Sistema Neuroendocrino

Med Microbiol Immunol (2008) 197:205–213
DOI 10.1007/s00430-008-0077-2

R EV IE W

The salivary glands as a privileged site of cytomegalovirus
immune evasion and persistence
Ann E. Campbell · Victoria J. Cavanaugh ·
Jacquelyn S. Slater

Received: 21 December 2007 / Published online: 8 February 2008
© Springer-Verlag 2008

Abstract The salivary glands (SG) provide a haven forpersistent cytomegalovirus replication, and in this regard
are a privileged site of virus immune evasion. The murine
cytomegalovirus (MCMV) model has provided insight into
the immunological environment of the SG and the unqiue
virus–host relationship of this organ. In response to MCMV
infection, a robust T cell-mediated immune response is elicited, comprised predominantly of CD8+ T cells thatphenotypically and functionally appear activated. However, they
fail to clear virus by an unknown evasion mechanism that is
independent of inhibitory NKG2A- or Programmed Death
1-mediated signaling. Virus is eventually eliminated from
the SG by eVector CD4+ T cells expressing antiviral cytokines. However, this mechanism is severely dampened by
high levels of the immunosuppressive cytokine IL-10,selectively expressed by SG CD4+ T cells.
Keywords Murine cytomegalovirus · T cells · IL-10 ·
NKG2A · PD-1

Introduction
Herpesviruses use a variety of strategies to maintain their
existence within the human population through transmission from acutely or latently-infected persons to noninfected individuals. Human cytomegalovirus (HCMV) is
eYciently spread through contact with mucosalsecretions
containing infectious virus. These include saliva, semen,

A. E. Campbell (&) · V. J. Cavanaugh · J. S. Slater
Department of Microbiology and Molecular Cell Biology,
Eastern Virginia Medical School, 700 W. Olney Road,
Norfolk, VA 23507, USA
e-mail: campbeae@evms.edu

vaginal and cervical secretions, and breast milk. There are
two features of mucosal shedding of HCMV that are ofparticular epidemiological signiWcance. One is the prolonged
secretion of virus in saliva (as well as urine) for months following an acute, primary infection. This is best exempliWed
by reports that as high as 78% of toddlers attending day
care centers in the United States shed HCMV in saliva for
up to 2 years, and consequently, seronegative day care providers and mothers of infected children haveannual infection rates ten times higher than expected [1–5].
A second feature is the intermittent and clinically unapparent excretion of reactivated virus in saliva, semen, cervical and vaginal secretions, and breast milk from latentlyinfected persons who have demonstrable immunity to
HCMV. HCMV DNA was detected in 8.7% of semen samples from healthy infertility patients, the highest among allsexually transmitted viruses [7], and in approximately 3%
of cryopreserved semen samples for sperm donations [8].
The percentage increases to as high as 65% in a subpopulation of HIV infected, antiretroviral therapy-naïve men [55].
HIV-infected persons, as well as transplant patients, also
excrete HCMV in saliva at elevated frequencies (31 and
45%, respectively) [13, 40]. Particularly high isthe frequency of HCMV isolated from or detected in breast milk
of lactating, seropositive mothers. In fact, nearly all HCMV
seropositive women will reactivate and shed virus during
lactation [53]. Breast-milk acquired HCMV is clearly a
source of transmission to the neonate who, if preterm, may
experience at least transient disease [38, 43, 44, 61].
From the above, it is evident that thevirus–host relationship at oral and genital mucosal sites is central to the pathogenesis and epidemiology of this clinically important virus.
Also evident is the fact that the immune response to the
virus at mucosal sites may be diVerent from systemic
immunity in that clearance is protracted and the incidence

123

206

of reactivation is high. This review will focus on the
immune...