Rotavirus capsid VP6 tubular and spherical nanostructures act as local adjuvants when co-delivered with norovirus VLPs

Ro a i us capsid VP6 ubula and sphe ical nanos uc u es ac as local adju an s when co-deli e ed wi h no o i us VLPs M. Malm, S. Heinim€ aki, T. Vesika i and V. Blaze ic Vaccine Resea ch Cen e , Uni e si y o Tampe e, Tampe e, Finland Accep ed o publica ion 7 Ap il 2017 Co espondence: Vesna Blaze ic, Bioka u 10, 33520 Tampe e, Finland. E-mail: [email p o ec ed] Summa y A subuni p o ein accine candida e based on no o i us (NoV) i us-like pa icles (VLPs) and o a i us (RV) VP6 p o ein agains acu e childhood gas oen e i is has been p oposed ecen ly. RV VP6 o ms di e en oligome ic nanos uc u es, including ubes and sphe es when exp essed in i o, which a e highly immunogenic in di e en animal models. We ha e shown ecen ly ha ecombinan VP6 nano ubes ha e an adju an e ec on immunogenici y o NoV VLPs in mice. In his s udy, we in es iga ed i he adju an e ec is dependen upon a VP6 dose o di e en VP6 s uc u al assemblies. In addi ion, local and sys emic adju an e ec s as well as equi emen s o an igen co-deli e y and co-localiza ion we e s udied. The magni ude and unc ionali y o NoV GII.4-speci ic an ibodies and T cell esponses we e es ed in mice immunized wi h GII.4 VLPs alone o di e en combina ions o VLPs and VP6. A VP6 dose-dependen adju an e ec on GII.4-speci ic an ibody esponses was obse ed. The adju an e ec was ound o be s ic ly dependen upon co-adminis a ion o NoV GII.4 VLPs and VP6 a he same ana omic si e and a he same ime. Howe e , he adju an e ec was no dependen on he ypes o oligome s used, as bo h nano ubes and nanosphe es exe ed adju an e ec on GII.4-speci ic an ibody gene a ion and, o he i s ime, T cell immuni y. These indings elucida e he mechanisms o VP6 adju an e ec in i o and suppo i s use as an adju an in a combina ion NoV and RV accine. Keywo ds: adju an , nanosphe es, nano ubes, no o i us VLP, o a i us VP6 In oduc ion No o i uses (NoV) and o a i uses (RV) a e he mos com- mon causes, a a e y young age, o childhood i al gas o- en e i is (GE) wo ldwide [1–3]. To elimina e he high isk o NoV and RV-caused gas oen e i is (GE), a accina ion soon a e bi h would be needed. Ne e heless, NoV in ec ions and disease a e no ye p e en able by accina ion, al hough NoV i us-like pa icles (VLPs)-based accines a e being s udied ex ensi ely. NoV VLPs a e non-in ec ious sel - assembled pa icles composed o VP1 capsid p o ein, highly esembling in ac i ions bo h s uc u ally and an igenically [4]. Cu en ly used RV accines a e based on li e a enua ed i uses [5,6] ha a e associa ed wi h po en ial sa e y issues, such as a isk o in ussuscep ion [7] and shedding and ans- mission o he accine s ains and easso men o yield mo e i ulen o ms [8], suppo ing he need o non-li e RV subuni accines. Ou g oup has ecen ly de eloped a combina ion accine candida e agains NoV and RV child- hood GE consis ing o NoV VLPs and RV VP6 [9,10]. The combina ion accine induced s ong ype-speci ic and c oss- eac i e humo al and cellula immuni y agains NoV and RV in a mouse model [9,11–13]. In gene al, highly pu i ied subuni accines wi h he bes sa e y p o ile ha e poo immunogenici y, and he magni ude and quali y o he immune esponses need o be enhanced by adju an s [14]. Howe e , due o he ulne abili y o in an s and young child en, i would be highly desi able o ha e an e icien accine wi hou adding ex e nal adju an s [15,16]. To his end, we ha e shown ha RV VP6 no only induced p o ec i e immuni y agains li e RV challenge in mice [13], bu i also ac ed as a s ong in- i o adju an on he gene a ion o an ibodies speci ic o NoV [17]. In a iple-laye ed RV pa icle he in e media e laye is o med by he VP6 p o ein (45 kD), si ua ed be ween he V C2017 The Au ho s. Clinical & Expe imen al Immunology published by John Wiley & Sons L d on behal o B i ish Socie y o Immunology, Clinical and Expe imen al Immunology, 189: 331–341 331 This is an open access a icle unde he e ms o he C ea i e Commons A ibu ion-NonComme cial-NoDe i s License, which pe mi s use and dis ibu ion in any medium, p o ided he o iginal wo k is p ope ly ci ed, he use is non-comme cial and no modi ica ions o adap a ions a e made. Clinical and Expe imen al Immunology ORIGINAL ARTICLE doi:10.1111/cei.12977 ou e mos laye consis ing o VP4 and VP7 p o eins and he inne co e p o ein VP2, which su ounds he double- s anded genome o RV [6,18,19]. RV VP6 is he mos abundan and immunogenic RV p o ein [20,21], which is highly conse ed among RV s ains [22,23]. VP6 o ms ime s o ganized in o hexagons and packed in o highe - o de s uc u al assemblies, e.g. VP6 nano ubes (VP6T) and nanosphe es (VP6S), when exp essed in i o [12,24–27] unde di e en condi ions [25,28]. The VP6 is highly immunogenic [9,12,29], and i has also been used as a ca ie o deli e y pla o m o he e ologous p o ein an i- gens and gene ically used epi ope-based accines, wi h imp o ed esponse o he o eign an igen [30–34]. VP6 is s able a di e en pH condi ions, and when deli e ed o ally i was a ge ed o in es inal cells, o e ing a p omising new deli e y pla o m o anspo pha maceu ical compounds o gas oin es inal ac [35]. In o de o elucida e he mechanisms o VP6 adju an ac ion, we ha e shown p e iously ha VP6T a e aken up e icien ly by mac ophages and dend i ic cells (DC) in i o, esul ing in ac i a ion and ma u a ion o hese an igen- p esen ing cells (APC). Also, VP6T we e shown o acili a e he in e naliza ion o co-deli e ed NoV VLPs o he APC [36]. We unde ook he p esen s udy o in es iga e equi emen s o co-localiza ion and co-deli e y o he VP6 wi h he NoV VLPs in i o and whe he VP6 wo ks as a local o sys emic adju an . In addi ion, adju an e ec o VP6T was compa ed o VP6S. Ma e ials and me hods Recombinan p o eins NoV VLPs and RV VP6 oligome ic p o eins we e p oduced in a baculo i us–insec cell exp ession sys em, as desc ibed in de ail elsewhe e [9,12,37]. NoV GII.4-1999 VLPs (GenBank e e ence s ain, Accession numbe AF080551) and VP6 an igens (Accession no. GQ477131) used o immuniza ions o animals we e highly pu i ied wi h mul i- s ep ch oma og aphic p ocedu es o a ious s eps o ul a- il a ion, as desc ibed p e iously [17,38]. The pu i ied VP6 was assembled in o nano ubes in phospha e-bu e ed saline (PBS) a pH 73–75 (Lonza, Ve ie s, Belgium) o nanosphe es in a 50 mM sodium ace a e bu e wi h 130 mM NaCl, pH 482 [38]. The concen a ion o he p o- eins was de e mined using Pie ce BCA p o ein assay (The mo Scien i ic, Wal ham, MA, USA). The pu i y o he p o eins was e i ied by Quan -i dsDNA B oad-Range Assay Ki (In i ogen, Ca lsbad, CA, USA; <10 ng dsDNA/10 lg o p o ein), sodium dodecyl sulpha e poly- ac ylamide gel elec opho esis (SDS-PAGE), BacPAK Rap- idTi e Ki [Clon ech Labo a o ies, Moun ain View, CA, USA; 0 plaque- o ming uni s (p u) li e BV/ml] and limulus amebocy e lysa e assay (Lonza, <01 endo oxin uni s/100 lg o p o ein), as desc ibed in de ail elsewhe e [9,38]. The VLPs and oligome ic VP6 nanos uc u es used o immu- niza ions we e con i med by nega i e-s aining ansmission elec on mic oscopy (TEM) using an FEI Tecnai F12 (Philips Elec on Op ics, Eindho en, he Ne he lands) a e nega i e s aining wi h 3% u anyl ace a e pH 46 o p o ein mo phology and in eg i y (Fig. 1a–c). NoV VLPs used o enzyme-linked immunoso ben assay (ELISA)-based ana- ly ical me hods, GII.4-1999, GII.4 New O leans (NO) 2009 (Accession no. GU445325) and GII.4 Sydney (SYD) 2012 (Accession no. AFV08795.1) we e pu i ied using suc ose g adien ul acen i uga ion, as desc ibed elsewhe e [37]. Syn he ic pep ides and i uses Fo quan i ica ion o NoV-speci ic T cell esponses by enzyme-linked immunospo (ELISPOT), an in e e on (IFN)-gassay GII.4 pep ide pool (Synpep ide Co. L d, Shanghai, China) con aining 76 syn he ic pep ides [18- me s, 11 amino acid (aa) o e lap], spanning he en i e 539aa sequence o GII.4-1999 NoV VP1 [39], was used. VP6-speci ic BALB/c mouse (H-2 d ) CD4 1 T cell epi ope 242 DGATTWYFNPVILRPNNV 259 [11,40], named R6-2, was syn he ized (P oimmune L d, Ox o d, UK) and used Fig. 1. S uc u e and in eg i y o he p o eins. Elec on mic oscopy images o baculo i us–insec cell sys em-p oduced no o i us (NoV) GII.4-1999 i us-like pa icles (VLPs) (a), o a i us (RV) VP6 nano ubes (b) and VP6 nanosphe es (c) examined by FEI Tecnai F12 elec on mic oscope (Philips Elec on Op ics) a e nega i e s aining wi h 3% u anyl ace a e, pH 46. Images obse ed a 323 000 (a,c) o 36800 (b) magni ica ion. M. Malm e al. 332 V C2017 The Au ho s. Clinical & Expe imen al Immunology published by John Wiley & Sons L d on behal o B i ish Socie y o Immunology, Clinical and Expe imen al Immunology, 189: 331–341 in RV VP6-speci ic ELISPOT assays. O albumin (OVA) 323–339 chicken egg albumin pep ide (aa 323 ISQAVHAA- HAEINEAGR 339 , ca . # ac-isq; In i ogen, San Diego, CA, USA) se ed as a nega i e con ol. Fo RV-speci ic ELISPOT assays human RV s ains Wa (G1P1A[8]) and bo ine RV s ain WC3 (G6P7[5]) we e p opaga ed in e al hesus monkey kidney (MA104) cells, as desc ibed p e iously [12], and Ridasc een V R Ro a i us ki (R-Biopha m AG, Da ms ad , Ge many; ca . C0901) wi h he in e nal VP6 s anda d was used o de e mine he VP6 amoun (ng/ml) in he cul u es. Immuniza ion o animals Female 7–8-week-old BALB/c mice ob ained om En igo Labo a o ies (Ho s , Limbu g, he Ne he lands) we e di ided o nine g oups (g oups I–IX) and immunized in amuscula ly (i.m.) a s udy weeks 0 and 3 wi h di e en doses o an igens dilu ed in 50 ml PBS, as shown in Table 1. A minimum o i e mice in each g oup was used. All an i- gens we e deli e ed in o he igh high muscle o each mouse excep g oup VI, which ecei ed GII.4 VLPs a he igh high and VP6T a he con ala e al le limb si e a he same ime. Fo immunizing g oups II–IV and VII, he VLPs and VP6 we e mixed p io o adminis a ion. To de ine he empo al equi emen s o VP6 adju an ac i i y, mice in g oup V ecei ed 10 mg VP6T i s , ollowed by 03 mg VLP 1 h la e a he same si e. Nai e mice ecei ing ca - ie only (g oup IX) we e used as nega i e con ols. Ani- mals we e anaes he ized be o e immuniza ion wi h se o lu ane inhala e and o eu hanasia wi h a mix u e o mede omidine (Do bene e , 1 mg/ml; Labo a o ios SYVA SA, Leon, Spain) and ke amine (Ke aminol e , 50 mg/ml; In e e In e na ional BV, Boxmee , he Ne he lands). Blood and spleens we e collec ed a he ime o e mina ion (a week 5) o he analysis o se ological and cell-media ed immune esponses, as desc ibed p e iously [41]. Se um samples we e s o ed a 2208C, and spleen cell suspensions we e s o ed in liquid ni ogen be o e use in he assays. All mice used we e nega i e o NoV GII.4-speci ic immuno- globulin (Ig)G an ibodies p io o he immuniza ions a week 0 (da a no shown). Mice wel a e was moni o ed h oughou he s udy and expe imen s we e pe o med in acco dance wi h he guidelines o he Finnish Na ional Ani- mal Expe imen Boa d. Se um IgG an ibody ELISA Se a o immunized and con ol mice we e analysed indi- idually by ELISA o de e mine NoV GII.4-1999, GII.4- NO, GII.4 SYD and VP6-speci ic IgG, IgG1 and IgG2a i es, as desc ibed p e iously [9,11]. B ie ly, 96-well pla es (Co ning Inc. Co ning, NY, USA) we e coa ed wi h 50 ng/ well o NoV VLPs o VP6 in PBS. Duplica es o wo old se ial dilu ions o se um samples we e incuba ed o 2 h a oom empe a u e and bound an ibodies we e de ec ed wi h ho se adish pe oxidase (HRP)-conjuga ed goa an i- mouse IgG (Sigma-Ald ich, S Louis, MO, USA), IgG1 (In i ogen) o IgG2a (In i ogen). Op ical densi y (OD) a 490 nm was measu ed by a mic opla e eade Vic o 2 1420 (Pe kin Elme , Wal ham, MA, USA) and a sample was conside ed posi i e i he OD was abo e he mean OD o con ol mice 13 s anda d de ia ions (s.d.) and >01. End- poin i es we e exp essed as he highes se um dilu ion gi ing a posi i e eading. Blocking assay (a su oga e neu aliza ion assay) To examine he abili y o se um an ibodies o block he binding o NoV VLPs o a pu a i e cellula his o-blood g oup an igen (HBGA) ecep o , wo di e en sou ces o HBGAs, pig gas ic mucin (PGM) ype III (Sigma Chemi- cals, S Louis, MO, USA) and human ype A sali a we e u ilized [42,43]. The blocking assays we e conduc ed wi h he p ocedu es desc ibed in de ail elsewhe e [44]. B ie ly, g oupwise pooled wo old se um dilu ions we e p eincu- ba ed wi h he 01mg/ml GII.4-1999 VLPs and he mix u es we e added o 96-mic owell pla es coa ed wi h 25mg/ml PGM o 1 : 3000 dilu ed human ype A sali a. Maximum Table 1. Expe imen al and con ol immuniza ion g oups G oup Dose o GII.4 VLP Dose and o m o VP6 Adminis a ion I03mg – i.m. II 03mg1mg VP6T i.m., co-adminis a ion III 03mg10mg VP6T i.m., co-adminis a ion IV 03mg30mg VP6T i.m., co-adminis a ion V03mg10mg VP6T i.m., VP6 i s ollowed by he VLP a e 1 h VI 03mg10mg VP6T i.m., le igh VP6, igh igh VLP, a he same ime VII 03mg10mg VP6S i.m., co-adminis a ion VIII 3 mg – i.m. IX (C l) – – i.m., ca ie only (PBS) VLP 5 i us-like pa icle; VP6 5 ecombinan VP6; i.m. 5in amuscula ; VP6T 5VP6 nano ubes; VP6S 5VP6 nanosphe es; C l 5con ol; PBS 5phospha e-bu e ed saline. Ro a i us VP6 ubula and sphe ical nanos uc u es as local adju an s V C2017 The Au ho s. Clinical & Expe imen al Immunology published by John Wiley & Sons L d on behal o B i ish Socie y o Immunology, Clinical and Expe imen al Immunology, 189: 331–341 333 VLP binding signal (OD) was de e mined in wells lacking se um. The bound VLPs we e de ec ed using human an i- NoV de ec ion se um and an i-human IgG conjuga e (No ex; The mo Fishe Scien i ic, F emon , CA, USA). Blocking index was calcula ed as ollows: 100% 2[OD (wells wi h se um)/OD (wells wi hou se um, maximum binding)] 3100%. Resul s a e exp essed as he blocking i e 50 (BT 50 ), a se um i e blocking 50% o he VLPs binding o he HBGAs [45]. No o i us-speci ic ELISPOT IFN-g To quan i y NoV GII.4-speci ic IFN-g-p oducing T cells o immunized mice, an ELISPOT assay was used [11]. Spleno- cy es om ei he indi idual o g oupwise pooled mice we e pla ed on Mul iSc eenHTS-IP il e pla es (Millipo e, Bille - ica, MA, USA) coa ed wi h an an i-mouse IFN-gmonoclo- nal an ibody (Mab ech AB, Nacka S and, Sweden) a 5 mg/ ml and blocked wi h 10% e al bo ine se um (FBS; Sigma- Ald ich). Cells we e s imula ed wi h GII.4-99 pep ide pool (2 mg/ml), GII.4-99 VLPs (5 mg/ml) o OVA pep ide (nega- i e con ol, 4 mg/ml). Cells incuba ed in cul u e medium (CM) only (RPMI 1640 supplemen ed wi h 10% e al bo ine se um, 100 U/ml penicillin, 100 mg/ml s ep omycin, 50 mM 2-me cap oe hanol and 2 mM L-glu amine; Sigma- Ald ich) and cells we e s imula ed wi h 10 mg/ml o T cell mi ogen concana alin A (ConA; Sigma-Ald ich, ca . C5275) se ed as a backg ound and iabili y con ol. A e o e nigh incuba ion (16–20 h) IFN-gwas de ec ed wi h bio inyla ed an i-mouse IFN-gmonoclonal an ibody (Mab ech AB, ca . 3321–6) and s ep a idin–alkaline phospha ase (ALP) conju- ga e (Mab ech AB, ca . 3310–10). The spo s we e de eloped wi h BCIP/NBT (5-b omo-4-chlo o-3-indolyl-phospha e) subs a e (Mab ech AB, ca . 3650–10) and coun ed by ImmunoSpo V R au oma ic CTL analyse (CTL-Eu ope GmbH, Bonn, Ge many). The esul s we e exp essed as mean spo - o ming cells (SFC)/10 6 splenocy es o duplica e wells. A posi i e esul was conside ed as an inc ease o wice o mo e abo e he con ol an igen SFC coun s. Backg ound coun s ne e exceeded 20 SFC/10 6 cells. RV VP6-speci ic IFN-gand IL-4 ELISPOT VP6-speci ic IFN-gand in e leukin (IL)-4 T cell esponses we e es ed using an ELISPOT assay s imula ing he spleno- cy es o immunized and con ol mice wi h VP6 p o ein (5 mg/ml), R6-2 pep ide (5 mg/ml), RV Wa, RV WC3 (05mg VP6/ml) o MA104 mock an igen. IFN-gELISPOT assay was pe o med simila ly o NoV-speci ic ELISPOT IFN-g,as desc ibed abo e. Fo quan i ica ion o IL-4 p oduc ion he Mul iSc eenHTS-IP il e pla es we e coa ed wi h an i- mouse IL-4 monoclonal an ibody (Mab ech AB, ca . 3311-3) a 5 mg/ml. G oupwise pooled cells we e pla ed and incu- ba ed wi h he an igens o 40–45 h. Bio inyla ed an i-mouse IL-4 an ibody (Mab ech AB, ca . 3311-6) ollowed by incu- ba ion wi h s ep a idin–ALP and BCIP/NBT subs a e we e used o de eloping he spo s. The spo s we e coun ed and he esul s we e exp essed as desc ibed abo e. S a is ics The s a is ical di e ences be ween independen g oups we e assessed by Mann–Whi ney U- es o se um IgG end- poin i es and ELISPOT assay esul s. Wilcoxon’s signed- ank es o compa ing wo ela ed samples was used o analysing he di e ence be ween IgG1 and IgG2a end-poin i es. S a is ical analyses we e pe o med using IBM SPSS s a is ics (SPSS, Chicago, IL, USA) e sion 23. P<005 was conside ed s a is ically signi ican . Resul s VP6 adju an e ec on NoV-speci ic se um an ibody esponses is dose-dependen In o de o de e mine he mechanism o adju an e ec o RV VP6 on NoV-speci ic an ibody esponses, mice we e immunized wi h a subop imal dose (03lg) o GII.4-1999 VLP alone o co-adminis e ed wi h di e en doses o VP6T. Immuniza ion wi h 03mg GII.4 VLP did no elici a signi ican se um IgG esponse, whe eas co- adminis a ion o 03mg GII.4 VLP wi h 10 mg VP6T esul ed in obus NoV-speci ic se um IgG le els (Fig. 2a), con i ming ou p e ious obse a ion [17]. The GII.4-spe- ci ic se um IgG i es we e simila (P50853) o he i es ob ained wi h a 3 lg dose o GII.4 VLP alone used as a pos- i i e con ol (Fig. 2a, g oup VIII). A 1 lg dose o VP6T co- adminis e ed wi h 03lg GII.4 VLP induced lowe an i- body esponses han a 10 lg dose o VP6T (Fig. 2a, g oups II and III, P<001), while a 30 lg dose induced simila esponses (Fig. 2a, g oups III and IV, P50425), indica ing ha he VP6 adju an e ec was dose-dependen , bu pla- eaued a e maximum e ec . VP6 adju an e ec on NoV-speci ic se um an ibody esponses is local and equi es co-deli e y The spa ial and empo al equi emen s o VP6 adju an ac i i y we e nex e alua ed. A equi emen o co- adminis a ion o 03mg GII.4 VLP 110 mg VP6T was es ed by deli e ing an igens a he same ime as a mix u e (Table 1, g oup III), deli e ing he VLPs 1 h ollowing he VP6T a he same si e (Table 1, g oup V) o deli e ing an i- gens a he same ime a he con ala e al si es (Table 1, g oup VI). Co-adminis a ion as a mix u e induced signi i- can ly highe se um NoV GII.4-speci ic IgG han injec ing VP6T and GII.4 VLPs sepa a ely in o he same si e wi h 1 h di e ence (Fig. 2a, g oups III and V, P50005). Rema k- ably, no VP6 adju an e ec on GII.4-speci ic se um IgG was seen when he an igens we e spa ially dissocia ed (Fig. 2a, g oups III and VI, P<0005). Hence, he adju an M. Malm e al. 334 V C2017 The Au ho s. Clinical & Expe imen al Immunology published by John Wiley & Sons L d on behal o B i ish Socie y o Immunology, Clinical and Expe imen al Immunology, 189: 331–341 e ec o VP6 is appa en ly exe ed locally a he si e o adminis a ion and no sys emically. VP6 adju an e ec on NoV-speci ic se um an ibody esponses is no dependen upon he ype o oligome ic s uc u es We u he de e mined i he e is a di e ence in VP6 adju- an e ec induced wi h VP6T o VP6S, as bo h ubula as well as sphe ical s uc u es we e shown p e iously o be highly immunogenic in mice [12]. When 10 lg VP6S we e co-deli e ed wi h he VLPs ins ead o he VP6T, compa able IgG i es we e obse ed (P50211) (Fig. 2a), indica ing ha bo h oligome ic con o ma ional s uc u es o VP6 ha e compa able adju an abili y. All nega i e con- ol mice ecei ing PBS only we e nega i e o NoV GII.4- speci ic an ibodies (da a no shown). VP6 p omo es unbiased Th1/Th2- ype esponses To de e mine i he VP6 has an adju an e ec on NoV GII.4-speci ic T helpe ype 1 (Th1)- ype and/o Th2- ype esponses, expe imen al g oups whe e he VP6 adju an e ec was seen (Fig. 2a, g oups III, IV and VII, espec i ely) Fig. 2. GII.4 geno ype-speci ic immunoglobulin (Ig)G an ibody i es. IgG (a), IgG1 (b) and IgG2a (c) end-poin i es we e de e mined om indi idual, se ially dilu ed se a o immunized mice in enzyme-linked immunoso ben assay (ELISA). Ba s ep esen log 10 geome ic mean i es wi h 95% con idence in e als. Fo nega i e mice se a, an a bi a y i e o 1 : 100 (hal he s a ing se um dilu ion, 1 : 200) was assigned. *Signi ican ly di e en (P<005) an ibody i es compa ed o g oup III (a) o compa ed o g oup I (b,c). Ro a i us VP6 ubula and sphe ical nanos uc u es as local adju an s V C2017 The Au ho s. Clinical & Expe imen al Immunology published by John Wiley & Sons L d on behal o B i ish Socie y o Immunology, Clinical and Expe imen al Immunology, 189: 331–341 335 we e es ed o GII.4-speci ic IgG1 and IgG2a an ibody sub ypes. Bo h GII.4-speci ic IgG1 (Fig. 2b) and IgG2a an ibodies (Fig. 2c) we e de ec ed cong uen ly o he IgG esponses seen in Fig. 2a, wi h signi ican ly highe end- poin i es induced when VP6 was co-adminis a ed wi h 03mg VLP (P<005 o all). The e we e no signi ican di - e ences in he sub ype-speci ic se um end-poin i es obse ed (P50125), indica ing balanced Th1- and Th2- ype esponses. VP6 adju an e ec on NoV-speci ic c oss- eac i e and blocking an ibodies The VP6 adju an e ec on c oss- eac i e se um IgG i es agains wo di e en a ian s o NoV GII.4 VLPs, GII.4 NO-2010 and GII.4 SYD-2012 VLPs was also in es iga ed (Fig. 3). C oss- eac i e IgG an ibodies ollowing immuni- za ion wi h 03mg GII.4 VLPs we e imp o ed signi ican ly (P<0005) by co-adminis a ion wi h 10 mg (g oup III) o 30 lg (g oup IV) o VP6T o 10 lg VP6S (g oup VII). The VP6 adju an e ec on NoV GII.4-speci ic blocking an i- bodies was in es iga ed u he . Human ype A sali a and PGM we e used as he HBGA sou ces o NoV GII.4 VLP binding. The e was no VLP binding blocking (BT 50 525) by he mouse se a immunized wi h he subop imal 03lg dose o VLP alone (Table 2). When VLPs we e co- adminis a ed wi h 10 o 30 lg o VP6T o VP6S, BT 50 o 100–200 we e obse ed (Table 2). As expec ed, due o he lack o GII.4-speci ic se um an ibodies, he 10 mgo VP6T Fig. 3. No o i us (NoV) c oss- eac i e immunoglobulin (Ig)G an ibodies. Indi idual, se ially dilu ed se a o immunized mice we e analysed o c oss- eac i e an i-GII.4 New O leans (NO) and an i-GII.4 Sydney (SYD) IgG end-poin i es. An a bi a y end-poin i e o 50 (hal he s a ing dilu ion, 1 : 100) was assigned o nega i e mice se a. Ba s ep esen log 10 -exp essed geome ic mean i es o each g oup wi h 95% con idence in e als. *Signi ican ly di e en (P<005) an ibody i es compa ed o g oup III. Table 2. Blocking an ibody i es in immunized mouse se a G oup Immuniza ion Sali a HBGA PGM HBGA BT 50 BT 50 I03mg VLP 25 a n. . II 03mg VLP 11mg VP6T 50 50 III 03mg VLP 110 mg VP6T 200 200 IV 03mg VLP 130 mg VP6T 200 100 V03mg VLP 110 mg VP6T (1 h) 50 50 VI 03mg VLP 110 mg VP6T (con ala e al si es) 25 n. . VII 03mg VLP 110 mg VP6S 100 100 VIII 3 mg VLP 400 n. . IX (C l) Ca ie only (PBS) 25 25 a BT 50 o 25 was assigned o samples lacking blocking a a se um dilu ion 1:50. HBGA 5his o-blood g oup an igen; PGM 5pig gas ic mucin; BT 50 5 i es wi h 50% blocking o no o i us (NoV) GII.4-1999 VLP binding; VLP 5 i us-like pa icle; n. . 5no es ed; VP6T 5VP6 nano ubes; VP6S 5VP6 nanosphe es; C l 5con ol; PBS 5phospha e-bu e ed saline. M. Malm e al. 336 V C2017 The Au ho s. Clinical & Expe imen al Immunology published by John Wiley & Sons L d on behal o B i ish Socie y o Immunology, Clinical and Expe imen al Immunology, 189: 331–341 and 03lg VLP adminis e ed a he con ala e al si e (g oup VI) as well as he con ol g oup (g oup IX) did no induce blocking an ibodies (Table 2). VP6 adju an e ec on NoV-speci ic T cell esponses NoV GII.4-speci ic T cell esponses we e analysed by s imu- la ing he splenocy es o immunized mice wi h GII.4-1999- speci ic pep ide pool and GII.4-1999 VLPs and measu ing IFN-gp oduc ion in he ELISPOT assay (Fig. 4). Immuni- za ion o mice wi h a 03mg dose o GII.4 VLPs alone did no induce de ec able IFN-gp oduc ion, bu when co- adminis a ed wi h 10 mg o VP6T o VP6S a conside able NoV GII.4-speci ic IFN-gsec e ion was obse ed (Fig. 4). The e was no signi ican di e ence in he o e all IFN-g p oduc ion by he T cells o mice immunized wi h VP6T o VP6S (P>0.05). Mice immunized wi h he 10 mg o VP6T and 03lg VLP a he con ala e al si es (g oup VI, Table 1) we e nega i e o all an igens es ed in he ELISPOT IFN-g(da a no shown), con i ming he local adju an e ec o VP6 as seen o he NoV GII.4-speci ic an ibody esponses (Fig. 2a). RV VP6-speci ic se um an ibody esponses Each g oup o mice immunized wi h VP6 (g oups II–VII, Table 1) a di e en doses o di e en oligome ic s uc u es de eloped RV VP6-speci ic se um IgG an ibodies, indica - ing success ul immuniza ion (Fig. 5a). In e es ingly, e y high VP6-speci ic IgG end-poin i es (GMT 60887) we e induced in mice immunized wi h GII.4 VLPs and VP6T a he con ala e al si es (g oup VI, Table 1), uling ou he possibili y ha a lack o VP6 adju an e ec obse ed in Fig. 4. No o i us (NoV) GII.4-speci ic in e e on (IFN)-g p oduc ion by T cells. NoV GII.4-1999-speci ic pep ide pool, GII.4- 1999 i us-like pa icles (VLPs) and o albumin (OVA) pep ide (nega i e con ol) we e used o s imula ing IFN-gp oduc ion o he mice immunized wi h 03mg VLP alone o co-adminis a ed wi h 10 mg VP6 nano ubes (VP6T) o nanosphe es (VP6S). Nega i e con ol mice (C l) we e immunized wi h phospha e- bu e ed saline (PBS) only. Resul s a e exp essed as he mean spo - o ming cells (SFC) pe 10 6 splenocy es o he duplica e wells wi h s anda d e o s o he mean. The expe imen s we e epea ed wo o mo e imes wi h simila esul s. Ro a i us VP6 ubula and sphe ical nanos uc u es as local adju an s V C2017 The Au ho s. Clinical & Expe imen al Immunology published by John Wiley & Sons L d on behal o B i ish Socie y o Immunology, Clinical and Expe imen al Immunology, 189: 331–341 337 his g oup migh esul om an unsuccess ul immuniza- ion wi h VP6. RV VP6-speci ic T cell esponses ELISPOT IFN-g(a hallma k o a Th1- ype esponse) and IL-4 (a hallma k o a Th2- ype esponse) we e used o ana- lyse RV VP6-speci ic T cell esponses in splenocy es o immunized mice o VP6 de i ed R6-2 pep ide epi ope, RV cell cul u e an igens (Wa and WC3) o VP6 p o ein. IFN- gp oduc ion o all an igens bu o a mock an igen was de ec ed in mice immunized by co-adminis a ion o 03lg NoV GII.4 VLPs and 10 lg VP6T o VP6S (Fig. 5b). Immuniza ion wi h VP6T induced somewha highe IFN-g p oduc ion han VP6S, bu he di e ence was no s a is i- cally signi ican (P50073). When he same cells we e es ed o VP6-speci ic IL-4 p oduc ion, a obus IL-4 esponse (up o 1016 SFC/10 6 cells) was obse ed in he g oup ha ecei ed 03mg NoV VLPs co-adminis e ed wi h 10 mg o VP6T (Fig. 5c). IL-4 p oduc ion was also induced in he g oup ha ecei ed 10 mg o VP6S (Fig. 5c) ins ead o VP6T; howe e , he esponse was signi ican ly highe in he g oup ha ecei ed VP6T (P50.008). No RV VP6- speci ic IFN-go IL-4 was sec e ed by he splenocy es o mice immunized wi h 03mg GII.4 VLP alone (da a no shown) o he con ol mice (Fig. 5b,c). Discussion We ha e demons a ed ecen ly an in- i o adju an e ec o VP6T on he immunogenici y o NoV VLPs [17]. In addi ion, in- i o s udies ha e shown ha VP6 induces APC ac i a ion and ma u a ion [36]. In he p esen s udy, he mechanism o VP6 adju an ac ion was in es iga ed u he o explo e he e ec o co-localiza ion o he wo an igens, NoV VLPs and VP6, ia co-deli e y as a mix u e o as sepa a e injec ions a he same o con ala e al injec- ion si es. An adju an is de ined as a compound ha enhances he immune esponse o accine an igens. The adju an s may be ega ded as depo s o ehicles o /and immunomodula- o y agen s. A depo e ec acili a es and imp o es deli e y o an igens o APCs. We he e o e in es iga ed i he VP6 unc ions as a ca ie o deli e y ehicle ha would necessi- a e he VLP and VP6 co-deli e y as a mix u e, ins ead o sepa a e injec ions. The esul s demons a ed clea ly ha he adju an e ec o VP6 is s ic ly dependen upon co- localiza ion o he VP6 wi h NoV GII.4 VLPs. To accom- plish he op imal VP6 adju an e ec on NoV VLP immu- nogenici y i was essen ial o co-adminis a e he VP6 wi h he NoV VLPs, as bo h empo al and spa ial dissocia ion o VP6 and NoV VLPs impai ed o comple ely abolished he adju an e ec . A 1-h empo al di e ence in he adminis- a ion o VP6T and NoV VLPs impai ed he adju an e ec d ama ically, u he suppo ing he ole o VP6 as a deli e y ehicle. I has been published ha pa icula e an i- gens a el e y quickly, in e ms o minu es, om he si e o injec ion o he local lymphoid issue [46]. Al hough his s udy does no ex end o con i m his, he esul s sug- ges ha RV VP6 and VLPs may o m agg ega es when co- o mula ed, VP6 unc ioning as a ca ie . Indeed, ou ecen ly published esul s showed ha NoV VLP up ake in o he APCs was inc eased when mixed wi h he VP6 [36]. In addi ion o ac ing ini ially as he deli e y ehicle o NoV VLPs he VP6 adju an e ec is p obably en o ced u he by APC ac i a ion. A e being deli e ed, VP6 may induce local cy okine and chemokine p oduc ion and ac i- a ion and ma u a ion o APC a he si e, as shown by ou ea lie in- i o s udies [36]. P oin lamma o y cy okines, such as umou nec osis ac o (TNF)-a, IL-6, IL-1 and g anulocy e mac ophage colony-s imula ing ac o (GM- CSF), enhance he adap i e immune esponse in se e al ways, including ec ui men o APC a he si e o injec ion, s imula ing hei ma u a ion and mig a ion o lymph nodes (LN) [47,48]. The e o e, he VP6 deli e y p obably leads o an inc eased numbe o NoV VLP-loaded APCs in he LN d aining he injec ion si e able o ac i a e an igen- speci ic T cells. Th ee ypes o DCs, con en ional and monocy e-de i ed in s eady s a e skele al muscles a e a ge ed by i.m. immuniza ion, which can encoun e he an igen and mig a e o LN [49]. Fu he mo e, VP6 ac i- a ion o T cells in lymphoid issues induces cy okine p oduc ion by hese cells, such as he obse ed IFN-g,a Th1- ype cy okine and IL-4, a Th2- ype cy okine. Pa a- c ine sec e ion o Th1 and Th2 cy okines by he VP6- speci ic T cells is o signi icance, as hese cy okines d i e p oli e a ion and di e en ia ion o co-deli e ed an igen; in his case NoV VLPs p imed nai e B and T lympho- cy es in o memo y cells [50,51]. The localized na u e o he VP6 adju an ici y is suppo ed by he obse a ion Fig. 5. Ro a i us (RV) VP6-speci ic an ibodies and T cell esponses. VP6-speci ic immune esponses we e de ec ed ollowing immuniza ion wi h 03mg no o i us (NoV) GII.4 i us-like pa icles (VLPs) oge he wi h VP6 nano ubes (VP6T) o nanosphe es (VP6S). Con ol (C l) mice ecei ed phospha e-bu e ed saline (PBS) only. (a) Se um an i-VP6 IgG o indi idual mice we e es ed in enzyme-linked immunoso ben assay (ELISA) and he mean i a ion cu es o each expe imen al g oup a e shown. Ba s ep esen log 10 geome ic mean i es wi h 95% con idence in e als. Fo nega i e mice se a, an a bi a y i e o 1 : 100 (hal he s a ing se um dilu ion, 1 : 200) was assigned. VP6-speci ic IFN-g(b) and in e leukin (IL)-4 (c) p oduc ion by T cells was es ed s imula ing he cells wi h VP6-speci ic R6-2 pep ide, RV Wa and WC3 cell cul u e an igens, mock an igen o ecombinan VP6 p o ein ( VP6). Resul s a e exp essed as he mean spo - o ming cells (SFC)/10 6 splenocy es o he duplica e wells wi h s anda d e o s o he mean. The expe imen s we e epea ed wo o mo e imes wi h simila esul s. 3 M. Malm e al. 338 V C2017 The Au ho s. Clinical & Expe imen al Immunology published by John Wiley & Sons L d on behal o B i ish Socie y o Immunology, Clinical and Expe imen al Immunology, 189: 331–341 ha VP6 adminis a ion a he con ala e al si e did no con ibu e o he NoV-speci ic immuni y, e en hough high VP6-speci ic immune esponses we e obse ed. Mo eo e , se um collec ed a 3 and 24 h pos -VP6T immuniza ion was nega i e o p oin lamma o y cy o- kines TNF-aand IL-6 by ELISA (da a no shown), a guing agains a sys emic adju an e ec . The esul s o he p esen s udy show ha VP6 p o- mo es bo h Th1- and Th2- ype esponses o NoV wi hou skewing he o e all immune esponse in any pa icula di ec ion. The adju an s a e mainly es ic ed o enhance ei he ype o he esponses, such as aluminium sal s, which p omo e mainly Th2- ype esponses o he co- adminis e ed accine an igens [52]. The pa icle size and shape a e shown o be cen al in he an igen in e naliza ion and anspo o LN and immune cell ac i a ion [48,53–55]. Addi ionally, he pa icle size and shape may play a ole in uning he adap i e immuni y, as od-shaped pa icles ha e been epo ed o induce Th2-biased esponses compa ed o sphe ical pa icles [56]. Howe e , al hough we obse ed mo e IL-4 p oduc ion by he VP6T- speci ic T cells compa ed o VP6S, bo h hese nanos uc- u es exe ed a simila adju an e ec on NoV GII.4-spe- ci ic an ibody esponses. Ou esul s show, o he i s ime, he VP6 adju an e ec on NoV-speci ic T cell immuni y. I has been pub- lished ha VP6 can po en ia e he se um an ibody esponse agains RV in a mouse model by an in e molecula help mechanism [57]. In he p esen s udy, as he VP6 is no conjuga ed o linked o he VLPs, bys ande T cell help is p obably an impo an mechanism. These cells may d i e p oli e a ion and di e en ia ion o NoV-an igen p imed B and T lymphocy es in o memo y cells and an ibody- sec e ing plasma cells ei he by cell- o-cell con ac (CD40-CD40L in e ac ion) [58] o soluble cy okine p o- duc ion [59]. Al oge he , his s udy shows clea ly ha co- adminis a ion o he VP6 wi h NoV VLPs is essen ial o he op imal adju an e ec on NoV-speci ic immune esponses and ha he VP6 ac s as a deli e y ehicle as well as immunomodula o . Sa e y is a majo conce n when i comes o adju an app o al o human use, especially o heal hy in an s [16]. The esul s o his s udy suppo he use o VP6 no only o p o ide p o ec ion agains RV in he p oposed NoV VLP–RV VP6 combina ion accine [9,10], bu also as an adju an o po en ia e he NoV VLP immunogenici y. Acknowledgemen s We g a e ully acknowledge Ki si Tamminen and he labo a- o y pe sonnel o he Vaccine Resea ch Cen e a he Uni e - si y o Tampe e o echnical assis ance gi en. No ex e nal unding was ecei ed o his s udy. Au ho con ibu ions M. M. conduc ed immuniza ions, sample acquisi ion and p ocessing and labo a o y analysis, including da a acquisi- ion, analysis and in e p e a ion (ELISA, blocking assays), w i ing he manusc ip and inal app o al o he e sion o be submi ed. S. H. conduc ed immuniza ions, sample acquisi ion, p ocessing and labo a o y analysis, da a acqui- si ion and analysis (p o ein and i us p oduc ion, ELI- SPOT assays). Final app o al o he e sion o be submi ed. T. V. is he head o he Vaccine Resea ch Cen e and conduc ed c i ical e iew o he pape and ga e inal app o al o he e sion o be submi ed. V. B. is he head o he labo a o y and co esponding au ho , and conduc ed he concep ion and designing he s udy, da a in e p e a- ion, d a ing and w i ing he pape , c i ical e ision o he pape o impo an in ellec ual con en and inal app o al o he e sion o be submi ed. Disclosu e All he au ho s decla e no disclosu es. Re e ences 1 Puus inen L, Blaze ic V, Salminen M, H€ am€ al€ ainen M, R€ as€ anen S, Vesika i T. No o i uses as a majo cause o acu e gas oen e - i is in child en in Finland, 2009–2010. Scand J In ec Dis 2011; 43:804–8. 2 Pa el MM, S eele D, Gen sch JR, Wecke J, Glass RI, Pa asha UD. Real-wo ld impac o o a i us accina ion. Pedia In ec Dis J 2011; 30:S1–5. 3 Dennehy PH. Vi al gas oen e i is in child en. Pedia In ec Dis J 2011; 30:63–4. 4 Jiang X, Wang M, G aham DY, Es es MK. Exp ession, sel - assembly, and an igenici y o he No walk i us capsid p o ein. J Vi ol 1992; 66:6527–32. 5 Vesika i T. Ro a i us accina ion: a concise e iew. Clin Mic o- biol In ec 2012; 18 (Suppl 5):57–63. 6 Desselbe ge U. Ro a i uses. Vi us Res 2014; 190:75–96. 7 Bu e y JP, Danchin MH, Lee KJ e al. In ussuscep ion ollowing o a i us accine adminis a ion: pos -ma ke ing su eillance in he Na ional Immuniza ion P og am in Aus alia. Vaccine 2011; 29:3061–6. 8 Dona o CM, Ch’ng LS, Boni ace KF e al. Iden i ica ion o s ains o Ro aTeq o a i us accine in in an s wi h gas oen e i- is ollowing ou ine accina ion. J In ec Dis 2012; 206:377–83. 9 Blaze ic V, Lappalainen S, Nu minen K, Huh i L, Vesika i T. No o i us VLPs and o a i us VP6 p o ein as combined accine o childhood gas oen e i is. Vaccine 2011; 29:8126–33. 10 Okada M, Mukai A, Nishino T, A inobu D, To H, Sa oh M. Pu i ica ion me hod o i us-like pa icles. 2013. Pa en appli- ca ion no. 2013-146240. Japan Pa en O ice, Japan. 11 Tamminen K, Lappalainen S, Huh i L, Vesika i T, Blaze ic V. T i alen combina ion accine induces b oad he e ologous immune esponses o no o i us and o a i us in mice. PLOS ONE 2013; 8:e70409. Ro a i us VP6 ubula and sphe ical nanos uc u es as local adju an s V C2017 The Au ho s. Clinical & Expe imen al Immunology published by John Wiley & Sons L d on behal o B i ish Socie y o Immunology, Clinical and Expe imen al Immunology, 189: 331–341 339