DC culture medium consisted of RPMI 1640 medium (Life Technologies, Carlsbad, Calif.) supplemented with 2 mM L-glutamine (Sigma), 1% nonessential amino acids (Life Technologies), 1% sodium pyruvate (Life Technologies), 50 μM 2-mercaptoethanol (Sigma), 50 μg of gentamicin (Life Technologies) per ml, and 10% fetal calf serum (Life Technologies).

All human specimens were obtained under informed consent, as approved by the University of Maryland Baltimore Institutional Review Board. Monocyte-derived DCs were generated as described previously (17), with minor modifications. Briefly, human peripheral blood mononuclear cells were isolated, from three independent normal healthy donors, by Ficoll-Hypaque density gradient centrifugation and were enriched for CD14⁺ monocytes by negative selection with a cocktail of monoclonal antibodies from StemCell Technologies (Vancouver, British Columbia, Canada), according to the instructions of the manufacturer. Typically, greater than 80% of the cells were CD14⁺ after enrichment (data not shown). The isolated monocytes were allowed to adhere to plastic by plating 10⁶ cells per/ml in RPMI 1640 medium for 2 h. Adherent monocytes were washed with RPMI 1640 medium and were then cultured for 6 days at 10⁶ cells per/ml in DC culture medium supplemented with 50 ng of recombinant GM-CSF (rGM-CSF, R&D Systems, Minneapolis, Minn.) per ml and 1,000 U of recombinant IL-4 (rIL-4; R&D Systems, Minneapolis, Minn.) per ml.

After 6 days in culture, MDDCs were pulsed with either 5 μg/ml of HIV-VLPs or 1 μg/ml of LPS for 4 and 8 hours, for gene microarray analysis, and for 16 hours for maturation and activation phenotype analysis.

DCs were incubated for 30 min at 4°C with murine monoclonal antibodies specific for CD80, CD83, CD86, and HLA-DR (BD Pharmingen, San Diego, CA), washed and then fixed with 2% paraformaldehyde for analysis with a FACScalibur flow cytometer (BD Pharmingen). Data analysis was carried out with FlowJo software (Tree Star Inc., San Carlos, CA). The fraction of MDDCs that responded by upregulation of activation markers on the cell surface was calculated by overlaying the histograms of treated and untreated MDDCs and Overton subtraction of the curves.

DCs were harvested, washed twice in PBS and lysed in 350 ul RLT buffer with fresh addition of 2-Mercaptoethanol per each well of the 6-well plate. Total RNA was isolated using RNeasy minikits (Qiagen), according to the manufacturer's protocol, and RNA quality and quantity was estimated by Agilent Bioanlayzer (Agilent Technologies, Palo Alto, CA) and NonoDrop. Amplified antisense RNA (aRNA) was obtained from total RNA (0.5–3 μg) via two round of in vitro transcription, according the protocol previously described by us 26. 6 ug of amplified test samples aRNA were labeled with Cy5 (Amersham) while the same amount of reference sample (pooled normal donor PBMCs) was labeled with Cy3. Test-reference sample pairs were mixed and co-hybridized to 17K cDNA microarrays 27.

Hybridized arrays were scanned at 10-μm resolution on a GenePix 4000 scanner (Axon Instruments) at variable PMT voltage to obtain maximal signal intensities with less than 1% probe saturation. Resulting jpeg and data files were deposited at microarray data base (mAdb) http://nciarray.nci.nih.gov and retrieved after median centered, filtering of intensity (>300) and spot elimination (bad and no signal). Data were further analyzed using Cluster and TreeView software 27 and Partek Pro software (Partek). Subsequent low-stringency filtering (80% gene presence across all experiments and at least one experiment with ratio fold change >3), 3,119 genes were selected for further analysis. Hierarchical cluster analysis was conducted on these genes according to Eisen et al. 28; differential expressed genes were visualized by Treeview and displayed according to the central method 29.

Immature MDDCs were obtained from three independent donors and, after 6 days of culture in IL-4- and GM-CSF-enrichment medium, were incubated with 5 μg/ml of baculovirus-expressed HIV-VLPs or 1μg/ml of LPS. In parallel, control MDDCs were loaded with PBS. After a 16 hr-induction, the expression of surface maturation/activation markers, such as CD80, CD83, CD86 and HLA-DR, was examined. The expression of all the four markers was upregulated by treatment with HIV-VLPs, compared to PBS (Fig. 1). The level of cytokines involved in the Th1/Th2 polarization was assessed in the supernatant of MDDCs loaded with HIV-VLPs or LPS. In particular, the TNF-α, IL-12 p70, IL-10 were produced at higher levels in HIV-VLP-loaded MDDCs compared to LPS-loaded MDDCs (data not shown).

These results indicate that baculovirus HIV-VLPs induce a specific MDDC maturation pathway, distinct from the one induced by the LPS.

Gene expression profiles were generated from treated and control MDDC harvested 4 and 8 hrs after stimulation. These time points were selected to evaluate a possible biphasic response, as described for LPS-induced DCs and mononuclear phagocytes (MPs) 2325. Amplified antisense RNA (aRNA) was obtained from total RNA extracts 26 and hybridized to a custom-made 17,000 (17K)-clone cDNA microarray chip enriched with genes relevant to immune function. Stringent filtering were further applied to eliminate genes with missing value in >20% of all the experiments and >3 fold change in at least one experiment. The remaining 3,119 genes were thus used for statistic analysis.

Unsupervised cluster analysis obtained on MDDCs stimulated for 4 and 8 hr with HIV-VLPs or LPS shows the segregation of the untreated (PBS) and the treated (HIV-VLPs or LPS) MDDCs in two independent clusters. However, within the treated MDDCs cluster, the samples subclustered according to the treatment (HIV-VLPs vs LPS), with the exception of a single HIV-VLP-treated sample which formed an independent cluster closer to the LPS-treated sample, indicating the induction of two distinct transcription machinery. Furthermore, while the HIV-VLP-treated MDDCs clustered according to the donor, the LPS-treated MDDCs clustered preferentially according to the treatment duration (4 or 8 hr) of stimulation (Fig. 2). This would suggest an individual donor susceptibility to HIV-VLP-treatment and a possible delayed transcription or secondary response to LPS treatment.

The differential gene expression in HIV-VLP-treated MDDCs, compared to either LPS- or PBS-treated MDDCs, was considered statistically significant only when supported by a p < 0.005.

Treatment-induced changes in gene profiling were analyzed using student t test, based on simple two-treatment comparison or one versus a combined two-treatment comparison. Considering only genes showing at least a 2-fold modulation (increase or reduction) in the transcriptional levels, it has been possible to identify unique and common genes in the profile induced by the different treatments. In particular, the HIV-VLP treatment, induced the upregulation of 140 and 72 genes as well as the downregulation of 108 and 46 genes, compared to PBS and LPS respectively. A sub-set of upregulated and downregulated genes are specific to the HIV-VLP treatment (Fig. 3), indicating that, besides transcriptional patterns shared with the LPS treatment, HIV-VLPs induce a specific reprogramming of the MDDCs transcriptional profile.

Visualization of the gene expression among different treatments revealed a transcriptional profile pattern in HIV-VLP-induced MDDCs distinct from either PBS or LPS (4 hr and 8 hr) induction (Fig. 4A). Furthermore, clustering analysis performed based on 217 genes selectively expressed (p < 0.005) in HIV-VLP induced MDDCs, compared to LPS treatment, confirmed a distinct transcriptional profiles signature according to the treatment, regardless of the length of stimulation (Fig. 4B).

Gene expression changes in response to stimuli are generally pathways directed. In order to analyse pathways modulated in HIV-VLPs induced MDDCs, we dissected genes according to cluster nodes derived on the basis of expression similarity. Supervised cluster analysis based on the 3,119 genes were further analyzed according to pattern recognition. Genes uniquely upregulated in MDDCs by HIV-VLPs only or by HIV-VLPs and LPS (either at 4 h or 8 h post-induction) are indicated in Fig. 5A and 5B respectively. Similarly, nodes were identified including genes down-regulated by HIV-VLPs only or by HIV-VLPs and LPS at 4 h post-induction (Fig. 5C top and bottom, respectively). Most of the genes present in these nodes are strictly associated to the functional and morphological changes characterizing the MDDCs maturation and activation process.

In this regard, the MDDCs transcriptional profile was analyzed on the basis of the cellular pathways modulated by the HIV-VLP-treatment, compared to either PBS or LPS treatment. In respect to the main focus of this study, only the pathways involved in the MDDCs-activation by HIV-VLPs have been evaluated in more detail. These can be divided into main super-families concerning the cell replication/death, the transcriptional control, the phagocytosis and pathogen recognition, the immunological function of dendritic cells, the migration and the control of the inflammatory responses (Table I).

The effects of the HIV-VLP treatment on MDDCs, compared to the PBS and LPS treatments, has been analyzed also considering the individual genes showing increased transcriptional levels. The presumed biological implication of selected genes involved in the specific functions of Dendritic Cells as professional APCs, has been evaluated in more detail according to their function (Table II).