Human cytomegalovirus (HCMV) infection causes severe clinical conditions in the face of immature or suppressed immune defense functions. The virus is, however, readily controlled by an immunocompetent host. Central to this process is the stimulation and maturation of dendritic cells (DC) (1, 2). Monocytederived, immature DC (iDC) are fully permissive to HCMV replication (3), but infection of these cells is restricted to strains expressing the viral genes UL128 to UL131A (4). Infection of iDC by HCMV and its impact on their maturation and function have been studied by several laboratories (5-10). There is consensus that T cell costimulatory molecules, such as CD40, CD80, and CD86 and MHC classes I and II, are downregulated on the surfaces of iDC following HCMV infection (2). As a result, iDC are impaired in their capacity to stimulate T cell responses (5-9), although this effect may depend on time after infection (4). With iDC function impaired, direct HCMV infection of iDC may not be sufficient to prime and sustain the vigorous T cell response against the virus that is seen during natural infection. Uptake and cross-presentation of HCMV antigen by uninfected iDC was identified as one mechanism to bypass HCMV-mediated immunosuppression of iDC function (11-13). Here we address whether exposure of iDC to subviral, noninfectious dense bodies (DB) of HCMV in the absence of viral infection is suitable for inducing activation and maturation, thereby rendering these cells competent for antigen presentation.(Part of this research was conducted by C. Sauer in partial fulfillment of the requirements for a doctoral degree from Johannes Gutenberg University, Mainz, Germany.)To investigate the impact of DB on iDC, we generated iDC from peripheral blood mononuclear cells (PBMC) of HCMV-seronegative individuals (14). Over 99% of the cells were CD3 negative, excluding lymphocyte contaminations (data not shown). Human foreskin fibroblasts (HFF; 3,6 ϫ 10 7 cells for each virus) were infected with either strain RV-HB5 (a derivative of the laboratory strain AD169) or the endotheliotropic strain RV-TB40/E. DB and virions were purified from 400 ml of culture supernatants at 7 days postinfection using glycerol-tartrate gradient ultracentrifugation (15). After high-speed sedimentation of the material collected from the bands, DB (RV-HB5) fractions were resuspended in 300 l of phosphate-buffered saline (PBS; 1 mg), virions from RV-HB5 were resuspended in 100 l of PBS (175 g), and virions from TB40/E were resuspended in 100 l of PBS (130 g). DB fractions were either used directly or exposed to UV light (DB UV ; 254 nm of UV light for 2 min) to inactivate residual infectious virus. HFF and iDC were incubated with 10 g of each of these fractions in order to determine infectivity (by staining for HCMV immediate early protein 1 [IE1] expression) or antigen uptake (by staining for the tegument protein pp65) ( Table 1). The results showed that DB contained very little virus (DB fraction) or no virus (DB UV fraction) that would infect iDC. By co...