lable at ScienceDirect

Clinical Nutrition xxx (2013) 1e5

Contents lists avai

Clinical Nutrition

journal homepage: ht tp: / /www.elsevier .com/locate/c lnu

Short communication

Production of interferon a and b, pro-inflammatory cytokines and theexpression of suppressor of cytokine signaling (SOCS) in obesesubjects infected with influenza A/H1N1

Elí Terán-Cabanillas, Maricela Montalvo-Corral, Erika Silva-Campa, Graciela Caire-Juvera,Silvia Y. Moya-Camarena, Jesús Hernández*

Coordinación de Nutrición, Centro de Investigación en Alimentación y Desarrollo, A.C., Hermosillo, Sonora, Mexico

a r t i c l e i n f o

Article history:Received 3 August 2013Accepted 16 October 2013

Keywords:ObesityInfluenzaTLRIFNSOCS3

* Corresponding author. Laboratorio de InmunologíCentro de Investigación en Alimentación y DesarrolloVictoria, Hermosillo, Sonora 83304, Mexico. Tel.: þ522892400.

E-mail address: jhdez@ciad.mx (J. Hernández).

0261-5614/$ e see front matter � 2013 Elsevier Ltd ahttp://dx.doi.org/10.1016/j.clnu.2013.10.011

Please cite this article in press as: Terán-Cabsuppressor of cytokine signaling (SOCS) in oj.clnu.2013.10.011

s u m m a r y

Background & aims: Obesity was recognized as an independent risk factor for morbidity and mortalityduring last influenza A/H1N1 pandemic. Mechanisms involved in the high mortality risk from obesityduring influenza A virus include reduced type I interferon production and delayed pro-inflammatoryresponse, which lead to a higher rate of morbidity and mortality in murine models. In this study, weevaluated the production of type I interferons, pro-inflammatory and anti-inflammatory cytokines inperipheral blood mononuclear cells (PBMCs) from obese and lean subjects with and without confirmedinfection of influenza A/H1N1. The expression levels of the suppressor of cytokine signaling-1 (SOCS1),SOCS3 and nuclear factor-kB were also evaluated.Methods: Cytokines were measured by real-time quantitative reverse transcription polymerase chainreaction (qRT-PCR) and/or by ELISA in PBMCs stimulated with toll like receptor-3 (TLR-3) and TLR-7ligands. The mRNA expression of SOCS1 and SOCS3 were evaluated by qRT-PCR.Results: The obese volunteers infected with influenza A/H1N1 showed a diminished ability to producetype I interferon in response to TLR-3 ligand. Interestingly, the pro-inflammatory response was alsoaffected in TLR-3 stimulated PBMCs. Obese influenza-free volunteers showed an increased basalexpression of SOCS3, but not SOCS1. During influenza infection, SOCS1 and SOCS3 expression was higherin the lean infected volunteers in contrast to those who were obese infected.Conclusions: These data suggest that obesity is related to TLR-3 impairment and explain, at least inpart, the inadequate immune response of obese individuals during infection with influenza A/H1N1virus.

� 2013 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.

1. Introduction

During the 2009 H1N1 influenza pandemic, obesity wasrecognized as an independent risk factor for increased influenzamorbidity and mortality.1 Mechanisms involved in the high mor-tality risk from obesity during influenza A virus infection havenbeen described in murine models and include delayed pro-inflammatory response and reduced type I interferon production,which lead to a higher rate of morbidity and mortality.2

a, Coordinación de Nutrición,, A.C., Km 0.6 Carretera a la662 2800010; fax: þ52 662

nd European Society for Clinical N

anillas E, et al., Production ofbese subjects infected with i

Type I IFNs are key cytokines involved in the early immuneresponse to viral infections and are secreted after the recognition ofviral components by intracellular receptors such as toll-like re-ceptor (TLR)-3 and TLR-7. Influenza virus is capable of inducingsuppressor of cytokine signaling (SOCS) expression, proteins thatare key regulators of type I interferons and pro-inflammatory cy-tokines,3 and thus evading the immune system.4

We have previously shown that PBMCs from obese individualshave a reduced ability to produce type I interferon and pro-inflammatory response after TLR stimulation. This aberrantresponse was related to increased SOCS3, but not SOCS1, transcriptbaseline levels in obese subjects.5 The aim of this study is toexamine the IFN-a/b, pro-inflammatory and anti-inflammatoryresponses, and SOCS1 and SOCS3 expression in obese and non-obese patients infected with the 2009 pandemic influenza A/H1N1.

utrition and Metabolism. All rights reserved.

interferon a and b, pro-inflammatory cytokines and the expression ofnfluenza A/H1N1, Clinical Nutrition (2013), http://dx.doi.org/10.1016/

E. Terán-Cabanillas et al. / Clinical Nutrition xxx (2013) 1e52

2. Methods

2.1. Experimental design and participants

This was a cross-sectional study in which peripheral bloodsamples (12 mL) were collected from fasting participants and bodymass index (BMI, kg/m2) was calculated. Participants were dividedinto four groups according to their BMI (obese, BMI � 30 and non-obese BMI < 30) and whether they were infected with influenza A/H1N1 as confirmed by Real-Time PCR: lean controls non-infected(n ¼ 4); obese control non-infected (n ¼ 9); lean influenza A/H1N1-infected (n ¼ 4); and obese influenza A/H1N1-infected(n ¼ 8). The protocol was approved by the Ethical Committee ofthe Centro de Investigación en Alimentación y Desarrollo, A.C., andvolunteers provided their written informed consent.

2.2. Procedures

Peripheral blood mononuclear cells (PBMCs) were isolated andstimulated with TLR-3 or TLR-7 ligand (Poly I:C and CL264,respectively; Invitrogen, San Diego, CA, USA) as previously re-ported.5 RNA extraction and Real-time qRT-PCR conditions andprimers for IFN-a2, IFN-a6, SOCS1 and SOCS3, and NF-kB have beenpreviously described.5 Type I IFNs and inflammatory cytokineswere quantified in supernatants of stimulated cells by ELISA,following the manufacturers’ recommendations.

2.3. Statistical analysis

Group comparisons were performed using the non-parametricManneWhitney U test. Analyses were performed using Stataversion 9.2 (StataCorp LP, College Station, TX, USA).

3. Results

3.1. SOCS3 and SOCS1 mRNA expression

To determine the effect of obesity in SOCS expression duringinfluenza A/H1N1 infection, we evaluated the basal transcript levelsof SOCS1 and SOCS3 in PBMCs from obese and lean patientsinfected or not with influenza A/H1N1 (Fig. 1). SOCS1 and SOCS3expression was significantly lower in the obese infected patientscompared with the lean infected patients (p < 0.05). SOCS1 andSOCS3 expression in lean influenza-infected patients was alsohigher compared with lean volunteers who were uninfected(p < 0.05).

0.00000.00050.00100.00150.0020

0.05

0.10

0.15

SOCS1 SOCS3

p<0.05

p<0.005

p<0.005

p<0.05p<0.05

p<0.005

Rel

ativ

e ex

pres

sion

/ β-a

ctin

Fig. 1. Basal mRNA expression of SOCS1 and SOCS3. qRT-PCR was used (as previouslydescribed in Materials and Methods) to evaluate the basal mRNA expression of SOCS1and SOCS3 from PBMCs obtained from obese (n ¼ 8, checkered bars) and lean (n ¼ 4,gray bars) patients infected with influenza A/H1N1, and obese (n ¼ 8, black bars) andlean (n ¼ 4, white bars) uninfected controls. Data represent the mRNA relativeexpression of SOCS1 or SOCS3 vs. a housekeeping gene (b-actin). Values represent themean � SEM; and significant differences are expressed.

Please cite this article in press as: Terán-Cabanillas E, et al., Production ofsuppressor of cytokine signaling (SOCS) in obese subjects infected with ij.clnu.2013.10.011

3.2. Effect of obesity on IFN-a expression after TLR stimulation ininfluenza A/H1N1 infected individuals

To determine the effect of obesity in the type I interferonresponse during an influenza A H1N1 infection, we evaluatedmRNA levels of IFN-a2 and IFN-a6 in TLR-3 and TLR-7 stimulatedPBMCs. When PBMCs were stimulated with the TLR-3 ligand, theobese influenza A/H1N1 infected participants showed decreasedIFNa2 transcripts levels compared to lean (p < 0.05) and obesecontrols (p < 0.05). Obese infected participants also showeddecreased IFNa6 levels compared to lean controls (p< 0.05) and, aswe have previously shown,5 lean influenza-free subjects showedhigher transcripts levels of both IFNa2 and IFNa6 compared toobese influenza-free volunteers (p < 0.05) after TLR-3 stimulationof PBMCs (Fig. 2). When PBMCs were stimulated with TLR-7 ligand,the obese influenza A/H1N1 infected subjects also showeddecreasedmRNA levels of IFNa2 and IFNa6 (p< 0.05). Interestingly,a statistical difference was found in the mRNA levels of IFNa6 be-tween lean infected and obese infected participants (p < 0.05).Again, the lean controls had increased IFNa2 and IFNa6 levels afterTLR-7 stimulation compared to obese controls (p < 0.05) (Fig. 2).

3.3. Type I interferon response by ELISA in TLR-3 and TLR-7 PBMCs

To evaluate if aberrant IFN-a mRNA expression correlated withtype I interferon production in obese influenza-infected in-dividuals, we evaluated both IFN-a and b production after stimu-lationwith TLR-7 and TLR-3, respectively. Lean-infected and obese-infected participants showed similar levels of IFN-a productionafter TLR-7 stimulation (170.36 � 35.01 and 170.52 � 8.15 pg/mL,respectively; p > 0.05). Consistent with interferon mRNA expres-sion, obese-infected patients showed lower levels of IFN-b pro-duction than lean-infected volunteers in PBMC supernatantsstimulated with the TLR-3 ligand (26.82 � 15.99 and65.58 � 33.07 IU/mL, respectively; p < 0.05) (Fig. 2).

3.4. Effect of obesity on NF-kB transcription factor mRNA levels

Obese and lean infected participants, as well as obese controls,showed diminished activation of NF-kB after TLR3 stimulationwhen compared to lean control subjects (p < 0.05). However, nodifference in NF-kB activation was found between obese and leaninfected subjects (p> 0.05) (Fig. 3). Regarding NF-kB expression inPBMCs stimulated with TLR-7 ligand, the obese infected and obesecontrols showed lower mRNA levels than lean controls (p < 0.05),but no difference was observed between lean infected and leancontrols and obese infected and obese controls (p > 0.05) (Fig. 3).

3.5. Pro-inflammatory and anti-inflammatory cytokine productionin PBMCs stimulated with TLR ligands

Finally, IL-6, IL-1b and IL-10 were measured in supernatantsfrom PBMC stimulated with TLR ligands for 20 h by ELISA. Obeseinfected participants showed a diminished capability to produce IL-6 and IL-1b after both TLR-3 and TLR-7 stimulation (p < 0.05)(Fig. 3). Interestingly, the reduction in IL-6 production was moresevere when cells were stimulated with TLR3 than with TLR7(Fig. 3). Similar results were observed with IL-1b, where the obeseinfected showed reduced production after TLR stimulation (Fig. 3).Interestingly, only the lean controls produced IL-10, in contrast toobese and lean infected participants, and obese controls (p < 0.05).When PBMC’s were stimulated through TLR-7, the obese influenza-infected patients had a decrease in IL-10 expression compared tolean and obese controls (p < 0.05). However, no statistical

interferon a and b, pro-inflammatory cytokines and the expression ofnfluenza A/H1N1, Clinical Nutrition (2013), http://dx.doi.org/10.1016/

IFN-α2 IFN-α6

Rel

ativ

e ex

pres

sion

/non

-sti

mul

ated

cel

ls

0

10

20

30

40

50 p< 0.05

TL

R3

TL

R7

0

20

40

60 p< 0.05

p< 0.05

0

5

10

15

p< 0.05

0

5

10

15p< 0.05

p< 0.05

0

20

40

60

80

100p< 0.05

IU/m

L

IFN-β

0

50

100

150

200

pg/m

LIFN-α

Fig. 2. Decreased type I interferon induction in TLR-3 ligand-stimulated PBMCs from obese A/H1N1 influenza infected patients. IFN-a expression was measured by qRT-PCR in TLR-stimulated PBMCs from obese infected (checkered bars, n ¼ 8); lean influenza A/H1N1 infected (gray bars, n ¼ 4); obese controls (black bars, n ¼ 9) and lean controls (white bars,n ¼ 4). IFN-b production in cell supernatants from PBMCs stimulated for 20 h with TLR-3 ligand measured by ELISA. Obese A/H1N1 influenza patients (n ¼ 8) and lean A/H1N1influenza participants (n ¼ 4), checkered and gray bars respectively. IFN-a production in cell supernatants from PBMCs stimulated with TLR-7 ligand measured by ELISA. Obese A/H1N1 influenza patients (n ¼ 8) and lean A/H1N1 influenza participants (n ¼ 4), checkered and gray bars respectively. Values are expressed as means � SEM.

E. Terán-Cabanillas et al. / Clinical Nutrition xxx (2013) 1e5 3

difference was found between lean influenza-infected subjects andthe obese and lean controls (p > 0.05) (Fig. 3).

4. Discussion

Elevated expression of SOCS3 had been associated with areduced production of interferon and had been proposed as amechanism of impaired response against influenza virus,4 the factthat the obese controls showed higher SOCS3 mRNA expressioncompared with the lean controls, might make them susceptible tohave an impaired type I interferon response during influenzainfection. The elevated expression of SOCS3 has also been observedin other viral infections, such Hepatitis C virus, which supports theparticipation of SOCS3 during viral infections.6 However, it wasinteresting to note that SOCS3 expression in obese patients infectedwith influenza A/H1N1 was not increased, similar to SOCS1expression. These data are consistent with previous results thatshowed that PBMCs from obese individuals did not increasinglyrespond to TLR-ligand stimulation, unlike PBMCs from healthycontrols.5 These data suggest that obese individuals differentiallyregulate SOCS3 after cell stimulation. This response could beinvolved in the immunomodulation of the response of obese,

Please cite this article in press as: Terán-Cabanillas E, et al., Production ofsuppressor of cytokine signaling (SOCS) in obese subjects infected with ij.clnu.2013.10.011

especially when cells are stimulated through TLR-3 ligands, as oc-curs during influenza virus infection.7

mRNA expression on IFN-a2 was lower in the obese infectedgroup than in obese controls, suggesting that the combination ofobesity and influenza infection affects the expression of IFN-a2. Thisscenario appeared to be different when cells were stimulated withTLR-7. In this case, no differences were observed with the combi-nation of obesity and influenza infection, even though a higherresponse was observed in the non-obese infected group. These datasuggest that obesity may affect the signaling of TLR-3 receptor,which, in turn, causes the reduced mRNA expression of interferon.This characteristic could be actively involved during the process ofinfluenza infection. Data obtained from the quantification of IFNsreleased to the supernatant support this conclusion. In sum, weobserved the reduced production of IFN when cells were stimulatedwith TLR-3 ligand, in contrast to the stimulus with TLR-7.

Consistent with our results, Smith and collaborators found thatDIO mice infected with A/PR8/34 influenza present minimal in-duction of IFN-a and IFN-b. This phenomenon is accompanied by adecrease in natural killer cytotoxicity and a delayed pro-inflammatory response, which results in increased mortality frominfection.2 Levels of INF-a are higher in cultures of macrophages48 h after infection with the A/PR/8/34 influenza strain than with

interferon a and b, pro-inflammatory cytokines and the expression ofnfluenza A/H1N1, Clinical Nutrition (2013), http://dx.doi.org/10.1016/

TLR-3 TLR-7R

elat

ive

expr

essi

on/

non-

stim

ulat

ed c

ells

0

2

4

6

8 p<0.05

0

2

4

6

8

10 p<0.05

-300

0

300

600

900

1200

TLR-3 TLR-7

p 0.05

p 0.05

p 0.05

p 0.05

p 0.05

pg/m

L

IL-6

-900

-600

-300

0

300

600

900

TLR-3 TLR-7

p 0.05

p 0.05

p 0.05

p 0.05

pg/m

L

IL-1β

-600

-300

0

300

600

TLR-3 TLR-7

p 0.05

p 0.05

pg/m

L

IL-10

Fig. 3. Decreased pro-inflammatory and anti-inflammatory response in obesity and influenza A/H1N1 infection. Obese A/H1N1 influenza infected (checkered bars, n ¼ 8), lean A/H1N1 influenza infected (gray bars, n ¼ 4), obese controls (black bars, n ¼ 9) and lean controls (white bars, n ¼ 4). NF-kB expression by qRT-PCR in PBMCs stimulated with TLR-3ligand. NF-kB expression by qRT-PCR in PBMCs stimulated with TLR-7 ligand. IL-6 production measured by ELISA in cell supernatants from TLR-3 and TLR-7 stimulated cells. IL-1bproduction measured by ELISA in cell supernatants from TLR-3 and TLR-7 stimulated cells. IL-10 production measured by ELISA in cell supernatants from TLR-3 and TLR-7 stimulatedcells. Cytokine production of stimulated cells was subtracted from cytokine production in non-stimulated cells. Values are expressed as means � SEM.

E. Terán-Cabanillas et al. / Clinical Nutrition xxx (2013) 1e54

the pandemic A/H1N1 virus,8 which suggests that the pandemic A/H1N1 influenza virus evades the immune system by a mechanismthat involves IFN down-regulation. The fact that the TLR-3 pathwayis altered by obesity suggests that obese individuals have a reducedcapability to mount an immune response during a viral infectionthat is regulated through this receptor and may explain, at least inpart, the association between obesity and higher susceptibly to A/H1N1 influenza infection.

Please cite this article in press as: Terán-Cabanillas E, et al., Production ofsuppressor of cytokine signaling (SOCS) in obese subjects infected with ij.clnu.2013.10.011

Whenwe analyzed pro-inflammatory cytokines (IL-1b and IL-6)we also observed the effects of impaired stimulation by TLR3-ligand induction. These effects are not only observed in the pro-duction of pro-inflammatory cytokines, but also in the anti-inflammatory response of IL-10. These data support the hypothe-sis that, in some way, obesity contributes to an altered pathway inthe signaling of TLR-3, and when obese individuals are infectedwith influenza A/H1N1 the effects are apparently worse.

interferon a and b, pro-inflammatory cytokines and the expression ofnfluenza A/H1N1, Clinical Nutrition (2013), http://dx.doi.org/10.1016/

E. Terán-Cabanillas et al. / Clinical Nutrition xxx (2013) 1e5 5

In conclusion, our results suggest that obese A/H1N1 influenza-infected volunteers present a diminished type I interferon and pro-inflammatory response in a TLR-3 dependent pathway. These re-sults could help to explain the association between obesity and theincreased morbidity and mortality caused by A/H1N1 influenzainfection.

Author contribution

Conceived and designed the experiments: ETC MMC GCC SMCJH. Performed the experiments: ETC MMC ESC. Analyzed the data:ETC MMC GCC JH. Contributed reagents/materials/analysis tools:GCC SMC JH. Wrote the paper: ETC MMC JH.

Conflict of interest

Authors declare that no competing interests exist.

Acknowledgments

This study was supported by the FONDO SECTORIAL SALUD-CONACYT grant 127013. We thank Monica Resendiz for her tech-nical assistance, Ricardo Pacheco M.D., Raul Pereida M.D, CarmenZamudio M.D, Andrés Mendoza M.D. and the administrative and

Please cite this article in press as: Terán-Cabanillas E, et al., Production ofsuppressor of cytokine signaling (SOCS) in obese subjects infected with ij.clnu.2013.10.011

medical staff of the hospitals. The participation of volunteers is alsoacknowledged.

References

1. Louie JK, Acosta M, Samuel MC, Schechter R, Vugia DJ, Harriman K, et al. A novelrisk factor for a novel virus: obesity and 2009 pandemic influenza A (H1N1). ClinInfect Dis 2011;52:301e12.

2. Smith AG, Sheridan PA, Harp JB, Beck MA. Diet-induced obese mice haveincreased mortality and altered immune responses when infected with influ-enza virus. J Nutr 2007;137:1236e43.

3. Pothlichet J, Chignard M, Si-Tahar M. Cutting edge: innate immune responsetriggered by influenza A virus is negatively regulated by SOCS1 and SOCS3through a RIG-I/IFNAR1-dependent pathway. J Immunol 2008;180:2034e8.

4. Pauli E-K, Schmolke M, Wolff T, Viemann D, Roth J, Bode JG, et al. Influenza Avirus inhibits type I IFN signaling via NF-kB-dependent induction of SOCS-3expression. PLoS Pathog 2008;4:e1000196.

5. Teran-Cabanillas E, Montalvo-Corral M, Caire-Juvera G, Moya-Camarena SY,Hernández J. Decreased interferon-a and interferon-b production in obesity andexpression of suppressor of cytokine signaling. Nutrition 2013;29:207e12.

6. Walsh M, Jonsson J, Richardson MM, Lipka G, Purdie D, Clouston A, et al. Non-response to antiviral therapy is associated with obesity and increased hepaticexpression of suppressor of cytokine signalling 3 (SOCS-3) in patients withchronic hepatitis C, viral genotype 1. Gut 2006;55:529e35.

7. Guillot L, Le Goffic R, Bloch S, Escriou N, Akira S, Chignard M, et al. Involvementof toll-like receptor 3 in the immune response of lung epithelial cells to double-stranded RNA and influenza A virus. J Biol Chem 2005;280:5571e80.

8. Ramírez-Martínez G, Cruz-Lagunas A, Jiménez-Alvarez L, Espinosa E, Ortíz-Quintero B, Santos-Mendoza T, et al. Seasonal and pandemic influenza H1N1viruses induce differential expression of SOCS-1 and RIG-I genes and cytokine/chemokine production in macrophages. Cytokine 2013;62:151e9.

interferon a and b, pro-inflammatory cytokines and the expression ofnfluenza A/H1N1, Clinical Nutrition (2013), http://dx.doi.org/10.1016/