Quantitative analysis of latent human cytomegalovirus

Abstract

Cytomegalovirus latency depends on an interaction with hematopoietic cells in bone marrow and peripheral blood. The distribution of viral DNA was investigated by PCR-driven in situ hybridization (PCR-ISH), and the number of viral genomes per cell was estimated by quantitative competitive PCR during both experimental and natural latent infection. During experimental latent infection of cultured granulocyte-macrophage progenitors, the viral genome was detected in >90% of cells at a copy number of 1 to 8 viral genomes per cell. During natural infection, viral genomes were detected in 0.004 to 0.01% of mononuclear cells from granulocyte colony-stimulating factor-mobilized peripheral blood or bone marrow from seropositive donors, at a copy number of 2 to 13 genomes per infected cell. When evaluated by reverse transcription-PCR-ISH, only a small proportion of experimentally infected cells (approximately 2%) had detectable latent transcripts. This investigation identifies the small percentage of bone marrow-derived mononuclear cells that become latently infected during natural infection and suggests that latency may proceed in some cells that fail to encode currently identified latent transcripts.

FIG. 1

(A to D) Photomicrographs showing detection of CMV DNA in experimentally infected GM-P culture by PCR-ISH. Cells from RC256-infected (A to C) or mock-infected (D) cultures were subjected to PCR-ISH, except that Taq DNA polymerase was omitted from the experiment in panel C. The solid arrow indicates a CMV DNA-positive cell, and the open arrow indicates a CMV DNA-negative cell. Magnification, ×1,280 (A) and ×3,200 (B to D). (E) DNA blot hybridization of total DNA extracted from CMV-infected (Inf, complete) or mock-infected (Mock, complete) GM-Ps after PCR amplification with IEP3A and IEP3B or with the omission of Taq DNA polymerase (Inf, −Taq and Mock, −Taq). The filter was probed with a ³²P-labelled probe derived from pON2810, washed, and exposed to X-ray film for 2 h. The predicted CMV-specific PCR product of 167 bp is indicated.

FIG. 2

PCR-ISH determination of the percentage of CMV DNA-positive cells from four separate GM-P cultures 2 to 3 weeks after infection with RC256 at a MOI of 3. Virus-infected (Inf) and mock-infected (Mock) cultures were analyzed in the presence of the full complement of reagents for CMV DNA detection (complete) or with the omission of Taq DNA polymerase (−Taq). Symbols: ⊡, experiment 1; , experiment 2; , experiment 3; ■, experiment 4.

FIG. 3

CMV DNA copy number determination by QC-PCR. Lysates of 1.3 × 10⁴ cells from an RC256-infected GM-P culture (MOI of 3, day 17 postinfection) were each analyzed in the presence of 3 × 10³ to 3 × 10⁵ copies of competitive template, a CMV ie1/ie2 cDNA plasmid pON2347. The copy numbers are indicated above the lanes. The positions of the 387-bp product from CMV genomic DNA and the 217-bp product from the cDNA competitive template are indicated by arrows. Cells from a mock-infected GM-P culture (Mock) or a sample without DNA (no DNA) were included as negative controls. The marker was a 100-bp ladder (Boehringer-Mannheim).

FIG. 4

(A to D) Photomicrographs showing detection of sense CLTs by RT-PCR-ISH in experimentally infected GM-P culture. Infected cells were subjected to RT-PCR-ISH (A and B), with controls omitting reverse transcriptase (C) or Taq DNA polymerase (D). Sense CLT-positive cells are arrowed. Magnification, ×1,280 (A) and ×3,200 (B to D). (E) RT-PCR-ISH determination of the percentage of sense CLT-positive cells from three separate GM-P cultures 2 to 3 weeks after infection with RC256 at a MOI of 3. Virus-infected (Inf) or mock-infected (Mock) cells were analyzed in the presence of the full complement of reagents for sense CLT detection (complete) or with the omission of either reverse transcriptase (−RTase) or Taq DNA polymerase (−Taq). ⊡, experiment 1; , experiment 2; ■, experiment 3.

FIG. 5

(A) Photomicrograph showing detection of CMV DNA in naturally infected mononuclear cells by PCR-ISH. A CMV DNA-positive cell is indicated by an arrow. Magnification, ×3,200. (B) CMV DNA copy number determination by QC-PCR in G-CSF-mobilized PB cells from a CMV-seropositive donor. Lysates of 5.4 × 10⁴ cells were each analyzed in the presence of 3, 10, or 30 copies of competitive template, a CMV ie1/ie2 cDNA plasmid, pON2347. The positions of the 387-bp product from CMV genomic DNA and the 217-bp product from the cDNA competitive template are indicated by arrowheads. A contamination control containing no DNA was included (lane c). (C) PCR-ISH determination of the percentage of CMV DNA-positive mononuclear cells from G-CSF-mobilized PB and BM autograft and allograft donors (12 donors were used altogether; see symbols below). Cells were analyzed in the presence of the full complement of reagents for CMV DNA detection (complete) or with the omission of Taq DNA polymerase (−Taq). The number of CMV genomes per CMV DNA-positive cell was determined by a combination of PCR-ISH and QC-PCR data and is shown as numbers above the bars in the graph. Symbols: , mobilized PB 1, autograft; , mobilized PB 2, autograft; , mobilized PB 3, autograft; , mobilized PB 4, autograft; ■, mobilized PB 5, autograft; ▧, mobilized PB 6, autograft; , bone marrow 1, autograft; , mobilized PB 7, allograft; ▨, mobilized PB 8, autograft; , mobilized PB 9, autograft; ▩, mobilized PB 10, autograft; , mobilized PB 11, allograft.