Low-passage clinical strains of HCMV, but not laboratory strains, establish a latent infection in cultured CD34⁺ hematopoietic progenitor cells. (A) Limiting dilution analysis was used to determine frequency of infectious center formation in CD34⁺ cells infected with low-passage (FIX and Toledo) or laboratory (Towne and AD169) strains at 10 to 14 dpi. The frequency is calculated as 1 over the number of cells required to yield one infectious center. The bars represent the average frequency of infectious center formation in the reactivation experiment (■) or the lysate control (▒) in at least 3 independent experiments for FIXwt, Toledo, and AD169 and two independent experiments for Towne. The standard error of the means is shown. The top number above each pair of bars reports the fold increase in infectious center formation in the reactivation compared with the lysate control. Using the paired Student t test, the probability (P) that the difference between the 2 averages is significant was calculated. Statistical significance is indicated by ** for P ≤ .01. All other pairs do not represent a significant difference (P > .05). The line marks the limit of detection (LD) for the assay. (B) CD34⁺ cells infected with Toledo or AD169 at a multiplicity of 5 PFU/cell were purified 20 hours later by FACS and seeded into long-term bone marrow cultures. Linearly amplified RNA was analyzed using the HCMV array at 20 dpi infection. All RNAs were hybridized in triplicate. Arabidopsis cDNA spots used as positive controls are circled. Cellular cDNA spots used for normalizing arrays are boxed. All other spots represent HCMV ORFs.