Translational Mini-Review Series on Infectious Disease: Congenital cytomegalovirus infection: 50 years on

Abstract

Cytomegalovirus (CMV) is the leading cause of congenital viral infection, with an incidence of 0·5–3% of live births worldwide. Clinical evidence has shown hearing and vision loss, mental retardation and sometimes death in affected newborns. Primary maternal CMV infection during gestation poses a 40% risk of intrauterine transmission in contrast to recurrent infection. European laboratories have made significant progress in the last decade in solving diagnostic problems linked to infection in pregnancy. With the advances in CMV serology, such as detection of anti-CMV IgM by enzyme immunoassays (EIA), confirmed by Western blot, together with seroconversion and anti-CMV IgG avidity evaluation in pregnant mothers, can help to identify recent infection. Preventative measures such as screening for CMV in the routine serological work-up of pregnant women have been introduced in countries such as Spain and Italy. The development of specific T cell-mediated immune responses in mothers, fetus and neonates is now emerging with regard to antigen-specific CD4 and CD8 T cells, differentiation status, proliferative and cytokine responses. A protective vaccine against CMV is a major public health priority and the study of vaccines in animal model systems has identified potential strategies for interrupting transmission and preventing disease in newborns. Congenital CMV infection has a variable outcome and therefore novel diagnostic methods are required to identify those at risk and therapeutic interventions are needed to improve the long-term prognosis of those infected. CMV was first isolated in 1957. We are now 50 years on, so procrastination is not an option.

Introduction

Human cytomegalovirus (HCMV) is a ubiquitous beta herpes virus which was first isolated in 1957 [1]. Humans are the only reservoir for HCMV. Primary infection with HCMV results in lifelong latency with immunological control of significant viral replication; however, excretion following reactivation throughout life is common and may result in viral transmission. Primary infection in an immunocompetent host is generally asymptomatic, although in addition to a viral syndrome a small minority may develop infectious mononucleosis and hepatitis [2]. As a consequence, HCMV severe disease is restricted to the immunocompromised, such as transplant recipients, HIV-infected individuals or the immunologically immature host. Primary HCMV infection during pregnancy carries the highest risk of intrauterine transmission, which can cause severe fetal damage including hepatosplenamegaly, jaundice, central nervous system (CNS) abnormalities and growth retardation [3]. Of the 90% of babies who are asymptomatic at birth, 30% may develop severe hearing deficits.

The virus

HCMV is a 235-kb double-stranded linear DNA virus [4] and generally has a seroprevalence of between 30 and 90% depending on socio-economic factors and geographical location [5]. Advances in laboratory techniques have shown that CMV replicates rapidly in vivo similar to that seen with human immunodeficiency virus (HIV) [6] and can infect many cell types; however, in in vitro culture, propagation of the virus occurs successfully in fibroblast [7,8]. As a consequence of repeated passaging in fibroblasts, laboratorystrains of CMV have lost some genes and are therefore not completely representative of wild-type strains [9]. Although there have been significant advances in our understanding of the immunobiology of HCMV, at present no vaccine is available.

Immunology of CMV infection

The interaction between strategies evolved by the virus to evade and avoid the host defence mechanisms and the host's ability to control viral infection is complex; however, animal models of CMV infection have helped in our understanding of these pathways. In an intact immune system HCMV infection can generally be kept under control, but complete clearance of the virus is rarely achieved and the viral genome remains at selected sites in a latent state. An important feature of HCMV is the ability to infect in vivo a broad spectrum of cells, including fibroblasts, epithelial, endothelial, macrophages and muscle cells [10].

Innate immune response

Early studies in the 1980s showed a substantial role for the protective function of natural killer (NK) cells in murine CMV infection. Mice which had a defect in NK function were more susceptible to MCMV [11,12] and the level of NK activity correlated with the degree of resistance in susceptible and resistant strains of mice [13,14]. Furthermore, depletion of NK cells increased the severity of MCMV disease [15]. The production of α and β interferon is an essential part of the host's non-specific response in the early stages of infection. Definitive evidence for the protective role of interferon was shown when administration of antibodies to α and β interferons reduced significantly the resistance of mice to MCMV infection and resulted in increased viral titres in the blood and liver [16,17].

Humoral host immune response

It is unclear whether the humoral response plays an important role in host defence against HCMV. A recent report has shown that the administration of HCMV-specific hyperimmune globulin to pregnant women significantly lowered the risk of congenital CMV infection and disease [18]. Furthermore, viral transmission from mother to fetus is increased if the maternal antibody response to HCMV is of low avidity or of poor neutralizing activity, indicative of a primary response [19], and primary HCMV infection is more frequent and severe in seronegative solid organ transplant recipients of a CMV-positive donor organ [20]. Clinical studies examining the kinetics of CMV specific antibody appearance have shown a beneficial effect in allogeneic bone marrow transplant recipients [21]. Conversely, the fetus can be infected by intrauterine transmission of HCMV in mothers known to have antibodies prior to pregnancy and, furthermore, the fetus can be infected by CMV in breast milk despite the presence of maternally derived passively acquired antibodies. Similarly, seropositive transplant patients can be reinfected with CMV from the donor, again suggesting that seropositivity does not in itself confer protection [22]. The presence of anti-CMV specific antibody is therefore a marker of previous infection rather than a measure of immunity [22]. Taken together, these findings suggest that infections during pregnancy or post-transplantation may be less severe in the presence of pre-existing antibodies. However, these specific antibodies may not prevent CMV infection but could reduce clinical manifestations [22].

Cellular immune response

Both human and murine studies have illustrated the importance of the cell-mediated immune response in CMV infection. Patients with deficiencies in cell-mediated immunity are at higher risk of CMV disease and T cell-deficient nude mice are also susceptible to murine CMV infection [16,23]. Studies in murine models have shown that a loss of T cell function coincided with increased reactivation and dissemination of viral infection [24]. To enhance its survival, HCMV has the ability to persist in a non-productive form or avoid the immune system due to its ability to induce a latent state of infection, its escape from CD8⁺ T cells by exploiting immunologically privileged tissues for replication and the expression of genes that interfere with the immune response [25].

Cellular immune responses are induced by dendritic cells (DC) that present antigens to CD4⁺ and CD8⁺ T cells in the lymph nodes which, after priming, re-enter the bloodstream. The strength of the immune response is governed by the stimulatory function of the DC which determines antigen processing, major histocompatibility complex (MHC) expression and co-stimulation. The exact role of DCs is limited, as direct analysis is difficult due to their low numbers in peripheral blood. Due to the emergence of new technologies, such as enzyme-linked immunospot (ELISPOT), MHC-peptide tetramers and predictive bioinformatics, profiling of T cell responses to a wide range of CMV antigens is now possible.

A direct role for CD8⁺ T cells was shown in studies where reconstitution of donor-derived HCMV-specific CD8⁺ T cells restored antigen specific cellular immunity and also protected from HCMV-associated clinical complications in patients undergoing allogeneic haematopoietic stem cell transplant [26,27]. Similarly, Ho and co-workers showed that in a murine model, passive transfer of T cells primed in vivo but restimulated in vitro could protect mice from MCMV infection [28]. Several studies showed that the Tc cells that could lyse HCMV-infected fibroblasts were MHC class I-restricted and were indeed CD8⁺ T cells [29,30]. Recent publications examining the specific antigenic target of these CD8⁺ T cells have shown a broad and diverse host response to a repertoire of antigens, although reactivity against pp65, pp55, IE-1, glycoprotein B and IE-2 form a substantial part of the response. This is supported by the finding that in healthy virus carriers approximately 10% of the total CD8⁺ memory T cell pool in the peripheral blood is specific for HCMV antigens, and this proportion can exceed 30% in some elderly individuals [31]. A similar scenario was found in the HCMV-specific CD4⁺ T cell population, suggesting that CD4⁺ T cells also play a crucial role in the control of HCMV infection. Kern et al. [32], examining the response of 40 donors to peptides derived specifically from the pp65 antigen, showed that 63% of normal healthy donors have a CD4 T cell response and 83% have a CD8 T cell response, indicating that this protein is an important target for both CD4 and CD8 T cells and that the response to CMV is high in the majority of individuals. The importance of CD4 T cells is highlighted further in a study examining the kinetics and characteristics of CMV-specific CD4⁺ and CD8⁺ T cells in the course of primary CMV infection in patients receiving renal transplants [33]. The authors showed that in asymptomatic individuals the CMV-specific CD4⁺ T cell response preceded the CMV-specific CD8⁺ T cell response; however, in symptomatic patients the CD4⁺ T cell response was delayed and detected only after anti-viral treatment. These findings suggest that the presence of functional and specific CD8⁺ T cell and antibody responses are not sufficient to control viral replication and that the formation of specific effector CD4⁺ T cells is essential for clearance of infection [33].

Prolonged viral shedding in urine and saliva (at least 12–29 months after acquisition of CMV) occurs in immunocompetent young children who acquire CMV, when compared to adults (6 months). This correlates with the decreased CMV-specific T helper 1 (Th1) response, as measured by the secretion of interferon (IFN)-γ and interferon (IL)-2 [34]. The authors suggest that CD4⁺ T cell immunity to HCMV may be generated in an age-dependent manner [34]. In a recent report, reduced IFN-γ levels were observed in congenitally infected neonates when compared to their matched mothers [35]. The presence of mature and functional CD8⁺ T cells in neonates in response to congenital HCMV infection, even as early as 28 weeks of gestation, has been reported [36]. Neonates expressed a late differentiation phenotype of low levels of CD28, CD27 and CD45RA as well as perforin-dependent cytotoxicity and IFN-γ production similar to infected adults [36].

Mechanisms of HCMV-induced subversion of the immune response

In efforts to design effective vaccines, it is of critical importance to understand how CMV induces and circumvents the host immune response. The main route of escape of CMV is by blocking the expression of MHC class I molecules which contain a cytotoxic T lymphocyte (CTL) target, which degrades MHC class II molecule proteins and prevents presentation of the viral antigen to CD4⁺ T lymphocytes [37]. Table 1 shows a list of known cell responses which are modulated by CMV. Recently, the discovery that pattern recognition receptors Toll-like receptor 2 (TLR2) and CD14 recognize CMV virus particles and thereby trigger inflammatory cytokine production has increased our understanding of the indirect pathological processes associated with CMV disease.

Table 1.

Mechanisms of CMV subversion of the host immune response.

Modulation of cell function	References
Mobilization of Ca2+ release into cytosol	[38]
Activation of phospholipase C and A2	[38]
Increased release of aracadonic acid and its metabolites	[38]
Induction of cellular oncogenes c-jun, c-fos, c-myc	[39]
Activation of transcription factors NF-κB, SP-1	[40]
Activation of map kinases, ERK1/2, p38	[40]
Inhibit macrophage activation	[41]
Down-regulate immune receptors; MHC classes I/II	[42]
Modulates antigen presentation
Interferes with antigen proteolysis, processing	[43]
ER retention, TAP binding, blocks peptide importation
UL83	[44]
Inhibits MHC class I export
US2, US3, US6, US11	[45]
Ligand decoys for NK receptors, UL18	[46]
Mimic chemokines/cytokines and their receptors, e.g.
UL144-TNF receptor, UL146, UL147- IL-8 like;	[47,48]
US28- chemokine receptor for fractalkine, US27, UL33	[49,50]
Induction of innate immune activation, e.g.
ISGs, inflammatory genes e.g. RANTES, IL-6, COX-2	[51,52]
TLR2 recognition; induction of inflammatory cytokines	[53]
Induction of co-stimulatory molecules, e.g. CD80 and CD86	[54]
Promotes cell adhesion and migration
Enhances ICAM-1 expression	[55,56]

NF-κB: nuclear factor kappa B; MHC: major histocompatibility complex; ER: endoplasmic reticulum; TAP: transporter associated with antigen processing; UL: unique long; US; unique short; NK: natural killer; ISG: interferon-stimulated gene family; RANTES: regulated upon activation normal T cell expressed and secreted; IL: interleukin; COX: cyclooxygenase; TLR: Toll-like receptor; ICAM-1: intercellular adhesion molecule-1; TNF: tumour necrosis factor.

Laboratory diagnosis of HCMV

Accurate and prompt diagnosis is essential in the prevention and monitoring of HCMV infection, especially in immunocompromised individuals. HCMV was first recognized as the cause of fetal death following a cytomegalic inclusion disease in 1957 when first in culture [1]. Diagnosis of HCMV can be performed by virological, serological, cellular and molecular techniques, as outlined in Table 2. The combination of these methods enables the discrimination between primary and reactivation, viral characterization, viral burden in blood and cerebrospinal fluid (CSF) compartments and the monitoring of viral excretion. Despite the range of assays available, it is still not possible to predict accurately which fetuses will be infected and the extent of clinical manifestations.

Table 2.

Laboratory assays currently used for the detection and antibody response to human cytomegalovirus (HCMV).

Test	Sample required	Method	Assay time
Virological
DEAFF test	Urine, throat washings, saliva	‘Shell vial’ assay using MRC5 fibroblast cell line	24–48 h
Serological
IgG, IgM	Serum/plasma	EIA/Western blotting	3–5 h
IgG avidity	Serum	Luminescence	1 h
Cellular
Antigenaemia	Peripheral blood, leucocytes	Immunofluorescence to detect pp65 matrix protein of CMV	5 h
Molecular
Nucleic acid amplification	Plasma, blood, amniotic fluid, urine	PCR-qualitative or quantitative-viral load	5 h

DEAFF: detection of early antigen fluorescent foci; PCR: polymerase chain reaction.

HCMV vaccine development

Currently, there is no licensed vaccine for CMV available, even though the first human study of a CMV vaccine using a laboratory-adapted strain ad 169 was conducted in the early 1970s [57]. Clinical evidence as shown above indicates that immunity to CMV appears to be advantageous to the host in preventing disease. Vaccines based upon the attenuated Towne strain has led to disappointing results in several trials conducted on both CMV-positive and -negative renal transplant recipients [58], adult women with children in daycare centres [59] and healthy immunocompetent individuals [60]. Although the virion envelope carries a range of glycoproteins, glycoprotein gB has been recognized as the primary target for neutralizing antibodies and dominates the humoral response. In the guinea pig CMV model, gB vaccination has a protective effect against congenital transmission [61,62]; however, results in humans have been disappointing [63]. Different delivery vehicles and regimens such as using an attenuated form of poxvirus, modified vaccinia Ankara and prime-boost are currently under investigation [64]. The use of subunit, recombinant and DNA vaccines is currently being evaluated.

In a recent report, Khanna and Diamond [65] suggest that perhaps a more feasible goal would be to examine strategies to prevent HCMV disease rather than infection. These authors suggest that strategies which reduce HCMV-related pathogenesis should be examined for the prevention of congenital HCMV infection, methods of inducing sufficient immunity in the prevention of infection of women at risk and vaccine strategies which would be effective to control reactivation of CMV in transplant patients and the limiting of disease development in these groups [65]. Although our knowledge of the virus–host relationship has increased, the pharmacological treatments available have severe limitations such as toxicity, development of resistance and low potency.

Conclusion

It is of concern that in many European countries the rate of seroprevalence of HCMV has fallen and continues to fall. To enhance prenatal diagnosis, recent studies emerging from Italy and France examined viral loads in amniotic fluid together with ultrasound analysis in an attempt to establish improved parameters to determine pregnancy outcome [66]. Viral loads have been shown to be good predictive and prognostic markers of disease severity. Detailed genotypic analysis in 56 women has shown that polymorphisms in the CMV encoded UL144 gene (truncated tumour necrosis factor receptor), in particular genotype C, may be associated with symptomatic outcome of congenital CMV infection [66]. Others have suggested that routine screening of Guthrie cards for HCMV by polymerase chain reaction will aid in identifying at-risk neonates who can then be monitored closely [67]. Recent reports of successful treatment with passive immunization during pregnancy to prevent or treat fetal infection suggests that prenatal therapy given at the time when most of the fetal damage may occur requires further study. The outcome of congenital CMV is highly variable, hence the study of new therapies remains a challenge.