pathogenicity for guinea pigs which manifested as100 lethality (between days 8 and 12) Thus only the three


amino acid changes in VP24 protein (positions 147 71 and154) one change in NP protein (position 566) and onechange in L-polymerase (position 236) may have an essentialcontribution to the gradually emerging virulence of EBOVin the course of GPA-P7 series [8] Of interest two identicalmutations in VP24 (positions 71 and 147) were discovered inthe abovementioned 8mc adapted virus strain [16]

In the latter study [18] recombinant viruses carryingvarious combinations of wild-type and guinea pig-adaptedgenes were generated by reverse genetic approach and theircontribution to the EBOV virulence in guinea pigs was eval-uated The membrane-associated protein VP24 was shownto play a critical role in EBOV pathogenesis and aminoacid changes in VP24 were shown essential and sufficient toconfer EBOV virulence in guinea pigs Indeed recombinantvirus stocks carrying only those 8mc mutations harboredin VP24 gene are causing a uniformed lethality in guineapigs Inoculation of animals with the viruses carrying NPandor L gene mutations did not show any sign of the diseaseexcept for the death of two guinea pigs in the group infectedwith recombinant virus carrying a mutation in NP gene(rEBOV-NP8mc F648L) Sequence analysis of virus fromthe blood of those dead animals revealed the additionalamino acid substitution L26F in the VP24 Inoculation ofguinea pigs with virus from such animals resulted in theiruniform death To evaluate the role of this mutation in thevirulence of guinea pig-adapted EBOV a recombinant viruswas generated that differed from the wild-type virus by asingle mutation in VP24 rEBOV-VP24L26F Strikingly allanimals infected with rEBOV-VP24L26F developed signsof EBOV disease approximately 4 days pi Importantly noadditional mutations were found in the viruses recoveredfrom guinea pigs inoculated with this recombinant virussuggesting that adaptivemutations inVP24 are necessary andsufficient for an increase in EBOV virulence in guinea pigsThe ldquo8mcrdquo mutation in NP (PherarrLeu(648)) is not strictlyrequired for the increase in pathogenicity but seems to lead tothe accelerated selection of EBOV variant in which a singlemutation in VP24 was able to confer lethality [18]

Of interest a similar major contribution of VP24 muta-tions in the EBOV adaptation to mice has been observedwith the requirement for the supportive mutations in NP[19] The role of VP24 and NP proteins in the virulenceof mouse-adapted EBOV strain however correlated withtheir ability to evade IFN-I-triggered antiviral responsesThe guinea pig-adapted and wild-type EBOV VP24 onthe contrary were similar in their ability to block IFN-I-stimulated response [18] which underlines the importance ofdifferent animal-adapted EBOVmodels in the dissection of acomplex phenotype of EBOV disease

A summary of all genetic determinants revealed in theguinea pig adaptation series of EBOV is provided in Table 2

4 Pathomorphological Findings inEBOV-Infected Guinea Pigs

Ebola virus infection in primates is associated with diffuseparenchymal necrosis in the liver without marked inflamma-tory changes [20] Unlike that Ryabchikova and colleagues

report the occurrence of granulomatous lesions in the liverto be the hallmark of the wild-type EBOV infection inguinea pigs [6] Those granulomas are proposed to playan important role in virus clearance Granuloma-like fociconsisting of the monocytes macrophages and neutrophilsappear at day 5 pi and increase in size and number thereafterNo involvement of hepatocytes was observed in this studyuntil days 7-8 when some of the perifocal hepatocytesdemonstrate lipid droplet accumulation and destruction oforganoids The other organs were only slightly involvedexcept for the spleen and lymph nodes in which some signsof immunosuppression (decreased number of mitoses lackof plasma cell formation and destruction of stroma cells andmacrophages) were found Virus reproduction occurred inthe mononuclear phagocytes of the liver only at later stagesof infection (on day 7) and was observed as the presence ofviral nucleocapsids and specific membranous structures inthe infected cells Most of the infected macrophages werelocated in the granuloma-like foci and contained virionswhich appeared eroded with their nucleoids indistinct Noreproduction of wild-type virus was detected in the otherorgans [21]

In our study we were able to observe the dynamics ofpathomorphologic changes in wild-type EBOV infection inguinea pigs [12] which was somewhat different from thepattern described above Electron microscopy revealed theoccurrence of EBO virions in the dilated Disse spaces asearly as 24 h pi Simultaneously a developing inflammatoryreaction in the periphery of liver acini was observed withballooning degeneration of hepatocytes showing disorgani-zation of organelles (swollen mitochondria) and necroticfoci containing phagocyting macrophages and virus particles(Figure 2(a)) However no virus replication centers wereobserved in the macrophages By day 5 the necrobioticchanges are progressing and involve the major part of liverparenchyma with the often occurrence of necrotic hepato-cytes numerous hemorrhages in the swollen subendothelialspace and centers of virus replication in the necrotic focialthough with only few virus particles visible (Figure 2(b))By day 7 of infection however the progression of necrobioticchanges ceases and the tissue structure stabilizes At thistime macrophages with numerous osmiophilic inclusionsnuclei containing viroplasm and numerous virions in theirneighborhood are seen (Figure 2(c)) The major differencefrom the course of wild-type EBOV infection in guineapigs observed by Ryabchikova et al [6] was the presence ofdamaged hepatocytes at days 5ndash7 pi

The gradual emergence of EBOV virulence to guineapigs in a series of passages is reflected in the changesof pathomorphological pattern of the infection At earlypassages (1ndash3) only quantitative differences in the numbers ofactivated and infected liver macrophages were observed Thepathomorphological pattern of infection changes at passage4 with greater numbers of liver macrophages involved aswell as the virus reproduction first detected in hepatocytesHowever only the formation of nucleocapsids not the propervirus budding occurs in hepatocytes at this stage makingit unclear whether the infection in these cells is productiveIn general the liver lesions are more severe than at earlier

Advances in Virology 7Ta

ble2Geneticchangeso

ccurrin

gin

different

adaptatio

nserie

sand

reverseg

eneticse

xperim

entsandtheirrele

vancetothed

evelo

pmento

fGPA

-EBO

Vvirulencelethality

ViralO

RFNucleotidep

osition

Aminoacid

substitution

Relevancetoadaptatio

nCom

ment

EBOV

-8mca

daptationserie

s

NP

1852

Non

eNo

Syno

nymou

s2410

Non

eNo

Syno

nymou

s2411

Pherarr

Leu(64

8)Unlikely

Con

servativemay

play

asup

portiver

oleinadaptatio

n

GP

6924

(insertion)

Fram

eshift

No

Clon

eswith

orwith

outtho

seGPmutations

dono

tdiffer

from

each

otherintheir

pathogenicity

inguinea

pigs

7228

Asprarr

Gly(397)

No

VP3

09595

Non

eNo

Locatedin

31015840-untranslated

region

VP2

410557

Metrarr

Ile(71)

Prob

ably

Atleaston

eoftho

sethreeV

P24mutations

indu

cesm

ajor

structural

rearrangem

entsin

EBOV-8m

cVP2

4which

arep

resented

asgelm

obilityshift

inele

ctroph

oresis

10784

Leurarr

Pro(147)

Prob

ably

10904

Thrrarr

Ile(187)

Prob

ably

L14038

Thrrarr

Ala(820)

Unlikely

Con

servativemay

play

asup

portiver

oleinadaptatio

nEB

OV-5Kadaptatio

nserie

s

VP2

410838

Asprarr

Gly(165)

No

Isacqu

iredin

thec

ourseo

fpassage

inNHPs

orL6

8cellsprio

rtoadaptatio

n10907

Hisrarr

Tyr(186)

Prob

ably

Acqu

ireddu

ringadaptatio

nliesc

lose

toEB

OV-8m

cVP2

4Th

rrarrIle

(187)

EBOV

-GPA

-P7serie

s

NP

1781

Asnrarr

His(438)

No

Isfoun

din

nonadapted

pre-GPA

clone

2043

Argrarr

Lys(525)

No

Isfoun

din

nonadapted

pre-GPA

clone

2092

Non

eNo

Syno

nymou

s2100

Leurarr

Arg(544

)No

Occurredatpassage1

after

pre-GPA

beforethe

acqu

isitio

nof

lethality

2164

Non

eNo

Syno

nymou

s

2166

Asnrarr

Ser(566)

Prob

ably

Occurredatpassage3

whenthefi

rstlethaloutcomew

asrecordedheterogenou

ssiteu

ntilfixed

atpassage5

con

comitant

with

40lethality

2188

Non

eNo

Syno

nymou

s2191

Non

eNo

Syno

nymou

s2209

Non

eNo

Syno

nymou

s2222

Non

eNo

Syno

nymou

s2233

Non

eNo

Syno

nymou

s2260

Non

eNo

Syno

nymou

s2261

Serrarr

Pro(598)

No

Occurredatpassage1

after

pre-GPA

beforethe

acqu

isitio

nof

lethality

2409

Serrarr

Phe(647)

No

Isfoun

din

nonadapted

pre-GPA

clone

VP4

02456

Asprarr

Asn(663)

No

Occurredatpassage1

after

pre-GPA

beforethe

acqu

isitio

nof

lethality

5576

Non

eNo

Syno

nymou

s

VP2

4

5597

Non

eNo

Syno

nymou

s10557

Metrarr

Ile(71)

Prob

ably

Occurredatpassage6

con

comitant

with

lethality

increase

to80

10784

Leurarr

Pro(147)

Prob

ably

Occurredatpassage3

whenthefi

rstlethaloutcomew


ssiteu

ntilfixed

atpassage5

con

comitant

with

40lethality

10805

Argrarr

Leu(154)

Prob

ably

Occurredatpassage7con

comitant

with

fulllethality

L12286

Valrarr

Ile(236)

Prob

ably


comitant

with

fulllethality

16821

Non

eNo

Syno

nymou

sRe

verseg

enetics

ofEB

OV-8mcm

utations

VP2

410422

Leurarr

Phe(26)

Definitely

Derived

from

theb

lood

ofan

anim

alw

hich

died

after

inoculationwith

anon

lethal

recombinant

rEBO

V-NP8m

cF6

48LvirusWhenintro

ducedinto

anotherw

isewild

-type

Zaire

EBOVgeno

meb

yreverseg

eneticsthismutationalon

econ

fersfull

lethality

toguinea

pigs


MP

Vpt

(a) Activated macrophage (MP) with phagosomes andpolymorphous virus particles (Vpt) in the necroticfocus of the liver Day 1 pi

(b) Necrotic focus in the liver day 5 pi Arrowsshow the virus replication centers in the celldebris

Vpt

VptVpl

(c) Cell fragment with osmiophilic inclusions(arrows) virus particles (Vpt) and viroplasm (Vpl) inthe nucleus Day 7 pi

Figure 2 Wild-type EBOV nonlethal infection in guinea pigs Electron microscopy of the liver

passages and the signs of spleen and kidney involvementbecome evident All observed morphological changes grad-ually become more prominent at passages 5-6 with liverfocal lesions becoming widespread and number of infectedcells and virions increasing With the acquisition of fulllethality of the adapted virus to guinea pigs (passage 8)the hepatic tissue becomes filled with infected hepatocytesand macrophages with intracellular nucleocapsids as well asnumerous virions Some cells contain structurally abnormalviroplasm and nucleocapsids Areas of liver necrosis contain-ing lysed hepatocytes macrophages and virions become all-pervading Other organs and tissues show greater severity oflesions as well [21]

In GPA-EBOV-infected guinea pigs we similarlyobserved the necrotic hepatocytes (Figure 3(a)) and virusparticles in subendothelial space (Figure 3(b)) as early as 24 hpi [12] On day 2 the tissue destruction is more prominentwith multiple hemorrhages and necrotic foci containingnumerous virus particles fibrin deposits and macrophages(Figure 3(c)) The centrilobular hepatocytes typically showthe signs of ballooning dystrophy while the perilobularhepatocytes become necrotic and contain virus particles(Figure 3(d)) This pattern indicates a progression of liver

damage on day 2 By day 5 the necrotic process involvesthe central parts of acini where numerous necrotic foci withfibrin impregnation (Figure 3(e)) and Kupffer cells withviroplasm and virus particles (Figure 3(f)) are detected Atday 7 the pronounced and overall destruction of liver tissueof animals infected with GPA-EBOV is visible with largeareas of necrosis involving both stromal and parenchymalcells (Figure 3(g)) Those necrotic lesions contain activelyphagocyting macrophages with small and large phagosomesviroplasm and virus particles (Figure 3(h)) as well ashepatocytes containing numerous viruses (Figure 3(i)) Alsoby this time the liver lesions involve massive hemorrhagesresulting in the occurrence of new areas of inflammationand the generalization of liver injury We believe that thelethality in GPA-EBOV-infected guinea pigs is thus dueto the triggering damage by the virus and the subsequentinvolvement of different areas of the liver resulting from theinflammatory reaction [12]

Day-by-day dynamics of pathomorphologic changesoccurring in the GPA-infected guinea pigs illustrated byConnolly and colleagues [7] is in a good agreement withstructural changes described above In animals infected withthe adapted fully lethal EBOV variant the viral RNA and


H

H

Er

T

(a) Fragment of liver acinusArrow points to a sinus Er ery-throcyte Vpt virus particle Tthrombocyte and H hepatocyte

(b) Arrow shows virus particles (Vpt) in suben-dothelial space Day 1 pi

H

Vpt

MPVpt

Vpt

(c) Destroyed macrophage (MP) debris fibrin deposits(arrow) virus particles (Vpt) and hepatocytes (H) incolliquative necrosis in the necrotic focus Day 2 pi

H

(d) Hepatocytes (H) in focal necrosis withtheir cytoplasm and perisinusoidal space filledwith numerous virus particles (arrow) Day 2pi

(e) Fibrin deposit (arrow) in the necrotic focusDay 5 pi

Vpt

Vpt

Vpt

CC

(f) Kupffer cell (CC) in the perisinusoidalspace Cytoplasm contains phagosomes virusparticles (Vpt) viroplasm and nucleocapsidsDay 5 pi

VptVpt

Vpt

Vpt

(g) Large necrotic focus in the liver with virusparticles (Vpt) found ubiquitously Day 7 pi

Vpl

Vpl

(h) Necrotic macrophage with cytoplasm con-taining viroplasm (Vpl) nucleocapsids andphagosomes Day 7 pi

VptVpl

Vpl

(i) Hepatocyte containing viroplasm (Vpl)and virus particles (Vpt) Day 7 pi

Figure 3 GPA-EBOV lethal infection in guinea pigs Electron microscopy of the liver

antigens are detected in the lymph nodemacrophages as earlyas 24 h pi The number of infected macrophages in whichthe typical EBOV intracytoplasmic inclusions representingviral nucleocapsids are seen by electronmicroscopy increases4ndash20-fold by 48 h pi confirming an ongoing productiveinfection From day 3 a developing multifocal lymphoidnecrosis and the depletion of lymph nodes of differentlocalization are observed which progressed by days 7ndash9Ultrastructurally the sinuses contain necrotic cellular debrisinfected macrophages and free virions while no evidence ofEBOV-infected lymphocytes in the lymph nodes is found

In the spleen the occurrence of single-cell necrosis in thered pulp is seen at day 2 pi progressing into a moreextensive necrosis of both splenic white and red pulp andthe depletion of hematopoietic elements in the red pulp byday 4 All signs of productive infection (viral antigens RNAand intracytoplasmic inclusions) are detected in the red pulpmacrophages and the antigen-presenting cells (dendritic andreticular cells) of some white pulp follicles The spread ofinfection is associated with numerous circulating monocytesand tissue macrophages containing EBOV inclusions withsome of infected cells already showing budding virions and


clusters of virions along their plasma membrane In theliver small foci of hepatocellular necrosis occur at days 2-3pi with EBOV-staining hepatocytes and Kupffer cells seenOn day 4 the ongoing infection in the liver spreads to thehepatocytes surrounding the necrotic foci and Kupffer cellssince they also begin to contain viral cytoplasmic inclusionsMost of these foci stain EBOV RNA- and antigen-positiveand the sinuses are filled with antigen-positive debris Freevirions in the Disse spaces and virus capsid inclusions arefound in Kupffer cells at day 5 Later at day 7 some peri-portal fibroblast-like cells become involved in the infectionand stained EBOV-positive The occurrence of the virion-containing fibrin depositions in subendothelial spaces andvirus particles in subepithelial connective tissue of the bileducts (both as free virions and cytoplasmic viral inclusionsin degenerate fibroblasts) probably marks the onset of asystemic spread of infection from the liver at day 9 In factat later stages of infection involvement of different organsand tissues as well as cellular targets is observed gastroin-testinal fibroblasts and macrophages adrenal cortical cellsand fibroblasts bladder epithelium interstitial and alveolarmacrophages the occasional fibroblasts and endothelial cellof pulmonary venules in the lungs and so forth [7 21]

An independent morphological study of the wild-typeand guinea pig-adapted EBOV infection reveals an inabilityof the nonadapted virus to reproduce in its primary targetthe macrophages [18] Peritoneal guinea pig macrophagesinfected in vitro by the recombinant virus obtained byreverse genetics and carrying the adaptive mutations inVP24 identical to those occurring in the 8mc adapta-tion series show characteristic inclusion bodies containingviral nucleocapsids Contrarily no typical viral nucleocap-sids but only massive protein inclusions were found inthe cells infected by the wild-type virus which correlatedwith the lack of infectious virus release from those cellsThis data suggested an efficient block of wild-type EBOVinfection already in the primary target cells probably atthe stage of VP24-mediated nucleocapsid assemblybudding[18]

Thus the morphologic patterns of wild-type and adaptedEBOV infection in guinea pigs are suggestive of the inabilityof the virus to establish productive infection in macrophagesand particularly in hepatocytes Failure tomassively produceinfectious wild-type EBO virions may result in the hostdefense successfully blocking further virus spread by lockingthe infection in granuloma-like inflammation foci with asubsequent virus clearance The adaptive mutations in theEBOV genome can restore the ability of a virus to reproducein its major targets macrophages and hepatocytes thusfacilitating systemic virus spread in the infected animals thatresults in high virulence On the other hand this mechanismof adaptation would present itself as the increase of bloodand liver virus titers in the adapted virus-infected animalsas compared to those inoculated with the wild-type viruswhich is not the case for two adaptation series K-5 and GPA-P7 There should be therefore an alternative mechanismof virulence increase during adaptation which is relevantto those series A morphological study of K-5 and GPA-P7 virus infections which has not been done as yet and

its comparison with the morphological patterns of infectiondescribed earlier would thus be of interest

5 Hematologic Immunologic andCoagulation Shifts in the Guinea PigInfections by Wild-Type and Adapted Virus

Wild-type EBOV infection causes serious hematologic aber-rations in its natural hosts including leukocytosis due tomarked neutrophilia and frequently lymphopenia in ter-minally ill primates Thrombocytopenia develops in acutelyill human patients and experimentally infected primatesexperiencing hemorrhagic syndrome Unlike that both neu-trophilia and lymphopenia can be observed in EBOV-infected guinea pigs as early as day 2 pi and are progressingover the course of the disease According to some reportsthe scale of those hematological shifts in individual infectedanimals can be as great as a gt35-fold increase in the numberof neutrophils and gt30-fold drop in the number of lym-phocytes This finding appears intriguing since lymphocytesare not infected by EBOV in vivo despite the progressionof lymphoid necrosis and they are resistant to infections bydifferent EBOV strains in vitro [7] The bystander lympho-cyte apoptosis an important feature of EBOV infection inprimates and mice has not been determined in guinea pigsinfected with adapted EBOV as well [2] The mechanisms oflymphoid necrosis and depletion occurring both in primatesand in guinea pigs remain unclear but are probably caused bythe events resulting from amassivemacrophage infection andchanging the microenvironment of lymphoid tissue Severethrombocytopenia (gt14-fold as compared to the preinfectionlevels) develops later on day 7 of infection (two days beforedeath at day 9) resembling the loss of platelets in terminallyill primates [7]

It appears though that the leukocyte counts should betaken with care since a wide-range variation of differentleukocyte types between the animals and their experimentalgroups is natural to guinea pigs (cf neutrophillymphocytepercentage in the reported group 2177 to the data from oneof our adaptation series 5738 [11]) Therefore the analysisof dynamics of hematologic indices at different passages ofadaptation and its correlation with the clinical course ofEBOV disease appears to be more valid

In one of our adaptation series GLA we specificallyaimed to study the evolution of different blood indicesin the course of adaptation [11] In this experiment theleukocyte counts remained stable throughout passages 1ndash3demonstrating a significant 3-fold decrease at next passage 4remaining at this lower level throughout the further adapta-tion This passage was also notable for the first occurrenceof eosinophils in blood samples and an abrupt (sim8-fold)although transient increase in monocyte and plasmacytecounts which returned to baseline values at the next passageAlso the amount of neutrophils dropped by 20 at thispassage (transiently to recover at passage 5) while thepercentage of young atypical and large probably activatedlymphocytes showed a transient increase at passage 4 (alsoto return to the baseline values at passage 5) Interestingly


Table 3 Serum and peritoneal TNF activity in EBOV-infected guinea pigs and rabbits

Virus Guinea pigs RabbitsMayinga (wild-type) 8mc guinea pig-adapted Mayinga (wild-type)

TNF activity serum (UmL) 34 plusmn 01 (day 7) 25 plusmn 016 (day 7)104 plusmn 36 (day 3)471 plusmn 137 (day 5)1398 plusmn 909 (day 9)

Virus titer serum (PFU) 103 104ndash105 NondetectableTNF activity peritoneal (UmL) Nondetectable Nondetectable 153 plusmn 21 (days 5ndash9)Virus titer peritoneal (PFU) 102ndash103 102ndash103 Nondetectable

these changes followed the acquirement of full lethality by theadapting EBOV at passage 3 [11]

The percentage of young thrombocytes demonstrated a3-fold increase at passage 2 and an overall 12-fold increaseby passage 6 The total thrombocyte count remained stableuntil it fell abruptly (2-fold) at passage 6 The occurrence ofyoung thrombocytes at early passages before the fully lethalEBOV has been selected indicates the important shifts in thethrombocytic hemostasis which resulted in a progressive lossof platelets observed in the next passages

An interesting finding was the change of erythrocytemorphology in the EBOV adaptation The percentage ofnormal red blood cells was progressively decreasing at everypassage concomitant with the occurrence of echinocytesand erythrocytes with inclusions which suggests the exit ofimmature erythrocytes from the bone marrow apparentlyto compensate for the deficiency of red blood cells in theorganism The occurrence of echinocytes in the peripheralblood has been noticed in the acute viral hepatitis and someother viral and inflammatory diseases [22] however theinvolvement of erythrocytes into the EBOV infection hasnever been reported earlier

The shifts of the peripheral blood counts were accompa-nied by the changes in bone marrow composition firstly bythe increase in the number of blast cells as early as passage2 reaching its peak of 65 pro mille by passage 5 but furtherdroppingThismay indicate compensatory activation of bonemarrow function with subsequent exhaustion Megakary-ocyte counts started growing at passage 2 and reached theirpeak at passage 5 clearly reflecting the upward dynamics ofyoung thrombocytes

Finally the phagocytic function of neutrophils was pro-gressively decreasing from passage 2 although the numberof neutrophils was variable at different passages Contrarilythe phagocytic function of peritoneal macrophages wasremaining stable for 6 passages and then dropping abruptly[11 23]

Although many of the hematologic indices were fluctuat-ing in the passage series or changing at later passages (whichallowed ruling out their causative role in the emerging lethal-ity of adapting EBOV) some were evidently concomitantwith the increasing pathogenicity or even appearing priorto the acquiring of full lethality by the adapted virus Thoseinclude the increase of the percentage of young thrombocytesalong with the growing number of blast cells and megakary-ocytes and the decrease of phagocytic activity of neutrophils

all commencing as early as passage 2 alongside the significantgrowth of blood and tissue virus titer and the occurrenceof the first lethal cases in the inoculated animals It is thustempting to hypothesize that the evolving wild-type EBOVacquires new characteristics triggering those early responsesin hematologic parameters [11] Those key parameters werealso changing consistently in our other adaptation seriesGPA-7 although it was rather different from the above GLAexperiment since the virus titer was not growing in thepassage series and the full lethality has been acquired laterat passage 5 versus passage 3 for GLA series [8] Howeverthe decreasing phagocytic activity of neutrophils was alsoan early response appearing at passage 2 and the trendof accumulation of atypical lymphocytes in the course ofadaptation was evident probably due to the progressingapoptotic and necrotic processes

A rather intriguing immunologic finding was the lackof correlation between the blood and peritoneal TNF levelsand the pathogenicity of EBOV infection in experimentalanimals The guinea pigs infected with wild-type Mayingaor lethal adapted 8mc EBOV demonstrated similar ratherlow levels of TNF activity in blood and no TNF activity inperitoneal exudates (Table 3) Contrarily when rabbits whichare completely resistant to EBOV infection were inoculatedwith a wild-type EBOV a rather high serumTNF activity wasdetected early (day 3 pi) and was progressively increasingto extraordinary high values later on (days 5 and 9 pi) Inrabbits TNF activity was also detected in peritoneal exudates(Table 3) [23] The previously published data report the ele-vated proinflammatory cytokine levels in the lethal wild-typeEBOV infections in humans and nonhuman primates [24 25]and in baboons infectedwith guinea pig-adaptedEBOVHighlevels of TNF-120572 production by peripheral bloodmacrophagesinfected in vitro by another filovirus Marburg virus wereproposed to be the major pathogenetic factor of Marburgvirus infection due to its ability to induce hemorrhagic shock[26] Thus our data on TNF activity in the rodent models ofEBOV infection (guinea pigs and rabbits) are contradictoryto the proposed causative role of TNF-induced hemorrhagesin primate infections

Since coagulopathies such as disseminated vascularcoagulation (DIC) or massive hemorrhages are the hallmarkof EBOV disease in primates [27] and considering the tightand clinically relevant cross talk between the coagulation andcomplement systems [28 29] it seems important to discuss


0

20

40

60

80

100

120

140

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

Com

plem

ent a

ctiv

ity (u

nits)

Days pi

Mayinga

AH3N2Aichi268 influenza virusVaccinated noninfectedIntact noninfected

Intact infected with Prevaccinated infected with 8mc

8mc

Figure 4 Complement activity in the intact and prevaccinatedguinea pigs infected with lethal GPA-EBOV-8mc strain or nonlethalwild-type EBOV Mayinga strain Animals infected with nonlethalinfluenza virus AAichi268 (H3N2) strain intact animals andguinea pigs vaccinated with inactivated nonlethal Mayinga strainEBOV while not infected were used as different controls of com-plement activity Slow lagging dynamics of complement activityin guinea pigs infected with nonlethal Mayinga strain EBOV iscontrasting to the rapid onsetdrop peak of complement activityin the intact and prevaccinated animals infected with a lethal 8mcstrain GPA-EBOV

the role of complement involvement in the experimentalEBOV infection in guinea pigs

The dynamics of complement activity in wild-typeEBOV-infected guinea pigs and in the intact or EBOV-preimmunized animals infected with a lethal adapted EBOVstrain demonstrates striking differences which can be impor-tant for the understanding of EBOV-induced coagulopathiesIn wild-type EBOV-infected animals the complement activ-ity shows a double-peak profile with a 2-day lag the firsthigher peak at days 4ndash7 and second lower peak at days 9-10Contrarily in animals infected with the adapted lethal 8mcstrain the complement activity demonstrates a rapid peak at24 h pi followed by an abrupt decline on day 2 return tothe baseline by days 4-5 and further drop to the backgroundlevels by days 9ndash11 followed by the animal death Finally theguinea pigs preimmunized with formalin-inactivated EBOVand infected by a lethal 8mc strain showed even a more steeprise of complement activity on day 1 and a more abrupt fallon day 2 pi with a further decline to near-zero values by day6 also followed by the animal death by day 8 (Figure 4)

The rate of complement activity growth in the lethallyinfected guinea pigs suggests the involvement of an alter-native pathway of complement activation In preimmunizedanimals the classic activation can additionally occur dueto the presence of anti-Ebola immunoglobulins which maybe the possible cause of higher complement activities thanin nonimmunized animals Alternatively slower kinetics ofcomplement activation in guinea pigs infected with nonlethalstrain indicates classic activation with the two peaks corre-sponding to IgM (before day 7) and IgG production (days7ndash14) Rapid complement activation by alternative pathwayappears to be a unique ability of the 8mc lethal EBOV strainto immediately hyperactivate complement production withsubsequent fast and complete exhaustion of its synthesis [30]Importantly the retrospective study of the sera from days1 and 5 and later of a single fatal case of a laboratory staffEBOV infection accident reveals the similar pattern of a sharpincrease and an immediate drop of complement activitysuggesting the relevance of this data for humans [12]

Interestingly the pattern of serum concentrations ofcirculating immune complexes (CICs) is closely resemblingthe kinetics of complement activity both in intact and inpreimmunized lethally infected animals with a typical rapidrise in the first day and an immediate decline to near-zero values by days 8-9 or 5-6 in intact or preimmunizedanimals respectively followed by the animal death SlowCIC and complement kinetics are also very similar in thenonlethal EBOV infections of guinea pigs The significanceof circulating immune complexes (CICs) in the pathogenesisof Ebola disease still remains to be studied [30]

Since no coagulation abnormalities are occurring in themouse-adapted EBOV model which is a general feature ofmouse models for acute viral infections [19] the guinea pig-adapted EBOV remains a unique nonprimate model for thestudy of EBOV-induced coagulopathies

6 Conclusion

Since the first isolation of wild-type EBOV strains theproblems with the animal model for the study of biologyand therapyprevention of this infection became evidentResistance of the most convenient rodent model animals(mice guinea pigs hamsters and rabbits) to EBOV infectionprompted the researchers to use adaptation of wild-typeEBOV strains to the small animal models Practically theEBOV adaptation to guinea pigs was the easier and the firstto be achievedThe lethality of the guinea pig-adapted EBOVvariant to the nonhuman primates remained unchanged [31]In the adaptation passage series of EBOV stocks differingin their passage histories described here and previouslythe emergence of the fully lethal GPA-EBOV variants wasreadily achieved in 7 passages or less EBOV adaptationto mice required much more efforts (M Bray personalcommunication) In our own attempt of EBOV adaptationto mice a paradoxical nonlethal strain was obtained whichdemonstrated high-level replication and persistence of virusin the animalrsquos liver while lacking any clinical symptoms of


the disease The historically first GPA-EBOV strain desig-nated 8mc was originally established for the vaccine devel-opment and heterologic immunoglobulin production [32]However it was soon realized that this variant not onlyrepresents a convenient small animal model but also offersthe opportunity to compare the genetic differences betweenthe wild-type and adapted EBOV strains and establish thecorrelation between the genetic determinants and the emerg-ing and evolving virulence and lethality in a new host [16]Also the comparison of pathomorphologic immunologichematologic and coagulation features of EBOV infection bythe wild-type and adapted virus allows one to establish themost important physiologic parameters which influence thecourse of EBOV disease [9 11 12 23]

We and others were able to demonstrate the mutations inEBOV VP24 protein to be the most important determinantsof the virus virulence and lethality [8 9 16 18]

The need to reproduce the consistency of processesoccurring in the course of adaptation and establish the detailsof adaptation process required the repeated adaptation seriesstarting from the virus stocks with different passage historiesincluding the cloned EBOV stocks

Thedifferences in the dynamics of hemolytic complementactivity found in the infections of guinea pigs by the wild-type and adapted EBOV strains allowed us to be able topredict the outcome of EBOV infection in the first hours pi[30]

Also our studies indicate strong suppression of neu-trophilic phagocytosis in the infections of guinea pigs withGPA-EBOV strain while only its mild activation on day 5of wild-type EBOV infection was observed This suggestedthat the reproduction of EBOV in macrophages on the oneside with concomitant suppression of neutrophils on theother results in the disturbance of neutrophilmacrophageinteraction as a driving force of a lethal EBOV disease

Another interesting discovery was the presence of youngthrombocytes in GPA-EBOV-infected animals The phe-nomenon whichwas lacking in thewild-type EBOV-infectedanimals together with the increase of megakaryocyte countsin the bone marrow can be the evidence of the activationof megakaryocytic hematopoiesis and the exit of nonmatureplatelets into the circulation

Previously the elevated levels of proinflammatorycytokines have been reported to occur in the course of EBOVdisease and contribute to the infection lethality in humansand NHPs due to their role in the development of EBOVcoagulopathies However in guinea pigs we paradoxicallyfound higher (while comparable) TNF-120572 levels in the animalsinfected with a nonlethal wild-type than in those infectedwith lethal adapted EBOV thus demonstrating lack of TNFinvolvement in GPA-EBOV pathogenesis

Conflict of Interests

The authors declare that they have no conflict of interests

Acknowledgment

The authors are grateful to Professor L V Shestopalova forher incredible help in the preparation of electron microscopyimages used in the preparation of this paper

References

[1] D Bente J Gren J E Strong and H Feldmann ldquoDiseasemodeling for Ebola and Marburg virusesrdquo Disease Models ampMechanisms vol 2 no 1-2 pp 12ndash17 2009

[2] V Wahl-Jensen L Bollinger D Safronetz F de Kok-MercadoD P Scott and H Ebihara ldquoUse of the Syrian hamster as anew model of Ebola virus disease and other viral hemorrhagicfeversrdquo Viruses vol 4 no 12 pp 3754ndash3784 2012

[3] H Feldmann S Jones H-D Klenk andH-J Schnittler ldquoEbolavirus from discovery to vaccinerdquo Nature Reviews Immunologyvol 3 no 8 pp 677ndash685 2003

[4] E T W Bowen G S Platt G Lloyd R T Raymond and D ISimpson ldquoA comparative study of strains of Ebola virus isolatedfrom southern Sudan and northern Zaire in 1976rdquo Journal ofMedical Virology vol 6 no 2 pp 129ndash138 1980

[5] A A Chepurnov I V Chernukhin V A Ternovoi et alldquoAttempts of creating a vaccine against Ebola feverrdquo VoprosyVirusologii vol 40 no 6 pp 257ndash260 1995

[6] E Ryabchikova L Kolesnikova M Smolina et al ldquoEbola virusinfection in Guinea pigs presumable role of granulomatousinflammation in pathogenesisrdquoArchives of Virology vol 141 no5 pp 909ndash921 1996

[7] B M Connolly K E Steele K J Davis et al ldquoPathogenesis ofexperimental Ebola virus infection in guinea pigsrdquoThe Journalof InfectiousDiseases vol 179 supplement 1 pp S203ndashS217 1999

[8] E Subbotina A Dadaeva A Kachko and A ChepurnovldquoGenetic factors of Ebola virus virulence in guinea pigsrdquo VirusResearch vol 153 no 1 pp 121ndash133 2010

[9] A A Chepurnov N M Zubavichene and A A DadaevaldquoElaboration of laboratory strains of Ebola virus and study ofpathophysiological reactions of animals inoculated with thesestrainsrdquo Acta Tropica vol 87 no 3 pp 321ndash329 2003

[10] M Bray K Davis T Geisbert C Schmaljohn and J HugginsldquoA mouse model for evaluation of prophylaxis and therapy ofEbola hemorrhagic feverrdquo Journal of Infectious Diseases vol 179supplement 1 pp S248ndashS258 1999

[11] A A Dadayeva L P Sizikova Y L Subottina and A AChepurnov ldquoHematological and immunological parametersduringEbola virus passages in guinea-pigsrdquoVoprosyVirusologiivol 51 no 4 pp 32ndash37 2006

[12] A A Chepurnov and L V Shestopalova Genetic and Patho-physiological Factors of Ebola Virus Virulence Nauka Novosi-birsk Russia 2010 (Russian)

[13] L V Gubareva J M Wood W J Meyer et al ldquoCodominantmixtures of viruses in reference strains of influenza virus due tohost cell variationrdquo Virology vol 199 no 1 pp 89ndash97 1994

[14] B Kissi H Badrane L Audry et al ldquoDynamics of rabiesvirus quasispecies during serial passages in heterologous hostsrdquoJournal of General Virology vol 80 no 8 pp 2041ndash2050 1999

[15] E TW Bowen G LloydW J Harris G S Platt A Baskervilleand E E Vella ldquoViral haemorrhagic fever in southern Sudanand northern Zaire Preliminary studies on the aetiologicalagentrdquoThe Lancet vol 309 no 8011 pp 571ndash573 1977


[16] V E Volchkov A A Chepurnov V A Volchkova V ATernovoj and H-D Klenk ldquoMolecular characterization ofguinea pig-adapted variants of Ebola virusrdquo Virology vol 277no 1 pp 147ndash155 2000

[17] V A Volchkova A A Chepurnov H Feldmann V TernovojH-D Klenk and V E Volchkov ldquoGenetic variability of Ebolavirus during host adaptationrdquo in Proceedings of the 10th Inter-national Conference on Negative Strand Viruses p 207 DublinIreland September 1997

[18] M Mateo C Carbonnelle O Reynard et al ldquoVP24 Is amolecular determinant of Ebola virus virulence in guinea pigsrdquoThe Journal of Infectious Diseases vol 204 supplement 3 ppS1011ndashS1020 2011

[19] H Ebihara A Takada D Kobasa et al ldquoMolecular determi-nants of Ebola virus virulence in micerdquo PLoS Pathogens vol 2no 7 article e73 2006

[20] A Baskerville S P Fisher-Hoch G H Neild and A BDowsett ldquoUltrastructural pathology of experimental Ebolahaemorrhagic fever virus infectionrdquo The Journal of Pathologyvol 147 no 3 pp 199ndash209 1985

[21] E I Ryabchikova S G Baranova V K Tkachev and A AGrazhdantseva ldquoThemorphological changes in Ebola infectionin guinea pigsrdquo Voprosy Virusologii vol 38 no 4 pp 176ndash1791993 (Russian)

[22] J S Owen D J Brown D S Harry et al ldquoErythrocyteechinocytosis in liver disease Role of abnormal plasma highdensity lipoproteinsrdquo The Journal of Clinical Investigation vol76 no 6 pp 2275ndash2285 1985

[23] A A Dadaeva L P Sizikova andA A Chepurnov ldquoFunctionalactivity of peritonealmacrophages in experimental Ebola feverrdquoVestnik Rossiiskoi Akademii Meditsinskikh Nauk no 8 pp 7ndash112004 (Russian)

[24] F Villinger P Rollin S Brar et al ldquoMarkedly elevated levels ofinterferon (INF)-120574 INF-120572 interleukin (IL)-2 IL-10 and tumornecrosis factor-120572 associated with fatal Ebola virus infectionrdquoThe Journal of Infectious Diseases vol 179 pp 188ndash191 1999

[25] L E Hensley H A Young P B Jahrling and T W GeisbertldquoProinflammatory response during Ebola virus infection ofprimate models possible involvement of the tumor necrosisfactor receptor superfamilyrdquo Immunology Letters vol 80 no 3pp 169ndash179 2002

[26] H Feldmann H Bugany F Mahner H-D Klenk D Drenck-hahn and H-J Schnittler ldquoFilovirus-induced endothelial leak-age triggered by infected monocytesmacrophagesrdquo Journal ofVirology vol 70 no 4 pp 2208ndash2214 1996

[27] M Bray and S Mahanty ldquoEbola hemorrhagic fever and septicshockrdquo Journal of Infectious Diseases vol 188 no 11 pp 1613ndash1617 2003

[28] U Amara D Rittirsch M Flierl et al ldquoInteraction between thecoagulation and complement systemrdquo Advances in Experimen-tal Medicine and Biology vol 632 pp 71ndash79 2008

[29] S Kurosawa andD J Stearns-Kurosawa ldquoComplement throm-boticmicroangiopathy anddisseminated intravascular coagula-tionrdquo Journal of Intensive Care vol 2 article 61 2014

[30] N M Zubavichene and A A Chepurnov ldquoThe complementdynamics in experimental Ebola virus infectionsrdquo VoprosyVirusologii vol 2 pp 21ndash25 2004

[31] G M Ignatiev A A Dadaeva S V Luchko and A A Chep-urnov ldquoImmune and pathophysiological processes in baboonsexperimentally infected with Ebola virus adapted to guineapigsrdquo Immunology Letters vol 71 no 2 pp 131ndash140 2000

[32] N M Kudoyarova-Zubavichene N N Sergeyev A A Chep-urnov and S V Netesov ldquoPreparation and use of hyperimmuneserum for prophylaxis and therapy of Ebola virus infectionsrdquoJournal of Infectious Diseases vol 179 supplement 1 pp S218ndashS223 1999

Submit your manuscripts athttpwwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Anatomy Research International

PeptidesInternational Journal of


Hindawi Publishing Corporation httpwwwhindawicom

International Journal of

Volume 2014

Zoology


Molecular Biology International

GenomicsInternational Journal of


The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014


BioinformaticsAdvances in

Marine BiologyJournal of



Signal TransductionJournal of


BioMed Research International

Evolutionary BiologyInternational Journal of



Biochemistry Research International

ArchaeaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Genetics Research International


Advances in

Virolog y

Hindawi Publishing Corporationhttpwwwhindawicom

Nucleic AcidsJournal of

Volume 2014

Stem CellsInternational



Enzyme Research



Microbiology


acquired [4ndash7] The guinea pig-adapted EBOV is causing alethal infection with minor manifestations in the first 4-5days and a subsequent rapid development of a highly febrilecondition resulting in the animal death on days 8ndash11 Firstdetected in lymph node macrophages as early as 24 h pithe virus spreads to the spleen and liver on day 2 and to theother organs and tissues further on The virus spread canbe accompanied with a progressive rise of tissue virus titers(from 17 to 48ndash64 log

10PFUg in different tissues including

spleen liver adrenals lungs kidneys and pancreas) ondays 1ndash9 of the infection and the peak viremia in blood isreached on day 7 with sim105 PFUmL [7] However in twoof our adaptation experiments an only modest [8] or evenzero increase in virus titer [9] between the nonlethal andlethal adapted EBOV was occurring A prolongation of theprothrombin time (PT) and the partial thromboplastin time(aPTT) is observed in the infected animals [1]

While resembling the course of EBOV infection inprimates in many aspects the EBOV disease in rodents hassome important differences Fever and maculopapular rashwhich are the typical signs of infection in primates are bothnot present in mice infected with mouse-adapted Ebola virus(MA-EBOV) [10] In guinea pigs infected with guinea pig-adapted Ebola virus (GPA-EBOV) only fever but not therash is present [5 7] Unlike in mice and similarly to Syrianhamster lethal EBOV infection in guinea pigs induces seriouscoagulation abnormalities including the drop of platelets andan increase in coagulation time however fibrin depositionsand disseminated intravascular coagulation (DIC) are notreadily observed in these animals [2 7 11] Occurrence ofhemorrhages in EBOV disease in guinea pigs is still dis-putable some researchers report that death of animals is notaccompanied by the visible signs of hemorrhage [6] howeverwe regularly observed typical hemorrhagic foci in the liverof at least part of the animals infected with GPA-EBOVin our experiments [12] Despite the severe lymphopenia inthe course of GPA-EBOV infection lymphocyte bystanderapoptosis which is an important feature of infection inprimates and mice is not generally observed in guinea pigs[7]

Here we present a comprehensive overview of the viro-logic pathomorphologic and hematologic data obtained inthe EBOV adaptation experiments in guinea pigs (includingfour passage series performed in our laboratory) and discussthe value of those findings obtained in the rodent models forthe general understanding of pathology of EBOV infection

2 Establishment of Guinea Pig-AdaptedLethal Strains of Ebola Virus and theStudy of Their Virulence and VirologicCharacteristics

Serial passage of different viruses in animals or cell culturescan result in the selection of mutants with different repro-duction characteristics and either higher or lower virulenceas compared to the original wild-type virus [13 14] Adultguinea pigs inoculated with the wild-type EBOV isolatedfrom primates typically develop a nonlethal mild febrile or

even asymptomatic disease However it has been previouslyshown that young guinea pigs may occasionally develop alethal EBOV infection and the serial passage of EBOV causesa substantial increase in lethality in these animals [4 15]In this review paper we summarize our own data on theestablishment of four independent adaptation series in guineapigs using the original stocks of EBOV subtype Zaire strainMayinga differing in their passage history [5 8 9 11] andcompare them with those from other labs where the EBOVadaptation series were successfully performed [6 7 10]

The original EBOV strain Mayinga stock of the firstadaptation series (8mc) was passaged 24 times in Vero E6cell line prior to the guinea pig adaptation experiments [5]while the stock of the second series (K-5) was passagedtwice in NHPs (green monkeys) followed by 15 passagesin human lung L68 cells [9] In the third series (GLA)the wild-type EBOV strain Mayinga was passaged twicein NHPs and further multiplied in Vero cell culture [11]The fourth series (GPA-P7) was started with an individualclone of strain Mayinga EBOV which was obtained by triple-cloningpassaging in Vero cells and was confirmed to causeno lethal disease prior to its adaptation to guinea pigs Thisvirus clone further used for the inoculation of animals wasdesignated as the precursor of guinea pig-adapted EBOV(pre-GPA clone) [8]

At first passage of 8mc adaptation series [5] a groupof animals (we used the groups of 6 animals in all ouradaptation experiments) was inoculated with virus derivedfrom a liver homogenate of a rhesusmacaque lethally infectedby Mayinga strain of EBOV and passaged 24 times in VeroE6 cells In some animals a nonlethal febrile conditiondeveloped with the body temperature rising up to 40∘CThe liver homogenates of the animals which demonstratedhigh-grade fevers and higher levels of liver virus titer wereused for the next passage inoculation With each passagethe febrile temperature was increasing (from 398ndash400∘Cat earlier passages to 415ndash420∘C at later passages) aswas the liver virus titer which showed the overall growthfrom 29 plusmn 04 log

10PFUmL to 55 plusmn 07 log

10PFUmL (as

determined by plaque forming assay in Vero cells) The firstlethal outcomes were detected at passage 3 and the lethalityand infectivity of the adapting virus isolate were graduallyincreasing until passage 8 In guinea pigs inoculated with thevirus isolates of late passage 8 the fever occurred at days 4ndash7 and represented a double-peaked pattern with the secondpeak at days 11ndash13 Fever episodes were associated with themaximum levels of tissue virus titersThe adapted virus strainwas designated EBOV-8mc and several clones were derivedfrom it and proved to have similar features as the nonclonedEBOV-8mc virus itself [5]

As mentioned above the original EBOV stock of K-5series was passaged twice in NHPs and further 15 times inL68 cell culture prior to its adaptation to guinea pig and itsinfectivity and virulence were evolving in the passage seriesThe biological titer of the virus harvest in PFU decreasedfrom the first to third passage and then began to increasereaching its plateau by passage 12 and further remainingstable However the virus infectivity in guinea pigs wasunchanged throughout the L68 passages Following the cell






10PFUmL) could



10PFUmL (in peri-
















K-5[9]

GLA[11]

GPA-P7[8] [6] [7] D-5

[9] [10]



ZEBOVMayingaZEBOV

MayingaZEBOV

MayingaZEBOV

MayingaZEBOV


NHPs +L68 cellline

passages




mdash mdash mdash




2955Liver

3535Plasma



6370 2950

2152Plasma


95117 4278


stock stock


without death)




stockstock


Clone 5Clone 4


Clone 2 Clone 3

Clone pre-GPA




Variant Ch-15

Clone 1






2 passages in NHP


Vero cell line

Strain 8mc























ble2Geneticchangeso

ccurrin

gin

different

adaptatio

nserie

sand

reverseg

eneticse

xperim

entsandtheirrele

vancetothed

evelo

pmento

fGPA

-EBO

Vvirulencelethality

ViralO

RFNucleotidep

osition

Aminoacid

substitution


nCom

ment

EBOV

-8mca

daptationserie

s

NP

1852

Non

eNo

Syno

nymou

s2410

Non

eNo

Syno

nymou

s2411

Pherarr

Leu(64

8)Unlikely

Con

servativemay

play

asup

portiver

oleinadaptatio

n

GP

6924

(insertion)

Fram

eshift

No

Clon

eswith

orwith

outtho

seGPmutations

dono

tdiffer

from

each

otherintheir

pathogenicity

inguinea

pigs

7228

Asprarr

Gly(397)

No

VP3

09595

Non

eNo

Locatedin


region

VP2

410557

Metrarr

Ile(71)

Prob

ably

Atleaston

eoftho

sethreeV

P24mutations

indu

cesm

ajor

structural

rearrangem

entsin

EBOV-8m

cVP2

4which

arep

resented

asgelm

obilityshift

inele

ctroph

oresis

10784

Leurarr

Pro(147)

Prob

ably

10904

Thrrarr

Ile(187)

Prob

ably

L14038

Thrrarr

Ala(820)

Unlikely

Con

servativemay

play

asup

portiver

oleinadaptatio

nEB

OV-5Kadaptatio

nserie

s

VP2

410838

Asprarr

Gly(165)

No

Isacqu

iredin

thec

ourseo

fpassage

inNHPs

orL6

8cellsprio

rtoadaptatio

n10907

Hisrarr

Tyr(186)

Prob

ably

Acqu

ireddu

ringadaptatio

nliesc

lose

toEB

OV-8m

cVP2

4Th

rrarrIle

(187)

EBOV

-GPA

-P7serie

s

NP

1781

Asnrarr

His(438)

No

Isfoun

din

nonadapted

pre-GPA

clone

2043

Argrarr

Lys(525)

No

Isfoun

din

nonadapted

pre-GPA

clone

2092

Non

eNo

Syno

nymou

s2100

Leurarr

Arg(544

)No

Occurredatpassage1

after

pre-GPA

beforethe

acqu

isitio

nof

lethality

2164

Non

eNo

Syno

nymou

s

2166

Asnrarr

Ser(566)

Prob

ably

Occurredatpassage3

whenthefi

rstlethaloutcomew


ssiteu

ntilfixed

atpassage5

con

comitant

with

40lethality

2188

Non

eNo

Syno

nymou

s2191

Non

eNo

Syno

nymou

s2209

Non

eNo

Syno

nymou

s2222

Non

eNo

Syno

nymou

s2233

Non

eNo

Syno

nymou

s2260

Non

eNo

Syno

nymou

s2261

Serrarr

Pro(598)

No

Occurredatpassage1

after

pre-GPA

beforethe

acqu

isitio

nof

lethality

2409

Serrarr

Phe(647)

No

Isfoun

din

nonadapted

pre-GPA

clone

VP4

02456

Asprarr

Asn(663)

No

Occurredatpassage1

after

pre-GPA

beforethe

acqu

isitio

nof

lethality

5576

Non

eNo

Syno

nymou

s

VP2

4

5597

Non

eNo

Syno

nymou

s10557

Metrarr

Ile(71)

Prob

ably

Occurredatpassage6

con

comitant

with

lethality

increase

to80

10784

Leurarr

Pro(147)

Prob

ably

Occurredatpassage3

whenthefi

rstlethaloutcomew


ssiteu

ntilfixed

atpassage5

con

comitant

with

40lethality

10805

Argrarr

Leu(154)

Prob

ably


comitant

with

fulllethality

L12286

Valrarr

Ile(236)

Prob

ably


comitant

with

fulllethality

16821

Non

eNo

Syno

nymou

sRe

verseg

enetics

ofEB

OV-8mcm

utations

VP2

410422

Leurarr

Phe(26)

Definitely

Derived

from

theb

lood

ofan

anim

alw

hich

died

after

inoculationwith

anon

lethal

recombinant

rEBO

V-NP8m

cF6

48LvirusWhenintro

ducedinto

anotherw

isewild

-type

Zaire

EBOVgeno

meb

yreverseg


econ

fersfull

lethality

toguinea

pigs


MP

Vpt



Vpt

VptVpl








H

H

Er

T



H

Vpt

MPVpt

Vpt


H



Vpt

Vpt

Vpt

CC


VptVpt

Vpt

Vpt


Vpl

Vpl


VptVpl

Vpl





























0

20

40

60

80

100

120

140

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

Com

plem

ent a

ctiv

ity (u

nits)

Days pi

Mayinga



8mc







6 Conclusion












Acknowledgment


References









































Volume 2014

Zoology






















Advances in

Virolog y



Volume 2014




Enzyme Research



Microbiology






10PFUmL) could



10PFUmL (in peri-
















K-5[9]

GLA[11]

GPA-P7[8] [6] [7] D-5

[9] [10]



ZEBOVMayingaZEBOV

MayingaZEBOV

MayingaZEBOV

MayingaZEBOV


NHPs +L68 cellline

passages




mdash mdash mdash




2955Liver

3535Plasma



6370 2950

2152Plasma


95117 4278


stock stock


without death)




stockstock


Clone 5Clone 4


Clone 2 Clone 3

Clone pre-GPA




Variant Ch-15

Clone 1






2 passages in NHP


Vero cell line

Strain 8mc























ble2Geneticchangeso

ccurrin

gin

different

adaptatio

nserie

sand

reverseg

eneticse

xperim

entsandtheirrele

vancetothed

evelo

pmento

fGPA

-EBO

Vvirulencelethality

ViralO

RFNucleotidep

osition

Aminoacid

substitution


nCom

ment

EBOV

-8mca

daptationserie

s

NP

1852

Non

eNo

Syno

nymou

s2410

Non

eNo

Syno

nymou

s2411

Pherarr

Leu(64

8)Unlikely

Con

servativemay

play

asup

portiver

oleinadaptatio

n

GP

6924

(insertion)

Fram

eshift

No

Clon

eswith

orwith

outtho

seGPmutations

dono

tdiffer

from

each

otherintheir

pathogenicity

inguinea

pigs

7228

Asprarr

Gly(397)

No

VP3

09595

Non

eNo

Locatedin


region

VP2

410557

Metrarr

Ile(71)

Prob

ably

Atleaston

eoftho

sethreeV

P24mutations

indu

cesm

ajor

structural

rearrangem

entsin

EBOV-8m

cVP2

4which

arep

resented

asgelm

obilityshift

inele

ctroph

oresis

10784

Leurarr

Pro(147)

Prob

ably

10904

Thrrarr

Ile(187)

Prob

ably

L14038

Thrrarr

Ala(820)

Unlikely

Con

servativemay

play

asup

portiver

oleinadaptatio

nEB

OV-5Kadaptatio

nserie

s

VP2

410838

Asprarr

Gly(165)

No

Isacqu

iredin

thec

ourseo

fpassage

inNHPs

orL6

8cellsprio

rtoadaptatio

n10907

Hisrarr

Tyr(186)

Prob

ably

Acqu

ireddu

ringadaptatio

nliesc

lose

toEB

OV-8m

cVP2

4Th

rrarrIle

(187)

EBOV

-GPA

-P7serie

s

NP

1781

Asnrarr

His(438)

No

Isfoun

din

nonadapted

pre-GPA

clone

2043

Argrarr

Lys(525)

No

Isfoun

din

nonadapted

pre-GPA

clone

2092

Non

eNo

Syno

nymou

s2100

Leurarr

Arg(544

)No

Occurredatpassage1

after

pre-GPA

beforethe

acqu

isitio

nof

lethality

2164

Non

eNo

Syno

nymou

s

2166

Asnrarr

Ser(566)

Prob

ably

Occurredatpassage3

whenthefi

rstlethaloutcomew


ssiteu

ntilfixed

atpassage5

con

comitant

with

40lethality

2188

Non

eNo

Syno

nymou

s2191

Non

eNo

Syno

nymou

s2209

Non

eNo

Syno

nymou

s2222

Non

eNo

Syno

nymou

s2233

Non

eNo

Syno

nymou

s2260

Non

eNo

Syno

nymou

s2261

Serrarr

Pro(598)

No

Occurredatpassage1

after

pre-GPA

beforethe

acqu

isitio

nof

lethality

2409

Serrarr

Phe(647)

No

Isfoun

din

nonadapted

pre-GPA

clone

VP4

02456

Asprarr

Asn(663)

No

Occurredatpassage1

after

pre-GPA

beforethe

acqu

isitio

nof

lethality

5576

Non

eNo

Syno

nymou

s

VP2

4

5597

Non

eNo

Syno

nymou

s10557

Metrarr

Ile(71)

Prob

ably

Occurredatpassage6

con

comitant

with

lethality

increase

to80

10784

Leurarr

Pro(147)

Prob

ably

Occurredatpassage3

whenthefi

rstlethaloutcomew


ssiteu

ntilfixed

atpassage5

con

comitant

with

40lethality

10805

Argrarr

Leu(154)

Prob

ably


comitant

with

fulllethality

L12286

Valrarr

Ile(236)

Prob

ably


comitant

with

fulllethality

16821

Non

eNo

Syno

nymou

sRe

verseg

enetics

ofEB

OV-8mcm

utations

VP2

410422

Leurarr

Phe(26)

Definitely

Derived

from

theb

lood

ofan

anim

alw

hich

died

after

inoculationwith

anon

lethal

recombinant

rEBO

V-NP8m

cF6

48LvirusWhenintro

ducedinto

anotherw

isewild

-type

Zaire

EBOVgeno

meb

yreverseg


econ

fersfull

lethality

toguinea

pigs


MP

Vpt



Vpt

VptVpl








H

H

Er

T



H

Vpt

MPVpt

Vpt


H



Vpt

Vpt

Vpt

CC


VptVpt

Vpt

Vpt


Vpl

Vpl


VptVpl

Vpl





























0

20

40

60

80

100

120

140

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

Com

plem

ent a

ctiv

ity (u

nits)

Days pi

Mayinga



8mc







6 Conclusion












Acknowledgment


References









































Volume 2014

Zoology






















Advances in

Virolog y



Volume 2014




Enzyme Research



Microbiology




K-5[9]

GLA[11]

GPA-P7[8] [6] [7] D-5

[9] [10]



ZEBOVMayingaZEBOV

MayingaZEBOV

MayingaZEBOV

MayingaZEBOV


NHPs +L68 cellline

passages




mdash mdash mdash




2955Liver

3535Plasma



6370 2950

2152Plasma


95117 4278


stock stock


without death)




stockstock


Clone 5Clone 4


Clone 2 Clone 3

Clone pre-GPA




Variant Ch-15

Clone 1






2 passages in NHP


Vero cell line

Strain 8mc























ble2Geneticchangeso

ccurrin

gin

different

adaptatio

nserie

sand

reverseg

eneticse

xperim

entsandtheirrele

vancetothed

evelo

pmento

fGPA

-EBO

Vvirulencelethality

ViralO

RFNucleotidep

osition

Aminoacid

substitution


nCom

ment

EBOV

-8mca

daptationserie

s

NP

1852

Non

eNo

Syno

nymou

s2410

Non

eNo

Syno

nymou

s2411

Pherarr

Leu(64

8)Unlikely

Con

servativemay

play

asup

portiver

oleinadaptatio

n

GP

6924

(insertion)

Fram

eshift

No

Clon

eswith

orwith

outtho

seGPmutations

dono

tdiffer

from

each

otherintheir

pathogenicity

inguinea

pigs

7228

Asprarr

Gly(397)

No

VP3

09595

Non

eNo

Locatedin


region

VP2

410557

Metrarr

Ile(71)

Prob

ably

Atleaston

eoftho

sethreeV

P24mutations

indu

cesm

ajor

structural

rearrangem

entsin

EBOV-8m

cVP2

4which

arep

resented

asgelm

obilityshift

inele

ctroph

oresis

10784

Leurarr

Pro(147)

Prob

ably

10904

Thrrarr

Ile(187)

Prob

ably

L14038

Thrrarr

Ala(820)

Unlikely

Con

servativemay

play

asup

portiver

oleinadaptatio

nEB

OV-5Kadaptatio

nserie

s

VP2

410838

Asprarr

Gly(165)

No

Isacqu

iredin

thec

ourseo

fpassage

inNHPs

orL6

8cellsprio

rtoadaptatio

n10907

Hisrarr

Tyr(186)

Prob

ably

Acqu

ireddu

ringadaptatio

nliesc

lose

toEB

OV-8m

cVP2

4Th

rrarrIle

(187)

EBOV

-GPA

-P7serie

s

NP

1781

Asnrarr

His(438)

No

Isfoun

din

nonadapted

pre-GPA

clone

2043

Argrarr

Lys(525)

No

Isfoun

din

nonadapted

pre-GPA

clone

2092

Non

eNo

Syno

nymou

s2100

Leurarr

Arg(544

)No

Occurredatpassage1

after

pre-GPA

beforethe

acqu

isitio

nof

lethality

2164

Non

eNo

Syno

nymou

s

2166

Asnrarr

Ser(566)

Prob

ably

Occurredatpassage3

whenthefi

rstlethaloutcomew


ssiteu

ntilfixed

atpassage5

con

comitant

with

40lethality

2188

Non

eNo

Syno

nymou

s2191

Non

eNo

Syno

nymou

s2209

Non

eNo

Syno

nymou

s2222

Non

eNo

Syno

nymou

s2233

Non

eNo

Syno

nymou

s2260

Non

eNo

Syno

nymou

s2261

Serrarr

Pro(598)

No

Occurredatpassage1

after

pre-GPA

beforethe

acqu

isitio

nof

lethality

2409

Serrarr

Phe(647)

No

Isfoun

din

nonadapted

pre-GPA

clone

VP4

02456

Asprarr

Asn(663)

No

Occurredatpassage1

after

pre-GPA

beforethe

acqu

isitio

nof

lethality

5576

Non

eNo

Syno

nymou

s

VP2

4

5597

Non

eNo

Syno

nymou

s10557

Metrarr

Ile(71)

Prob

ably

Occurredatpassage6

con

comitant

with

lethality

increase

to80

10784

Leurarr

Pro(147)

Prob

ably

Occurredatpassage3

whenthefi

rstlethaloutcomew


ssiteu

ntilfixed

atpassage5

con

comitant

with

40lethality

10805

Argrarr

Leu(154)

Prob

ably


comitant

with

fulllethality

L12286

Valrarr

Ile(236)

Prob

ably


comitant

with

fulllethality

16821

Non

eNo

Syno

nymou

sRe

verseg

enetics

ofEB

OV-8mcm

utations

VP2

410422

Leurarr

Phe(26)

Definitely

Derived

from

theb

lood

ofan

anim

alw

hich

died

after

inoculationwith

anon

lethal

recombinant

rEBO

V-NP8m

cF6

48LvirusWhenintro

ducedinto

anotherw

isewild

-type

Zaire

EBOVgeno

meb

yreverseg


econ

fersfull

lethality

toguinea

pigs


MP

Vpt



Vpt

VptVpl








H

H

Er

T



H

Vpt

MPVpt

Vpt


H



Vpt

Vpt

Vpt

CC


VptVpt

Vpt

Vpt


Vpl

Vpl


VptVpl

Vpl





























0

20

40

60

80

100

120

140

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

Com

plem

ent a

ctiv

ity (u

nits)

Days pi

Mayinga



8mc







6 Conclusion












Acknowledgment


References









































Volume 2014

Zoology






















Advances in

Virolog y



Volume 2014




Enzyme Research



Microbiology






















ble2Geneticchangeso

ccurrin

gin

different

adaptatio

nserie

sand

reverseg

eneticse

xperim

entsandtheirrele

vancetothed

evelo

pmento

fGPA

-EBO

Vvirulencelethality

ViralO

RFNucleotidep

osition

Aminoacid

substitution


nCom

ment

EBOV

-8mca

daptationserie

s

NP

1852

Non

eNo

Syno

nymou

s2410

Non

eNo

Syno

nymou

s2411

Pherarr

Leu(64

8)Unlikely

Con

servativemay

play

asup

portiver

oleinadaptatio

n

GP

6924

(insertion)

Fram

eshift

No

Clon

eswith

orwith

outtho

seGPmutations

dono

tdiffer

from

each

otherintheir

pathogenicity

inguinea

pigs

7228

Asprarr

Gly(397)

No

VP3

09595

Non

eNo

Locatedin


region

VP2

410557

Metrarr

Ile(71)

Prob

ably

Atleaston

eoftho

sethreeV

P24mutations

indu

cesm

ajor

structural

rearrangem

entsin

EBOV-8m

cVP2

4which

arep

resented

asgelm

obilityshift

inele

ctroph

oresis

10784

Leurarr

Pro(147)

Prob

ably

10904

Thrrarr

Ile(187)

Prob

ably

L14038

Thrrarr

Ala(820)

Unlikely

Con

servativemay

play

asup

portiver

oleinadaptatio

nEB

OV-5Kadaptatio

nserie

s

VP2

410838

Asprarr

Gly(165)

No

Isacqu

iredin

thec

ourseo

fpassage

inNHPs

orL6

8cellsprio

rtoadaptatio

n10907

Hisrarr

Tyr(186)

Prob

ably

Acqu

ireddu

ringadaptatio

nliesc

lose

toEB

OV-8m

cVP2

4Th

rrarrIle

(187)

EBOV

-GPA

-P7serie

s

NP

1781

Asnrarr

His(438)

No

Isfoun

din

nonadapted

pre-GPA

clone

2043

Argrarr

Lys(525)

No

Isfoun

din

nonadapted

pre-GPA

clone

2092

Non

eNo

Syno

nymou

s2100

Leurarr

Arg(544

)No

Occurredatpassage1

after

pre-GPA

beforethe

acqu

isitio

nof

lethality

2164

Non

eNo

Syno

nymou

s

2166

Asnrarr

Ser(566)

Prob

ably

Occurredatpassage3

whenthefi

rstlethaloutcomew


ssiteu

ntilfixed

atpassage5

con

comitant

with

40lethality

2188

Non

eNo

Syno

nymou

s2191

Non

eNo

Syno

nymou

s2209

Non

eNo

Syno

nymou

s2222

Non

eNo

Syno

nymou

s2233

Non

eNo

Syno

nymou

s2260

Non

eNo

Syno

nymou

s2261

Serrarr

Pro(598)

No

Occurredatpassage1

after

pre-GPA

beforethe

acqu

isitio

nof

lethality

2409

Serrarr

Phe(647)

No

Isfoun

din

nonadapted

pre-GPA

clone

VP4

02456

Asprarr

Asn(663)

No

Occurredatpassage1

after

pre-GPA

beforethe

acqu

isitio

nof

lethality

5576

Non

eNo

Syno

nymou

s

VP2

4

5597

Non

eNo

Syno

nymou

s10557

Metrarr

Ile(71)

Prob

ably

Occurredatpassage6

con

comitant

with

lethality

increase

to80

10784

Leurarr

Pro(147)

Prob

ably

Occurredatpassage3

whenthefi

rstlethaloutcomew


ssiteu

ntilfixed

atpassage5

con

comitant

with

40lethality

10805

Argrarr

Leu(154)

Prob

ably


comitant

with

fulllethality

L12286

Valrarr

Ile(236)

Prob

ably


comitant

with

fulllethality

16821

Non

eNo

Syno

nymou

sRe

verseg

enetics

ofEB

OV-8mcm

utations

VP2

410422

Leurarr

Phe(26)

Definitely

Derived

from

theb

lood

ofan

anim

alw

hich

died

after

inoculationwith

anon

lethal

recombinant

rEBO

V-NP8m

cF6

48LvirusWhenintro

ducedinto

anotherw

isewild

-type

Zaire

EBOVgeno

meb

yreverseg


econ

fersfull

lethality

toguinea

pigs


MP

Vpt



Vpt

VptVpl








H

H

Er

T



H

Vpt

MPVpt

Vpt


H



Vpt

Vpt

Vpt

CC


VptVpt

Vpt

Vpt


Vpl

Vpl


VptVpl

Vpl





























0

20

40

60

80

100

120

140

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

Com

plem

ent a

ctiv

ity (u

nits)

Days pi

Mayinga



8mc







6 Conclusion












Acknowledgment


References









































Volume 2014

Zoology






















Advances in

Virolog y



Volume 2014




Enzyme Research



Microbiology


MP

Vpt



Vpt

VptVpl








H

H

Er

T



H

Vpt

MPVpt

Vpt


H



Vpt

Vpt

Vpt

CC


VptVpt

Vpt

Vpt


Vpl

Vpl


VptVpl

Vpl





























0

20

40

60

80

100

120

140

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

Com

plem

ent a

ctiv

ity (u

nits)

Days pi

Mayinga



8mc







6 Conclusion












Acknowledgment


References









































Volume 2014

Zoology






















Advances in

Virolog y



Volume 2014




Enzyme Research



Microbiology


H

H

Er

T



H

Vpt

MPVpt

Vpt


H



Vpt

Vpt

Vpt

CC


VptVpt

Vpt

Vpt


Vpl

Vpl


VptVpl

Vpl





























0

20

40

60

80

100

120

140

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

Com

plem

ent a

ctiv

ity (u

nits)

Days pi

Mayinga



8mc







6 Conclusion












Acknowledgment


References









































Volume 2014

Zoology






















Advances in

Virolog y



Volume 2014




Enzyme Research



Microbiology

























0

20

40

60

80

100

120

140

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

Com

plem

ent a

ctiv

ity (u

nits)

Days pi

Mayinga



8mc







6 Conclusion












Acknowledgment


References









































Volume 2014

Zoology






















Advances in

Virolog y



Volume 2014




Enzyme Research



Microbiology











Acknowledgment


References









































Volume 2014

Zoology






















Advances in

Virolog y



Volume 2014




Enzyme Research



Microbiology


























Volume 2014

Zoology






















Advances in

Virolog y



Volume 2014




Enzyme Research



Microbiology

review article adapted lethality: what we can learn...

Documents