1

Differential kinetics of human cytomegalovirus (HCMV) load and antibody responses in

primary infection of the immunocompetent and immunocompromised host

Giuseppe Gerna1*

, Daniele Lilleri1,2

, Chiara Fornara1, Francesca Bruno

1, Elisa Gabanti

1, Ilaria

Cane1, Milena Furione

3, M. Grazia Revello

4

1Laboratori Sperimentali di Ricerca, Area Trapiantologica, Fondazione IRCCS Policlinico San

Matteo, Pavia, Italy

2Institute for Research in Biomedicine, Bellinzona, Switzerland

3Struttura Semplice Virologia Molecolare, Struttura Complessa Microbiologia e Virologia,

Fondazione IRCCS Policlinico San Matteo, Pavia, Italy

4Struttura Complessa Ostetricia e Ginecologia, Fondazione IRCCS Policlinico San Matteo, Pavia,

Italy

Running title: antibody response to primary HCMV infection

*Corresponding author at the Laboratori Sperimentali di Ricerca, Area Trapiantologica, Fondazione

IRCCS Policlinico San Matteo, 27100 Pavia, Italy. Tel.: +39 0382 503314; fax: +39 0382 502648.

Email: g.gerna@smatteo.pv.it

Content Category: Animal viruses, Large DNA

Figure number: 7

Suppl. Figure number: 1

Table number: 1

Summary words: 249

Text + figure legends words: 3703

JGV Papers in Press. Published October 14, 2014 as doi:10.1099/vir.0.070441-0

2

Summary

The comparative long-term kinetics of human cytomegalovirus (HCMV) load and HCMV-specific

antibody responses in the immunocompetent and immunocompromised solid-organ transplanted

host during primary HCMV infection was investigated. On the whole, 40 immunocompetent

subjects and 17 transplanted patients were examined for viral load as well as for IgG antibody

responses to HCMV glycoproteins gH/gL/pUL128L, gH/gL and gB, and neutralizing antibodies in

ARPE-19 epithelial cells and human fibroblasts. In parallel, the CD4+ and CD8

+ HCMV-specific T-

cell responses were determined by cytokine flow cytometry. Transplanted patients reached

significantly higher viral DNA peaks, which persisted longer than in immunocompetent subjects.

The ELISA-IgG responses to the pentamer, gH/gL and gB were significantly higher in primary

infections of the immunocompetent until six months after onset, then the two antibody levels

overlapped from six to 12 months. Antibody levels neutralizing infection of epithelial cells were

significantly higher in transplanted patients after six months, persisting up to a year after

transplantation. This trend was not observed for antibodies neutralizing infection of human

fibroblasts, which showed higher titers in the immunocompetent over the entire 1-year follow-up. In

conclusion, in immunocompromised patients the viral load peak was much higher, while the

neutralizing antibody response exceeded that detected in the immunocompetent host starting six

months after onset of follow-up, often concomitantly with a lack of specific CD4+ T-cells. In this

setting, the elevated antibody response occurred in the presence of differentiated follicular helper T-

cells in blood, which decreased in number as did antibody titers upon reappearance of HCMV-

specific CD4+ T-cells.

KEYWORDS: human cytomegalovirus, primary HCMV infection, HCMV load, antibody

response, immunocompetent individuals, immunocompromised patients.

Abbreviations: HCMV, human cytomegalovirus; gH/gL/pUL128L, glycoprotein complex

(pentamer) including glycoprotein H (gH), glycoprotein L (gL) and the pUL128 locus (L)

consisting of three genes (UL128, UL130, and UL131); gH/gL, glycoprotein complex consisting of

gH and gL; gB, glycoprotein B; Nt Ab-ARPE-19 cells, neutralizing antibodies preventing infection

of epithelial cells ARPE-19; Nt-Ab-HELF, neutralizing antibodies preventing infection of human

embryonic lung fibroblasts.

3

INTRODUCTION

In the last few years, multiple significant steps forward have been made in the diagnosis of human

cytomegalovirus (HCMV) infections both for viral DNA quantification by real-time PCR (recently

standardized with reference to the WHO International Standard; Furione et al., 2012), and antibody

response determination to the HCMV envelope glycoprotein complexes gH/gL/pUL128L (the

pentameric complex), gH/gL, and gB as well as neutralizing antibodies (Genini et al., 2011; Lilleri

et al. 2012, 2013).

In 2004, in collaboration with the group of U. Koszinowksi (Munchen, Germany), it was

documented for the first time that HCMV UL128-131 genes are indispensable for virus growth in

endothelial cells and virus transfer to leukocytes (Hahn et al.,2004). Subsequently, it was shown

that the UL128, UL130 and UL131 locus (thus referred to as UL128L) gene products formed a

pentameric complex with gH/gL (gH/gL/pUL128/pUL130/pUL131), which was required for

infection of both endothelial and epithelial cells (Wang and Shenk,2005; Ryckman et al., 2008;

Revello and Gerna, 2010). Multiple antigenic sites of the pentamer have been shown to elicit very

potent neutralizing mAbs with high frequency unlike gB inducing a predominant number of

binding, but non-neutralizing mAbs (Macagno et al., 2010; Potzsch et al., 2011; Fouts et al., 2012). In

parallel, gB-based vaccines have shown only partial protection against vertical HCMV transmission

in pregnant women and against HCMV infection in transplanted patients (Pass et al., 2009;

Griffiths et al., 2011). In parallel, neutralizing antibodies have been shown to display far higher

titers when measured in epithelial rather than fibroblast cells (Gerna et al., 2008; Cui et al., 2008).

During follow-up of a series of primary HCMV infections in the immunocompetent (pregnant and

non-pregnant) host and the solid-organ immunocompromised transplanted patient, (HCMV-

seronegative patients receiving organs from seropositive donors, thus developing primary HCMV

infections), some differential features in the kinetics of viral load and antibody responses in the two

clinical conditions, were preliminarily observed. These features were relevant to: i) viral DNA

appearance in blood of the immunocompetent subject which elicited an immediate antibody

response, whereas in the immunocompromised patient the antibody response was delayed by

several months; ii) the earlier antibody response in immunocompetent subjects reached a higher

level with respect to immunocompromised patients until three months or more (according to

4

different assays) after onset of infection; afterwards, the antibody response in immunocompromised

patients appeared to overlap that of immunocompetent subjects; iii) very high antibody levels in the

immunocompromised host reaching titers of 1:1x105-1:1x10

6 were often observed in the absence of

HCMV-specific CD4+ T-cells. This study was conducted to verify the consistency and

reproducibility of these preliminary observations

The main objective of this study was to comparatively investigate the kinetics of HCMV load and

different types of antibody responses to HCMV glycoprotein complexes (gB, gH/gL/pUL128L and

gH/gL) and neutralizing antibodies (determined in both ARPE-19 epithelial cells and human

embryonic lung fibroblast -HELF- cells) in the immunocompetent subject and the

immunocompromised solid-organ transplanted patient. Both parameters were evaluated with

reference to the development of HCMV-specific T-cell immune responses, which behaved very

differently in the two patient populations examined (Lozza et al., 2005; Lilleri et al., 2009).

RESULTS

Clinical characteristics and management of HCMV infection in the two patient populations

studied

All transplanted patients (n=17) received one or more courses of pre-emptive therapy with

ganciclovir or valganciclovir for an overall median duration of 78 (17-179) days after reaching a

level of 300,000 DNA copies/ml blood (or in the presence of an organ localization, shown by virus

detection in organ biopsy by PCR). In all transplanted patients, HCMV disappeared from blood

after antiviral therapy. Immunocompetent subjects were either asymptomatic (n=10 pregnant

women), symptomatic (n=13: 3 pregnant women and 10 non-pregnant subjects) or

paucisymptomatic (n=17 pregnant women). None of them received antiviral treatment.

HCMV load in primary HCMV infections

The onset of HCMV infection was determined on the basis of criteria reported in the Methods for

immunocompetent (pregnant and non-pregnant) subjects, and with reference to the day of

transplantation for transplanted patients. Within the limits of these criteria, the median first viral

DNA detection (Fig.1a) occurred significantly earlier (P=0.03) in immunocompetent subjects (22,

range 2-68, days) than in transplanted patients (29, range 19-52, days). In addition, the median peak

5

DNA level (copies/ml whole blood) was significantly higher (p<0.01) in transplanted patients

(189,800, range 5,000-774,000 copies/ml) than in immunocompetent individuals (56, range 25-

6762) (Fig. 1b). Finally, the median persistence of viral DNA (Fig. 1c, Kaplan-Meier curves) in

blood of transplanted patients (235, range 120-670, days) was significantly higher (p<0.01, Log

rank test) than in immunocompetent subjects (85, range 33-194, days).

Kinetics of HCMV load in relation to the antibody and T-cell responses

First viral DNA detection occurred earlier in immunocompetent with respect to transplanted

patients. In parallel, the antibody response occurred significantly (p<0.01) earlier in the

immunocompetent independently of the antibody assay performed (Fig. 2a). Similarly, the HCMV-

specific CD4+ and CD8

+ T cell

responses were delayed in transplant recipients with respect to the

immunocompetent host. However, the CD8+ T cell response appeared at a time comparable to that

of the antibody response, whereas the CD4+ T cell response appeared later (Fig 2a).

Concomitantly, while the median viral DNA peak (Fig. 2b) was reached significantly later in

transplanted patients (45, range 27-132, vs 22, range 5-48, days), the median peak of the antibody

response was also reached significantly later in transplanted patients with four of the five types of

antibodies determined, with the exception of the ELISA IgG antibodies to the pentamer, whose

levels were reached in the two populations at a comparable time. Similarly, the T-cell response

peaks (for both HCMV-specific CD4+ and CD8

+ T-cells) were reached later in transplant recipients

(Fig 2b).

Kinetics of the ELISA IgG antibody response to the three major HCMV glycoprotein

complexes in the immunocompetent and the immunocompromised host during primary

HCMV infection

The kinetics of the IgG antibody response to the pentamer gH/gL/pUL128L, the dimer gH/gL, and

gB glycoprotein complexes in the immunocompetent and the immunocompromised patient were

significantly different (p<0.0001), as shown in Figures. 3a, 4a, and 5a, respectively. When

considered individually, the antibody responses to the three glycoprotein complexes were

significantly higher in the immunocompetent host until 180 days after onset. Subsequently, at days

181-360, the three types of antibody response observed in both the immunocompetent and the

6

immunocompromised host substantially overlapped until 12 months (Fig. 3b, 4b, and 5b). However,

a fair number of serum samples taken at both time intervals (91-180 and 181-360 days post-

transplant) were 4-16fold higher in titer than the relevant highest antibody titers of the

immunocompetent subjects. This was true for antibody titers to all three major glycoprotein

complexes, although the difference in antibody titer between the two populations studied was less

pronounced for antibodies to gB (Figures. 3b, 4b, and 5b).

Kinetics of the neutralizing antibody response as determined in ARPE-19 epithelial cells and

in human embryonic lung fibroblasts (HELF) in the immunocompetent and the

immunocompromised host

The kinetics of the neutralizing antibody response determined in ARPE-19 epithelial cells was

significantly different (p<0.0001) in the two study populations (Fig. 6a). Similarly, the kinetics of

the antibody response in the two patient populations was significantly different when the

neutralizing activity was measured in HELF cells (Suppl. Fig. 1). However, the neutralizing activity

in ARPE-19 cells was higher in the immunocompetent until 90 days after onset of follow-up, then it

became higher in the immunocompromised patient after 180 days (Fig. 6a). The above reported

finding of a markedly higher titer to the three major glycoprotein complexes in the

immunocompetent as compared to the immunocompromised host until 180 days follow-up was

partially confirmed for antibodies neutralizing infection of ARPE-19 epithelial cells, but only until

90 days after transplantation. Subsequently, at 91-180 days titers were overlapping, while at 181-

360 days after transplantation, neutralizing antibody titers were significantly higher in transplanted

patients (Fig. 6b). The great majority of transplanted patients showed neutralizing antibody titers 4-

16 fold higher as compared with immunocompetent subjects (Fig. 6a).

Kinetics of precursor CCR7lo

PD-1hi

CXCR5+ CD4

+ follicular helper T-cells (Tfh) in

immunocompetent and immunosuppressed patients

Antibody responses to primary HCMV infection in immunocompetent individuals started early

concomitantly with the appearance of differentiated Tfh cells in blood, an aliquot of which (around

30%) expressed ICOS. Once the antibody titer reached the plateau, both precursor Tfh and ICOS+

Tfh-cells decreased in number (Fig. 7a).

7

In Fig. 7b, the case of a heart-transplanted patient with primary HCMV infection is reported, in

whom, in the absence of HCMV-specific CD4+ T-cells, IgG and neutralizing antibody responses

appeared later than those currently observed in immunocompetent individuals, but again

concomitantly with the differentiation of circulating Tfh precursor CCR7lo

PD-1hi

CXCR5+ ICOS

+

T-cells. In Table 1, the kinetics of blood Tfh cells and different types of antibody responses are

reported for a heart-transplanted patient, showing the neutralizing antibody increase concomitantly

with the increase in ICOS+ Tfh-cells and the neutralizing antibody titer decrease upon ICOS

+ Tfh-

cell decrease, following reconstitution of HCMV-specific CD4+ T-cells.

DISCUSSION

The main objective of this study was to investigate the differential kinetics of HCMV load, and IgG

and neutralizing antibody responses in the immunocompetent and the immunocompromised host

affected by primary HCMV infection. The IgG antibody responses to the glycoprotein complexes

representing the pentamer (gH/gL/pUL128L), the dimer gH/gL (Hahn et al., 2004; Ryckman et al.,

2008), and gB were investigated. In addition, the neutralizing antibody response was determined in

parallel by inoculating virus-serum mixtures onto ARPE-19 epithelial cells and HELF cells. The

double neutralization assay was based on the recent discovery that HCMV requires the pentamer to

infect epithelial/endothelial/dendritic cells (Gerna et al., 2005; Ryckman et al., 2008), while

infection of HELF cells requires the presence of gH/gL/gO on the HCMV envelope (Zhou et al.,

2013) . It has been shown and repeatedly confirmed that neutralizing titers determined in epithelial

cells are by far higher than those obtained in HELF cells (Gerna et al., 2008; Cui et al., 2008).

The HCMV viral load measured in whole blood (expression of the enhanced replication rate

occurring in transplanted patients still lacking immune control of viral infection) was markedly

higher in immunocompromised patients due to the delayed development of the T-cell-mediated

immune response caused by the immunosuppressive therapy. On the other hand, in the

immunocompetent individual, the rapid mounting of the T-cell response provided a timely help to

B-cells for antibody production, and allowed a prompt control of viral replication (Lilleri et al.,

2009), thus reducing the antigen exposure and the prolonged stimulation of the antibody response.

On the contrary, in transplanted patients the antibody response was delayed and poorly developed

until 90 days after transplantation. At 181-360 days post-transplant, the antibody response in

8

transplanted patients increased first overlapping and sometimes exceeding that of

immunocompetent subjects, as was the case for neutralizing antibodies preventing infection of

epithelial cells. This process occurred mostly in the absence of HCMV-specific CD4+ T-cells in

peripheral blood as detected by cytokine flow cytometry. Under these conditions, we found that

precursor CD4+ Tfh cells circulating in peripheral blood (PD-1

hi CCR7

lowCXCR5

+ ICOS

+)

increased before the antibody response in these patients, thus suggesting their CD4+ T-cell helper

function for B-cells. In several of transplanted patients examined in this study, the antibody

response reached titers never observed in immunocompetent individuals, i.e. titers ranging from

1:1x105 to 1:1x10

6 for both HCMV-specific ELISA IgG antibodies to viral glycoprotein complexes,

and neutralizing antibodies preventing infection of epithelial cells. A likely explanation for the

higher antibody titers observed in these patients at later time points may reside in the prolonged

viremia, reflecting an extended period of virus replication, and, thus, antigenic stimulation. High

titers persisted for several months until reappearance of specific CD4+ T-cells, when, due to virus

replication control, titers started dropping concomitantly with a decrease in Tfh cells, until they

reached in most cases levels comparable to those of immunocompetent subjects. This mechanism

appeared to be responsible for the delayed and enhanced antibody response of transplanted patients

during primary HCMV infection.

It has recently been shown that circulating precursor CCR7lo

PD-1hi

CXCR5+ CD4

+ T-cells are

indicative of Tfh cell activity, which promotes effective antibody responses upon antigen

reexposure (He et al., 2013). This has been shown in humans and mice. These precursor cells may

rapidly differentiate into mature Tfh cells after encountering HCMV to promote a potent antibody

response. In both immunocompetent and immunocompromised patients, the kinetics of CCR7lo

PD-

1hi

CXCR5+ T-cells somewhat paralleled the profile of ICOS

+ T-cells, thus suggesting that this co-

stimulator is required to keep precursor Tfh cells at an efficient level to become mature Tfh cells to

promote the antibody response. In our study on primary HCMV infections, we can speculate that in

transplanted patients the long delay observed between virus appearance in blood and antibody

response in the absence of HCMV-specific CD4+ T-cells is required for the differentiation of

CCR7hi

PD-1lo

ICOS+ T-cells outside germinal centers. Following exposure to the antigen, these

precursor T-cells rapidly differentiate into mature Tfh cells to promote an effective antibody

response.

9

In conclusion, in transplant recipients undergoing primary HCMV infection the antibody response

is delayed with respect to immunocompetent subjects. However, transplanted patients develop a

high antibody response (especially antibodies neutralizing HCMV infection of epithelial cells)

concomitantly with or slightly earlier than HCMV-specific CD8+ T cell appearance, but before

reconstitution of HCMV-specific CD4+ T cells. The significance of this high antibody response and

the interaction of humoral and cell-mediated immunity in protection from HCMV infection deserve

further clarification in future studies. However, the delayed appearance of high antibody titers in

transplant recipients, in whom primary HCMV infection is often severe, may suggest that vaccines

capable of inducing high antibody titers (Griffiths et al., 2011), or passively transferred monoclonal

antibodies might have a role in protecting these patients.

METHODS

Immunocompetent and immunocompromised patient populations. The immunocompetent

subject population affected by primary HCMV infection included 30 pregnant women (aged 20 to

40 yrs) and 10 non-pregnant subjects in the same age range, including 7 males and 3 females. The

immunocompromised transplanted patient population included 17 D+/R

- solid-organ transplant

recipients (8 kidney, 7 heart, 1 single lung and 1 heart-lung) with primary HCMV infection with a

median age of 52 (range 13-75) years.

Diagnosis of primary HCMV infection. In the group of immunocompetent subjects with primary

HCMV infection, diagnosis was ascertained along two major lines (Revello and Gerna, 2002;

Revello et al., 2011). In 10 non-pregnant subjects diagnosis was based on the presence of HCMV-

related clinical symptoms that presumably indicated onset of infection (high fever lasting 2-4 weeks

in the absence of other identifiable causes and in association with one of the following

symptoms/signs: lymphoadenopathy, sore throat, arthromyalgias, headache, elevated liver enzymes

levels, lymphocytosis). Virological and serological assays then confirmed a suspected diagnosis. On

the other hand, in the group of seronegative women entering pregnancy, only 3/30 women displayed

overt clinical symptoms, thus raising the suspect of a primary infection, while in 17/30 women mild

or non-specific symptoms prompted the performance of viral tests. In the remaining 10 cases, the

detection of primary infection was by chance during routine pregnancy follow-up (asymptomatic

primary infection). In the latter cases, the date of infection onset was determined in the middle

10

interval between the last negative and the first positive serological result. In the group of 17

immunocompromised patients, the onset of infection was dated with the day of transplantation.

Viral and serological assays. Conventional diagnostic assays performed in this study for the

diagnosis of primary HCMV infection were ELISAs for determination of HCMV-specific IgG and

IgM antibody by ETI-CYTOK-G and ETI-CYTOK-M, respectively (DiaSorin, Saluggia, Italy)

according to manufacturer’s instructions. In addition, the HCMV-specific avidity index (AI) was

measured with an in-house developed ELISA, as reported (Revello et al., 2010; Furione et al.,

2013). When clinical symptoms suggested the onset of primary infection, diagnosis was later

confirmed by serological and viral assays, whereas in paucisymptomatic/asymptomatic women,

primary infection was suspected based on the presence of at least two of the four following criteria:

IgG seroconversion, presence of IgM antibody, low IgG AI, and presence of viral DNA in blood

(DNAemia) (Revello et al., 2011). Infection onset timing was determined in symptomatic infections

based on both symptoms/signs and viral assays, whereas in asymptomatic infections, it was based

on IgG and IgM antibody kinetics in association with low AI. The unconventional assays were

ELISAs for determination of IgG antibodies to the pentamer (gH/gL/pUL128L), gH/gL and gB, as

reported (Lilleri et al., 2012). The three glycoprotein complexes were captured in 96-well

polystyrene microplates (Corning) with an in-house developed murine anti-gH mAb (mH1P73), or

an anti-gB mAb (HCMV 37, Abcam, Cambridge, UK). Net OD was determined by subtracting OD

of serum incubated in the absence of antigen from OD of serum incubated in the presence of

antigen.

Neutralizing assays were performed in 96-well microtiter plates by inoculating virus-serum dilution

mixtures in duplicate onto monolayers of epithelial (ARPE-19) or fibroblast (HELF) cells, as

reported (Lilleri et al., 2013). HCMV isolate VR1814 was used for assays in ARPE-19 cells, while

laboratory strain AD169 was used for assays in HELF. The differential neutralizing activity of

human sera when tested in ARPE-19 or HELF cells was previously reported (Gerna et al., 2008).

Quantitative DNAemia results were expressed as HCMV DNA copies/ml blood (Gerna et al.,

2006). When requested, results were also translated into IU/ml with reference to the WHO

11

International HCMV DNA Standard (Furione et al., 2012). Viral DNA was quantified weekly in

transplant recipients, and mostly monthly in immunocompetent individuals.

T-cell assays. HCMV-specific T-cells were determined ex-vivo (Lozza et al., 2005) following

stimulation with autologous monocyte-derived (Sallusto and Lanzavecchia, 1994) HCMV-infected

(VR1814) immature dendritic cells. Peripheral blood mononuclear cells (PBMCs) were tested for

frequencies of IFN-γ producing CD4+ and CD8

+ T-cells by cytokine flow cytometry (Lilleri et al.,

2009). PBMCs were washed, then incubated with Live/Dead Fixable Violet Dye (Invitrogen) and

V500 (clone RPA-T8)-conjugated anti-CD8 (BD Biosciences, San Jose, CA, USA) for cell surface

staining. Cells were then washed and permeabilized (FACS Permeabilizing Solution, BD

Biosciences, San Jose, CA, USA) and incubated with an intracellular mix of the following mAbs:

PerCP-Cy5.5TM

(clone UCHT1)-conjugated anti-CD3, APC-Cy7TM

(clone RPA-T4)-conjugated

anti-CD4, PE-Cy7TM

(clone B27)-conjugated anti-IFN-γ (BD Biosciences). Finally, cells were

washed, resuspended in 1% paraformaldehyde and analyzed with a FACSCanto II Flow cytometer

(BD Biosciences). As a routine, 1-2x105 viable lymphocytes were collected, and at least 2.5x10

4

CD3+CD4

+ and CD3

+CD8

+ T-cells were analyzed. The frequency of CD4

+ and CD8

+ T-cells

producing IFN-γ was calculated by subtracting the value of the sample incubated with mock-

infected DCs from the test value. Absolute CD3+CD4

+ and CD3

+ CD8

+ T-cells were determined on

whole blood samples with a direct immunofluorescence flow cytometry method (TruCOUNT tubes,

BD Biosciences). The total number of HCMV-specific CD4+ and CD8

+ T-cells was calculated by

multiplying the percentages of HCMV-specific T-cells positive for IFN-γ by the relevant absolute

CD4+

and CD8+ T-cell counts.

Cell surface staining of Tfh cells was done by incubating thawed PBMCs with a mix of the

following mAbs: PECy7TM

(clone EH12.1)-conjugated anti-CD279 (PD-1), BV-510 (clone

RF8B2)-conjugated anti-CXCR5, APC-Cy7

(clone RPA-T4)-conjugated anti-CD4 (BD

Biosciences), APC (cloneISA-3)-conjugated anti-CD278 (ICOS) (eBiosciences Inc., San Diego,

CA, USA), BV421 (clone G043H7)-conjugated anti-CD197 (CCR7) (BioLegend Inc., San Diego,

CA, USA). Finally, cells were washed, resuspended in 1% paraformaldeyde and analyzed.

12

Statistical analysis. Non-linear regression models were used to express the kinetics of different

serological parameters (IgG antibodies to gH/gL/pUL128L, gH/gL and gB; neutralizing antibodies

titers in ARPE-19 and HELF cells). The curves were compared by the extra sum-of-square F test

(GraphPad 5.0 Software). Medians of the glycoprotein complex-specific and neutralizing antibody

titers at different time intervals as well as HCMV load of different patient group populations were

compared by the Mann-Whitney U-test. Time to HCMV disappearance was calculated with the

Kaplan-Meier curves, which were compared by the Log-rank test.

ACKNOWLEDGEMENTS

This study was supported by the Fondazione Carlo Denegri (FCD), Turin, Italy (to G: Gerna),

Fondazione CARIPLO (grant 93043/A), Milan, Italy (to G. Gerna), and Ministero della Salute,

(grant GR-2010-2311329), Rome, Italy (to D. Lilleri). The Sponsors had no role in the design and

performance of the study, collection, management, analysis and interpretation of the data, review or

approval of the manuscript. We are indebted to Daniela Sartori for editing the manuscript, and to

Laurene Kelly for revision of the English. All the authors have seen and approved the final version

of the manuscript.

CONFLICT OF INTEREST

None.

ETHICAL APPROVAL

The research protocol was approved by the Bioethics Committee of the Fondazione IRCCS

Policlinico San Matteo. All participants provided written informed consent.

13

REFERENCES

Cui, X., Meza, B.P., Adler, S.P. & McVoy, M.A. (2008). Cytomegalovirus vaccines fail to induce

epithelial entry neutralizing antibodies comparable to natural infection. Vaccine 26, 5760-5766.

Fouts, A.E., Chan, P., Stephan, J.P., Vandlen, R., & Feierbach, P. (2012). Antibodies against

gH/gL/UL128/UL130/UL131 complex comprise the majority of the anti-cytomegalovirus (anti-

CMV) neutralizing antibody response in CMV hyperimmune globulin. J Virol 86, 7444-7447.

Furione, M., Rognoni, V., Sarasini, A., Zavattoni, M., Lilleri, D., Gerna, G. & Revello, M.G.

(2013). Slow increase in IgG avidity correlates with prevention of human cytomegalovirus

transmission to the fetus. J Med Virol 85, 1960-1967.

Furione, M., Rognoni, V., Cabano, E. & Baldanti, F. (2012). Kinetics of human cytomegalovirus

(HCMV) DNAemia in transplanted patients expressed in international units as determined with the

Abbott RealTime CMV assay and an in-house assay. J Clin Virol 55, 317-322.

Genini, E., Percivalle, E., Sarasini, A., Revello, M.G., Baldanti, F. & Gerna, G. (2011). Serum

antibody response to the gH/gL/pUL128-131 five-protein complex of human cytomegalovirus

(HCMV) in primary and reactivated HCMV infections. J Clin Virol 52, 113-118.

Gerna, G., Percivalle E., Lilleri, D., Lozza, L., Fornara, C., Hahn, G., Baldanti, F., & Revello,

M.G., (2005). Dendritic-cell infection by human cytomegalovirus is restricted to strains carrying

functional UL 131-128 genes and mediates efficient viral antigen presentation to CD8 T cells. J

Gen Virol 86; 275-284.

Gerna, G., Sarasini, A., Patrone, M., Percivalle, E., Fiorina, L., Campanini, G., Gallina, A.,

Baldanti, F. & Revello, M.G. (2008). Human cytomegalovirus serum neutralizing antibodies block

virus infection of endothelial/epithelial cells, but not fibroblasts, early during primary infection. J

Gen Virol 89, 853-865.

Gerna, G., Vitulo, P., Rovida, F., Lilleri, D., Pellegrini, C., Oggionni, T., Campanini, G.,

Baldanti, F. & Revello, M.G. (2006). Impact of human metapneumovirus and human

citomegalovirus versus other respiratory viruses on the lower respiratory tract infections of lung

transplant recipients. J Med Virol 78, 408-416. Erratum in: J Med Virol (2008) 80, 1869.

14

Griffiths, P.D., Stanton, A., McCarrel E., Smith, C., Osman, M., Harber, M., Davenport, A.,

Jones, G., Wheeler, D.C. & other authors (2011). Cytomegalovirus glycoprotein-B vaccine with

MF59 adjuvant in transplant recipients: A phase 2 randomized placebo-controlled trial. Lancet

377,1256-1263

Hahn, G., Revello, M.G., Patrone, M., Percivalle, E., Campanini, G., Sarasini, A., Wagner,

M., Gallina, A., Milanesi, G. & other authors (2004). Human cytomegalovirus UL131-128 genes

are indispensable for virus growth in endothelial cells and virus transfer to leukocytes. J Virol 78,

10023-10033.

He, J., Tsai, L.M., Leong, Y.A., Hu, X., Ma, C.S., Chevalier, N., Sun, X., Vandenberg, K.,

Rockman, S. & other authors (2013). Circulating precursor CCR7lo

PD-1hi

CXCR5⁺ CD4⁺ T cells

indicate Tfh cell activity and promote antibody responses upon antigen reexposure. Immunity 39,

770-781.

Lilleri, D., Zelini, P., Fornara, C., Comolli, G., Revello, M.G. & Gerna, G. (2009). Human

cytomegalovirus-specific CD4+ and CD8

+ T cell responses in primary infection of the

immunocompetent and the immunocompromised host. Clin Immunol 131, 395-403.

Lilleri, D., Kabanova, A., Lanzavecchia, A. & Gerna, G. (2012). Antibodies against

neutralization epitopes of human cytomegalovirus gH/gL/pUL128-130-131 complex and virus

spreading may correlate with virus control in vivo. J Clin Immunol 32, 1324-1331.

Lilleri, D., Kabanova, A., Revello, M.G., Percivalle, E., Sarasini, A., Genini, E., Sallusto, F.,

Lanzavecchia, A., Corti, D. & other authors (2013). Fetal human cytomegalovirus transmission

correlates with delayed maternal antibodies to gH/gL/pUL128-130-131 complex during primary

infection. PLoS One 8, e59863.

Lozza, L., Lilleri, D., Percivalle, E., Fornara, C., Comolli, G., Revello, M.G. & Gerna, G.

(2005). Simultaneous quantification of human cytomegalovirus (HCMV)-specific CD4+ and CD8

+

T cells by a novel method using monocyte-derived HCMV-infected immature dendritic cells. Eur J

Immunol 35, 1795-1804.

15

Macagno, A., Bernasconi, N.L., Vanzetta, F., Dander, E., Sarasini, A., Revello, M.G., Gerna, G.,

Sallusto, F. & Lanzavecchia, A. (2010). Isolation of human monoclonal antibodies that potently neutralize

human citomegalovirus infection by targeting different epitopes on the gH/gL/UL128-131A complex. J Virol

84,1005-1013.

Pass, R.F., Zhang, C., Evans A., Simpson, T., Andrews, W., Huang M.L., Corey, L., Hill, J., Davis, E.

& other authors. (2009). Vaccine prevention of maternal cytomegalovirus infection. N Engl J Med 360,

1191-1199.

Potzsch, S., Spindler, N., Wiegers, A.K., , T., Rucker, P., Sticht, H., Grieb, N., Baroti, T., Weisel, F. &

other authors. (2011). B cell repertoire analysis identifies new antigenic domains on glycoprotein B of

human cytomegalovirus which are target of neutralizing antibodies. PLoS Pathog 7, e1002172.

Revello, M.G., Fabbri, E., Furione, M., Zavattoni, M., Lilleri, D., Tassis, B., Quarenghi, A.,

Cena, C., Arossa, A. & other authors (2011). Role of prenatal diagnosis and counseling in the

management of 735 pregnancies complicated by primary human cytomegalovirus infection: a 20-

year experience. J Clin Virol 50, 303-307.

Revello, M.G. & Gerna, G. (2002). Diagnosis and management of human cytomegalovirus

infection in the mother, fetus, and newborn infant. Clin Microbiol Rev 15, 680-715.

Revello, M.G. & Gerna, G. (2010). Human citomegalovirus tropism for endothelial/epithelial

cells: Scientific background and clinical implications. Rev Med Virol 20, 136-155

Revello, M.G., Genini, E., Gorini, G., Klersy, C., Piralla, A. & Gerna, G. (2010). Comparative

evaluation of eight commercial human cytomegalovirus IgG avidity assays. J Clin Virol 48, 255-

259.

Ryckman, B.J., Rainish, B.L., Chase, M.C., Borton, J.A., Nelson, J.A., Jarvis, M.A. &

Johnson, D.C. (2008). Characterization of the human cytomegalovirus gH/gL/UL128-131 complex

that mediates entry into epithelial and endothelial cells. J Virol 82, 60-70.

16

Sallusto, F. & Lanzavecchia, A. (1994). Efficient presentation of soluble antigen by cultured

human dendritic cells is maintained by granulocyte/macrophage colony-stimulating factor plus

interleukin 4 and downregulated by tumor necrosis factor alpha. J Exp Med 179, 1109-1118.

Wang, D. & Shenk, T. (2005). Human cytomegalovirus virion protein complex required for

epithelial and endothelial cell tropism. Proc Natl Acad Sci USA 102, 18153-18158.

Zhou, M., Yu, Q., Wechsler, A. & Ryckman, B.J. (2013). Comparative analysis of gO isoforms

reveals that strains of human cytomegalovirus differ in the ratio of gH/gL/gO and gH/gL/UL128-

131 in the virion envelope. J Virol 87, 9680-9690.

17

Table 1. Kinetics of total and HCMV-specific CD4+, CD8

+ and follicular helper T-cells in a heart-transplanted patient showing a sharp increase in

Nt antibody titers at 189 days and their decrease at 350 days after transplantation upon increase/decrease in precursor and differentiated Tfh cells.

Nt, neutralization test; SOTR, solid-organ transplant recipient. *Precursor Tfh cells.

†Diffentiated Tfh cells.

Patient #,

age, transplant

organ, HCMV

serology

Days

after

trans-

plant

HCMV

DNA

(copies/ml)

CD4/µl CD8/µl

CD4

HCMV-

spec/µl

CD8

HCMV-

spec/µl

Antibody titer

%CXCR5

+

PD1hi

CCR7low

(of CD4

+ )

*

% ICOS+

(of CD4+

CXCR5+

PD-1hiCCR7

low)† ARPE-19

Nt

HELF-

Nt

Pentamer

IgG

gH/gL

IgG

gB

IgG

SOTR-054, 0 0 956 233 0.3 1.7 <40 <5 <50 105 <50

33 yrs, 19 0 467 135 0.1 0.3 40 <5 <50 164 <50

heart, 48 3,100 261 73 0.1 0.1 40 <5 <50 <50 <50 0.04 3.51

D+/R- 58 320,000 218 116 0.1 0.2 40 <5 <50 <50 <50

91 0 958 373 0.1 0.7 40 <5 <50 <50 <50 0.03 19.64

114 100 832 281 0.2 3.3 40 <5 <50 <50 91

148 700 830 479 0.7 14.6 40 <5 <50 <50 107

189 13,700 699 2,126 0.2 23.6 1,280 5 <50 <50 6,400 0.14 16.96

223 700 1,099 3,266 0.8 261.6 40,960 20 3,200 100 25,600

253 0 930 3,322 0.9 69.8 163,840 40 12,800 400 102,400 0.08 11.86

321 0 793 3,513 0.6 337.6 5,120 640 12,800 12,800 25,600

350 0 841 3,753 0.8 67.6 5,120 1,280 25,600 25,600 25,600 0.04 0.00

18

Figure Legends

Fig. 1. Human cytomegalovirus (HCMV) DNA (a) appearance, (b) peak, and (c) persistence in

blood of 40 immunocompetent subjects and 17 solid-organ transplanted patients. Median values

with interquartile ranges are shown. Mann Whitney U test (a, b) and log-rank test (c) were adopted

for statistical analysis.

Fig.2. (a) Time (days) to 1st antibody detection after transplantation in the immunocompromised

transplanted patients and after presumed onset of infection in immunocompetent subjects. (b) Time

(days) required to reach the peak antibody titer after reaching the DNA peak in the transplanted and

the immunocompetent patient. Median values with interquartile ranges are shown. Mann-Whitney

U test was adopted for statistical analysis.

Fig. 3. (a) Kinetics of the ELISA IgG antibody response to the pentamer gH/gL/ in transplanted

patients compared with immunocompetent individuals during 1-year follow-up: non-linear

regression curves are shown and compared with the extra sum-of-square F test. (b) IgG titers at

different time intervals after onset of infection: median values with interquartile ranges are shown.

Mann-Whitney U test was adopted for statistical analysis.

Fig. 4. (a) Kinetics of the ELISA IgG antibody response to gH/gL in the two populations examined

during 1-year follow-up: non-linear regression curves are shown and compared with the extra sum-

of-square F test. (b) IgG titers at different time intervals after onset of infection: median values with

interquartile ranges are shown. Mann-Whitney U test was adopted for statistical analysis.

Fig. 5. (a) Kinetics of the ELISA IgG antibody response to gB in the two populations studied during

1-year follow-up: non-linear regression curves are shown and compared with the extra sum-of-

square F test. (b) IgG titers at different time intervals after onset of infection: median values with

interquartile ranges are shown. Mann-Whitney U test was adopted for statistical analysis.

Fig. 6. (a) Kinetics of the median neutralizing activity preventing infection of ARPE-19 epithelial

cells by the HCMV isolate VR1814 in the two populations during 1-year follow-up: non-linear

regression curves are shown and compared with the extra sum-of-square F test. (b) Neutralizing

titers at different time intervals after onset of infection: median values with interquartile ranges are

shown. Mann-Whitney U test was adopted for statistical analysis.

Fig. 7. Primary HCMV infections in (a) an immunocompetent subject and (b) a transplanted patient

are shown. (Top panel): kinetics of total and HCMV-specific CD4 + and CD8

+ T-cells as well as

viral load are shown. (Middle panel): Kinetics of indicated antibody titers as well as percentages of

19

CXCR5+ PD-1

hi CCR7

low CD4

+ T-cells are reported. (Lower panel): The percentage of ICOS

+

among CXCR5+ PD-1

hi CCR7

low CD4

+ T-cells is reported.

20

21

22

23

24

25

26

27