Human coronavirus reinfection dynamics: lessons for SARS-CoV-2

Abstract

In the current SARS-CoV-2 pandemic a key unsolved question is the quality and duration of acquired immunity in recovered individuals. This is crucial to solve, however SARS-CoV-2 has circulated for under five months, precluding a direct study. We therefore monitored 10 subjects over a time span of 35 years (1985-2020), providing a total of 2473 follow up person-months, and determined a) their antibody levels following infection by any of the four seasonal human coronaviruses, and b) the time period after which reinfections by the same virus can occur. An alarmingly short duration of protective immunity to coronaviruses was found by both analyses. We saw frequent reinfections at 12 months post-infection and a substantial reduction in antibody levels as soon as 6 months post-infection.

Full text


1

Humancoronavirusreinfectiondynamics:lessonsforSARS‐CoV‐2

Authors
ArthurW. D.Edridge1,JoannaKaczorowska1, AlexisC.R.Hoste2,Margreet Bakker1,Michelle Klein1,
Maarten F. Jebbink1, Amy Matser3, Cormac M. Kinsella1, Paloma Rueda2, Maria Prins3,4, Patricia
Sastre2,MartinDeijs1,LiavanderHoek1,*

Affiliations
1. Laboratory of Experimental Virology, Department of Medical Microbiology and Infection
Prevention,AmsterdamInfection&ImmunityInstitute,AmsterdamUMC,UniversityofAmsterdam,
Meibergdreef15,1105AZAmsterdam,TheNetherlands.
2.INGENASA,InmunologíayGenéticaAplicadaS.A.,Av.delaInstitución Libre de Enseñanza,39, 
28037Madrid,Spain
3.DepartmentofInfectiousDiseases,PublicHealthServiceofAmsterdam,NieuweAchtergracht100,
1018WTAmsterdamTheNetherlands.
4. Amsterdam UMC, University of Amsterdam, Department of Infectious Diseases, Amsterdam
Infection& ImmunityInstitute,AmsterdamUMC,University ofAmsterdam,Meibergdreef15,1105
AZAmsterdam,TheNetherlands.

*Correspondingauthor


ABSTRACT

InthecurrentSARS‐CoV‐2pandemicakeyunsolvedquestionisthequalityanddurationofacquired
immunity in recovered individuals. This is crucial to solve, however SARS‐CoV‐2 has circulated for
underfivemonths,precludingadirectstudy.Wethereforemonitored10subjectsoveratimespanof
35years(1985‐2020), providinga total of2473follow upperson‐months,and determineda)their
antibodylevelsfollowinginfectionbyanyofthefourseasonalhumancoronaviruses,andb)thetime
periodafterwhichreinfectionsbythesameviruscanoccur.Analarminglyshortdurationofprotective
immunitytocoronaviruseswasfoundbybothanalyses.Wesawfrequentreinfectionsat12months
post‐infectionandsubstantialreductioninantibodylevelsassoonas6monthspost‐infection.

ONESENTENCESUMMARY:Coronavirusprotectiveimmunityisshort‐lasting



. CC-BY 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted May 18, 2020. .https://doi.org/10.1101/2020.05.11.20086439doi: medRxiv preprint


2

MAINTEXT

SARS‐CoV‐2isanovelcoronavirusresponsibleforanongoingpandemic.Itsrapidtransmissionismost
probably caused bythefactthatthe virus entered a grossly naive, thus highly susceptible, human
population, combinedwiththecapacity ofthevirusto transmit duringtheasymptomaticphaseof
infection.Sincenopharmaceuticalinterventionsareuniversallyavailableorapplied,currentpolicies
tolimitthespreadofSARS‐CoV‐2revolvearoundcontainment,socialdistancing,andtheassumption
thatrecovered patientsdevelopprotectiveimmunity.Itisstillunclearwhetherprotectiveimmunity
isindeed inducedafterinfection, andforhowlong.Thedurationofprotectionwillimpactnotonly
theoverallcourseofthecurrentpandemic,butalsothepost‐pandemicperiod.Todate,noconcrete
evidenceofreinfectionbySARS‐CoV‐2isavailable,norisanyexampleofreinfectionbySARS‐CoV‐1or
MERS‐CoV,yetthisislikelyinfluencedbytherecentemergenceofSARS‐CoV‐2andthelimitedscale
of the SARS‐CoV‐1 and MERS‐CoV epidemics. The limited available data on potential protective
immunityagainstcoronavirusreinfectionisderivedfromoneexperimentalinfectionstudyin
volunteers(1),fromwhichitisgenerallyassumedthatreinfectionbycoronavirusescanoccur.Natural
coronavirusinfectionsandsusceptibilitytoreinfectionhavenotbeen investigatedthusfar. If they
occur,reinfectionswillprobablybedictatedbytwovariables:exposuretothevirus,andthequality
ofsustainedimmunity(2).
Even though it is not possible to investigate SARS‐CoV‐2 reinfections yet, the seasonal
coronavirusesmayserveasareliablemodel.Therearefourspeciesofseasonalcoronaviruses,HCoV‐
NL63,HCoV‐229E,HCoV‐OC43,andHCoV‐HKU1.Allareassociatedwithmostlymildrespiratorytract
infections. However, aside from being etiological agents of common cold, the four viruses are
biologically dissimilar. Two belong to the genus Alphacoronavirus, and two to the genus
Betacoronavirus.The virusesuse characteristicreceptormoleculestoentera targetcell,andbased
onreceptordistributiontheydonotallenterthesameepithelialcelltypeinthelungs(3).Giventhis
variability,theseasonalcoronavirusesarethemostrepresentativeviru sgroupfr omwhichtoco nclude
generalcoronaviruscharacteristics,particularlycommondenominatorslikehostprotectiveimmunity
and susceptibility to reinfection. Since most people experience their first seasonal coronavirus
infectionduringearlychildhood(4–6),reinfectionslaterinlifecanbeinvestigated.
Theaimofthisstudywastoinvestigatethetimeperiodbetweencoronavirusreinfectionsand
thedynamicsofantibodydeclinefollowinginfection.Theseparameterswereassessedbymeasuring
theimmuneresponsetoeachindividualseasonalcoronavirus,overanextendedperiod.Wefollowed
healthysubjectsformultipledecadesatregularintervals.AsIgGlevelsonlyincreaseaftersuccessful
infection and—importantly—remain constant after unsuccessful viral challenge (1), increased
antibodylevelsidentifyaninfection.Weusedthecarboxylterminalofthenucleocapsidprotein(NCt)
as the target antigen in ELISA, as this protein is immunogenic, relatively low in interspecies
conservation,yethighinintraspeciesconservationwhencomparedtotheSpikeprotein(4,7–10).

Frequencyofinfectionsandreinfections.
Fromaprospectivecohortstudyfollowingadultmales(seeM&Mand (11)), ten subjects were
randoml yselected.Foll ow‐upof subject sstartedin1985 and,besidesagapinfollowupbetween1997
and2003,continueduntil2020atregularintervals(every3monthspriorto1989,andevery6months
afterwards,Table1).Thecumulativeperiodatwhichsubjectswerecontinuouslyfollowedtotaled
morethan200person‐years(2473months).Atstartofthestudy,subjectagerangedfrom27to40
years; by the end of follow‐up , subjects were 49 to 66 years old.Thestudywasapprovedbythe
MedicalEthicsCommitteeoftheAmsterdamUniversityMedicalCenter of the University of
Amsterdam,theNetherlands(MEC07/182).
. CC-BY 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted May 18, 2020. .https://doi.org/10.1101/2020.05.11.20086439doi: medRxiv preprint


3

Infections by seasonal coronaviruses were defined as 1.4‐fold increase in antibody levels
occurringbetweentwovisits(1),andadditionallythestandard deviationsofthemeanatbothvisits
notoverlapping.Atotalof132events,rangingfrom3to22persubject,metthesecriteriaandwere
thusconsideredinfections(Table1).Theincidenceratesofseasonalcoronavirusreinfectionsper100
person‐yearsof follow‐upwas15.3 (95%CI10.0‐23.4);20.1(95% CI14.2‐28.7);16.4(95%CI 10.4‐
26.1);6.3(95%CI2.4‐16.8),forHCoV‐NL63,HCoV‐229E,HCoV‐OC43andHCoV‐HKU1respectively.To
test whether serological infection criteria represented symptomatic infections , we compared self‐
reported influenza like illnesses in the interval directly preceding the antibody peaks. Indeed,
reportingoffeverandcoughweresignificantlyassociated(Fisher’sexacttestp=0.031,supplementary
tableS1).

Table1.Studysubjectsandseasonalcoronavirusinfectionsduringfollowup.
Subject Year Age
Continuous
follow‐up
period
CoronavirusInfections
Start End Start End Months Years Total* NL63 229E HKU1 OC43
1 1985 2017 32 64 265 22.1 14 4 3 1 6
2 1985 2019 30 64 310 25.9 15 5 6 3 1
3 1985 2020 29 64 340 28.3 6 2 3 0 1
4 1985 2010 33 59 230 19.2 22 3 12 2 5
5 1985 2010 27 53 232 19.3 9 5 3 1 0
6 1985 1997 37 49 144 12.0 3 1 1 0 1
7 1985 2003 32 49 138 11.5 13 3 5 1 4
8 1986 2014 34 62 256 21.3 10 3 5 0 2
9 1985 2010 40 75 342 28.6 22 7 5 3 7
10 1985 2011 35 60 233 19.4 18 4 6 2 6
Total  2473 205.6 132 37 49 13 33


Medianreinfectiontimesof33(IQR18–60),31(IQR15–42),27(IQR21–49),and46(IQR
36–68)monthswerefoundforHCoV‐NL63,HCoV‐229E,HCoV‐OC43andHCoV‐HKU1,respectively,
and30(IQR18–54)monthsforallvirusescombined(Fig.1A).Therewasnostatisticallysignificant
differencebetweentheinfectionintervallengthsoftheindividualviruses(Kruskal‐Wallistest,P=0.74).
Inafewcases,re‐infectionsoccurredasearlyas6months(twotimesforHCoV‐229Eandonetimefor
HCoV‐OC43)and9months(twotimesforHCoV‐NL63).Themostfrequentobservedreinfectiontime
was 12 months. For reinfections occurring as early as 6 months, we observed no reduction in
antibodies between infections (Fig. 1A, white circles). At longer infection intervals, intermediate
reductionsinantibodylevelswereobserved.
Theability todetect short‐termreinfectionsislimitedbythesamplinginterval.Importantly
though,noreinfectionwasobservedatthefirstsubsequentfollow‐upvisitaftera 3monthinterval
(Fig.1A).Wedidobserveseveralreinfectionsatsubsequentvisitswitha6‐monthinterval,suggesting
that reinfections within 6 month s do not occur. To further support this point, weexamined ratios
betweenantibodylevelsbetweentwovisits,bothfor3and6monthfollow‐ups.AsshowninFig.1B,
only ratios below 1 were found in the observations with every 3month‐sampling,andwecan
thereforesafelyconcludethatreinfectionsoccurfrom6monthson,yetmostprobablynotearlier.

. CC-BY 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted May 18, 2020. .https://doi.org/10.1101/2020.05.11.20086439doi: medRxiv preprint


4


Figure1Infectionandreinfectioncharacteristics,andwaningimmunity for seasonal coronaviruses.(A) The
intervaldurationsbetweenreinfections.Onlyreinfectionswhich were observed within a continuous follow‐u p
periodareshown.Whitedotsindicatereinfectionsforwhichnointermediatedecreaseinantibodylevelscouldbe
observed. Black vertical lines describe median reinfection times. (B) Changes in antib ody levels post‐infection
relativetothefollow‐upintervalduration.Eachcirclerepresentsaninfection. Thex‐axisdescribesthetimeuntil
thenext follow‐upvisit post‐infection.The y‐axisdescribesthechange inantibody levelatthesubsequentvisit.
Largercircles representsahigherratio‐riseinantibodylevelsattheinitialinfection.Thehorizontallineindicates
theborderbetweenincreases(>1.0)ordecreases(<1.0)inantibodylevelsatthenextstudyvisit(C)Kaplan‐Meier
curve showing decline of antibodies post infe ction (100%, 75% and 50%). The visit at which the infection wa s
establishedwas countedas timepoint0,caseswere subsequentlyfollowed. Anevent isdefinedwhen antibody
levels drop below the indicated level. When the antibody level did not decrease to the indicated levels, an
observationwascensoredatthelasttimepointofactivefollow‐up.


Antibodydynamicsafterinfection.
Onsomeoccasions,antibodyrisesreachedhighlevels(uptoratio11),representedinFig.1Baslarger
circles,yetthesehighvalueswereneversustainedatthenextstudyvisit(Fig.1B).Toinvestigatethese
dynamicsfurther,wecalculatedthetimeuntila50%,75%,orfullreturnofantibodylevelstobaseline
(pre‐infectionantibodylevels).AscanbeseeninFig.1C,within6monthspost‐infectionthemajority
ofpeoplelost50%oftheirantibodies,whereasafterayearthemajoritylost75%.Acompletereturn
tobaselinelevelsoccurredwithin4yearsforhalfofinfections(Fig.1C).

. CC-BY 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted May 18, 2020. .https://doi.org/10.1101/2020.05.11.20086439doi: medRxiv preprint


5

Simultaneousinfectionsbyalpha‐andbetacoronaviruses.
AlthoughourELISAtestsusingtheC‐terminalpartoftheNproteinwerecautiouslydesignedtobe
specificforeachindividualvirus,wecannotruleoutthatacertaindegreeofantibodycross‐reactivity
occurred.Wethereforeinvestigatedhowofteninfectionscoincided,sincecross‐reactivitymayhave
ledtofalselabelingofinfections. We observed no significant simultaneous infections for
Alphacoronavirus(HCoV‐NL63orHCoV‐229E)alongsideBetacoronavirus(HCoV‐HKU1orHCoV‐OC43),
however, we did see that infections by the betacoronaviruses HCoV‐OC43 and HCoV‐HKU1 often
coincided (38.5%, Table 2).Likewise,foralphacoronaviruses,HCoV‐229Einfectionscoincided with
HCoV‐NL63infectionsin59.5%ofthecases,andviceversain44.9%ofthecases.Hence,thereisarisk
thatweoverestimated the numberinfectionsandthus reinfections. We thereforere‐analyzedthe
datawithamorestringentdefinitionofinfection,includingonlythestrongestantibodyriseinduced
bya BetacoronavirusorAlphacoronavirusatagiventimepoint.Underthisdefinitionwestillfound
infectionintervalscomparabletotheoriginaldata(supplementaryFig.S2),withminimuminfection
intervals as short as 6 months and frequent reinfections at 12 months, although the number of
reinfectionswasobviouslyreduced.
Table2.Coincidingcoronavirusinfections
Simultaneousinfections(%)
Infection HCoV‐NL63 HCoV‐229E HCoV‐OC43 HCoV‐HKU1
HCoV‐NL63 NA 44.9 3.0 7.7
HCoV‐229E 59.5 NA 18.2 23.1
HCoV‐OC43 2.7 12.2 NA 38.5
HCoV‐HKU1 2.7 6.1 15.2 NA


Broadly acting antibodies recognizing SARS‐CoV‐2. In theory, antibodies induced by (repeated)
coronavirus infections may have broad coronavirus‐recognizing characteristics. We therefore
performedanadditionalELISAonall10subjects,thistimeusingthecompleteNproteinofSARS‐CoV‐
2, to allow detection of broadly reacting antibodies. Visual inspection suggested that broadly
recognizing antibodies wer e produced, and were most likely induced by combined infections with
HCoV‐NL63andHCoV‐HKU1(subjects2,9and10,supplementaryFig.S1).

Coronavirusinfectionsinchangingseasons.
To date it is uncertain whether SARS‐CoV‐2 will share the same winter prevalence peak that is
observedforseasonalcoronavirusesinnon‐equatorialcountries.However,itisimportanttoconsider
thatwinterpreferenceof seasonal coronaviruses has only beendetermined by testingrespiratory
samplesofpeoplethatexperienceddisease(12).Samplingandstorageisthereforedictatedbyhaving
symptomsandnotbystudyprotocol.Ifco ronavirusspreadcontinuesunabatedinsummer,yetpeople
rarely display symptoms (e.g. because of higher vita min D levels) and are therefore not sampl ed,
infectionswillremainundetected.Ourserologicalstudyisuniquebecauseitavoidsthissamplingbias.
TheNetherlandshas a typicaltemperateclimate,andour study samples werecollectedatregular
intervals.Thesamplingofeachsubjectstartedatdifferenttimesoftheyear,and,becauseofthe3or
6monthsregimeofvisits,sampleswerecollectedthroughoutallseasons.Consequently,wecanfor
thefirsttimevisualizeinanunbiasedmannertheseasonalityofcoronavirusinfections.Asshownin
Fig.2,May,June,July,AugustandSeptemberindeedshowthelowestprobabilityofinfectionsforall
fourseasonalcoronaviruses(Wilcoxonsigned‐ranktest,p=0.005).
. CC-BY 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted May 18, 2020. .https://doi.org/10.1101/2020.05.11.20086439doi: medRxiv preprint


6




Fig.2Seasonalityofinfections.Theprevalenceof thefour seasonal coronavirusesshown as theprobability of infection
onsetforaspecificmonth.Themonthsbetweenandincludingthemonthsofthefollow‐upvisitsatwhichtheinfectionwas
establishedandthepriorfollow‐upvisitwerecounted.Thevaluepermonthwasdividedbythenumberofmonthsbetween
follow‐upvisits.





. CC-BY 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted May 18, 2020. .https://doi.org/10.1101/2020.05.11.20086439doi: medRxiv preprint


7

DISCUSSION

Weshow,forthefirsttime,thatreinfectionswithallseasonal coronaviruses occur in nature.The
majorityofreinfectionsoccurredwithin3years.IfSARS‐CoV‐2willbehavelikeaseasonalcoronavirus
inthefuture,asimilarpatternmaybeexpected.However,thistimespanbetweeninfectionsdoesno t
indicatethatan individual’sprotectiveimmunitylastsforthesameperiodoftime,asreinfectionis
alsodependentonre‐exposure.Infact,basedontheminimuminfectionintervalsandthedynamics
ofantibodywaningthatweobserved,theprotectiveimmunity maylast aslittle as6 to12 months.
Recently Kissler et al.modeledthe protective immunity andreinfectiondynamicsHCoV‐OC43 and
HCoV‐HKU1 and estimated a 45 week period of protective immunity(13).Ourserologicalstudy
confirmsthisprediction.
WhenweviewourfindingsinlightofthecurrentcontrolactionstakenforSARS‐CoV‐2,itis
clearthatcoronavirusreinfectionriskiskeytopublichealthpolicy.Herewerevealariskthatinthe
nearfuture,serologybasedteststhatmeasurepreviousinfectionsforSARS‐CoV‐2mayhavelimited
use if that infection has occurred >1 year prior to sampling. Additionally, vaccine studies should
anticipate that sustained protective immunity may be uncertain for coronaviruses, and repeated
yearlyorhalf‐yearlyvaccinationsmaybeneededtocircumventongoingtransmission.
ThereisongoingdiscussionregardingherdimmunityforSARS‐CoV‐2control.Herdimmunity
occurswhenathresholdproportion ofapopulation isimmunetoacertainpathogen,andprotects
evennon‐immune individuals against theinfectionbylimitingoverall spread. This effect has been
observedforavarietyofviruses,mainlythosethatarepartiallycontrolledviavaccinationprograms,
suchashepatitisAvirus(14),influenzaAvirus(15), andhuman papillomavirus(16).Inthecaseof
SARS‐CoV‐2,achievingherdimmunitymaybechallengingduetorapidlossofprotectiveimmunity.It
wasrecentlysuggestedthatrecoveredindividualsshouldreceiveaso‐called“immunitypassport”(17)
whichwouldallowthemtorelaxsocialdistancingmeasuresandprovide governmentswith dataon
herdimmunitylevels inthepopulation. However,asprotective immunitymaybelostby6 months
post infection, the prospe ct of reaching functional herd immunity by natural infection seems very
unlikely.
It is generally assumed that exposure to each seasonal coronavirus is not equal. The
BetacoronavirusHCoV‐HKU1is consideredtohave thelowestprevalence (4,13)andwefoundthe
same. We also observed that antibody levels to HCoV‐HKU1 were low over the whole range (all
samplestested).WhetherthisindicatesHCoV‐HKU1infectionelicitsapoorinductionofantibodiesis
intriguing,yetnot answerableinourstudy. Weuseda relativelysmallpartoftheNproteinasthe
antigeninourtests,tomostspecificallyidentifyinfectionsbyeachvirusindividually.Thissmallpart
oftheproteinmaynotcontainthemostimmunodominantepitopesofHCoV‐HKU1.Aneutralization
testwouldideallybedonetosolvetheissueofprotectiveimmunityraisedbyHCoV‐HKU1;however,
unfortunatelynoculturingsystemsuitableforneutralizationtestingisavailableforHCoV‐HKU1(18).
WenoticedthreesubjectstocarryantibodiesrecognizingSARS‐CoV‐2N protein atcertain
timepoints.ItisunlikelythattheyhadbeeninfectedwithaSARS‐CoV‐2‐likevirusin1985(subject10),
1992(subject2),or 2006(subject 9),and wethereforesuggestthatbroadly actingantibodiesmay
havebee ninduce dbycoin cidin ginfect ionsofanAlpha‐andaBetacoronavirus(inoursub ject(s) HCoV‐
HKU1andHCoV‐NL63).Toexplorethisfindingwelookedat thegenetic distanceandconsequently
aminoacid differencesinthestructuralproteinofthevariouscoronaviruses(supplementaryTable
S2).Notably,SARS‐CoV‐2Nproteinhasonly32%and34%identityontheaminoacidlevelwiththeN
proteinofHCoV‐OC43andHCoV‐HKU1respectively,andonly26%and24%identitywithHCoV‐NL63
andHCoV‐229Erespectively. Similarly,thedistancebetweenAlphacoronavirusandBetacoronavirus
Nproteinislarge (only 24% to26%aminoacid identity). Still, we cannot exclude the presenceof
conserved(conformational) epitopesin HCoV‐HKU1andHCoV‐NL63 Nproteinthat mayresultin a
. CC-BY 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted May 18, 2020. .https://doi.org/10.1101/2020.05.11.20086439doi: medRxiv preprint


8

more broadly acting antibody response, due to simultaneous exposure in concurrent infections.
Additionalscreening,includingmoresubjects,isrequiredforconfirmation.
Ourstudywassubjecttolimitations.Oneistheinabilitytosequencethevirusgenomeduring
infection.Intheory,strainvariationcouldplayaroleinsusceptibilitytoreinfection.HCoV‐NL63,HCoV‐
OC43,andHCoV‐HKU1allshowdifferentco‐circulatinggeneticclusters(19–21).ForHCoV‐229E,there
arenomajorgeneticsubtypesknown,likethecurrentSARS‐CoV‐2situation(22,23).Itcouldbethat
ourobservationofshort‐termimmunityisinfluencedbygenotypevariation,andastudyonprotective
immunitywouldthereforeideallyallowsequencingofre‐infectingstrainsfromrespiratorymaterial;
however,thisisintractableinanaturalinfectionstudybecausevirussheddinginreinfectionscanbe
asshortasoneday(1).Anotherlimitationisthatthestudiessubjectswereallmales.ForCOVID‐19,
andalsoHCoV‐NL63,menhaveahigherincidenceofdisease(25,26),anditisthereforeofinterestto
determinethedynamicsofprotectiveimmunityalsoinacohortofhealthywomen.
In conclusion, seasonal human coronaviruses have little in common, apart from causing
commoncold.Still,theyallseemtoinduceashort‐lastingimmunitywithrapidlossofantibodies.This
maywellbeageneraldenominatorforhumancoronaviruses.


REFERENCES
1.K.A.Callow,Effectofspecifichumoralimmunityandsomenon‐specificfactorsonresistance
ofvolunteerstorespiratorycoronavirusinfection.Epidemiology&Infection.95,173–189
(1985).
2.L.vanderHoek,G.Ihorst,K.Sure,A.Vabret,R.Dijkman,M.deVries,J.Forster,B.Berkhout,
K.Uberla,BurdenofdiseaseduetohumancoronavirusNL63infectionsandperiodicityof
infection.JournalofClinicalVirology.48,104–108(2010).
3.L.vanderHoek,Humancoronaviruses:whatdotheycause?Antiviraltherapy.12,651(2007).
4.R.Dijkman,M.F.Jebbink,E.Gaunt,J.W.A.Rossen,K.E.Templeton,T.W.Kuijpers,L.vander
Hoek,ThedominanceofhumancoronavirusOC43andNL63infectionsininfants.Journalof
ClinicalVirology.53,135–139(2012).
5.W.Zhou,W.Wang,H.Wang,R.Lu,W.Tan,Firstinfectionbyallfournon‐severeacute
respiratorysyndromehumancoronavirusestakesplaceduringchildhood.BMCInfectious
Diseases.13,1–8(2013).
6.H.Hofmann,K.Pyrc,L.vanderHoek,M.Geier,B.Berkhout,S.Pöhlmann,Humancoronavirus
NL63employsthesevereacuterespiratorysyndromecoronavirusreceptorforcellularentry.
ProceedingsoftheNationalAcademyofSciencesoftheUnitedStatesofAmerica.102,7988–
7993(2005).
7.E.G.Severance,I.Bossis,F.B.Dickerson,C.R.Stallings,A.E.Origoni,A.Sullens,R.H.Yolken,
R.P.Viscidi,Developmentofanucleocapsid‐basedhumancoronavirusimmunoassayand
estimatesofindividualsexposedtocoronavirusinaU.S.metropolitanpopulation.Clinical
andVaccineImmunology.15,1805–1810(2008).
8.P.Sastre,R.Dijkman,A.Camuñas,T.Ruiz,M.F.Jebbink,L.vanderHoek,C.Vela,P.Rueda,
DifferentiationbetweenhumancoronavirusesNL63and229Eusinganoveldouble‐antibody
sandwichenzyme‐linkedimmunosorbentassaybasedonspecificmonoclonalantibodies.
ClinicalandVaccineImmunology.18,113–118(2011).
. CC-BY 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted May 18, 2020. .https://doi.org/10.1101/2020.05.11.20086439doi: medRxiv preprint


9

9.R.Dijkman,M.F.Jebbink,N.B.elIdrissi,K.Pyrc,M.A.Müller,T.W.Kuijpers,H.L.Zaaijer,L.
vanderHoek,HumancoronavirusNL63and229Eseroconversioninchildren.Journalof
ClinicalMicrobiology.46,2368–2373(2008).
10.C.Lehmann,H.Wolf,J.Xu,Q.Zhao,Y.Shao,M.Motz,P.Lindner,Alineimmunoassayutilizing
recombinantnucleocapsidproteinsfordetectionofantibodiestohumancoronaviruses.
DiagnosticMicrobiologyandInfectiousDisease.61,40–48(2008).
11.W.P.H.vanBilsen,A.Boyd,M.F.S.vanderLoeff,U.Davidovich,A.Hogewoning,L.vander
Hoek,M.Prins,A.Matser,DivergingtrendsinincidenceofHIVversusothersexually
transmittedinfectionsinHIV‐negativeMSMinAmsterdam.AIDS.34(2020)(availableat
https://journals.lww.com/aidsonline/Fulltext/2020/02010/Diverging_trends_in_incidence_of
_HIV_versus_other.16.aspx).
12.E.R.Gaunt,A.Hardie,E.C.J.Claas,P.Simmonds,K.E.Templeton,Epidemiologyandclinical
presentationsofthefourhumancoronaviruses229E,HKU1,NL63,andOC43detectedover3
yearsusinganovelmultiplexreal‐timePCRmethod.Journalofclinicalmicrobiology.48,
2940–2947(2010).
13.S.M.Kissler,C.Tedijanto,E.Goldstein,Y.H.Grad,M.Lipsitch,Projectingthetransmission
dynamicsofSARS‐CoV‐2throughthepostpandemicperiod.Science(2020),
doi:10.1126/science.abb5793.
14.D.Smith,C.Huynh,A.J.Moore,A.Frick,C.Anderson,M.Porrachia,B.Scott,S.Stous,R.
Schooley,S.Little,HerdImmunityLikelyProtectedtheMenWhoHaveSexWithMeninthe
RecentHepatitisAOutbreakinSanDiego,California.ClinicalInfectiousDiseases.68,1228–
1230(2019).
15.E.Q.Mooring,S.Bansal,Increasingherdimmunitywithinfluenzarevaccination.
Epidemiology&Infection.144,1267–1277(2016).
16.R.L.Cameron,K.Kavanagh,J.Pan,J.Love,K.Cuschieri,C.Robertson,S.Ahmed,T.Palmer,K.
G.J.Pollock,Humanpapillomavirusprevalenceandherdimmunityafterintroductionof
vaccinationprogram,Scotland,2009–2013.Emerginginfectiousdiseases.22,56(2016).
17.A.Bendavid,etal.,https://www.medrxiv.org/content/10.1101/2020.04.14.20062463v1
(2020).
18.K.Pyrc,A.C.Sims,R.Dijkman,M.Jebbink,C.Long,D.Deming,E.Donaldson,A.Vabret,R.
Baric,L.vanderHoek,R.Pickles,CulturingtheUnculturable:HumanCoronavirusHKU1
Infects,Replicates,andProducesProgenyVirionsinHumanCiliatedAirwayEpithelialCell
Cultures.JournalofVirology.84,11255–11263(2010).
19.P.C.Y.Woo,S.K.P.Lau,C.C.Y.Yip,Y.Huang,H.‐W.Tsoi,K.‐H.Chan,K.‐Y.Yuen,
Comparativeanalysisof22coronavirusHKU1genomesrevealsanovelgenotypeand
evidenceofnaturalrecombinationincoronavirusHKU1.Journalofvirology.80,7136–7145
(2006).
20.K.Pyrc,R.Dijkman,L.Deng,M.F.Jebbink,H.A.Ross,B.Berkhout,L.vanderHoek,Mosaic
structureofhumancoronavirusNL63,onethousandyearsofevolution.Journalofmolecular
biology.364,964–973(2006).
. CC-BY 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted May 18, 2020. .https://doi.org/10.1101/2020.05.11.20086439doi: medRxiv preprint


10

21.S.K.P.Lau,P.Lee,A.K.L.Tsang,C.C.Y.Yip,H.Tse,R.A.Lee,L.‐Y.So,Y.‐L.Lau,K.‐H.Chan,P.
C.Y.Woo,MolecularepidemiologyofhumancoronavirusOC43revealsevolutionofdifferent
genotypesovertimeandrecentemergenceofanovelgenotypeduetonatural
recombination.Journalofvirology.85,11325–11337(2011).
22.D.Chibo,C.Birch,Analysisofhumancoronavirus229Espikeandnucleoproteingenes
demonstratesgeneticdriftbetweenchronologicallydistinctstrains.Journalofgeneral
virology.87,1203–1208(2006).
23.S.M.J.Farsani,R.Dijkman,M.F.Jebbink,H.Goossens,M.Ieven,M.Deijs,R.Molenkamp,L.
vanderHoek,Thefirstcompletegenomesequencesofclinicalisolatesofhumancoronavirus
229E.VirusGenes.45,433–439(2012).
25. P.Mo,XingY,XiaoY,DengL,ZhaoQ,WangH,XiongY,ChengZ,GaoS,LiangK,LuoM,Chen
T,SongS,MaZ,ChenX,ZhengR,CaoQ,WangF,ZhangY.Clinicalcharacteristicsofrefractory
COVID‐19pneumoniainWuhan,China.ClinInfectDis.2020Mar16.pii:ciaa270.doi:
10.1093/cid/ciaa270.
26. L.vanderHoek,SureK,IhorstG,StangA,PyrcK,JebbinkMF,PetersenG,ForsterJ,Berkhout
B,UberlaK.CroupisassociatedwiththenovelcoronavirusNL63.PLoSMed.2005
Aug;2(8):e240.




Acknowledgments
TheauthorsgratefullyacknowledgetheAmsterdamCohortStudies(ACS)onHIVinfectionandAIDS,acollaborationbetween
the Public Health Service of Amsterdam, the Amsterdam UMC of the University of Amsterdam, Sanquin B lood Supply
Foundation,MedicalCenter JanvanGoyen, andthe HIVFocusCenteroftheDC‐Clinics.ItispartoftheNetherlandsHIV
MonitoringFoundation andfinanciallysupportedbytheCenterforInfectiousDiseaseControlof theNetherlandsNational
InstituteforPublicHealthandtheEnvironment.TheauthorsthankallACSparticipantsfortheircontribution,aswellasthe
ACS study nurses, data‐managers, and lab technicians. This work was supported by a grant from the European Union's
Horizon2020researchandinnovationprogramme,undertheMarieSkłodowska‐CurieActionsgrantagreementno.721367
(HONOURs),AmsterdamUMCfundingconnectedtoHONOURs,andtheAmsterdamUMCPhDscholarshipofA.W.D.Edridge.
Authorcontributions:A.W.D.E:Conceptualization,Writing– originaldraft,review andediting,Investigation,Visualization,
Formal Analysis, Validation;  J.K.: Investigation, Writing  – original draft, review and editing, A.C.R.H.: Resources; M.B.:
Investigation,Resources;M.K.: Investigation;M.F.J.: Investigation,Methodology;A.M.: FormalAnalysis;C.M.K.:Writing–
originaldraft,reviewandediting,FormalAnalysis; P.R.:Resources; M.P.:Resources;P.S.: Resources;M.D.:Investigation,
Methodology;L.v.d.H.:Conceptualization,Writing–originaldraft,Writing‐reviewandediting,Supervision.Competing
interests:Authorsdeclarenocompetinginterests.


. CC-BY 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted May 18, 2020. .https://doi.org/10.1101/2020.05.11.20086439doi: medRxiv preprint