ucosal and systemic anti-GAG immunity induced by neonatal immunizationith HIV LAMP/gag DNA vaccine in mice
driana Letícia Goldonia, Milton Maciel Jr. e, Paula Ordonhez Rigatoa, Orlando Piubelli a,yro Alves de Britoa, Andrea Meloc, Ernesto Torres Marquesc,d, Joseph Thomas Augustb,lberto José da Silva Duartea, Maria Notomi Satoa,∗
Laboratory of Dermatology and Immunodeficiency – LIM56, School of Medicine, University of São Paulo, BrazilDepartment of Pharmacology and Molecular Sciences, The Johns Hopkins School of Medicine, Baltimore, USALaboratório de Virologia e Terapia Experimental do Centro de Pesquisas Aggeu Magalhães-CPqAM, Fiocruz, BrazilCenter for Vaccine Research, Department of Infectious Diseases and Microbiology, USACenter for Vaccine Development, University of Maryland, Baltimore, USA
r t i c l e i n f o
rticle history:eceived 13 July 2010eceived in revised form 25 August 2010ccepted 26 August 2010
eywords:NA vaccinesIViceucosaleonates
a b s t r a c t
Vaccines capable of inducing mucosal immunity in early postnatal life until adulthood, protecting earlysexual initiation, should be considered as strategies to vaccination against HIV. The HIV-1 GAG proteinas a chimera with the lysosome-associated membrane protein (LAMP/gag), encoded by a DNA vaccine, istargeted to the endosomal/lysosomal compartment that contains class II MHC molecules and has beenshown to be immunogenic in adult mice. Assuming that one such strategy could help to overcome theimmunological immaturity in the early postnatal period, we have evaluated the systemic and mucosalimmunogenicity of LAMP/gag immunization in neonatal mice. Intranasal immunization with LAMP/gagvaccine induced higher levels of sIgA and IgG anti-GAG antibodies in intestinal washes than did the gagvaccine. The combination of ID injections and the IN protocol with the chimeric vaccine promoted theincrease of Ab levels in sera. Both vaccines induced splenic IFN-�− secreting cells against GAG peptide
pools, as well as in vivo cytotoxic T lymphocyte (CTL) function, and increased the percentage of CD8+T cells to the immunodominant class I peptide in gut and spleen. However, only the chimeric vaccinewas able to enhance Th1/Th2 cytokine secretion in response to class II GAG peptide and to enhance IL-4-secreting cells against GAG peptides and p24 protein stimuli. Long-lasting humoral and cellular responseswere detected until adult age, following neonatal immunization with the chimeric vaccine. The LAMP/gag
duced an
vaccination was able to inessential to elicit sustaine
ntroduction
The estimated average number of new HIV-1 infections inhildren under the age of 15 has been reported as 430,000
240,000–610,000] (UNAIDS 2008). Women and girls continue toe disproportionately affected by HIV in sub-Saharan Africa. In theine most affected countries of southern Africa, HIV prevalencemong young women, aged 15–24 years, was on average three
ononuclear spleen cells; sIgA, secretory IgA; nIm, non-immunized; Ag, antigen;b, antibody.∗ Corresponding author at: Laboratório de Dermatologia e Imunodeficiência, Fac-ldade de Medicina da Universidade de São Paulo, Instituto de Medicina Tropical,rédio II, Av Dr Enéas de Carvalho Aguiar, 500, 3◦ andar, 05403-000, São Paulo, Brazil.el.: +55 11 3061 7499; fax: +55 11 3081 7190.
times higher than among men of the same age group (Gouws etal. 2008). Therefore, strategies to develop vaccines able to inducelong-lasting protection against HIV, from early life until adulthood,to protect early sexual initiation and adolescent pregnancy shouldbe considered as vital measures to decrease HIV transmission.
Natural transmission of HIV is through the mucosal surface, suchas the vaginal and gastrointestinal mucosa. These mucosal sites areimportant virus reservoir, in which depletion of the memory CD4+T lymphocytes causes profound impact in hosts (Grossman et al.2006; Jiang et al. 2005; Mehandru et al. 2004). Depletion of CD4+T lymphocytes in the gut is mediated, at least in part, through thebinding of the HIV envelope gp120 protein to the �4�7 integrinmolecule on CD4+ T cells, which otherwise promotes T cells homing
to the gut (Arthos et al. 2008). Also, gp120 binds to the mucosaldendritic cells (DCs) through a C-type lectin (DC-SIGN), syndecan-3 and DCIR, leading to further CD4+ T infection and depletion (deWitte et al. 2007; Geijtenbeek and van Kooyk 2003; Lambert et al.2008). Therefore, induction of mucosal immunity not only by CD8+
but also by CD4+ T lymphocytes is crucial for the development offfective vaccination strategies.
It has been substantially demonstrated that DNA vaccine encod-ng the HIV-1 gag gene as a chimera with the lysosome-associated
embrane protein (LAMP), which is engaged in antigen traffickingo compartments that contain class II MHC molecules (Marques etl. 2003; Rowell et al. 1995; Ruff et al. 1997), is able to enhance0-fold the response of CD4+ T cells, increase epitope recognitiony T and B cells, enhance 4–5-fold the response of CD8+ T cells,nd to elevate by 50–100-fold antibodies (Abs) titers, as comparedo BALB/c adult mice immunized with gag vaccine (Arruda et al.006; Marques et al. 2003). Also, the chimera LAMP/gag vaccineas able to develop long-lasting memory B and T cells (de Arruda
t al. 2004). In Rhesus macaques, immunization with LAMP/gagNA vaccine by intramuscular route induced T and B cell response,
ncluding anti-GAG sIgA Abs in external secretions (Chikhlikar et al.006). Also in this animal model, the LAMP/gag chimera inducedD4+ T cell responses with effector memory phenotype (CD28−
D45RA−), whereas the native gag induced central effector mem-ry phenotype (CD28+ CD45RA−). In addition, the chimera inducedaster humoral responses than did the gag vaccine (Valentin et al.009).
Vaccination in the early postnatal period is a challenge dueo the high susceptibility to tolerance after Ag exposure and tommunological immaturity. There are qualitative and quantitativemmunological differences between neonates and adult counter-arts (Adkins et al., 2004), such as the impaired activation of T cellsy antigen-presenting cell (APC), skewed Th2 immune responsend low antibody production (Adkins 1999; Dadaglio et al. 2002).owever, when neonates are appropriately stimulated, they maychieve an adequate immune response that is comparable to adultice (De Brito et al. 2009; Siegrist 2001). Therefore, developing
accines able to direct antigen trafficking to the compartments con-aining class II MHC molecules, allowing CD4+ T cell activation, maye essential to trigger immunogenicity in the neonatal period.
aterials and methods
lasmids
HIV-gag plasmids were constructed by cloning the p55gagequence from the HXB2 strain of HIV (nucleotides 1-1503; Gen-ank K03455; HIV sequence Database, 1997, Los Alamos Nationalaboratory Theoretical Biology and Biophysics, Los Alamos, NM)nto the mammalian expression vector, pITR, which contains aMV promoter and ITR sequences from the adeno-associated virusanking the expression elements. The mouse LAMP-1 sequenceGenBank J03881) was also cloned in the same vector. TheAMP/gag constructed was made by inserting the p55gag sequenceetween the luminal domain and the transmembrane and cytoplas-ic tail of LAMP-1, as previously described (Marques et al. 2003).
ice immunization protocol
BALB/c mice of both sexes (8–10-weeks old) obtained from thenimal facilities of the Animal Laboratory (CEMIB, UNICAMP, Sãoaulo, Brazil) were mated and offspring were used. All experimentsere approved by the Ethics Committee for Animal Research of
he Institute of Biomedical Sciences. Some experiments were per-ormed with intranasal (IN) administration of 20 �g of the DNA
accines LAMP/gag, gag or pITR at days 7, 21 and 31 after birth. Thismmunization schedule (3 IN doses with either one of the three DNAaccines) was associated with two doses of 5 �g of LAMP/gag, gagr pITR by intradermal (ID) route at days 7 and 21, and mice werevaluated at 38 d.o.. Mice were evaluated for immunization lasting
ogy 216 (2011) 505–512
effect 8 months after the neonatal schedule, and boosted with thesame DNA vaccines received at neonatal period, 7 days before theanalysis.
Intestinal washes
Mouse small intestine samples were washed with 1.5 mL ofphosphate-buffered saline. The suspension was centrifuged at200 × g for 15 min at 4 ◦C and the supernatant was treated with pro-tease inhibitor cocktail (Calbiochem, CA, USA) and kept at −20 ◦Cuntil further analysis.
Antibody response by ELISA
Anti-GAG IgA and IgG Abs and total IgA and IgG immunoglobu-lin (Ig) were evaluated by enzyme-linked immunosorbent assay(ELISA). Briefly, 96-well microplates (Greiner Bio-One, FL, USA)were coated with 5 �g/mL HIVIIIB lysate (ABS, Rockville, MD),1 �g/mL HIV-1 p24 (kindly donated by Dr. Luis Carlos de Souza Fer-reira, University of São Paulo, Brazil) or 3 �g/mL goat anti mouseIgG or IgA (Southern Biotechnology Associates Inc, AL, USA) andincubated overnight at 4 ◦C. The plates were blocked with PBS con-taining 0.5% gelatin or 1% bovine serum albumin (BSA, Sigma, MO,USA) for 1 h at 37 ◦C, washed with PBS and incubated with theserial dilutions from samples or standards for 2 h at 37 ◦C. Plateswere then washed, incubated with biotinylated antibody anti-�1,anti-�2a, anti-� or anti-� (Southern) during 1 h at 37 ◦C. Later, theplates were incubated with streptoavidin peroxidase (Sigma) for1 h at 37 ◦C. The reaction was developed by tetramethylbenzidinesubstrate (Zymed, CA, USA) and stopped with 1 M sulphuric acid;plates were read at 450 nm using ELISA microplate reader (Molec-ular Devices, CA, USA). Total IgA and IgG levels were correlated topurified protein standards.
Detection of IFN-� and IL-4 producing cells by ELISPOT
Spleen and mesenteric lymph nodes (MLN) were obtained asep-tically and mashed at cell strainer (BD PharMingen, CA, USA) andmononuclear spleen cells (MSC) were obtained after cell cen-trifugation on Ficoll–Hypaque solution. Cells were washed twiceand resuspended in RPMI 1640 with 1% FetalClone III (HyCloneIII, UT, USA). Cell viability was greater than 90%. The number ofIFN-� and IL-4 producing cells from MLN, and from spleen, wasassessed by ELIspot assays (BD PharMingen), according to themanufacture’s protocol. Briefly, 96-well microplates with ELISPOTmicroplates (Millipore, MA, USA) were incubated with 5 �g/mLanti-IFN-� or anti-IL-4 at 4 ◦C overnight. After blocking with RPMI-1640 medium containing 10% fetal calf serum (FCS) for 2 h atroom temperature, cells (5 × 105 spleen cells/well or 1 × 106 MLNcells/well) were cultured in the presence of medium alone, anti-mouse CD3 antibodies (1 �g/mL, PharMingen), 12 pools (10 �g/mL)containing 123 peptides of 15-mer HIV-GAG with 11-amino acidoverlap between sequential peptides of the HIV-1 HXB2GAG pep-tide complete set [National Institutes of Health (NIH) Research andReference Reagent Program] and p24 (2 �g/mL) during 16 h (IFN-�) or 32 h (IL-4) at 37 ◦C and 5% CO2. HIV-1 Consensus Subtype BGAG (15-mer) peptide pools were organized as follows: pool 1, aa1–51; 2, aa 41–91; 3, aa 81–131; 4, aa 121–171; 5, aa 161–211; 6,aa 201–251; 7, aa 241–291; 8, aa 281–331; 9, aa 321–371; 10, aa361–411; 11, aa 401–451 and 12, aa 441–500.
The plates were washed and incubated with biotinylated
anti-IFN-� or anti-IL-4 for 2 h at room temperature, followedby HRP-conjugated avidin incubation for 1 h. After, the reactionwas developed with 3-amino-9-ethylcarbazole substrate (Cal-biochem). The data was evaluated considering the number ofspot forming cells (SFC)/1 × 106 for each stimulus counted in
A.L. Goldoni et al. / Immunobiology 216 (2011) 505–512 507
F oduction in neonate mice. Groups of BALB/c mice received IN doses of 20 �g of gag orL testinal washes were collected to evaluate the presence of sIgA (1:200) and IgG (1:160)A five to eight animals per group. *P ≤ 0.05 compared to gag immunized mice.
i(avc
C
mpcfdmi
Q
2id
P
cawXa
I
pitCTiAwmmwef
%
Table 1Immunoglobulin and TGF-� levels in intestinal washes of neonates mice immunizedwith DNA vaccines by IN route.
a Neonatal mice were non-immunized (nIm) or immunized by IN route with 20 �gof plasmid DNA vaccine encoding GAG or LAMP/GAG. Total sIgA and IgG concentra-
nal washes in both groups of IN immunized mice compared to
ig. 1. Intranasal LAMP/gag DNA immunization induces intestinal anti-GAG Ab prAMP/gag DNA vaccine at 7-, 21- and 31-days old. Seven days after the last dose, inbs against HIVIIIB lysate by ELISA. The data represent the mean value ± S.E.M. from
mmunospot imager analyzer using the software ImmunoSpot 3.2C.T.L ImmunoSpot® S4 Analyzer, C.T.L., OH, EUA). The number ofntigen-specific SFC was calculated by subtracting the backgroundalues. Cut-off value was determined as less than 10 SFC/1 × 106
ells to IFN-� and 20 SFC/1 × 106 cells to IL-4.
ytokine determination
MSC (4 × 106 cells/mL) cultures were distributed in 96-wellicroplates (Costar, MA, USA) with RPMI 10% FCS or 10 �g/mL GAG
eptide pool number 5 (aa 161–211) or number 7 (aa 241–291)ontaining class I or class II peptide, respectively, and incubatedor 72 h at 37 ◦C and 5% CO2. Supernatants were taken for cytokineetermination by cytometric bead array, using mouse Th1/Th2 andouse inflammation kits (BD Pharmingen, CA, USA), and analyzed
n BD FACSCaslibur cytometer (BD Biosciences).
uantification of Ag-specific CD8+ T cells and in vivo CTL assay
Groups of immunized mice received two IP injections of0 �g/mL MHC I-restricted GAG epitope AMQMLKETI65–73 (NIH)
n 50 �L of water-in-oil emulsion adjuvant (TiterMax® Gold) at 10ays interval and analyzed 5 days later.
entamer bindingMLN cells and MSC were harvested and incubated with PE-
onjugated H-2Kd/AMQMLKETI pentamer and FITC conjugatednti-CD8 (ProImmune, FL, USA) for 30 min. After wash, the cellsere analyzed by flow cytometry (250000 events in Coulter EpicsL flow cytometer) and the results were expressed as the percent-ge of pentamer+/CD8+ T cells.
n vivo CTLMLN and MSC isolated from non-immunized BALB/c mice were
rocessed and used as target cells. The cells were equally dividednto two populations, with high (10 �M) or low (1 �M) concen-ration of carboxyfluorescein succinimidyl ester (CFSE, Invitrogen,A, USA) in RPMI, and incubated in the dark for 30 min at 37 ◦C.he high-CFSE population was pulsed with 2 �g/mL GAG H-2Kd
mmunodominant peptide AMQMLKETI for 1 h at 37 ◦C and 5% CO2.fter incubation, cells were washed and 3 × 107 cells in 200 �Lere intravenously (IV) injected in immunized and control ani-als. MLN cells and spleen cells were harvested from recipientice 18 h after injection. The relative amount of transferred cellsas determined by the CFSE-positive cells detected by flow cytom-
try (FACSCalibur). Percent specific lysis was determined by theollowing formula
specific lysis = 100 −{[
(% CFSE-high population/% CFSE-low populat(% CFSE-high population/% CFSE-low popu
tion and TGF-� from intestinal washes were determined by ELISA. The data representthe mean value ± S.E.M from 5–8 animals per group.
* P ≤ 0.05 compared to gag immunized mice.# P ≤ 0.05 compared to non-immunized mice.
Statistical analysis
Data are expressed as mean values ± standard errors of allexperiments. A P value ≤ 0.05 was considered to be statistically sig-nificant as determined by Kruskall–Wallis test with Dunn’s post testor Mann–Whitney test.
Results
Mucosal and systemic anti-GAG Ab response to LAMP/gag DNAvaccine in neonatal period
We have verified the immunogenicity of the chimeric LAMP/gagDNA vaccine to induce mucosal immunity in the early neona-tal period. Neonatal BALB/c mice (7 d.o.) received intranasal (IN)doses of 20 �g DNA vaccines at 7-, 21- and 31-days old, and thepresence of anti-GAG antibody was subsequently evaluated in thesmall intestinal washes. Fig. 1 shows that IN immunization withLAMP/gag at neonatal period led to increased levels of mucosalanti-GAG sIgA and IgG Abs as compared to the gag immunizedgroup.
Intranasal immunization with 20 �g DNA vaccines showed thatmucosal sIgA levels increased by at least 2-fold in both vaccinatedmice groups, compared to non-immunized (nIm) group (Table 1).Moreover, total IgG immunoglobulin amounts were significantlyenhanced in the intestinal washes of LAMP/gag immunized mice,compared to the gag vaccinated group, decreasing the sIgA/IgGratio. Levels of TGF-� were measured for its relevance as a reg-ulatory cytokine in the gut and its role in the IgA switching. Thedata showed a remarkable increase in TGF-� levels in the intesti-
ion) from immunized mouselation) from naive mouse
]× 100
}non-immunized mice (Table 1).
508 A.L. Goldoni et al. / Immunobiology 216 (2011) 505–512
0.0
0.5
1.0
1.5
2.0 pITR
gagLAMP/gag
IgG1
*
*
21 31 38 d-o
O.D
. (4
50n
m)
0.0
0.5
1.0
1.5
2.0IgG2a
21 31 38 d-o
*
F mice received three IN doses of 20 �g of either DNA vaccine and two doses of 5 �g of thes 8 d.o. Antibodies IgG1 (1:1350) and IgG2a (1:1350) against HIV-1 p24 were evaluated byE P ≤ 0.05 compared to gag immunized mice.
ttDmheaistLcir
Hw
tmWcGstnp
iporp12biio(ft(
LMtM
Fig. 3. Neonatal immunization with DNA vaccines induces IFN-� and IL-4 secretingcells to GAG peptides from gut and spleen. BALB/c mice received three doses of 20 �gof either DNA vaccine by IN route and two doses of 5 �g of the same DNA vaccineby ID route at 7 and 21-days old. Seven days after the last IN dose, spleen cells were
ig. 2. Systemic anti-GAG Ab production induced by LAMP/gag vaccination. BALB/came DNA vaccine by ID route at 7 and 21-days old. Mice were bled at 21, 31 and 3LISA. The data represent the mean value ± S.E.M. from 9 to 10 animals per group. *
None of the immunization protocols using only IN adminis-ration induced systemic Ab production; thus we associated thewo ID injection protocol along with the IN immunization for bothNA vaccines. The ID route is the usual choice for adult BALB/cice immunization with chimeric DNA vaccine, when cellular and
umoral responses are desirable (Marques et al. 2003; de Arrudat al. 2004). The schedule of three IN administration (20 �g each)nd two ID injections (5 �g each) showed that only the LAMP/gagmmunization was able to induce anti-p24 IgG1 and IgG2a Abs inerum, in which case, IgG1 appeared first and more intensivelyhan IgG2a (Fig. 2). These results show that DNA vaccination withAMP-mediated antigen targeting to the endosomal/lysosomalompartment plays an essential role at neonatal period ands necessary to trigger both mucosal and systemic humoralesponses.
igh CD4+ T threshold in mucosal and systemic immunizationith LAMP/gag DNA vaccine
We have further verified whether the LAMP/gag and gag neona-al immunization with IN–ID associated protocol had induced
ucosal and/or systemic anti-GAG CD4+ T and CD8+ T responses.e evaluated spleen cells for the frequency of spot forming
ells (SFC), producing IFN-� and IL-4 by ELISPOT against theAG peptide pools (n = 13) – encompassing the entire p55GAGequence – and against p24. For the MLN cells, we evaluatedhe number of IFN-�-secreting SFC against the immunodomi-ant class I restricted GAG peptide (aa 197–211) and against24.
A broader IFN-� T-cell response was verified for LAMP/gagmmunized mice, whose spleen cells recognized five GAG peptideools, whereas spleen cells from the gag immunized mice rec-gnized only two pools (Fig. 3A). Among the GAG peptide poolsecognized by spleen cells from the LAMP/gag vaccinated mice,ools number 4 (aa 121–171), 7 (aa 241–291), 8 (aa 281–331) and0 (aa 361–411) included peptides presented by class II MHC H-Kd molecules, which were previously described to be recognizedy CD4+ T cells (Arruda et al. 2006). The five peptide pools induc-
ng IFN-� secreting spleen cells from LAMP/gag immunized micencluded four pools containing class II MHC binding peptide andne pool containing peptides presented by class I MHC moleculesFig. 3A). Interestingly, IL-4 SFC spleen cells were only observedor LAMP/gag immunized mice, whose spleen cells responded tohe pools containing class II MHC binding peptide and p24 proteinFig. 3B).
At mucosal compartment, both DNA vaccines, gag andAMP/gag, were able to induce similar frequency of IFN-�-secretingLN cells against p24 or GAG peptide (Fig. 3C). Against p24, in par-
icular, similar profiles of IFN-� SFC were generated by splenic andLN cells, and to a lesser extent by MLN cells (Fig. 3C).
evaluated by ELISPOT assay for (A) IFN-� and (B) IL-4 SFC against pools of peptidesfrom GAG protein and p24, and MLN cells were evaluated for (C) IFN-� SFC stim-ulated by class I peptides (aa 197–211) and p24 protein. The data represent meanvalue ± S.E.M. from 10 or 4 experiments for IFN-� and IL-4, respectively, with poolsof cells from two to three animals per group. *P ≤ 0.05 compared to gag immunizedmice.
A.L. Goldoni et al. / Immunobiology 216 (2011) 505–512 509
Fig. 4. Immunization with LAMP/gag enhanced in vitro cytokine secretion to class I and class II GAG peptides stimulation. BALB/c mice received three doses of 20 �g of eitherD ne byc 211) ob secreg
tnscotitcCic
Ai
tsit
l(mgfftt(prv
Lv
nabb
m
NA vaccine by IN route combined with two doses of 5 �g of the same DNA vacciultured for 72 h in the presence of the GAG peptide pool containing class I (aa 161–ead array. The results represent the mean value ± S.E.M. of the difference betweenroup. *P ≤ 0.05 compared to gag immunized mice.
Using cytometric bead array, we evaluated T-cell activationhrough the production of spleen cytokines in response to poolumber 5 (aa 161–211) or number 7 (aa 241–291) stimuli, pre-enting class I and class II MHC peptides, respectively. The in vitroellular evaluation of the immunized groups, with either LAMP/gagr gag, showed similar high levels of IFN-� secretion in responseo class I stimulus (Fig. 4). It is noteworthy that only the chimerammunized mice showed significant CD4+ T activation, leadingo increased levels of TNF-�, IFN-�, IL-2 and IL-6 secretion, uponlass II stimulus as compared to the gag immunized group (Fig. 4).ytokine secretion by spleen cells from the gag immunized mice
nduced by class II peptides was barely detected as so was theontrol from the pITR immunized mice.
nti-GAG CD8+ T cell response in mucosal and systemicmmunization with LAMP/gag DNA vaccine
To address the anti-GAG CD8+ T cell response, following neona-al DNA vaccination, the frequency of MHC pentamer+ CD8+ T cellspecific to the immunodominant class I peptide (aa 197–211) andn vivo cytotoxicity were assessed by flow cytometry, labeling thearget cells with caboxyfluorescein succinimidyl ester (CFSE).
A similar frequency of antigen-specific CD8+ T cells stimu-ated by MHC Class I pentamer from MLN (Fig. 5A) and spleenFig. 5B) cells was detected for gag and LAMP/gag immunized
ice groups. The in vivo cytotoxic T cell activity against the tar-et cells pulsed with peptide 197–211 revealed that they wererom the gut compartment in both immunized groups differentlyrom the non-immunized group (Fig. 5C and E). On the other hand,he CTL function by spleen cells was increased only in those fromhe chimera vaccine group but not from the non-immunized miceFig. 5D and F). The results showed that the IN/ID immunizationrotocol with either DNA vaccine equally induced CD8+ T mucosalesponse; whereas in the case of systemic response, the LAMP/gagaccine was more effective.
ong-term response following neonatal gag and LAMP/gagaccination
We assessed whether the neonatal vaccination schedule begin-ing at 7 d.o. would induce lasting immune response until the adult
ge. After 8 months of neonatal immunization, mice were boostedy IN/ID routes to assess specific Ab and cellular responses fromoth compartments MLN and spleen cells.
As expected, the humoral response of IgG1 Ab to GAG wasaintained at high levels only in the LAMP/gag immunized group
ID route at 7 and 21-days old. Seven days after the last IN dose, splenocytes werer class II stimulus (aa 241–291); cytokine secretion was determined by cytometricted cytokine upon stimulus and spontaneous production from 9 to 13 animals per
(Fig. 6A). The frequency of IFN-� secreting MLN and spleen cells,stimulated with p24 or 197–211 peptide pool, was similar in bothgag and LAMP/gag immunized groups (Fig. 6C and D). Cytokineevaluation upon class I stimuli showed similar production of IFN-�and IL-6 between LAMP/gag and gag vaccination; whereas uponclass II stimuli, only cells from the LAMP/gag immunized miceinduced these cytokines (Fig. 6E and F).
Discussion
The present study showed that immunization with theLAMP/gag DNA vaccine in neonatal mice was able to inducemucosal and systemic long-lasting immune response through T andB cells. The immunogenic role of the chimera vaccine in the neona-tal period was a consequence of the CD4+ T activation, enhancingrecognition of the GAG peptides and thus creating favorable con-ditions to cytokine secretion and anti-GAG antibody productionupon stimulation by class II GAG peptides. Both DNA vaccineswere able to generate mucosal GAG-specific CD8+ T cells and CTLresponse; however, the in vivo systemic CTL function was mainlyraised among the LAMP/gag immunized group.
The vaccination by intranasal route in the neonatal period over-came the undesirable effect of antigenic tolerance, in which caseonly the chimeric vaccine induced both anti-GAG sIgA and IgG Ablevels, as determined in intestinal washes. Besides the Ag-specificAb, increased levels of total sIgA in the gut were triggered byboth vaccines, but only the chimeric vaccine enhanced IgG levels.Vaccine administration by IN route led to an efficient Ag presenta-tion and T–B cell activation, resulting in cytokine production andincrease in the polyclonal IgG level, probably due to a bystandereffect. Moreover, we verified that TGF-� levels in gut washes fromLAMP/gag and gag immunized groups were 22-fold higher thanfrom non-immunized mice, what may explain the increased levelsof sIgA found in this compartment, following IN immunization withboth DNA vaccines. TGF-� is essential in the regulation of IgA iso-type switching, as an anti-inflammatory cytokine that plays a keyrole in the maintenance of the immune system homeostasis, andin the generation and function of CD4 + CD25+ regulatory T cells(Cazac and Roes 2000; Huber et al. 2004).
Gag-specific mucosal immune responses of intramuscularlyimmunized Rhesus macaques with human LAMP/gag chimera vac-
cine showed that HIV-1-specific IgA antibodies were present in theplasma and external secretions, including nasal washes (Chikhlikaret al. 2006). However, in mouse, we had to include two ID injectionsin the IN protocol to simultaneously induce systemic and mucosalanti-GAG IgG response. The ID route had already been shown to effi-
510 A.L. Goldoni et al. / Immunobiology 216 (2011) 505–512
Fig. 5. Systemic and mucosal CD8+ T cell response after neonatal immunization with DNA vaccine. BALB/c mice received three doses of 20 �g of either DNA vaccine by INroute combined with two doses of 5 �g of the same DNA vaccine by ID route at 7- and 21-days old. Seven days after the last IN dose, the animals received two IP injectionsof AMQMLKETI peptide with adjuvant. (A and B): pentamer+ CD8+ H-2Kd/AMQMLKETI T cells in MLN and spleen were analyzed by flow cytometry. The data represent themean value ± S.E.M. from 5 to 10 animals per group; (C and D): CTL function: Target cells were stained with two different concentration of CFSE; the CFSEhigh target cellswere pulsed with AMQMLKETI peptide. Target cells were injected by IV route into vaccinated (bold line) or control (thin line) mice. After 18 h, MLN cells and splenocytesw ntagefi LN anc
cmDsta(aoso(giitvmI(
ere harvested and analyzed by flow cytometry to evaluate the in vivo killing perceve to seven animals per group. E and F histograms show in vivo CTL response by Mompared to non-immunized (nIm) mice.
iently induce high antibody and cellular immune response in adultice (de Arruda et al. 2004; Marques et al. 2003). Naked plasmidNA is a safe vaccine vector, for it is neutral, nontoxic, and should be
afe for use in newborns (Donnelly et al. 1997). DNA vaccines havehe advantage of attenuated live vaccines in their immunogenicity,nd a level of biological safety similar to that of inactivated vaccinesDonnelly et al. 1997). Furthermore, in utero fetal immunization byDNA vaccine delivered into the amniotic fluid in the oral cavityf fetal lambs has been showed to be immunogenic providing aafe and effective method for preventing or reducing the high riskf vertical disease transmission during pregnancy and after birthGerdts et al. 2000). However, DNA vaccines are much less immuno-enic than the adenoviral vector (Bessis et al. 2004), which is able tonduce Ab in serum and mucosal secretions upon intranasal or oralmmunization in neonatal mice (Xiang et al. 2003). Furthermore,
he absence of humoral response, when the LAMP/gag and gag DNAaccines were given by IN route only, cannot be explained by theouse immunological immaturity, in view of the fact that the same
N protocol did not induce systemic response in adult mice eitherdata not shown). Combination of IN doses with ID injections gen-
of target cells pulsed with peptide. The data represent the mean value ± S.E.M. fromd spleen cells, respectively, from mice immunized with chimeric vaccine. *P ≤ 0.05
erated high levels of anti-p24 IgG Ab in serum, mainly of the IgG1subclass – followed by the IgG2a – due to the capacity of the LAMPprotein to direct endogenous proteins to compartments containingclass II MHC molecules.
The intranasal delivery of genetic vaccines has been shownto be limited to induce adequate mucosal antibody responses,thus requiring the use of adjuvants, such as the chemokine adju-vant CCL27/CTACK (Kraynyak et al. 2010), lipid adjuvant (Braveet al. 2008), ligants of TLR (Deml et al. 1999; Kojima et al. 2002),cytokines, or cytokine genes (Dale et al. 2004; Rollman et al. 2004).We have shown herein that the chimeric vaccine LAMP/gag wasable to efficiently induce systemic and mucosal antibody responses;this vaccine efficiently enhanced anti-GAG antibody production –as a consequence of the appropriate antigen presentation – andactivated CD4+ T cells, stimulating a broad clonal expansion of B
cells.
The high degree of CD4+ T cell activation in mouse was obtainedonly with the LAMP/gag immunization, as shown by the broadenrecognition of the GAG peptide pools by IFN-� SFC, of the responseto p24 protein by IL4-secreting cells, and the cytokine production
A.L. Goldoni et al. / Immunobiology 216 (2011) 505–512 511
Fig. 6. Neonatal DNA vaccination strategy induces long-term immune response. BALB/c neonatal mice received three IN doses of 20 �g of either DNA vaccine associatedwith two doses of 5 �g of the same DNA vaccine by ID route at 7- and 21-days old. 8 months after the first IN/ID dose, the groups received another IN/ID dose. After sevendays cells from MLN and spleen were harvested. (A and B): Anti-p24 IgG1 (1:1350) and IgG2a (1:1350) Ab against HIV-1 p24 were evaluated by ELISA. The data represent themean value ± S.E.M. from six to eight mice per group. (C): IFN-� SFC from MLN stimulated by the immunodominant class I peptide (aa 197–211) and p24 were evaluated byE d to pe nocyt( by cytb als pe
s�ohnes
rrSAoifStvtcgr
LISPOT assay. (D): Splenic IFN-� SFC against pools of peptides from GAG protein anxperiments with pools of cells from two to three animals/per group. (E and F): Spleaa 161–211) or class II stimulus (aa 241–291); cytokine secretion was determinedetween secreted cytokine upon stimuli and spontaneous production from six anim
timulated by class II GAG peptides. The cytokines IL-2, IFN-�, TNF-, and IL-6 are essential factors for Th proliferation, differentiationf Th1 cells, and generation of CTL and plasma cells. These findingsave evidenced that CD4+ T cell activation after LAMP/gag immu-ization included both Th1 and Th2 responses, which are crucialvents for vaccine immunogeniticy in the neonatal period and foretting out immunological memory response.
Immunization of adult mice with the chimeric vaccine by IDoute has been shown to increase by 10-fold the CD4+ T cellesponse, and to promote a broader spectrum of IFN-� secretingFC against GAG peptides in comparison with the gag vaccine (derruda et al. 2004; Marques et al. 2003). The combined protocolsf IN and ID routes presented a surprising effect with both vaccinemmunizations. We detected by ELISPOT assay with spleen cellsrom the chimeric vaccinated mice that the magnitude of IFN-�FC recognizing GAG peptides was not exacerbated as it was forhe gag vaccinated group. This effect is probably due to the acti-
ation of a regulatory mechanism that prevents exacerbation ofhe immune response triggered by the LAMP protein trafficking toompartments containing class II MHC molecules, contrary to theag vaccine that stimulates the CD8+ T response only. The gut envi-onment is essentially an anti-inflammatory milieu that favors an
24 evaluated by ELISPOT assay. The data represent the mean value ± S.E.M. from 4es were cultured for 72 h in the presence of the GAG peptide pool containing class Iometric bead array. The results represent the mean value ± S.E.M. of the differencer group. *P ≤ 0.05 compared to gag immunized mice.
increase of TGF-� levels upon IN vaccination. In the gut mucosalcompartment, the Th2 cytokines play a key role in maintaining themicroenvironment equilibrium (Neurath et al. 2002), and may bepreventive for Th1 response activation, avoiding the developmentof an inflammatory milieu. Thereby, the activation of Th2 response,as seen by the increased frequency of IL-4 SFC against GAG peptides,was induced only by the chimeric vaccine.
The similar magnitude of IFN-� SFC in response to GAG pep-tides observed for both vaccines in neonates is in sharp contrastto the results obtained for adult mice immunized with ID injec-tions of the chimera vaccine, in which the frequency of IFN-� SFCto GAG peptides was much higher than in animals immunized withgag vaccine (de Arruda et al. 2004; Marques et al. 2003). On theother hand, neonate mice subjected to the ID injected LAMP/gagvaccine did present T-cell response of a similar magnitude to theadult counterparts (data not shown).
It is remarkable that the effects induced by the chimeric vac-
cine lasted, at least, 8 months after the neonatal immunization.This result shows that only the LAMP/gag vaccine induced thegeneration of memory cells in mice immunized early in the neona-tal phase, which is indispensable for the long-lasting protectiveresponse in both mucosal and systemic sites. In adult mouse, this
ong-lasting response triggered by the chimeric vaccine had alreadyeen observed (de Arruda et al. 2004); taking together with ourndings in neonates, these results emphasize the importance of aNA vaccine construction having an adequate delivery system fornhancing the transfection of APC favoring CD4+ T cells activation,o as to generate long-life memory T and B cells.
In conclusion, immunization with the genetic vaccines throughN–ID combined routes during neonatal phase was able to generatemmunologic response in lymphoid organs in mucosa and system-cally. The expression of the LAMP/gag chimeric protein enhancedhe response to the class II GAG peptides, through Th1/Th2 cytokinenduction. Unlike the gag vaccine, which activates preferentiallyD8+ T cells because it presents epitopes in the context of class IHC molecules only, a DNA vaccine expressing a LAMP-mediated
ntigen targeting to the endosomal/lysosomal compartments playsn essential role in the neonatal period and has proved to be effi-ient to trigger both mucosal and systemic humoral responses. TheAMP/gag vaccination was able to overcome the relative immuno-ogic immaturity in early life, favoring T helper activation, anssential step to induce sustained and long-lasting response.
cknowledgments
We thank Vilma dos Anjos Mesquita for the dedicated ani-al care, Noemia Orii for valuable technical assistance and Dr.abriela Ribeiro-dos-Santos for reviewing the manuscript. We
hank Prof. Luis Carlos Ferreira for providing p24 Gag from HIV-. This work was supported by Ministério da Saúde do Brasil –rograma Nacional de HIV/AIDS/DST (914BRA1101), FAPESP andIM-56/HCFMUSP.
eferences
dkins, B., 1999. T-cell function in newborn mice and humans. Immunol. Today 20,330.
rruda, L.B., Sim, D., Chikhlikar, P.R., Maciel Jr., M., Akasaki, K., August, J.T., Mar-ques, E.T., 2006. Dendritic cell–lysosomal-associated membrane protein (LAMP)and LAMP-1–HIV-1 gag chimeras have distinct cellular trafficking pathways andprime T and B cell responses to a diverse repertoire of epitopes. J. Immunol. 177,2265.
rthos, J., Cicala, C., Martinelli, E., Macleod, K., Van Ryk, D., Wei, D., Xiao, Z., Veen-stra, T.D., Conrad, T.P., Lempicki, R.A., McLaughlin, S., Pascuccio, M., Gopaul, R.,McNally, J., Cruz, C.C., Censoplano, N., Chung, E., Reitano, K.N., Kottilil, S., Goode,D.J., Fauci, A.S., 2008. HIV-1 envelope protein binds to and signals through inte-grin alpha4beta7, the gut mucosal homing receptor for peripheral T cells. Nat.Immunol. 9, 301.
essis, N., GarciaCozar, F.J., Boissier, M.C., 2004. Immune responses to gene therapyvectors: influence on vector function and effector mechanisms. Gene. Ther. 11(Suppl. 1), S10.
rave, A., Hallengard, D., Schroder, U., Blomberg, P., Wahren, B., Hinkula, J., 2008.Intranasal immunization of young mice with a multigene HIV-1 vaccine incombination with the N3 adjuvant induces mucosal and systemic immuneresponses. Vaccine 26, 5075.
azac, B.B., Roes, J., 2000. TGF-beta receptor controls B cell responsiveness andinduction of IgA in vivo. Immunity 13, 443.
hikhlikar, P., de Arruda, L.B., Maciel, M., Silvera, P., Lewis, M.G., August, J.T., Marques,E.T., 2006. DNA encoding an HIV-1 Gag/human lysosome-associated membraneprotein-1 chimera elicits a broad cellular and humoral immune response inRhesus macaques. PLoS One 1, e135.
adaglio, G., Sun, C.M., Lo-Man, R., Siegrist, C.A., Leclerc, C., 2002. Efficient in vivopriming of specific cytotoxic T cell responses by neonatal dendritic cells. J.
Immunol. 168, 2219.
ale, C.J., De Rose, R., Wilson, K.M., Croom, H.A., Thomson, S., Coupar, B.E., Ramsay,A., Purcell, D.F., Ffrench, R., Law, M., Emery, S., Cooper, D.A., Ramshaw, I.A., Boyle,D.B., Kent, S.J., 2004. Evaluation in macaques of HIV-1 DNA vaccines containingprimate CpG motifs and fowlpoxvirus vaccines co-expressing IFNgamma or IL-12. Vaccine 23, 188.
ogy 216 (2011) 505–512
de Arruda, L.B., Chikhlikar, P.R., August, J.T., Marques, E.T., 2004. DNA vac-cine encoding human immunodeficiency virus-1 Gag, targeted to the majorhistocompatibility complex II compartment by lysosomal-associated mem-brane protein, elicits enhanced long-term memory response. Immunology 112,126.
De Brito, C.A., Goldoni, A.L., Notomi-Sato, M., 2009. Immune adjuvants in early life:targeting the innate immune system to overcome impaired adaptive response.Immunotherapy 1, 883.
de Witte, L., Bobardt, M., Chatterji, U., Degeest, G., David, G., Geijtenbeek, T.B., Gallay,P., 2007. Syndecan-3 is a dendritic cell-specific attachment receptor for HIV-1.Proc. Natl. Acad. Sci. USA 104, 19464.
Deml, L., Schirmbeck, R., Reimann, J., Wolf, H., Wagner, R., 1999. ImmunostimulatoryCpG motifs trigger a T helper-1 immune response to human immunodefi-ciency virus type-1 (HIV-1) gp 160 envelope proteins. Clin. Chem. Lab. Med. 37,199.
Geijtenbeek, T.B., van Kooyk, Y., 2003. DC-SIGN: a novel HIV receptor on DCs thatmediates HIV-1 transmission. Curr. Top. Microbiol. Immunol. 276, 31.
Gerdts, V., Babiuk, L.A., van Drunen Littel-van den, H., Griebel, P.J., 2000. Fetal immu-nization by a DNA vaccine delivered into the oral cavity. Nat. Med. 6, 929.
Gouws, E., Stanecki, K.A., Lyerla, R., Ghys, P.D., 2008. The epidemiology of HIV infec-tion among young people aged 15–24 years in southern Africa. AIDS 22 (Suppl.4), S5.
Grossman, Z., Meier-Schellersheim, M., Paul, W.E., Picker, L.J., 2006. Pathogenesis ofHIV infection: what the virus spares is as important as what it destroys. Nat.Med. 12, 289.
Huber, S., Schramm, C., Lehr, H.A., Mann, A., Schmitt, S., Becker, C., Protschka, M.,Galle, P.R., Neurath, M.F., Blessing, M., 2004. Cutting edge: TGF-beta signal-ing is required for the in vivo expansion and immunosuppressive capacity ofregulatory CD4+ CD25+ T cells. J. Immunol. 173, 6526.
Jiang, J.Q., Patrick, A., Moss, R.B., Rosenthal, K.L., 2005. CD8+ T-cell-mediated cross-clade protection in the genital tract following intranasal immunization withinactivated human immunodeficiency virus antigen plus CpG oligodeoxynu-cleotides. J. Virol. 79, 393.
Kojima, Y., Xin, K.Q., Ooki, T., Hamajima, K., Oikawa, T., Shinoda, K., Ozaki, T., Hoshino,Y., Jounai, N., Nakazawa, M., Klinman, D., Okuda, K., 2002. Adjuvant effect ofmulti-CpG motifs on an HIV-1 DNA vaccine. Vaccine 20, 2857.
Kraynyak, K.A., Kutzler, M.A., Cisper, N.J., Khan, A.S., Draghia-Akli, R., Sardesal,N.Y., Lewis, M.G., Yan, J., Weiner, D.B., 2010. Systemic immunization withCCL27/CTACK modulates immune responses at mucosal sites in mice andmacaques. Vaccine 28, 1942.
Lambert, A.A., Gilbert, C., Richard, M., Beaulieu, A.D., Tremblay, M.J., 2008. The C-type lectin surface receptor DCIR acts as a new attachment factor for HIV-1 indendritic cells and contributes to trans- and cis-infection pathways. Blood 112,1299.
Marques Jr., E.T., Chikhlikar, P., de Arruda, L.B., Leao, I.C., Lu, Y., Wong, J., Chen, J.S.,Byrne, B., August, J.T., 2003. HIV-1 p55Gag encoded in the lysosome-associatedmembrane protein-1 as a DNA plasmid vaccine chimera is highly expressed, traf-fics to the major histocompatibility class II compartment, and elicits enhancedimmune responses. J. Biol. Chem. 278, 37926.
Mehandru, S., Poles, M.A., Tenner-Racz, K., Horowitz, A., Hurley, A., Hogan, C., Boden,D., Racz, P., Markowitz, M., 2004. Primary HIV-1 infection is associated withpreferential depletion of CD4+T lymphocytes from effector sites in the gastroin-testinal tract. J. Exp. Med. 200, 761.
Neurath, M.F., Finotto, S., Glimcher, L.H., 2002. The role of Th1/Th2 polarization inmucosal immunity. Nat. Med. 8, 567.
Rollman, E., Hinkula, J., Arteaga, J., Zuber, B., Kjerrstrom, A., Liu, M., Wahren, B.,Ljungberg, K., 2004. Multi-subtype gp160 DNA immunization induces broadlyneutralizing anti-HIV antibodies. Gene Ther. 11, 1146.
Rowell, J.F., Ruff, A.L., Guarnieri, F.G., Staveley-O’Carroll, K., Lin, X., Tang, J., August,J.T., Siliciano, R.F., 1995. Lysosome-associated membrane protein-1-mediatedtargeting of the HIV-1 envelope protein to an endosomal/lysosomal compart-ment enhances its presentation to MHC class II-restricted T cells. J. Immunol.155, 1818.
Ruff, A.L., Guarnieri, F.G., Staveley-O’Carroll, K., Siliciano, R.F., August, J.T., 1997. Theenhanced immune response to the HIV gp160/LAMP chimeric gene product tar-geted to the lysosome membrane protein trafficking pathway. J. Biol. Chem. 272,8671.
Siegrist, C.A., 2001. Neonatal and early life vaccinology. Vaccine 19, 3331.UNAIDS, 2008. Report on the global AIDS epidemic 2008. UNAIDS, 1.Valentin, A., Chikhlikar, P., Patel, V., Rosati, M., Maciel, M., Chang, K.H., Silvera,
DNA vaccines producing HIV-1 Gag and LAMP/Gag chimera in rhesus macaquesreveals antigen-specific T-cell responses with distinct phenotypes. Vaccine 27,4840.
