j our nal homep ag e: www.elsev ier .com/ locate /vacc ine
renatal vitamin A deficiency impairs adaptive immune responses toentavalent rotavirus vaccine (RotaTeq®) in a neonatal gnotobioticig model
ukumar Kandasamy1, Kuldeep S. Chattha1, Anastasia N. Vlasova, Linda J. Saif ∗
ood Animal Health Research Program, Department of Veterinary Preventive Medicine, Ohio Agricultural Research and Development Center, The Ohio Stateniversity, 1680 Madison Avenue, Wooster, OH 44691-4096, USA
r t i c l e i n f o
rticle history:eceived 9 July 2013eceived in revised form 6 December 2013ccepted 12 December 2013vailable online 29 December 2013
eywords:itamin Aitamin A deficiencyntibody responsesotaTeq®
ytokinesotavirusiarrhea
a b s t r a c t
Vitamin A deficiency (VAD) is associated with increased childhood mortality and morbidity in impover-ished Asian and African countries, but the impact of VAD on rotavirus (RV) vaccine or infection is poorlyunderstood. We assessed effects of gestational and dietary induced pre- and post-natal VAD and vitaminA supplementation on immune responses to a pentavalent rotavirus vaccine, RotaTeq® in a neonatalgnotobiotic pig model. Vaccine efficacy was assessed against virulent G1P[8] human rotavirus (HRV)challenge. VAD and vitamin A sufficient (VAS) piglets were derived from dietary VAD and VAS sows,respectively. VAD piglets had significantly lower levels of hepatic vitamin A compared to that of VASpiglets. RotaTeq®-vaccinated VAD piglets had 350-fold higher fecal virus shedding titers compared tovaccinated VAS piglets post-challenge. Only 25% of vaccinated non-vitamin A supplemented VAD pigletswere protected against diarrhea compared with 100% protection rate in vaccinated non-supplementedVAS piglets post-challenge. Intestinal HRV specific immune responses were compromised in VAD piglets.Vaccinated VAD piglets had significantly lower ileal HRV IgG antibody secreting cell (ASC) responses (pre-challenge) and duodenal HRV IgA ASC responses (post-challenge) compared to vaccinated VAS piglets.Also, intestinal HRV IgA antibody titers were 11-fold lower in vaccinated VAD compared to vaccinated VASpiglets post-challenge. Persistently elevated levels of IL-8, a pro-inflammatory mediator, and lower IL-10responses (anti-inflammatory) in vaccinated VAD compared to VAS piglets suggest more severe inflam-
matory responses in VAD piglets post-challenge. Moreover higher IFN-� responses pre-challenge wereobserved in VAD compared to VAS piglets. The impaired vaccine-specific intestinal antibody responsesand decreased immunoregulatory cytokine responses coincided with reduced protective efficacy of theRV vaccine against virulent HRV challenge in VAD piglets. In conclusion, VAD impaired antibody responsesto RotaTeq® and vaccine efficacy. Oral supplementation of 100,000 IU vitamin A concurrent with RVvaccine failed to increase the vaccine efficacy in VAD piglets.
. Introduction
Rotavirus (RV) is a leading cause of morbidity and mortalityn children worldwide. Two oral human RV vaccines, RotaTeq®
Merck and Co., Inc.) and Rotarix® (GlaxoSmithKline Biologicals)re currently licensed to prevent severe RV disease. RotaTeq® is
pentavalent human-bovine reassortant RV vaccine that contains
he dominant circulating human RV parent strain G and P types;1, G2, G3, G4 and P1A[8], and G6 and P[7] from bovine rotavirusarent strain [1].
Rotavirus is a major cause of severe diarrhea and mortality inchildren in impoverished countries, where RV vaccine efficacy islowest. The efficacy of RotaTeq® against severe RV gastroenteritis inAfrica is 39% [2] and 48.5% in Bangladesh and Vietnam [3]. In com-parison, the efficacy of RotaTeq® is >90% in developed countriessuch as the US and Finland [4,5]. Several factors were attributedto the lower efficacy of RV vaccines, such as higher prevalence ofmalnutrition, vitamin A deficiency (VAD), higher RV disease bur-den, presence of diverse RV strains, and interference by maternalantibodies.
Vitamin A deficiency is prevalent in economically poor countries[6–9]. According to the World Health Organization (WHO), an esti-
mated 250 million preschool-age children in low income countriesare affected by VAD. A recent study revealed that 66% of youngchildren were affected by VAD in Guinea-Bissau [6]. The WHOhas recommended supplemental vitamin A (50,000–100,000 IU)
or children at immunization contacts to overcome vitamin A defi-iency in these countries [10]. Vitamin A and its metabolite retinoiccid (RA) play a significant role in development of gastrointesti-al immune responses to antigens or vaccines. VAD is associatedith reduced immunogenicity of different vaccines such as diph-
heria [11] measles [12] and tetanus toxoid [13]. Retinoic acid haseen shown to mediate mucosal IgA production [14] and intesti-al homing of IgA+ plasma cell [15] and T-cells [16]. In additiono mediating these responses, vitamin A is also essential for nor-
al T-cell responses in the gut mucosa [16]. VAD alters cytokineesponses such as polarization toward increased T-helper cell type-(Th1) cytokine responses in infants [17] and mice [18]. In addition,reatment of T cells with RA enhanced Th2 development and sup-ressed Th1 development. Thus, vitamin A is involved in regulationf intestinal mucosal immunity. Based on these previous observa-ions, we hypothesized that vitamin A may act as an adjuvant andnhance efficacy of the pentavalent RotaTeq® vaccine and inducencreased RV immunity in infants.
The WHO has recommended inclusion of RV vaccines in nationalmmunization programs worldwide. The impact of VAD on RVnfection and on efficacy of RotaTeq vaccine is unknown. More-ver the modulation by supplemental vitamin A of RV immunityn neonates has not been investigated. Neonatal gnotobiotic (Gn)iglets are susceptible to HRV induced diarrhea and their mucosal
mmune systems and gut physiology are similar to that of infants.etinol metabolism in pigs is comparable with that in humans [19].
n addition, vitamin A content in liver and milk of pigs is similaro that in humans [19–21]. Because of the similarity in vitamin A
etabolism, pigs have been previously used to evaluate vitamin supplementation programs in lactating women [19,22] and chil-ren [23]. Also piglets are outbred and similar in size to newborn
nfants. In this study, we used Gn piglets to investigate effects ofestational and dietary induced pre- and post-natal vitamin A defi-iency and supplemental vitamin A (as recommended by WHO atmmunization contacts) on RotaTeq® vaccine, virulent HRV chal-enge and the adaptive immune responses induced.
. Materials and methods
.1. Experimental design
This study was approved by The Ohio State University Insti-utional Animal Care and Use Committee. Cesarean-derived Gniglets from near-term sows were maintained in sterile isola-ors as described previously [24]. VAD and vitamin-A sufficientVAS) piglets were derived from 2 paired VAD and VAS sowsboth Landrace × Yorkshire White × Duroc crossbred), respectively,s described previously [25]. VAD pigs were defined on the basisf their origin (from VAD sows) and hepatic vitamin A levels.riefly, pregnant VAD and VAS sows were kept on vitamin Aestricted or supplemented diets, respectively, for approximately0 days (until piglet derivation) starting from gestation day 34.AS sows were given an additional 500,000 IU injectable vita-in A (NDC50989-178-12; Vedco) during gestation to maintain
itamin A levels. Gestational and dietary induced VAD and VASiglets were fed commercial ultra-high temperature-treated ster-
le bovine milk (Parmalat) containing low amounts of vitamin for maintaining low levels of vitamin A in VAD piglets. TheAD and VAS piglets were assigned to one of the following fourroups: 3XRotaTeq® vaccinated (Vac); 3XRotaTeq® vaccinated anditamin A supplemented (100,000 U) (Vac + VitA), unvaccinated
ontrol (Control), and unvaccinated and vitamin A supplemented100,000 IU) controls (Control + VitA). RotaTeq® vaccine with andithout supplemental vitamin A was given orally at 6- (Post inoc-lation day 0, [PID0]), 16- (PID10) and 28- (PID21) days of age.
e 32 (2014) 816– 824 817
Vitamin A supplemented groups received oral supplementation ofretinyl palmitate (Sigma–Aldrich, R3375) at a dosage of 100,000 IUwith each vaccine dose.
Serum samples were collected to assess cytokine and HRV-specific antibody responses at multiple time-points. Subsets of pigswere euthanized at PID28/post-challenge day 0 (PCD0) to assesspre-challenge responses and the remaining were challenged with105 FFU of virulent HRV Wa strain (VirHRV) and were euthanizedat PID35/PCD7 to assess post-challenge immune responses. Mono-nuclear cells (MNC) were isolated from euthanized pigs to measureHRV specific antibody secreting cell (ASC) responses in intestinaltissues [25,26].
2.2. Clinical signs and virus shedding
VirHRV challenged piglets were examined daily post-challengeto assess diarrhea and fecal HRV shedding, as previously described[27].
2.3. B cell and cytokine responses and vitamin A levels
HRV specific IgA and IgG antibody responses were measured byELISA as described previously [27,28], with slight modifications.Enumeration of IgA and IgG isotype-specific HRV ASC was per-formed by enzyme-linked immunosorbent spot (ELISPOT) assay aspreviously described [25,27], with slight modifications. RotaTeq®
vaccine was used as antigen in both HRV specific ELISA and ELISPOTassays. Serum IFN�, IFN�, TGF�, IL4, IL12, IL10 and IL8 levels atPID0, PID2, PID6/7, PID28/PCD0, PID30/PCD2 and PID35/PCD7 weremeasured by ELISA as described previously [25,29,30]. Hepatic andserum vitamin A levels were assessed as described previously [25].
2.4. In vitro stimulation with RA and RV antigen
Frozen splenic MNC (in 90% FBS/10% dimethylsulfoxide [DMSO])from four unrelated RV vaccinated Gn piglets (7-week old) werethawed and washed immediately to remove residual DMSO. In eachculture, 2 × 106 cells were stimulated with semi-purified group ARV antigen (12 �g/ml), alone or in combination with RA at 10 nm,100 nm and 1000 nm or mock stimulated for 6 days. Followingincubation, supernatants were collected to measure total IgA titersby ELISA and cells were harvested and washed to determine fre-quency of IgA+ B cells (CD79�+IgA+) by flow cytometry. Briefly,for ELISA, goat anti-pig IgA antibody (cat #A100-102A, Bethyl) wasused as coating antibody. The HRP conjugated goat anti-pig IgAantibody (cat #A100-102AP, Bethyl) was used as detection anti-body. Standard curves were generated using pig reference serum(RS10-107, Bethyl) to extrapolate the IgA levels in the samples.For flow cytometry, the stimulated MNC were stained with mousemonoclonal anti-porcine IgA (Clone K61 1B4, Serotec) followed byanti-mouse IgG1-Allophycocyanin (BD Biosciences, CA) antibodies.Subsequently, samples were permeabilized and stained intracellu-larly for porcine cross-reactive anti-mouse CD79� antibody (CloneAT107-2, Serotec, NC). The samples were acquired and analyzedusing Accuri C6 flow cytometer and C6 flow sampler software,respectively.
2.5. Statistical analysis
Log10 transformed isotype-specific ELISA antibody titers wereanalyzed using one-way ANOVA followed by Duncan’s multiplerange test. The area under the curve (AUC) for diarrhea sever-
ity and shedding was calculated as previously explained [31].The HRV-specific ASC, cytokine concentrations, AUC, total IgAconcentration and frequencies of CD79�+IgA+ cells were com-pared among groups using the Kruskal–Wallis rank sum test. All
818 S. Kandasamy et al. / Vaccine 32 (2014) 816– 824
Fig. 1. (a) Hepatic vitamin A in VAD and VAS piglets, pre- and post-challenge. Statistically significant differences (p ≤ 0.05) were observed between vitamin A supplemented(††) and non-supplemented (†) groups, between pre- and post-challenge (*) levels for VAS piglets and between VAD (A) and VAS groups (B). The data are shown regardlessof vaccination status because there was no effect of vaccination on serum or hepatic vitamin A levels. (b) Serum vitamin A levels in VAD and VAS piglets at PID0 andP lemeno is of vl
sIn
3
3
dvstacHte
TD
N
ID28/(PCD35/PCD7). The data are shown regardless of vaccination or vitamin A suppn serum or hepatic vitamin A levels. Data represent mean ± SEM. One-way analysevels.
tatistical analyses were performed using SAS program (SASnstitute, NC) or Graphpad prism. Differences were considered sig-ificant at P < 0.05.
. Results
.1. VAD piglets had significantly lower levels of hepatic vitamin A
Throughout the study, VAD piglets from sows with gestationalietary induced VAD did not show any overt clinical effects ofitamin A deficiency such as blindness, hyperkeratinization ofkin and conjunctival xerosis and they gained weight similar tohe VAS piglets from VAS sows. RV vaccination did not haveny effect on hepatic vitamin A levels; therefore data for vac-
inated and non-vaccinated Gn piglets were analyzed together.epatic vitamin A levels for VAD groups were significantly lower
han counterpart VAS groups (Fig. 1a). Hepatic vitamin A lev-ls of non-supplemented piglets were significantly lower than
able 1iarrhea scores and virus shedding titers in VAD and VAS piglets vaccinated with RotaTe
– number of piglets/group.a Only portion of piglets shed virus and determination of mean days to onset of virus s* Determined by cell culture immunofluorescence infectivity assay and expressed as fl
** Area under the curve (AUC) as measure for cumulative diarrhea scores and fecal shed# Fecal consistency was scored as follows: 0 = normal; 1 = pasty. 2. Semi-liquid; 3 = liqui
## Protection rate = [1 − (percentage of treatment group piglets with diarrhea/percentag†† Means in the same column, with same letters differed significantly, (Kruskal–Wallis r††† No significant difference was found among groups (One way ANOVA followed by Dun
tation status because there was no effect of vaccination/vitamin A supplementationariance followed by Duncan’s multiple range test, were used to compare vitamin A
those of supplemented piglets, respectively on PID28/PCD0 andPID35/PCD7 (Fig. 1a). All VAS piglets (with or without supple-mental vitamin A) and vitamin A supplemented VAD piglets hadsignificantly lower and a trend of lower hepatic vitamin A levels atpost-challenge time-point compared to pre-challenge time-point,respectively (Fig. 1a). Supplementation of vitamin A to VAD pigletsresulted in hepatic vitamin A levels similar to non-supplementedVAS piglets at pre- and post-challenge time-points suggesting thatsupplementation helped to restore hepatic vitamin A reserves inVAD piglets (Fig. 1a). Mean serum vitamin A levels of VAS piglets(Fig. 1b) were within the normal range defined for nursing piglets(0.4–1 IU/ml) and remained at similar levels throughout the exper-iment (Fig. 1b). In contrast VAD pigs had only 67% of mean serumvitamin A levels of VAS pigs at the first dose of RV vaccine (PID0).
In VAD piglets, during the course of the experiment, serum vita-min A levels increased ∼1.4-fold (Fig. 1b), suggesting that uniformamounts of vitamin A from the milk diet may have contributed tothis increase.
q® vaccine with or without vitamin A supplementation post-challenge.
hedding is not applicable.uorescent focus forming units (FFU).ding.d. Piglets with >1 fecal scores were considered diarrheic.e of control piglets with diarrhea)] × 100.ank sum test, p ≤ 0.05).can’s multiple range test, p ≤ 0.05).
S. Kandasamy et al. / Vaccine 32 (2014) 816– 824 819
Fig. 2. Intestinal HRV IgA IgG antibody responses. Geometric mean titers (GMT) of HRV IgA antibody titers in (a and b) small-intestinal contents (SIC) of Gn piglets vaccinatedw ® D0) and ns mu
3a
pcts((vpFi
Fm5
ith RotaTeq vaccine with or without vitamin A supplementation at pre- (PID28/PCay. Bars with different letters (A and B) differ significantly between groups (Dunca
.2. VAD impaired RotaTeq® protective efficacy against diarrheand virus shedding post-challenge
Vaccinated VAS groups (Vac: 100%, Vac + VitA: 50%) had higherrotection rates against HRV induced diarrhea compared to vac-inated VAD groups (Vac 25%, Vac + VitA: 0%) (Table 1). Similaro diarrhea scores, lower virus shedding titers and higher virushedding protection rates were observed in vaccinated VAS groupsVac: 66.6%, Vac + VitA: 50%) compared to vaccinated VAD groupsVac 0%, Vac + VitA: 0%) (Table 1). Moreover, overall mean peak
irus shedding titers were >350-fold higher in vaccinated VADiglets (180,281 FFU/ml) relative to vaccinated VAS piglets (508FU/ml). The AUC for shedding was 440-fold and 96-fold highern Vac and Vac + VitA VAD groups compared to counterpart
ig. 3. Small intestinal HRV HRV-specific IgA (a and b) and IgG (c and d) ASC responses iock with or without vitamin A supplementation at pre- (PID28/PCD0) and post-HRV ch
× 105 MNC. Data were analyzed using the Kruskal–Wallis test, P < 0.05. PID-Post-inocula
d post-HRV challenge (PID35/PCD7). PID-Post-inoculation day; PCD-Post-challengeltiple range t test on log10 transformed data, P < 0.05). Error bars indicate SEM.
vaccinated VAS groups, respectively (Table 1). The three dosesof supplemental vitamin A at each vaccine time-point did nothave any beneficial effects on RotaTeq® vaccine induced protec-tion against diarrhea or VirHRV shedding. Therefore, we pooleddata (diarrhea scores and fecal HRV shedding) from Vac andVac + VitA groups belonging to VAD and VAS pigs to deter-mine the impact of pre- and post-natal vitamin A deficiency onHRV vaccine and VirHRV challenge. The vaccinated VAD grouphad a higher percentage of pigs with diarrhea and HRV fecalshedding compared to the vaccinated VAS group (Supplemen-
tary Table S1). Our findings suggest that gestational and dietaryinduced pre- and post-natal vitamin A deficiency may compro-mise the protective efficacy of the RotaTeq® vaccine in the pigletmodel.
n duodenum and ileum of VAS and VAD pigs vaccinated with RotaTeq® vaccine orallenge (PID35/PCD7). Data represent the mean numbers of HRV-specific ASC pertion day. PCD-Post-challenge day.
820 S. Kandasamy et al. / Vaccine 32 (2014) 816– 824
020406080
100120
pg/m
l
020406080
100120
pg/m
l
0
100
200
300
400
500
pg/m
l
0
100
200
300
400
500
pg/m
l
020406080
100120140
pg/m
l
020406080
100120140
pg/m
l
IFNγ IL12 IL4
VAS
VAD
(a)
(b)
(c) (e)
(d) (f)
Vac(VAS n=3-7, VAD n=4-7)
Vac+VitA(VAS n=4-7, VAD n=3-7)
Control(VAS n=4-6, VAD n=4-6)
Control+VitA(VAS n=4-6, VAD n=3-6)
* *
****
F d d) av ultiplf etwee
3r
aVepwVw
3v
nSopAt(rlcs
3p
tmla
ig. 4. Serum cytokine responses in VAS and VAD piglets. IFN-� (a and b), IL-12 (c anaccinated with RotaTeq® vaccine with or without vitamin A supplementation at mor the same cytokine. Different asterisks symbols indicate significant differences b
.3. VAD piglets had decreased intestinal HRV IgA antibodyesponses
In intestinal contents, Vac and Vac + VitA VAS groups had 11-nd 3-fold higher IgA HRV antibody titers compared to Vac andac + VitA VAD groups post-challenge, respectively although differ-nces did not differ statistically likely due to low numbers of pigser group (Fig. 2b). No consistent effect of supplemental vitamin Aas observed on intestinal HRV IgA antibody titers in vaccinatedAD and VAS groups suggesting no adjuvant effect of vitamin Ahen supplemented at vaccine time-points (Fig. 2a and b).
.4. VAD significantly decreased small intestinal HRV IgA ASC inaccinated piglets
Overall, vitamin A status (VAD vs VAS) affected ileal and duode-al IgA and IgG HRV ASC responses both pre- and post-challenge.ignificantly lower ileal HRV IgA and IgG ASC responses werebserved in the Vac VAD group compared to the Vac VAS groupre-challenge (Fig. 3a and c). Significantly lower duodenal HRV IgASC were observed in the Vac and Vac + Vit A VAD groups compared
o the counterpart Vac and Vac + Vit A VAS groups, respectivelyFig. 3b), which coincided with lower intestinal HRV IgA antibodyesponses post-challenge. Also ileal HRV IgG ASC were significantlyower in the Vac VAD group compared to the Vac VAS group post-hallenge (Fig. 3d). No effect of vitamin A supplementation on HRVpecific IgA ASC was observed post-challenge (Fig. 3b).
.5. VAD modulated serum Th1 and Th2 cytokines pre- andost-challenge
All VAS piglets had lower serum IFN� pre-challenge compared
o VAD piglets irrespective of vaccination or vitamin A supple-
entation. VAD groups had 2–3-fold higher mean serum IFN�evels compared to VAS piglets at PID0, PID2 and PID6 (Fig. 4and b) suggesting a pro-inflammatory environment associated with
nd IL-4 (e and f) levels were measured in serum samples from VAS and VAD pigletse time-points. Circles represent comparisons between the VAS and the VAD groupsn indicated groups at the same time-point for the same cytokine.
decreased vitamin A levels. Interestingly the concentration of IL12,also a Th1 cytokine was significantly higher (p ≤ 0.05) in vaccinatedVAS piglets as compared to vaccinated VAD piglets, irrespective ofvitamin A supplementation at PID2 and PID6 (Fig. 4c and d). VAScontrol piglets had a trend (p = 0.12) for higher serum IL12 levels(310 ± 79) as compared with VAD control piglets (Control: 190 ± 72,Control + VitA: 214 ± 83) at PCD2 and they showed a delayed (atPCD7) increase in serum IFN� post-challenge as compared to VADpiglets (at PCD2) (Supplementary Table S2). The Vac + VitA VASgroup had 2.5- and 5-fold higher IL4 (Th2 cytokine) at PCD2 ascompared to Vac + VitA and Vac VAD groups, respectively (Fig. 4eand f, Supplementary Table S2) confirming higher Th2 cytokinesin former group, which also had higher hepatic vitamin A levelscompared to other groups.
3.6. Regulatory/Th3 and innate cytokines pre- and post-challenge
IL10 was significantly higher (p ≤ 0.05) in Vac + VitA VAS pigletsat PID2 compared to Vac VAS group and vaccinated VAD piglets,irrespective of supplementation (Fig. 5a and b). Moreover, post-challenge the VAS vaccinated groups with or without supplementalvitamin A had 6- and 2.5-fold higher serum IL10 levels compared tothe VAD vaccinated groups with or without vitamin A supplemen-tation, respectively, (Supplementary Table S3) suggesting a role ofvitamin A deficiency in compromising anti-inflammatory and theimmunoregulatory environment. However, no significant differ-ences in TGF� levels were observed between VAS and VAD piglets(Fig. 5c and d). The innate anti-viral cytokine, IFN� was weaklyinduced by the vaccine at PID2 in the VAD vaccinated piglets (Vac:6.5 ± 3.5. Vac + VitA: 7.2 ± 3.4), but was not observed in the VAS vac-cinated groups (Fig. 5e and f). Comparable levels of IFN� betweenVAD and VAS control piglets were observed coinciding with no
differences in fecal HRV shedding between control groups post-challenge. The pro-inflammatory cytokine IL8 reached peak levelsat PCD2 for all the VAS groups, whereas VAD groups reached peakat PCD7, except VAD non-vitamin A supplemented control piglets
S. Kandasamy et al. / Vaccine 32 (2014) 816– 824 821
VAS
VAD
0
100
200
300
400pg
/ml
0
100
200
300
400
pg/m
l
0
200
400
600
800
1000
pg/m
l
0
200
400
600
800
1000
pg/m
l
050
100150200250300350
IU/m
l
050
100150200250300350
IU/m
l
IL10 TGFβ IFNα
0
20
40
60
80
100
pg/m
l
0
20
40
60
80
100
pg/m
l
VAS VADIL8
Vac(VAS n=3-7, VAD n=4-7)
Vac+VitA(VAS n=4-7, VAD n=3-7
Control(VAS n=4-6, VAD n=4-6)
Control+VitA(VAS n=4-6, VAD n=3-6)
)c()a(
(b) (d)
(e)
(f)
)h()g(
*
**
Fig. 5. Serum cytokine responses in VAS and VAD piglets. IL-10 (a and b), TGF-� (c and d), IFN-� (e and f) and IL-8 (g and h) levels were measured in serum samples fromV ntatioV iffere
(m
3
tsfoictfpFfig
4
m
AS and VAD piglets vaccinated with RotaTeq® with or without vitamin A supplemeAD groups for the same cytokine. Different asterisks symbols indicate significant d
Fig. 5g and h) suggesting that the kinetics of innate IL8 responsesay depend on vitamin A status.
.7. Exogenous RA increases total IgA titers in pig MNC
RA is a specific IgA isotype switch factor and has been showno increase IgA production [32,33]. To confirm whether RA plays aimilar role in pigs, we conducted in vitro studies with splenic MNCrom RV vaccinated pigs. Significantly higher total IgA levels werebserved in culture supernatants from RV antigen stimulated MNCn the presence of different doses (10, 100 and 1000 nm) of RA asompared to that of RV stimulated only or mock (Fig. 6) MNC. Fur-her, the higher IgA levels also coincided with significantly higherrequencies of CD79�+IgA+ B cells in RV ± RA supplemented sam-les as compared to RV stimulated only or mock (Supplementaryig. S1) samples. These preliminary findings indicate that in vivo dif-erences in RA (vitamin A) levels may result in observed differencesn RV specific IgA antibody titers between VAD and VAS vaccinatedroups.
. Discussion
Vitamin A plays a significant role in generation of protectiveucosal immunity. VAD is associated with increased susceptibility
n at multiple time-points. Circles represent comparisons between the VAS and thences between indicated groups at the same time-point for the same cytokine.
to diseases and impaired immunity in children in developingcountries [34,35]. Clinical studies showed beneficial effects of vita-min A supplementation in preventing both morbidity and mortalityin children in developing countries [35]. However, the impact ofVAD on RV vaccine efficacy or infection is still largely unknown. Inthis study we showed that gestational and dietary induced VADimpaired immune responses to a licensed RV vaccine and thevaccine-induced protection in a neonatal Gn piglet model againstchallenge with a G1P[8] HRV (Wa strain). Supplementation of100,000 IU vitamin A with each dose of RV vaccine failed to enhanceRV vaccine efficacy in VAD piglets. However, bolus supplemen-tation of vitamin A was able to restore hepatic vitamin A levelsin VAD pigs equal to that of non-vitamin A supplemented VAS pigs.
VAD piglets exhibited subclinical vitamin A deficiency as indi-cated by lower hepatic vitamin A reserves and lack of overtsymptoms of clinical vitamin A deficiency as seen in humans suchas poor growth, birth of abnormal piglets, hyperkeratinization ofskin and xerophthalmia [36]. Serum vitamin A levels are not goodindicators of deficiency unless hepatic reserves are critically low[37]. In our study, we observed low serum vitamin A levels in VAD
piglets as compared to VAS piglets at early vaccination time-point(PID0) and there was an increase in vitamin A levels in VAD pigletsover the course of experiments (due to milk vitamin A) indicat-ing a positive 30-d dose response, which has been used to identify
822 S. Kandasamy et al. / Vaccin
0
2
4
6
8
10
12
14To
tal I
gA (µ
g/m
l)
*
*
Fig. 6. Total IgA concentrations in culture supernatants of in vitro stimulated splenicmononuclear cells. Splenic mononuclear cells (n = 4 pigs) were stimulated with RVantigen (12 �g/ml), RV(12 �g/ml) ± RA (10 nm, 100 nm or 1000 nm) or mock for 6days and culture supernatants were collected to determine concentrations of totalIg
ssdvc(
imdcpirpstaiR
cwiRgtrovlmh[aAsho
gA. Different asterisks symbols indicate significant differences between indicatedroups. Data were analyzed using the Kruskal–Wallis test, P ≤ 0.05.
ubclinical vitamin A deficiency in human studies [38,39]. It is con-idered that a majority of children exhibit subclinical vitamin Aeficiency, which is not accurately diagnosed by measuring serumitamin A at one time-point. In this study we replicated VAD ofhildren from developing countries and studied effects of prenatalgestational and dietary) VAD on RotaTeq vaccine efficacy.
We showed that normal hepatic vitamin A levels may play a rolen generation of protective RV vaccine induced immunity in a Gn pig
odel of HRV disease. A lower protection rate against RV inducediarrhea and higher fecal HRV shedding in vaccinated VAD groupsompared to vaccinated VAS groups post-challenge suggests thatrenatal VAD impaired induction of protective immunity. Indeed,
ntestinal HRV IgA antibody responses and intestinal IgA ASC (cor-elates of protection to RV infection) were lower in vaccinated VADiglets compared to vaccinated VAS piglets post-VirHRV challengeuggesting that prenatal and dietary VAD may affect anamnes-ic responses, without affecting induction of primary HRV specificntibody responses. Thus, VAD, common in children from develop-ng countries, may result in impaired memory B cell responses tootaTeq® contributing to lower vaccine efficacy.
A higher number of intestinal IgA and IgG ASC in the VAS vac-inated groups compared to the VAD vaccinated groups coincidedith higher hepatic vitamin A levels in the former groups suggest-
ng a role for vitamin A in eliciting gut immune responses to oralotaTeq® vaccine. The reduced HRV IgA ASC responses in the VADroups also coincided with reduced intestinal HRV IgA antibodyiters in the VAD piglets indicating that effector and plasma B cellesponses were affected by VAD. Similar to our study, impairmentf antigen specific intestinal IgA antibody responses to oral choleraaccine was also observed in VAD rats [40]. Retinoic acid, a metabo-ite of vitamin A, is essential for various functions of B cells including
aturation [41], antibody isotype switching [42,43], intestinal-oming of IgA ASC [16] and generation of antibody responses44,45]. Specialized CD103+ (�E�7) dendritic cells in the gut associ-ted lymphoid tissue (GALT) have the ability to metabolize vitamin
and synthesize retinoic acid (RA). These DCs facilitate IgA classwitching and antibody production by B cells and also imprint gutoming receptor expression on effector B cells [14,46,47]. Previ-usly, we have shown that the frequencies of intestinal CD103+
e 32 (2014) 816– 824
expressing DCs were significantly lower in the VAD vaccinatedgroups compared to VAS groups [25] and this may affect the kinet-ics and magnitude of B cell responses to RotaTeq® vaccine in VADvaccinated animals. Moreover, previous studies have shown thatgut homing B cells are involved in inducing protective immunityagainst RV infection [48,49]. Collectively, our findings suggest thatone or more of the following factors may have caused significantlydecreased intestinal HRV IgA ASCs and intestinal IgA antibody titersto RotaTeq® vaccine in VAD piglets compared to VAS piglets: (a)defective intestinal homing of activated B cells; (b) impaired Bcell-intrinsic functions such as altered signaling process in B cellsrequired for induction of plasma cells [50] and/or (c) reduced classswitching to IgA effector B cells [51] under VAD conditions. Furtherresearch is required to investigate the mechanisms involved. In ourstudy, we also observed higher IL10 (anti-inflammatory) cytokinelevels in vaccinated VAS pigs compared to vaccinated VAS pigs afterfirst immunization, which may also contribute to higher HRV IgAresponses [41] observed in these pigs.
Th1 cytokine responses to RV vaccine and/or RV challenge werealso dependent on host vitamin A status. Levels of circulatingIFN�, constitutively, after RV vaccination and after RV challenge(in control groups at PCD2 only) were higher in VAD piglets com-pared to VAS piglets indicating altered IFN� responses in deficientgroups. Higher IFN� levels pre-challenge (at early time-points)suggests dysregulation of immune homeostasis and a Th1 bias asobserved in a VAD mouse model in previous studies [52,53]. In VADmice, higher constitutive expression of IFN� was also observed,but unlike in our study, supplementation of RA in mice allevi-ated the aberrant constitutive expression of IFN� [53]. The higherIFN� responses in control VAD piglets at early time-points post-challenge may result from the higher RV replication in this group(indicated by fecal shedding) suggesting impaired innate immu-nity [28]. Interestingly serum levels of IL12, another Th1 cytokinewere significantly lower in vaccinated VAD pigs compared to vacci-nated VAS pigs, a reverse trend as compared to IFN�. Although thesefindings are incongruent with the currently accepted view, thesedifferential effects of VAD on IFN-� and IL-12 cytokine expressionlevels have also been reported in VAD rats [54]. Our study and pre-vious studies have shown that vitamin A affects Th1–Th2 balanceand vitamin A deficiency results in Th1-biased immune responses[17,55]. In agreement with effects of vitamin A, we observed higheramounts of Th2 cytokine, IL4 in the vaccinated and vitamin A sup-plemented VAS groups compared to the vaccinated VAD groupspost-challenge. In general, our results concur with previous find-ings in which VAD results in altered cytokine responses in humans[17,55] or in animal models [53,56]. Overall, in our study, loweranti-inflammatory IL10 and higher pro-inflammatory IL8 and IFN�levels in VAD animals may contribute to more severe RV infectionand diarrhea in these animals.
In our study, supplementation with 100,000 IU vitamin A (doseas per WHO recommendations) concurrent with each vaccinationdose had no effects on improving either antibody responses toRotaTeq® vaccine or ameliorating the severity of diarrhea and virusshedding in VAD piglets post-challenge. In contrast, it appears thatsupplementation of vitamin A may result in increased diarrheaseverity post-challenge. At present, we do not have an explanationfor this observation. However, in our previous study [25] wherewe used a monovalent RV vaccine, we did not observe any trendfor increased diarrhea severity in vitamin A supplemented VADand VAS vaccinated or non-vaccinated animals compared to theircounterpart non-supplemented groups. Reasons behind the lackof beneficial effects of vitamin A supplementation are unknown,
but one or more of the following factors might be involved: (a)insufficient/non-optimal dose of supplemental vitamin A; (b) tim-ing of vitamin A supplementation relative to RV vaccine regimenand/or (c) intrinsic limitations of immune cells from gestational
Vaccin
prsppgttAiIbim[
mtwpmebV
nppovV
A
eJgRgtO
A
fj
R
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
S. Kandasamy et al. /
renatal VAD hosts to respond to supplemental vitamin A such aseduced expression of RAR� receptors, etc. [25]. Reduced expres-ion of RAR� receptor may play a critical role in induction ofrotective immune responses in VAD groups because Vac + VitAigs had hepatic vitamin A levels similar to those of Vac VASroup; however, they had lower immune responses comparedo the latter groups. Regarding dosage of vitamin A supplemen-ation, a randomized trial conducted in India showed that vitamin
supplementation to both mothers at 3–4 weeks postpartum andnfants enhanced antibody responses to poliovirus type 1 [57].n our case, modification in timing of supplementation such aseginning supplementation prior to initiation of RV vaccination,
ncreasing/decreasing the dose of vitamin A, or supplementing vita-in A to the mothers might have improved protective immunity
58].These preliminary studies were conducted in a germfree pig
odel to observe clear interactions between VAD/vitamin A andhe host immune system without confounding microflora, whichould add an extra variable complicating data analysis and inter-retation. The commensal bacteria are involved in vitamin andineral metabolism including vitamin A and in the future we will
xtend our studies for vitamin A supplementation and RV vaccinesy using defined commensal colonized or humanized (microflora)AD piglets [59].
In conclusion, VAD compromised RV vaccine efficacy in our Gneonatal piglet model. Thus, it may be possible that the higherrevalence of VAD in children in developing countries may be aotential factor affecting the efficacy of oral RV vaccines. Resultsf this study also suggest that optimization of dose and timing ofitamin A supplementation may be required to alleviate potentialAD associated reduced efficacy of RV vaccines in children.
cknowledgements
We gratefully acknowledge the technical assistance of Dr. Juli-tte Hanson, Rich McCormick, Lindsey Good, Ozkan Timurkan,oshua Amimo, and Kyle T. Scheuer. This work was supported byrants from Merck & Co. Inc. (ID#:35907) and Ohio Agriculturalesearch and Development Center (OARDC) research enhancementrant (2009013), The Ohio State University (Linda J. Saif-PI, Anas-asia N. Vlasova-Co-PI), and state and federal funds allocated to theARDC.
ppendix A. Supplementary data
Supplementary data associated with this article can beound, in the online version, at http://dx.doi.org/10.1016/.vaccine.2013.12.039.
eferences
[1] Heaton PM, Goveia MG, Miller JM, Offit P, Clark HF. Development of apentavalent rotavirus vaccine against prevalent serotypes of rotavirus gas-troenteritis. J Infect Dis 2005;192(September (Suppl. 1)):S17–21.
[2] Armah GE, Sow SO, Breiman RF, Dallas MJ, Tapia MD, Feikin DR, et al. Efficacy ofpentavalent rotavirus vaccine against severe rotavirus gastroenteritis in infantsin developing countries in sub-Saharan Africa: a randomised, double-blind,placebo-controlled trial. Lancet 2010 Aug 21;376(9741):606–14.
[3] Zaman K, Dang DA, Victor JC, Shin S, Yunus M, Dallas MJ, et al. Efficacy of pen-tavalent rotavirus vaccine against severe rotavirus gastroenteritis in infants indeveloping countries in Asia: a randomised, double-blind, placebo-controlledtrial. Lancet 2010 Aug 21;376(9741):615–23.
[4] Vesikari T, Karvonen AV, Majuri J, Zeng SQ, Pang XL, Kohberger R, et al. Safety,efficacy, and immunogenicity of 2 doses of bovine-human (UK) and rhesus-
rhesus-human rotavirus reassortant tetravalent vaccines in Finnish children. JInfect Dis 2006 Aug 1;194(3):370–6.
[5] Boom JA, Tate JE, Sahni LC, Rench MA, Hull JJ, Gentsch JR, et al. Effectiveness ofpentavalent rotavirus vaccine in a large urban population in the United States.Pediatrics 2010 Feb;125(2):e199–207.
[
e 32 (2014) 816– 824 823
[6] Danneskiold-Samsoe N, Fisker AB, Jorgensen MJ, Ravn H, Andersen A, Balde ID,et al. Determinants of vitamin a deficiency in children between 6 months and2 years of age in Guinea-Bissau. BMC Public Health 2013;13:172.
[7] West Jr KP. Extent of vitamin A deficiency among preschool chil-dren and women of reproductive age. J Nutr 2002;132(September (9Suppl.)):2857S–66S.
[8] Singh V, West Jr KP. Vitamin A deficiency and xerophthalmia amongschool-aged children in Southeastern Asia. Eur J Clin Nutr 2004;58(October(10)):1342–9.
[9] Wedner SH, Ross DA, Congdon N, Balira R, Spitzer V, Foster A. Validation ofnight blindness reports among children and women in a vitamin A deficientpopulation in rural Tanzania. Eur J Clin Nutr 2004;58(March (3)):409–19.
10] Organization WH. Global prevalence of vitamin A deficiency in populations atrisk 1995–2005: WHO global database on vitamin A deficiency. Geneva: WHO;2009. p. 283.
11] Rahman MM, Mahalanabis D, Hossain S, Wahed MA, Alvarez JO, Siber GR,et al. Simultaneous vitamin A administration at routine immunization con-tact enhances antibody response to diphtheria vaccine in infants younger thansix months. J Nutr 1999;129(December (12)):2192–5.
12] Bhaskaram P, Rao KV. Enhancement in seroconversion to measles vaccinewith simultaneous administration of vitamin A in 9-months-old Indian infants.Indian J Pediatr 1997;64(July–August (4)):503–9.
13] Ma Y, Ross AC. Toll-like receptor 3 ligand and retinoic acid enhance germinalcenter formation and increase the tetanus toxoid vaccine response. Clin VaccineImmunol 2009;16(October (10)):1476–84.
14] Mora JR, Iwata M, Eksteen B, Song SY, Junt T, Senman B, et al. Genera-tion of gut-homing IgA-secreting B cells by intestinal dendritic cells. Science2006;314(November (5802)):1157–60.
15] Hammerschmidt SI, Friedrichsen M, Boelter J, Lyszkiewicz M, Kremmer E,Pabst O, et al. Retinoic acid induces homing of protective T and B cells to thegut after subcutaneous immunization in mice. J Clin Invest 2011;121(August(8)):3051–61.
16] Iwata M, Hirakiyama A, Eshima Y, Kagechika H, Kato C, Song SY. Retinoicacid imprints gut-homing specificity on T cells. Immunity 2004;21(October(4)):527–38.
17] Wieringa FT, Dijkhuizen MA, West CE, van der Ven-Jongekrijg J, van der MeerJW, Muhilal. Reduced production of immunoregulatory cytokines in vitaminA- and zinc-deficient Indonesian infants. Eur J Clin Nutr 2004;58(November(11)):1498–504.
18] Cantorna MT, Nashold FE, Hayes CE. In vitamin A deficiency multiple mech-anisms establish a regulatory T helper cell imbalance with excess Th1 andinsufficient Th2 function. J Immunol 1994;152(February (4)):1515–22.
19] Penniston KL, Valentine AR, Tanumihardjo SA. A theoretical increase in infants’hepatic vitamin a is realized using a supplemented lactating sow model. J Nutr2003;133(April (4)):1139–42.
20] Schmitz HH, Poor CL, Wellman RB, Erdman Jr JW. Concentrations of selectedcarotenoids and vitamin A in human liver, kidney and lung tissue. J Nutr1991;121(October (10)):1613–21.
21] Howells LC, Livesey CT. A survey of vitamin A concentrations in the liver offood-producing animals. Food Addit Contam 1998;15(January (1)):10–8.
22] Valentine AR, Tanumihardjo SA. One-time vitamin A supplementation of lac-tating sows enhances hepatic retinol in their offspring independent of dosesize. Am J Clin Nutr 2005;81(February (2)):427–33.
23] Surles RL, Mills JP, Valentine AR, Tanumihardjo SA. One-time graded doses ofvitamin A to weanling piglets enhance hepatic retinol but do not always preventvitamin A deficiency. Am J Clin Nutr 2007;86(October (4)):1045–53.
24] Meyer RC, Bohl EH, Kohler EM, Procurement. Maintenance of germ-free seine for microbiological investigations. Appl Microbiol 1964;12(July):295–300.
25] Vlasova AN, Chattha KS, Kandasamy S, Siegismund CS, Saif LJ. Prenatallyacquired vitamin A deficiency alters innate immune responses to humanrotavirus in a gnotobiotic pig model. J Immunol 2013;190(9):4742–53.
26] Ward LA, Rosen BI, Yuan L, Saif LJ. Pathogenesis of an attenuated and a virulentstrain of group A human rotavirus in neonatal gnotobiotic pigs. J Gen Virol1996;77(July (Pt 7)):1431–41.
27] Yuan L, Ward LA, Rosen BI, To TL, Saif LJ. Systematic and intestinalantibody-secreting cell responses and correlates of protective immunity tohuman rotavirus in a gnotobiotic pig model of disease. J Virol 1996;70(May(5)):3075–83.
28] Parreno V, Hodgins DC, de Arriba L, Kang SY, Yuan L, Ward LA, et al. Serumand intestinal isotype antibody responses to Wa human rotavirus in gnotobi-otic pigs are modulated by maternal antibodies. J Gen Virol 1999;80(June (Pt6)):1417–28.
29] Azevedo MS, Yuan L, Pouly S, Gonzales AM, Jeong KI, Nguyen TV, et al. Cytokineresponses in gnotobiotic pigs after infection with virulent or attenuated humanrotavirus. J Virol 2006;80(January (1)):372–82.
30] Chattha KS, Vlasova AN, Kandasamy S, Rajashekara G, Saif LJ. Divergentimmunomodulating effects of probiotics on T cell responses to oral atten-uated human rotavirus vaccine and virulent human rotavirus infection in aneonatal gnotobiotic piglet disease model. J Immunol 2013;191(September(5)):2446–56.
31] Chattha KS, Vlasova AN, Kandasamy S, Rajashekara G, Saif LJ. Divergentimmunomodulating effects of probiotics on T cell responses to oral atten-uated human rotavirus vaccine and virulent human rotavirus infection in aneonatal gnotobiotic piglet disease model. J Immunol 2013;191(September(5)):2446–56.
[response to bacterial antigens and the maintenance of natural killer cells. Clin
24 S. Kandasamy et al. / V
32] Seo G-Y, Jang Y-S, Kim H-A, Lee M-R, Park M-H, Park S-R, et al. Retinoic acid,acting as a highly specific IgA isotype switch factor, cooperates with TGF-�1 toenhance the overall IgA response. J Leukoc Biol 2013;94(2):325–35.
33] Mora JR, Iwata M, Eksteen B, Song S-Y, Junt T, Senman B, et al. Generationof gut-homing IgA-secreting B cells by intestinal dendritic cells. Sci Signal2006;314(5802):1157.
34] Demissie T, Ali A, Mekonnen Y, Haider J, Umeta M. Demographic and health-related risk factors of subclinical vitamin A deficiency in Ethiopia. J Health PopulNutr 2009;27(October (5)):666–73.
35] Semba RD. Vitamin A and immunity to viral, bacterial and protozoan infections.Proc Nutr Soc 1999;58(August (3)):719–27.
37] Wasantwisut E. Recommendations for monitoring and evaluating vita-min A programs: outcome indicators. J Nutr 2002;132(September (9Suppl.)):2940S–2S.
38] Ferraz IS, Daneluzzi JC, Vannucchi H, Jordao Jr AA, Ricco RG, Del CiampoLA, et al. Detection of vitamin A deficiency in Brazilian preschool childrenusing the serum 30-day dose-response test. Eur J Clin Nutr 2004;58(October(10)):1372–7.
39] Underwood BA. Methods for assessment of vitamin A status. J Nutr1990;120(November (Suppl. 11)):1459–63.
40] Wiedermann U, Hanson LA, Holmgren J, Kahu H, Dahlgren UI. Impaired mucosalantibody response to cholera toxin in vitamin A-deficient rats immunized withoral cholera vaccine. Infect Immun 1993;61(September (9)):3952–7.
41] Ertesvag A, Naderi S, Blomhoff HK. Regulation of B cell proliferation and differ-entiation by retinoic acid. Semin Immunol 2009;21(February (1)):36–41.
42] Tokuyama H, Tokuyama Y. Retinoids enhance IgA production bylipopolysaccharide-stimulated murine spleen cells. Cell Immunol1993;150(September (2)):353–63.
43] Tokuyama H, Tokuyama Y. The regulatory effects of all-trans-retinoic acid onisotype switching: retinoic acid induces IgA switch rearrangement in cooper-ation with IL-5 and inhibits IgG1 switching. Cell Immunol 1999;192(February(1)):41–7.
44] Semba RD, Muhilal, Scott AL, Natadisastra G, Wirasasmita S, Mele L, et al.Depressed immune response to tetanus in children with vitamin A deficiency.J Nutr 1992;122(January (1)):101–7.
45] Pasatiempo AM, Bowman TA, Taylor CE, Ross AC. Vitamin A depletion andrepletion: effects on antibody response to the capsular polysaccharide of
Streptococcus pneumoniae, type III (SSS-III). Am J Clin Nutr 1989;49(March(3)):501–10.
46] Berlin C, Berg EL, Briskin MJ, Andrew DP, Kilshaw PJ, Holzmann B, et al. Alpha 4beta 7 integrin mediates lymphocyte binding to the mucosal vascular addressinMAdCAM-1. Cell 1993;74(July 16 (1)):185–95.
[
e 32 (2014) 816– 824
47] Kantele JM, Arvilommi H, Kontiainen S, Salmi M, Jalkanen S, Savilahti E, et al.Mucosally activated circulating human B cells in diarrhea express homingreceptors directing them back to the gut. Gastroenterology 1996;110(April(4)):1061–7.
48] Williams MB, Rose JR, Rott LS, Franco MA, Greenberg HB, Butcher EC. Thememory B cell subset responsible for the secretory IgA response and protec-tive humoral immunity to rotavirus expresses the intestinal homing receptor,alpha4beta7. J Immunol 1998;161(October (8)):4227–35.
49] Youngman KR, Franco MA, Kuklin NA, Rott LS, Butcher EC, Greenberg HB. Corre-lation of tissue distribution, developmental phenotype, and intestinal homingreceptor expression of antigen-specific B cells during the murine anti-rotavirusimmune response. J Immunol 2002;168(March (5)):2173–81.
50] Morikawa K, Nonaka M. All-trans-retinoic acid accelerates the differentiationof human B lymphocytes maturing into plasma cells. Int Immunopharmacol2005;5(December (13–14)):1830–8.
51] Rousset F, Garcia E, Defrance T, Peronne C, Vezzio N, Hsu DH, et al.Interleukin 10 is a potent growth and differentiation factor for activatedhuman B lymphocytes. Proc Natl Acad Sci U S A 1992;89(March (5)):1890–3.
52] Carman JA, Hayes CE. Abnormal regulation of IFN-gamma secretion in vitaminA deficiency. J Immunol 1991;147(August (4)):1247–52.
53] Cantorna MT, Nashold FE, Hayes CE. Vitamin A deficiency results in a primingenvironment conducive for Th1 cell development. Eur J Immunol 1995;25(June(6)):1673–9.
54] Yang Y, Yuan Y, Tao Y, Wang W. Effects of vitamin A deficiency onmucosal immunity and response to intestinal infection in rats. Nutrition2011;27(February (2)):227–32.
55] Jason J, Archibald LK, Nwanyanwu OC, Sowell AL, Buchanan I, Larned J, et al.Vitamin A levels and immunity in humans. Clin Diagn Lab Immunol 2002;9(May(3)):616–21.
56] Stephensen CB, Jiang X, Freytag T. Vitamin A deficiency increases the in vivodevelopment of IL-10-positive Th2 cells and decreases development of Th1cells in mice. J Nutr 2004;134(October (10)):2660–6.
57] Bahl R, Bhandari N, Kant S, Molbak K, Ostergaard E, Bhan MK. Effect ofvitamin A administered at expanded program on immunization contacts onantibody response to oral polio vaccine. Eur J Clin Nutr 2002;56(April (4)):321–5.
58] Ross AC. Vitamin A deficiency and retinoid repletion regulate the antibody
Immunol Immunopathol 1996;80(September (3 Pt 2)):S63–72.59] Resta SC. Effects of probiotics and commensals on intestinal epithelial phys-
iology: implications for nutrient handling. J Physiol 2009;587(September (Pt17)):4169–74.
