UBC Faculty Research and Publications

CMV and Immunosenescence: from basics to clinics Solana, Rafael; Tarazona, Raquel; Aiello, Allison E; Akbar, Arne N; Appay, Victor; Beswick, Mark; Bosch, Jos A; Campos, Carmen; Cantisán, Sara; Cicin-Sain, Luka; Derhovanessian, Evelyna; Ferrando-Martínez, Sara; Frasca, Daniela; Fulöp, Tamas; Govind, Sheila; Grubeck-Loebenstein, Beatrix; Hill, Ann; Hurme, Mikko; Kern, Florian; Larbi, Anis; López-Botet, Miguel; Maier, Andrea B; McElhaney, Janet E; Moss, Paul; Naumova, Elissaveta; Nikolich-Zugich, Janko; Pera, Alejandra; Rector, Jerrald L; Riddell, Natalie; Sanchez-Correa, Beatriz; Sansoni, Paolo; Sauce, Delphine; van Lier, Rene; Wang, George C; Wills, Mark R; Zieliński, Maciej; Pawelec, Graham Oct 31, 2012

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-12979_2012_Article_130.pdf [ 513.06kB ]
JSON: 52383-1.0221622.json
JSON-LD: 52383-1.0221622-ld.json
RDF/XML (Pretty): 52383-1.0221622-rdf.xml
RDF/JSON: 52383-1.0221622-rdf.json
Turtle: 52383-1.0221622-turtle.txt
N-Triples: 52383-1.0221622-rdf-ntriples.txt
Original Record: 52383-1.0221622-source.json
Full Text

Full Text

REVIEW Open AccessCMV and Immunosenescence: from basicsto clinicsRafael Solana1*, Raquel Tarazona2, Allison E Aiello3, Arne N Akbar4, Victor Appay5, Mark Beswick6, Jos A Bosch7,30,Carmen Campos1, Sara Cantisán1, Luka Cicin-Sain8, Evelyna Derhovanessian9, Sara Ferrando-Martínez10,Daniela Frasca11, Tamas Fulöp12, Sheila Govind13, Beatrix Grubeck-Loebenstein14, Ann Hill15, Mikko Hurme16,Florian Kern17, Anis Larbi18, Miguel López-Botet19, Andrea B Maier20, Janet E McElhaney21, Paul Moss6,Elissaveta Naumova22, Janko Nikolich-Zugich23, Alejandra Pera1, Jerrald L Rector24, Natalie Riddell4,Beatriz Sanchez-Correa2, Paolo Sansoni25, Delphine Sauce5, Rene van Lier26, George C Wang27, Mark R Wills28,Maciej Zieliński29 and Graham Pawelec9AbstractAlone among herpesviruses, persistent Cytomegalovirus (CMV) markedly alters the numbers and proportions ofperipheral immune cells in infected-vs-uninfected people. Because the rate of CMV infection increases with age in mostcountries, it has been suggested that it drives or at least exacerbates “immunosenescence”. This contention remainscontroversial and was the primary subject of the Third International Workshop on CMV & Immunosenescence whichwas held in Cordoba, Spain, 15-16th March, 2012. Discussions focused on several main themes including the effects ofCMV on adaptive immunity and immunosenescence, characterization of CMV-specific T cells, impact of CMV infectionand ageing on innate immunity, and finally, most important, the clinical implications of immunosenescence and CMVinfection. Here we summarize the major findings of this workshop.IntroductionThe impact of cytomegalovirus on the immune systemand its relevance for the decline of immune functionwith ageing was discussed by international experts dur-ing the Third International Workshop on CMV &Immunosenescence held in Cordoba, Spain, 15-16thMarch, 2012 (local organizer, Prof. R. Solana). Thisfollowed two previous Workshops held in Tubingen,Germany in 2009, and Cambridge, UK in 2010, the out-comes of which were summarized in this Journal byPawelec et al. [1] and Wills et al. [2]. This commentarysummarizes the major issues discussed at the ThirdWorkshop in this series with an emphasis on thosequestions raised in the previous meetings that were leftopen. The meeting ended with a session of perspectivesand closing remarks that included a discussion summaryand several action items (Figure 1), and will be followedby a 4th Workshop to be organized by Prof. P. Sansoniin Parma, Italy, 25-27th March, 2013.Biomarkers of Immunosenescence and CMV infectionPawelec (Tübingen, Germany) reviewed the recentadvances in immunosenescence defined as the deleteri-ous age-associated changes to immunity observed in allmammals studied so far. It was suggested that a betterdesignation might be immune frailty as a continuousvariable rather than a discrete state. The clinical impactof the observed age-associated changes in componentsof innate and adaptive immunity is mostly not clear inhumans, and controversial data exist regarding themechanisms of immunosenescence and the identifica-tion of new markers. In longitudinal studies of Swedishoctogenarians and nonagenarians (OCTO and NONA)an immune risk profile (IRP) was proposed that wasassociated with increased mortality. The IRP was presentin approximately 15% of individuals at baseline and 4year mortality was almost double in individuals with IRPcompared with non-IRP individuals. The main limita-tions of the IRP were emphasized. Thus, it has been* Correspondence: rsolana@uco.es1Immunology Unit, Instituto Maimónides de Investigación Biomédica deCórdoba (IMIBIC)-Reina Sofia University Hospital-University of Cordoba,Cordoba, SpainFull list of author information is available at the end of the articleIMMUNITY & AGEING© 2012 Solana et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the CreativeCommons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, andreproduction in any medium, provided the original work is properly cited.Solana et al. Immunity & Ageing 2012, 9:23http://www.immunityageing.com/content/9/1/23shown to be relevant only in a small sample of very eld-erly people in one particular country but we do notknow if these results can be extrapolated to larger andyounger samples in different countries. The initial IRPanalysis lacked the sophistication of modern immuneanalysis and it is unknown at which age it may becomerelevant and its association with underlying diseases isunclear. In addition, the IRP does not take into accountmany other clinical factors, such as nutritional status,psychological stress or inflammatory status that may berelevant, as discussed intensively during this workshop.Recent reports have shown that signs of inflammationsuch as increased CRP or IL-6 are independent of theIRP and must be accounted for separately. In the Na-tional Health and Examination Survey (NHANES) IIIstudy, combined CMV-seropositivity and higher CRPlevels were associated with increased mortality. A rela-tionship between functional ability in older people andthe immune system has been documented in severalstudies: frail elderly individuals have higher CMV titers,elevated IL-6 and lower responses to influenza vaccin-ation, but the interrelations between these are not clear.It is also not clear if CMV reactivates more often in theelderly with or without frailty and if higher levels of anti-bodies against CMV are associated with lower survival.Studies of CMV prevalence in different populations haveshown high variability. In elderly Chinese Singaporeans,CMV prevalence is very high (99%) as shown by Larbi(Singapore). Since 2003, the Singapore Longitudinal Age-ing Study (SLAS) recruiting individuals over 55 years ofage has characterized over 4000 study participants for nu-tritional, behavioural, metabolic, social, and biologicalparameters. The extensive clinical information collectedover the years enables us to now identify immune para-meters associated with chronic conditions (eg. diabetes,hypertension, high cholesterol). A global approach includ-ing immune monitoring and bioinformatics should enablethe identification of immune correlates of longevity andco-morbidities, in a systems biology approach. Data pre-sented highlighted that immunological history in Asianelderly (eg. Dengue, H. pylori, CMV, EBV) is differentfrom in other parts of the world. This may be of major im-portance for the identification of other or additional driv-ing forces than CMV leading to immunosenescence.CMV IgG levels are different in CMV-seropositive elderlydepending on the presence of co-morbidities (eg. diabetes)suggesting that stratification of elderly individuals basedon any parameter related to chronic conditions should beconsidered. Correlations between CMV IgG and inflam-matory markers were presented and also shown to correl-ate with the frequency of differentiated CD8+CD28-CD27- T cells. The data presented suggest that betterstratification of the elderly should be performed in orderto understand the role CMV in healthy ageing and howother conditions may synergize or conflict with CMV-induced effects at the immunological level.Figure 1 CMV and immunosenescence: Open questions. The relevance of the following questions on the role of CMV infection onimmunosenescence and inflamm-aging were highlighted: 1) the need to standardize the panel of mAbs used to asses lymphocyte subsetsalterations, 2) the role of each lymphoid subset in anti-CMV response, 3) the significance of CMV-induced inflammation and 4) the complexity ofCMV infection in humans.Solana et al. Immunity & Ageing 2012, 9:23 Page 2 of 9http://www.immunityageing.com/content/9/1/23Derhovanessian (Tübingen, Germany) presented dataon familial longevity as illustrated by the Leiden Longev-ity study (LLS), which includes 450 families in which off-spring enjoy a standardized mortality rate 30% lowerthan their partners from the general population. Off-spring from long-lived families had a significantly lowernumber of late-stage, possibly terminally, differentiatedCD8+ T cells (CD45RA−CCR7−CD27−CD28−) andlatent CMV infection did not have the same impacton the percentage of naive (CD45RA+CCR7+CD27+CD28+) and late-differentiated effector memory CD8+T-cells as it has in the rest of the population. Thus, thedecrease in naïve and accumulation of late-differentiated“senescent” CD8+ T cells that is commonly taken asa hallmark of immunosenescence was not seen inthese subjects. CMV-associated pro-inflammatory status(assayed as CRP levels) in CMV-seropositive offspringfrom long-lived families was also lower than in the gen-eral population, but no differences in the cellularresponses to CMV in vitro were found. However, ana-lysis of the serological response to CM2, a fusion proteincontaining the C-terminal portion of viral proteinpUL44 and a highly reactive fragment of pUL57, asso-ciated with active infection, revealed that the percentageof individuals with detectable levels of IgM and IgG anti-bodies to CM2 was lower in the offspring compared totheir partners; in parallel, a lower percentage of naïveCD8+ cells and higher percentage of late-differentiatedCD8+ cells was seen in subjects with CM2-binding IgGand IgM antibodies. No differences were found in naïveCD4 cells. These data suggest either a lower reactiva-tion rate of CMV in offspring predisposed for familiallongevity, or better immune control of the virus onreactivation, and may help to explain the absenceof CMV-associated markers of immunosenescence inthese individuals [3].The identification of phenotypes that could be used toanticipate which individuals are at higher risk for immu-nosenescence and mortality (“biomarkers”) remains elu-sive. The analysis of 2 year survival in a cohort ofdonors over 65 from the south of Spain (Sevilla, Spain)confirmed that CD4/CD8 ratios below 1 as well as mar-kers of inflammation (neutrophilia, high CRP levels,IL-6) and thymic function (indirectly calculated in per-ipheral PBMC DNA using the sj/β-TREC ratio, [4]), wasassociated with increased risk of death from any cause(Ferrando-Martinez, Seville, Spain). Multivariate analysisshowed that lower thymic function, higher CRP levels,and presence of neutrophilia were independently asso-ciated with time to death in this cohort. Based on theseresults the use of the “CRT” index (CRP and Thymicfunction) was proposed to define a biomarker profile toidentify individuals at higher risk of death [5]. Therewere some limitations to this study regarding the role ofCMV because most of the elderly in Spain are CMV-seropositive, as in Singapore. We therefore need a moresophisticated analysis than mere sero-positivity or –negativity, because it is clear that the way that an indi-vidual deals with the infection is very important (as illu-strated in the LLS by Derhovanessian, mentioned above).This question was approached by Hurme (Tampere,Finland) who presented results on transcriptomic ana-lysis of CMV reactivation in seropositive nonagenar-ians. The presence of CMV DNA was found not tocorrelate with T cell subset distribution or with the levels ofinflammatory markers. In the Vitality study, a cohort of131 nonagenarians, a genome-wide gene expressionarray was performed. The results showed that 55 geneswere upregulated and 65 genes downregulated. Themost highly upregulated gene is the H3F3C gene in thePKA pathway. It is questioned whether the presence ofCMV DNA is due to passive release, rather than due toan active, productive infection and the possibility of aninefficient immune elimination.Is immunosenescence treatableCicin-Sain (Braunschweig, Germany) and Nikolich-Zugich(Tucson, Arizona, USA) analyzed cause-effect relation-ships between CMV infection and age-associated changesin the immune system using an experimental murinemodel of life-long CMV infection. Several changes consist-ent with the development of an IRP (e.g. enrichment of ef-fector memory CD8 cells) were observed in mice infectedwith MCMV but not in control mice infected with non-persistent virus or with other herpesviruses. MCMV infec-tion resulted in loss of CD8 T-cell functional activityagainst other viruses that correlated with the accumula-tion of MCMV-specific EM cells, suggesting that thesecells may compete with the responses against novel anti-gens [6,7]. In general, therefore, these results in mice aresimilar to those seen in humans. It may therefore be ap-propriate to use this animal model to investigate theeffects of anti-viral agents on immunosenescence inthe elderly.To this end, Beswick and Moss (Birmingham, UK) pre-sented a model using elderly mice with latent CMV in-fection to analyze whether antiviral therapy can reversethe development of immune senescence. Administeringvalaciclovir antiviral treatment for 12 months signifi-cantly reduced the frequency of MCMV-specific T-cellsin 18 month-old mice with pre-existing memory infla-tion and the residual immune response was less highlydifferentiated. Furthermore, this treatment lead to a res-toration of the frequency of naive CD8+ T cells, andimproved the de novo immune response to Influenzachallenge as seen by improved survival and a higher fre-quency of Influenza- specific lymphocytes in the medias-tinal lymph nodes. In addition, MCMV infection withSolana et al. Immunity & Ageing 2012, 9:23 Page 3 of 9http://www.immunityageing.com/content/9/1/23the attenuated tsm5 virus (mutated DNA-polymerase)did not elicit memory inflation and therefore theseinvestigators concluded that lytic viral reactivation is keyto the accumulating MCMV-specific immune response.If these results could be translated into clinical practise,the implications would be very exciting.Nikolich-Zugich (Tucson, Arizona, USA) also pre-sented data on CMV infection and T cell ageing in miceshowing that lifelong CMV infection leads to anincreased mortality in aged mice, reduced immune re-sponse to infections and reduced polyfunctionality oflymphocytes. In humans, a search for the IRP and the ef-fect of CMV in a cross-sectional analysis of a mixed UScohort showed that the absolute loss of naïve cells wasdue to ageing as it occurred in both CMV+ and CMV-individuals, whereas the increase of CD8+ EM cells anda decrease of the total CD8+ T cell pool were due toCMV infection. There is some controversy as to the ef-fect of CMV in this context, because some studies haveshown that a significant age-associated deficit of naïveCD8+ T cells is not seen in CMV-seronegatives, or atleast not to anything like the same extent. A moredetailed consideration of the impact of CMV on theseparameters is therefore required.The effect of ageing and CMV on adaptive immunityA diverse T cell repertoire is required to develop effect-ive pathogen-specific immunity, and it is therefore im-portant to assess virus-specific T cells and thoserecognizing other pathogens in elderly people. Wang(Baltimore, Maryland, USA) presented a study on CMV-specific repertoire diversity and antibody levels in youngand older adults. Using a single-cell strategy for clonoty-pic analysis of the TCRαβ repertoire of CD8+ T cells,they analyzed the diversity and magnitude of the CMV-specific CD8+ T-cell response. It was found that TCRαβdiversity, but not the size of the T-cell response, was in-versely related to antibody levels against CMV, which inturn was associated with the detectability of circulatingviral DNA. These results indicate that the CMV-specificCD8+ TCRαβ repertoire diversity may be more import-ant than the size of the CD8+ T cell response in viralcontrol [8].A major question remains whether age impacts on thebreadth, frequency and stability of CD4 and CD8 T cellspecificities in healthy donors, and to what extent CMVis responsible for any such effects? Wills (Cambridge,UK) tried to answer that question by analyzing the re-sponse towards multiple CMV antigens. It was foundthat IE-1-, pp65-, US3- and pp71-specific CD8+ T cellsare able to control the dissemination of virus in vitro.The ability of CMV-specific CD8+ T cells to controlvirus dissemination was not affected by donor age.Young, middle aged and old individuals had good CTLresponses and produced IFN-γ. CD4 cells target the la-tent proteins UL138 and LUNA; a proportion of this re-sponse is mediated by Th1 cells while other cellsinvolved in of the response secrete the immunosup-pressive cytokines IL-10 and TGF-β. These resultsemphasize the importance of maintaining effectiveimmunosurveillance against CMV at any age. This workchallenges the opinion that the immune response toCMV in the elderly is focused on a small number of pro-teins and is characterized by lower functionality com-pared to younger individuals. Kern (Brighton, UK)presented an analysis of T cell response to 19 differentCMV targets previously identified to be the most rele-vant CD4 and CD8 T-cell target antigens in the CMVproteome [9]. Using multi-parameter flow cytometryand intracellular cytokine staining (ICS) they evaluatedmultiple functions and phenotype markers in parallel,including IL-2, TNF-α, IFN-γ, CD40L, degranulation,and the memory surface markers, CD45RA and CD27.No significant age-associated differences with respect toCD4+ T cell responses were found in their preliminarydata analysis, but there was an apparent increase in thesize of the CD8+ T cell response in older age (manu-script in preparation). Those individuals with the great-est single protein-specific responses had retained broadresponses in terms of target recognition. In addition,there was no apparent reduction of functional breadth,neither in CD4+ nor CD8+ T-cells. This seems to indi-cate that responses do not concentrate on fewer specifictarget proteins as a consequence of aging. These resultsseem to support the notion that neither the quantity northe quality of the T cell mediated anti-CMV response isaffected by ageing. Reports published several years agoseemed to indicate that with respect to very select speci-ficities, there may be a functional loss [10]. This is likelyowed to differences in technology, where staining ofCD8 T-cell with tetramers detects non-functional cellsthat will not be detected using ICS. However, the resultsfrom Kern’s lab clearly show that CMV-specific poly-functional T-cells do not decrease in absolute numberswith advancing age, suggesting that non-functional T-cells might occur in addition to but not in the place offully functional T-cells [11]. Riddell and Akbar (London,UK) analyzed the avidity of CMV-specific CD8 T cells inthe elderly. CMV infection induces the accumulation ofCMV-specific CD45RA+ memory CD8+ T cells. Tetra-mer binding avidity correlates inversely with CD45RAexpression whereas high TCR avidity is associated withenhanced effector function (CD107 expression, IFN-γand TNF-α production). The functional characterizationof HLA-A*0201-restricted pp65-specific CD8+ T cellsthat accumulate with age showed that they had low avid-ity. The causes of this accumulation and implications forimmunosenescence are open questions.Solana et al. Immunity & Ageing 2012, 9:23 Page 4 of 9http://www.immunityageing.com/content/9/1/23There is limited quantitative data on clonal diversity andits evolution over time after primary infection, or on cellphenotypes in immune tissues other than blood. Van Lier(Amsterdam, The Netherlands) presented data on TCR di-versity in healthy donors and kidney transplant recipientsusing high throughput sequencing of CDR 3 regions. Theclonal distribution of CMV-specific clones is relativelystable. IE-1 responses differ in that they are morerestricted. CMV-specific cells represented a minor fractionof the CD8+ T cell pool found in the lymph nodes. In con-trast to peripheral blood, pp65-reactive CD8+ T cells fromlymph nodes resemble central memory-like cells [12]. It issuggested that CMV-specific clones in lymph nodes can berecruited into the circulating pool upon CMV reactivation.Sansoni (Parma, Italy) explored anti-CMV CD8+ T cellresponses in a cohort of CMV- seropositive elderly indivi-duals. The results indicate that absolute numbers of anti-CMV CD8+ T cells did not significantly change with age.Increasing age correlates with the loss of naïve CD8+ Tcells, but not with the expansion of CD28+ memory orCD28− effector CD8+ T cells. By contrast, the magnitudeof anti-CMV responses is not correlated with naïve CD8+T cells, but strongly correlates with the accumulation ofantigen-experienced CD8+ T cells. These results suggestthat there is a dichotomy between age and anti-CMVresponses acting as independent factors subverting thenaïve pool and EM cell subset respectively.The accumulation of large numbers of CD8+ effectorT cells, frequently CMV-specific, may hamper the ef-fect of vaccination and exacerbate the development ofage-related diseases. Grubeck-Loebenstein (Innsbruck,Austria) reported that resting CD8+ CD28− effector Tcells, that are more prone to undergo apoptosis follow-ing DNA damage, can be rescued by cytokines. Thus,CMV-specific CD8+ CD28− T cells may survive and ac-cumulate in the bone marrow, specifically in the elderly,due to high IL-15 and IL-6 production. It is suggestedthat in elderly individuals, CD8+ CD28− T cells mayrepresent a useful line of defence against pathogens suchas CMV and may compensate for the loss of naïve andearly memory T cells. In another set of experimentsaimed at analyzing in vitro the production of cytokinesby fibroblasts in young and elderly individuals it wasobserved that CMV infection induced the production ofIL-6 and IL-8, particularly in old age. Thus, lifelong in-fection with CMV may contribute to age-related inflam-matory processes, referred to as inflammageing. Clearly,CMV infection affects many cell types and has wide-ranging direct and indirect effects also on innate im-mune mechanisms [13,14].HCMV infection, ageing and innate immunityLopez-Botet (Barcelona, Spain) analyzed NK cell responsesto CMV. CMV- seropositivity is associated with a variableand persistent increase of NKG2Cbright NK cells inhealthy adults and children. NKG2Cbright NK cells donot co-express NKG2A, display lower levels of NKp30and NKp46 activating receptors and include highernumbers of cells bearing LILRB1 and KIR inhibitoryreceptors specific for HLA class I molecules. The pro-portions of NKG2C+ NK cells and the NKG2Cbrightphenotypic profile appeared comparable in HCMV+individuals of different ages (median: 19, 49 and 70years). CMV infection has been reported to induce amarked NKG2Cbright differentiation and expansion inkidney and stem cell transplantation recipients and inimmunodeficient patients. It is hypothesized that theinteraction of the activating CD94/NKG2C receptor witha ligand on infected cells, together with cytokine-mediated signaling (e.g. IL-15), may induce the differen-tiation and expansion of this NK cell subset.Several age-related changes in NK cells were reportedby Solana (Córdoba, Spain). In elderly individuals, thepercentage of NK cells is increased. The analysis of NKcell subsets in the elderly shows an increase of the moremature CD56dim NK cells and a decrease of CD56brightNK cells. The expression of CD16 allows the identifica-tion of a novel CD56−CD16+ subset that is increased inthe elderly. Analysis of activating receptors shows thatNKp30, NKp46 and DNAM-1 are significantly decreasedin the elderly. Decreased per-cell NK cell cytotoxicityand a decreased capacity of NK to collaborate in DCmaturation (NK-DC crosstalk) are observed in elderlyindividuals. The remodeling of NK cell subsets togetherwith the decreased expression of NCRs and DNAM-1may contribute to explaining the functional alterationsfound in NK cells from the elderly [15,16].A key component of the innate immune system ismannose-binding lectin (MBL). Naumova (Sofia, Bulgaria)studied gene polymorphisms that have been described tocorrelate with MBL production capacity [17]. The low se-cretor haplotypes are very common worldwide, makingMBL-deficiency the most common form of immune defi-ciency. MBL deficiency is frequently asymptomatic, butmay become symptomatic when associated with other im-mune system stressors. The relevance of two innate im-munity gene systems KIR and MBL2 for successful ageingwas discussed. MBL2-deficient haplotypes are associatedwith high levels of anti-CMV antibodies in the elderly.Extrinsic factors affecting immunosenescence: role of CMV?Bosch (Amsterdam, The Netherlands) analyzed theeffects of CMV on immunity in young adults, and therole of life style and psychological stress on the immuneresponse to CMV. The analysis of a cohort of 160 healthyuniversity students showed that CMV-seropositive indivi-duals had a reduction in their CD4:CD8 ratio, an increaseof EMRA T cells, elevated plasma IL-6 and lower responseSolana et al. Immunity & Ageing 2012, 9:23 Page 5 of 9http://www.immunityageing.com/content/9/1/23to influenza vaccination. These data suggested that rudi-mentary signs of immune senescence can already beobserved in CMV-infected young adults. In another co-hort study (Augsburg EADS, 950 factory workers, mostlymales) an association of low income and low educationstatus with CMV-seropositivity was observed. Moreover,in CMV-seropositives, the increase of late differentiatedCD8 T cells was positively correlated with lower incomeand lower education status.Rector, (Birmingham, UK), analyzing the same cohort,found that smokers were 79% more likely to be CMV+.Other lifestyle factors, such as alcohol consumption andexercise, were not significant. Cardiovascular risk factorslike triglycerides and several psychological factors suchas sleep disturbances, vital exhaustion, depression, andself-assessed mental health were positively correlatedwith CMV titers. These findings raise the possibility thatthe associations between morbidity/mortality and CMVobserved in older adults may be secondary to life styleand socio-economic correlates of CMV infection inyoung and middle-aged adults, setting the stage forpoorer health in later life. The analysis of differentiatedEM and EMRA CD8β cells showed that, this populationdecreased with increasing age in those that were CMV+,but not in those that were CMV–.Clinical implications of latent HCMV infection andimmunosenescenceThe impact of measurement errors on the relationshipbetween CMV infection and clinical outcome wasexplored by Wang (Baltimore, Maryland, USA) by astrategy of data simulation based on data and regressioncoefficients from the Women’s Health and Ageing Stud-ies. The results from this study showed that measure-ment errors lead to a significant underestimation of theeffect of CMV infection on chronic disease and mortalityrisk that should be considered in these studies.It is well-known that elderly individuals generally havea worse response to vaccination than the young. Govind(Cranfield, UK) presented data from the MARKAGEEuropean Study to establish biomarkers of human age-ing. In order to identify those individuals who will notrespond effectively to vaccination, the group aims toidentify a set of biomarkers of ageing. Absolute quantifi-cation of herpes viral load achievable in urine (CMV,HHV6a and 6b are detectable) may represent a usefulnon-invasive diagnostic method for immune competencein older individuals.McElhaney (Vancouver, Canada) presented data onbiomarkers of “inflammageing”, the chronic elevation ofinflammatory mediators that weakens the immune sys-tem as we age. A key role in inflammageing may beplayed by chronic CMV infection that stimulates andthereby exhausts the immune system, as previouslymentioned. In response to influenza vaccination, a highproportion of potentially senescent CD8+ T cells do notco-express Granzyme B (GrzB) and Perforin. Using thebaseline level of GrzB, a biomarker called bGrzB has beendeveloped. The level of bGrzB increases with age and ishigher in CMV+ than in CMV− donors. GrzB is co-localized with CD8+ T cells in atherosclerotic lesions lead-ing to plaque instability. In heart failure, GrzB is releasedfrom CD8+ T cells most probably due to the chronic in-flammatory stimulus associated with this disease.Frasca (Miami, Florida, USA) assessed activation-induced cytidine deaminase (AID) as a biomarker of Bcell function. AID is crucial for somatic hypermutationand class-switch recombination. A reduced serum re-sponse of the elderly to influenza vaccination assessedby ELISA and hemagglutination inhibition assays iscommonly observed. In vitro AID responses to CpGwere also decreased with age and correlated with serumresponse. The age-associated defects in B cell functionmay be due to the increased levels of systemic TNF-α(which are positively correlated with levels of CMVIgG), which induce more TNF-α production by B cellsand this pre-activated status of the B cells renders themrefractory to undergo in vitro class switching.The NHANES III study on nutrition and health inthe USA between 1988 and 1994 was a cross-sectional,multistage, stratified, clustered probability sample ofthe US civilian non-institutionalized population thatincluded CMV serology. This information was used byHill (Portland, Oregon, USA) to test the associationbetween CMV seropositivity and mortality, using Coxlogistic regression. Consistent with the results of Aiellomentioned below who had previously analyzed the samedataset, it was found that CMV produces a modestincreased risk for all-cause mortality. The impact ofCMV was largely explained by an increase in cardiovas-cular (CVD) deaths. CMV’s main impact was seen inindividuals aged 55–75 at the time of the survey, and par-ticularly in the 55–65 year age group. CMV imposed littleincreased risk of either all-cause or CVD mortality in themost elderly (aged 75–90). High cholesterol was asso-ciated with CVD mortality only in the younger age group.These results may reflect a selective loss from thepopulation of individuals with the highest risk of CVDmortality due to unknown or unmeasured factors.Aiello (Ann Arbor, Michigan, USA) presented data onCMV, stress and immune markers of ageing. The socialgradient in health may act in part through stress path-ways. Reactivation of herpesviruses is considered a hall-mark of psychosocial stress. Using data from NHANESIII, Aiello and colleagues found that those with lower in-come and education levels were more likely to be CMVseropositive and have higher IgG antibody titers againstCMV than those with higher income and education [18].Solana et al. Immunity & Ageing 2012, 9:23 Page 6 of 9http://www.immunityageing.com/content/9/1/23Social patterning of infection and immune responseagainst CMV may reflect increased likelihood of expos-ure, higher dose of infection and/or poorer immuno-logical control of CMV. CMV seropositivity as well aselevated CMV IgG antibody titer have, in turn, beenassociated with increased risk for all-cause mortality[19,20]. Using data on persons 18 years of age and olderfrom the Detroit Neighborhood Health Study, Aielloet al. examined the possible immunological mechanismswhich may link CMV to mortality. Elevated CMV IgGantibody titer was associated with an increased ratio oflate differentiation-stage EMRA T cells to naïve cellseven after controlling for age, medication use and co-infection with herpes simplex virus-1. Therefore, im-munological ageing due to elevated immune responseagainst CMV may be a novel biologic pathway by whichpsychosocial stressors impact risk for mortality.Persistent infection with CMV is thought to be a keyfactor for the amplification of the inflammation asso-ciated with ageing. Maier (Leiden, The Netherlands) pre-sented an analysis of the innate immune capacity,measured by LPS induced cytokine production capacityof whole blood, and CMV infection in three differentcohorts, the Leiden Longevity Study, Prosper Study andLeiden 85-Plus Study. However, levels of proinflamma-tory and anti-inflammatory cytokines produced uponLPS stimulation showed no correlation with CMV seros-tatus or with CMV IgG levels in these three cohorts.However, cytokine production capacity from CMV– orCMV+ donors showed high interindividual variabilityand reflected survival propensities and metabolic disease.Analyses on comorbidities and CMV infection showedhigher prevalence of diabetes, higher non-fasting glucoselevels and glycosylated haemoglobin (HbA1c) in CMV+oldest old participants, but there was no correlation withCMV IgG titers [21].The possible relationship between diabetes mellitus(DM) Type 2, elderly, frailty and CMV have been analyzedby Fülop (Sherbrooke, Canada). CMV-seropositivity ismore prevalent in diabetic elderly subjects. It seems thatchronic disease is a more important determinant for frailtythan the CMV status. CMV+ elderly DM patients haveless putatively senescent cells than healthy CMV-seropositives but express more CD57 at the single celllevel which is more accentuated in CD8+ T cells. After in-fluenza vaccination, the highest level of GrzB is found inCMV− DM patients that had the lowest level of poten-tially senescent cells. Diabetes seems to suppress the per-centage of these CD8+ T cells.The possibility that infection by CMV or other herpes-viruses could be related to inflammatory diseases hasbeen explored by Sauce (Paris, France) who analyzed thecapacity to respond to CMV and EBV by Systemic LupusErythematosus (SLE) active and inactive patients. A dis-crepancy between CMV and EBV response was observedin SLE patients. Whereas CMV responsiveness was notaltered, the response to EBV was depressed, with lowerIFN-γ, TNF-α, MIP1β and IL-2 secretion and lowerFigure 2 Age and CMV infection are major driving forces contributing to the deterioration of innate and adaptive immunity.Age-associated decrease of adaptive immunity is termed immunosenescence. The deregulation of innate immunity is associated withinflammageing. Immunosenescence and inflammageing play a significant role in the pathogenesis of different clinical situations that can leadto increased risk of frailty and death in the elderly.Solana et al. Immunity & Ageing 2012, 9:23 Page 7 of 9http://www.immunityageing.com/content/9/1/23cytotoxicity. This dysfunctionality can be due to anexhausted T cell phenotype (PD-1hi) since PD-1 blockingrestores responsiveness [22].CMV is still regarded as being the most significant in-fectious pathogen in the solid organ transplant recipient,and in spite of the improvements in surveillance andtreatment, it continues to be a major cause of morbidityand mortality after transplantation and is associated withlower graft survival. Appay (Paris, France) focused onthe role of CMV in the pathogenesis of acute rejectionin lung transplantation. An association between cellularimmune activation (i.e. expression of CD38 on T cells)and acute rejection was found. Levels of total CD38+CD8+ T cells correlated with the frequency of CMV-specific cells. The development of pro-inflammatoryCMV-specific CD8+ T cell immune responses may ex-plain the relationship between CMV infection and acutelung rejection. It is suggested that potent CMV prophy-laxis should be given to all CMV-seropositive patients toprevent the occurrence of acute rejection. CD8+ T cellactivation levels in peripheral blood correlate well withCD8 responses in the lung and may predict the risk ofacute rejection. Zielinski (Gdansk, Poland) analyzed CMVinfection in renal transplantation patients. In a retrospect-ive study, it was observed that elderly kidney recipients,CMV-positive with high numbers of CD28− T cells andshort telomeres, had fewer episodes of acute rejection.Preliminary data suggest that CMV challenge has a strongimpact on the immune system after allotransplantationin the elderly. Solana and Cantisan (Cordoba, Spain)reported an analysis of CD45RA+ (EMRA) CMV-specificCD8 T cells in relation to CMV replication parameters insolid organ transplantation. It was observed that EMRAincrease by 4% for each log of viral replication. Replicationis associated with continuous and constant increases inpercentages of CD28− cells. The impact of CMV on indu-cing expansion of CD28− CMV-specific CD8 T cells isseen mainly in young individuals [23].Future directions and concluding remarksSignificant advances in the understanding of the changesof immune system associated with CMV infection andtheir possible significance for immunosenescence havebeen made over the past few years in many areas ofmedicine, as reported here.In this workshop, many important questions have beenaddressed and other new questions been raised. A panel ofmarkers to assess T cell phenotypes in T cells subsets instudies of immunosenescence and CMV infection was pro-posed and discussed, as well as the significance of the IRPphenotype, markers of thymic function or inflammationon survival in the elderly. The role of CMV in im-munosenescence has been confirmed in experimental mod-els, opening new perspectives to explore possible therapiesaimed at reversing immunosenescence. However, a majorquestion still remains concerning the mechanisms drivingthe homeostatic fluctuations of CD8 T cells during latencyof CMV infection, and how age impacts on the breadth,frequency, phenotype and function of CD4 and CD8 T cellspecificities in healthy donors. Although primarily affectingthe T cell compartment, evidence is accumulating in sup-port of a significant effect of CMV infection and ageing oninnate immunity as well, in particular NK cells.Results from translational research of CMV infectionand immunosenescence in clinical conditions such astransplantation, cancer, immunodeficiency and auto-immune and inflammatory diseases, supports the notionthat CMV can affect their evolution and prognosis by in-ducing a process of “early” immunosenescence (Figure 2).Furthermore, other extrinsic factors can also contribute,together with CMV, to the age-associated deteriorationof the immune system.However, there are still many open questions about theimmune response to HCMV itself in ageing and about therole of CMV in early immunosenescence (Figure 1):a) It is likely that humans can be infected with multipleCMV strains. How does this affect their interactionswithin the immune system and their effect onimmunosenescence?b) What is the real impact of the complexity of CMVinfections on the processes of T cell homeostasis anddifferentiation in the course of ageing?c) It is necessary to test the function of T cells in ageingand CMV infection before they are labeled as“senescent”, “exhausted” or in other manner inadequate.d)What is the contribution of CMV- inducedinflammation to inflammageing? What are the bestmarkers that could be used to define CMV-inducedsystemic inflammation?It was proposed that the organization of a 4th work-shop is vital for the field to move forwards, to answerthese and other questions. We are grateful to PaoloSansoni who agreed to organize it in Parma (Italy) in 2013.Competing interestsThe authors declare that they have no competing interests.Authors’ contributionsAll authors attended the Workshop, participated in the discussion, saw andcommented on the text published here. All authors read and approved thefinal manuscript.AcknowledgementsThe Workshop was partially supported by the University of Cordoba and theSpanish Ministry of Science (SAF2011-16169-E).Author details1Immunology Unit, Instituto Maimónides de Investigación Biomédica deCórdoba (IMIBIC)-Reina Sofia University Hospital-University of Cordoba,Cordoba, Spain. 2Immunology Unit, Department of Physiology, University ofSolana et al. Immunity & Ageing 2012, 9:23 Page 8 of 9http://www.immunityageing.com/content/9/1/23Extremadura, Caceres, Spain. 3University of Michigan, Department ofEpidemiology, Center for Social Epidemiology and Population Health, AnnArbor, Michigan, USA. 4Division of Infection and Immunity, University CollegeLondon, London, UK. 5Infections and Immunity, Institut National de la Santéet de la Recherche Médicale (INSERM), Paris, France. 6School of CancerSciences, University of Birmingham, Birmingham, UK. 7University ofAmsterdam, Amsterdam, The Netherlands. 8Department of Vaccinology andApplied Microbiology, Helmholtz Centre for Infection Research,Braunschweig, Germany. 9Department of Internal Medicine II, Center forMedical Research, University of Tübingen, Tübingen, Germany. 10Laboratoryof Molecular Immune-Biology, Hospital General Universitario GregorioMarañón, Madrid and Laboratory of Immunovirology; Infectious DiseasesService, IBiS, Seville, Spain. 11Department of Microbiology and Immunology,University of Miami Miller School of Medicine, Miami, Florida, USA.12Research Center on Aging, Sherbrooke, Canada. 13Regenerative MedicineGroup, Cranfield Health, Cranfield University, Cranfield, UK. 14Institute forBiomedical Aging Research, University of Innsbruck, Innsbruck, Austria.15Department of Molecular Microbiology and Immunology, Oregon Healthand Science University, Portland, Oregon, USA. 16University of Tampere,Medical School, Tampere, Finland. 17Brighton and Sussex Medical School,Brighton, UK. 18Singapore Immunology Network, Singapore, Singapore.19IMIM (Hospital del Mar Research Institute), Univ. Pompeu Fabra, Barcelona,Spain. 20Department of Gerontology and Geriatrics, Leiden University MedicalCenter, Leiden, The Netherlands. 21University of British Columbia, Vancouver,Canada. 22Department of Clinical Immunology, University HospitalAlexandrovska, Sofia, Bulgaria. 23Department of Immunobiology and theArizona Center on Aging, University of Arizona College of Medicine, Tucson,Arizona, USA. 24University of Birmingham, Birmingham, UK. 25Department ofInternal Medicine and Biomedical Sciences, University of Parma, Parma, Italy.26Experimental Immunology, Academic Medical Center, Amsterdam, TheNetherlands. 27Division of Geriatric Medicine and Gerontology, Biology ofHealthy Aging Program, Johns Hopkins University School of Medicine,Baltimore, Maryland, USA. 28University of Cambridge Department of MedicineAddenbrookes Hospital, Cambridge, UK. 29Department of ClinicalImmunology and Transplantology, Medical University of Gdansk, Gdansk,Poland. 30Mannheim Institute of Public Health, Social and PreventiveMedicine (MIPH), University of Heidelberg, Mannheim Medical Faculty,Mannheim, Germany.Received: 3 October 2012 Accepted: 26 October 2012Published: 31 October 2012References1. Pawelec G, Akbar A, Beverley P, Caruso C, Derhovanessian E, Fulop T, et al:Immunosenescence and Cytomegalovirus: where do we stand after adecade? Immun Ageing 2010, 7:13.2. Wills M, Akbar A, Beswick M, Bosch JA, Caruso C, Colonna-Romano G, et al:Report from the Second Cytomegalovirus and ImmunosenescenceWorkshop. Immun Ageing 2011, 8:10.3. Derhovanessian E, Maier AB, Hahnel K, Zelba H, de Craen AJ, Roelofs H,et al: Lower proportion of naive peripheral CD8+ T cells and anunopposed pro-inflammatory response to human Cytomegalovirusproteins in vitro are associated with longer survival in very elderlypeople. Age (Dordr) 2012, doi:10.1007/s11357-012-9425-7. in press.4. Ferrando-Martinez S, Franco JM, Ruiz-Mateos E, Hernandez A, Ordonez A,Gutierrez E, et al: A reliable and simplified sj/beta-TREC ratioquantification method for human thymic output measurement.J Immunol Methods 2010, 352:111–117.5. Ferrando-Martinez S, Romero-Sanchez MC, Solana R, Delgado J, Munoz-Fernandez MA, la RR D, et al: Thymic function failure and C-reactiveprotein levels are independent predictors of all-cause mortality inhealthy elderly humans. Age (Dordr) 2011, doi:10.1007/s11357-011-9341-2.in press.6. Cicin-Sain L, Brien JD, Uhrlaub JL, Drabig A, Marandu TF, Nikolich-Zugich J:Cytomegalovirus infection impairs immune responses and accentuatesT-cell pool changes observed in mice with aging. PLoS Pathog 2012,8:e1002849.7. Smithey MJ, Li G, Venturi MP, Davenport MP, Nikolich-Zugich J: Life-longpersistent viral infection alters the naïve T-cell pool, impairing CD8 T-cellimmunity in late life. J Immunol 2012, in press.8. Wang GC, Dash P, McCullers JA, Doherty PC, Thomas PG: T cell receptoralphabeta diversity inversely correlates with pathogen-specific antibodylevels in human cytomegalovirus infection. Sci Transl Med 2012, 4:128ra42.9. Sylwester AW, Mitchell BL, Edgar JB, Taormina C, Pelte C, Ruchti F, et al:Broadly targeted human cytomegalovirus-specific CD4+ and CD8+ Tcells dominate the memory compartments of exposed subjects.J Exp Med 2005, 202:673–685.10. Ouyang Q, Wagner WM, Zheng W, Wikby A, Remarque EJ, Pawelec G:Dysfunctional CMV-specific CD8(+) T cells accumulate in the elderly.Exp Gerontol 2004, 39:607–613.11. Lachmann R, Bajwa M, Vita S, Smith H, Cheek E, Akbar A, et al:Polyfunctional T cells accumulate in large human cytomegalovirus-specific T cell responses. J Virol 2012, 86:1001–1009.12. Remmerswaal EB, Havenith SH, Idu MM, van Leeuwen EM, van DonselaarKA, Ten BA, et al: Human virus-specific effector-type T cells accumulate inblood but not in lymph nodes. Blood 2012, 119:1702–1712.13. Herndler-Brandstetter D, Landgraf K, Tzankov A, Jenewein B, Brunauer R,Laschober GT, et al: The impact of aging on memory T cell phenotypeand function in the human bone marrow. J Leukoc Biol 2012, 91:197–205.14. Wolf J, Weinberger B, Grubeck-Loebenstein B: The immunoregulatoryeffects of CMV-infection in human fibroblasts and the impact on cellularsenescence. Immun Ageing 2012, 9:1.15. Sanchez-Correa B, Gayoso I, Bergua JM, Casado JG, Morgado S, Solana R, etal: Decreased expression of DNAM-1 on NK cells from acute myeloidleukemia patients. Immunol Cell Biol 2012, 90:109–115.16. Gayoso I, Sanchez-Correa B, Campos C, Alonso C, Pera A, Casado JG, et al:Immunosenescence of human natural killer cells. J Innate Immun 2011,3:337–343.17. Naumova E, Ivanova M, Pawelec G, Constantinescu I, Bogunia-Kubik K,Lange A, et al: 'Immunogenetics of Aging': report on the activities of the15th International HLA and Immunogenetics Working Group and 15thInternational HLA and Immunogenetics Workshop. Tissue Antigens 2011,77:187–192.18. Dowd JB, Aiello AE, Alley DE: Socioeconomic disparities in theseroprevalence of cytomegalovirus infection in the US population:NHANES III. Epidemiol Infect 2009, 137:58–65.19. Simanek AM, Dowd JB, Pawelec G, Melzer D, Dutta A, Aiello AE: Seropositivityto cytomegalovirus, inflammation, all-cause and cardiovascular disease-related mortality in the United States. PLoS One 2011, 6:e16103.20. Roberts ET, Haan MN, Dowd JB, Aiello AE: Cytomegalovirus antibodylevels, inflammation, and mortality among elderly Latinos over 9 yearsof follow-up. Am J Epidemiol 2010, 172:363–371.21. Chen S, de Craen AJ, Raz Y, Derhovanessian E, Vossen AC, Rudi WG, et al:Cytomegalovirus seropositivity is associated with glucose regulation inthe oldest old. Results from the Leiden 85-plus Study. Immun Ageing2012, 9:18.22. Larsen M, Sauce D, Arnaud L, Fastenackels S, Appay V, Gorochov G:Evaluating cellular polyfunctionality with a novel polyfunctionality index.PLoS One 2012, 7:e42403.23. Cantisan S, Torre-Cisneros J, Lara R, Zarraga S, Montejo M, Solana R: Impactof Cytomegalovirus on Early Immunosenescence of CD8+ TLymphocytes After Solid Organ Transplantation. J Gerontol A Biol Sci Med2013, 68:1–5.doi:10.1186/1742-4933-9-23Cite this article as: Solana et al.: CMV and Immunosenescence: frombasics to clinics. Immunity & Ageing 2012 9:23.Solana et al. Immunity & Ageing 2012, 9:23 Page 9 of 9http://www.immunityageing.com/content/9/1/23


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items