|Home | About | Journals | Submit | Contact Us | Français|
To define the ratio of simian immunodeficiency virus (SIV) RNA molecules to infectious virions in plasma, a ramp-up-stage plasma pool was made from the earliest viral RNA (vRNA)-positive plasma samples (collected approximately 7 days after inoculation) from seven macaques, and a set-point-stage plasma pool was made from plasma samples collected 10 to 16 weeks after peak viremia from seven macaques; vRNA levels in these plasma pools were determined, and serial 10-fold dilutions containing 1 to 1,500 vRNA copies/ml were made. Intravenous (i.v.) inoculation of a 1-ml aliquot of diluted ramp-up-stage plasma containing 20 vRNA copies infected 2 of 2 rhesus macaques, while for the set-point-stage plasma, i.v. inoculation with 1,500 vRNA copies was needed to transmit infection. Further, when the heat-inactivated set-point-stage plasma pool was mixed with ramp-up-stage virions, infection of inoculated macaques was blocked. Notably, 2 of 2 animals inoculated with 85 ml of a pre-ramp-up plasma pool containing <3 SIV RNA copies/ml developed SIV infections characterized by high levels of viral replication, demonstrating that “vRNA-negative” plasma collected from macaques in the pre-ramp-up stage is infectious. Furthermore, there is a high ratio of infectious virions to total virions in ramp-up-stage plasma (between 1:1 and 1:10) and a lower ratio in set-point-stage plasma (between 1:75 and 1:750). Heat-inactivated chronic-stage plasma can “neutralize” the highly infectious ramp-up-stage virions. These findings have implications for the understanding of the natural history of SIV and human immunodeficiency virus infection and transmission.
Understanding the virology of the earliest stages of human immunodeficiency virus (HIV) infection is critical to the development of interventions to prevent mucosal or parenteral HIV exposure from becoming disseminated infection of mucosal and systemic lymphoid tissues. The first 1 to 2 weeks of typical HIV infections are characterized by an “eclipse” stage following viral infection but prior to the development of detectable systemic viremia (13). During this period, HIV replication is initially localized but then progresses to active replication in local lymphoid tissues, with dissemination, including seeding of gut-associated lymphoid tissue (21, 30). This eclipse period is followed by a 2- to 4-week “ramp-up” period of uncontrolled viral replication in all lymphoid tissues, particularly at mucosal sites, which results in high plasma viral RNA (vRNA) levels and significant acute depletion of CCR5+ CD4+ T cells (13, 21, 30, 44). This ramp-up stage of viremia, which is often associated with a clinically apparent retroviral syndrome, is followed over succeeding weeks by the development of humoral and cellular virus-specific immune responses, which, along with acute depletion of target cell populations, is believed to contribute to a typically observed decline in plasma virus levels from peak values and to the establishment of a quasi-equilibrium between virus and host that is associated with a stable “set point” level of plasma viremia. Signs and symptoms of late-stage HIV disease eventually develop, along with rising plasma vRNA levels and clinically significant immunodeficiency (28, 29). This pattern of infection is also typical of experimental infection of rhesus macaques with pathogenic simian immunodeficiency virus (SIV) (16), making this system an excellent model for HIV transmission and pathogenesis studies.
Recognition of the viremic window period that precedes seroconversion in acute HIV infection led to the replacement of serological screening of blood donations by nucleic acid testing (NAT) to prevent transfusion-associated transmission of HIV. Prior to the use of NAT, HIV was transmitted via seronegative window-stage donations at moderate rates, depending on the incidence of HIV infection in donor populations and the sensitivity of the serological assay employed (4, 8, 40). For practical and economic reasons, NAT is often performed using minipools (MP), consisting of pooled specimens derived from different donors, with follow-up testing of individual donors from positive MP. Even after the implementation of NAT, HIV has been transmitted by transfusions of blood obtained from donations that were determined to have viral loads of fewer than 100 to 500 vRNA copies/ml, the sensitivity limit of MP-NAT assays (9). These MP-NAT breakthrough transmission cases have led to recommendations to move to individual-donation NAT (sensitivity limits, 5 to 30 vRNA copies/ml) and/or to implement pathogen reduction and/or inactivation procedures to eliminate the infectivity of low-level viremic units missed by NAT and serological screening.
The observation of HIV transmission by donations with very low plasma vRNA levels suggests that plasma virions in the pre-ramp-up stage of HIV infection may be particularly infectious, especially compared to set-point-stage plasma virions, for which larger inocula, as measured by viral RNA copy numbers, appear to be required for transmission by blood transfusion (6, 37). Understanding the relative infectiousness of plasma virions at different stages of HIV infection/exposure would provide more confidence in assessing the safety of blood donations but would also yield significant insights into potentially critical biological differences between transmitted viruses and the viral variants that develop during chronic infection. The significance of previous efforts to characterize the infectivity of HIV during the preseroconversion stage of infection based on in vitro (tissue culture) or animal (chimpanzee) transmission experiments has been difficult to interpret due to the questionable relevance of these model systems (34, 38). We used an SIV/rhesus macaque model of HIV infection to address these issues effectively.
The rhesus macaques (Macaca mulatta) used in these studies were housed at the California National Primate Research Center in accordance with the regulations of the Association for Assessment and Accreditation of Laboratory Animal Care. The experiments were approved by the Institutional Animal Use and Care Committee of the University of California, Davis. All animals were negative for antibodies to HIV type 2, SIV, type D retrovirus, and simian T-cell lymphotropic virus type 1 at the time the study was initiated. When necessary, animals were anesthetized with ketamine hydrochloride (10 mg/kg; Parke-Davis, Morris Plains, NJ) or 0.7 mg/kg tiletamine HCl and zolazepam (Telazol; Fort Dodge Animal Health, Fort Dodge, IA) injected intramuscularly.
A cell-free stock of SIVmac251 (UCD-2/02) produced by short-term expansion of a previous virus stock (SIVmac251 UCD-2/00) in Staphylococcus endotoxin A (SEA)-stimulated rhesus monkey peripheral blood mononuclear cells (PBMC) was used for these studies (25). This SIVmac251 stock contains approximately 109 vRNA copies/ml and 105 50% tissue culture infection doses (TCID50)/ml when titered on CEMx174 cells. For vaginal inoculation, the stock was diluted 100-fold to produce an inoculum containing 103 TCID50/ml, and 1 ml was introduced atraumatically into the vaginal canal using a needleless, 1-ml tuberculin syringe. The animals were inoculated twice in one day, with a 4-h interval between inoculations. This SIV inoculation regimen was performed weekly for 13 weeks or until experimental necropsy after a predetermined number of vaginal SIV inoculations. The viral status of the animals was not determined until after the inoculation series was complete. Blood samples were collected twice weekly, just prior to each inoculation and 4 days after inoculation, using published methods (23).
PBMC were isolated from heparinized blood using lymphocyte separation medium (ICN Biomedicals, Aurora, OH). PBMC samples were frozen in 10% dimethyl sulfoxide (Sigma, St. Louis, MO)-90% fetal bovine serum (Gemini BioProducts, Calabasas, CA) and stored in liquid nitrogen until analysis by immunological and virological assays (32).
The TCID50 of ramp-up- and set-point-stage plasma pools were determined in CEMx174 cell cultures that were maintained for 1 week without medium change, and then aliquots of media were assayed weekly for the presence of SIV major core protein (p27) by an antigen capture enzyme-linked immunosorbent assay (ELISA) (26).
Because no virus could be isolated from the ramp-up-stage plasma pool on CEMx174 cells, a CCR5-negative cell line, titration of this plasma pool on primary rhesus PBMC was also attempted. PBMC were isolated from blood of SIV-naïve rhesus macaques as described above. The cells were stimulated with SEA (0.5 μg/ml) (Toxin Technologies, Sarasota, FL) and cultured in complete RPMI 1640 containing recombinant human interleukin-2 (50 U/ml; Chiron Inc., Emeryville, CA) until the cell numbers had doubled (7 to 10 days). To titrate the virus in plasma pools, the activated PBMC were resuspended in RPMI at a concentration of 2 × 106/ml, and 100 μl of the cell suspension was added to wells of a 96-well plate. One-hundred-microliter volumes of serial 10-fold dilutions of the plasma pools (10−1 to 10−5) were added directly to the wells in quadruplicate.
The individual plasma samples from the donor animals that were combined to produce the six plasma pools were analyzed for vRNA by a quantitative branched-DNA assay (7). The levels of virion-associated RNA in these samples are reported as vRNA copy numbers per milliliter of plasma. The detection limit of this assay is 125 vRNA copies/ml of plasma.
Virion-associated SIV RNA levels in plasma samples from the recipient animals inoculated with the plasma pools and in the donor plasma pools used for inoculation were measured by use of a real-time reverse transcription-PCR (RT-PCR) assay based on detection of a highly conserved sequence in Gag, as described in detail elsewhere; results are reported as vRNA copies per milliliter of plasma (7). The per-reaction threshold quantification limit (95% assurance) of this assay is 30 copies of viral RNA per reaction, as determined from repeated testing of a dilution series of an RNA standard, with due consideration for the Poisson distribution in sampling. The relevant in vivo threshold quantification limit is then a function of the volume of plasma processed for RNA extraction and the number of aliquots into which the final RNA preparation is divided for replicate testing—in this instance, three. An ultrasensitive modification of this assay, based on the processing of 5 ml of plasma per sample, was used to determine the levels of virion-associated SIV RNA in the plasma pools. The threshold quantification limit for this modified assay format was established at 3 copies of vRNA per ml of plasma (5 copies/reaction, using one-third of 5 ml, or 1.67 ml of plasma equivalent per replicate reaction).
Blood samples were collected at frequent intervals, and the percentages of CD3+ CD4+ T cells and CD3+ CD8+ T cells within the lymphocyte population were determined by flow cytometric analysis using a FACSCalibur instrument (Becton Dickinson Immunocytometry Systems, Milpitas, CA) and rhesus macaque-reactive antibodies against CD3 (clone SP34), CD4 (clone M-T477), and CD8 (clone SK1) (all from Pharmingen/Becton Dickinson, San Diego, CA).
SIV-specific immunoglobulin G (IgG) binding antibody titers in plasma were measured using ELISA plates coated with detergent-disrupted SIVmac251 as previously described (23). Prior to the determination of antibody titers, plasma samples were screened for the presence of anti-SIV antibodies using a 1:100 dilution of plasma with the same ELISA protocol used to determine antibody titers, described below. The results of the screening assay were calculated using the ratio of the change in optical density (ΔOD) to the cutoff value (CO), where ΔOD is defined as the difference between the mean OD of a dilution of sample tested in two antigen-coated wells and the mean OD of the same dilution of sample tested in two antigen-free (control) wells. The CO is the mean ΔOD plus 3 standard deviations of duplicate wells containing plasma from 12 randomly selected seronegative adult female rhesus macaques. If the ΔOD/CO ratio for a sample was greater than 2, the sample was considered to be positive.
An additional analysis was used to determine anti-SIV antibody titers in antibody-positive plasma. Ninety-six-well microtiter plates (Nunc Immunoplate II Maxisorp; Applied Scientific, South San Francisco, CA) were coated with whole pelleted SIVmac251 (Advanced Biologics Inc., Columbia, MD) at 5 μg/ml in 0.1 M Na2CO3/NaHCO3 buffer (pH 9.6) and were blocked with 4% nonfat powdered milk. Plasma samples were serially diluted (1:4) in duplicate, and the plates were incubated overnight at 4°C. The initial dilution of serum was 1:10,000 for the SIV-specific IgG assay. Antibody binding was detected using a 1:2,000 dilution of peroxidase-conjugated goat anti-monkey IgG(Fc) (100 μl per well) (Nordic Laboratories, San Juan Capistrano, CA) for 1 h at 37°C. Plates were developed with o-phenylenediamine dihydrochloride (Sigma Chemical Co., St. Louis, MO) for 5 min and stopped with H2SO4 before the OD at 490 nm was read. For each plasma sample, the end point titer of anti-SIV antibodies was defined as the reciprocal of the last dilution giving a ΔOD greater than 0.2, where ΔOD is defined as the difference between the mean ODs of two antigen-coated and two antigen-free (control) wells.
Neutralization was measured as the reduction in luciferase reporter gene expression after multiple rounds of virus replication in 5.25.EGFP.Luc.M7 cells (3). This cell line is a genetically engineered clone of CEMx174 that expresses multiple SIV and HIV type 1 entry receptors (CD4, CCR5, CXCR4, GPR15/Bob) (33). The cells also possess Tat-responsive reporter genes for luciferase (Luc) and green fluorescent protein. Cells were maintained in growth medium (RPMI 1640, 12% heat-inactivated fetal bovine serum, 50 μg gentamicin/ml) containing puromycin (0.5 μg/ml), G418 (300 μg/ml), and hygromycin (200 μg/ml) to preserve the CCR5 and reporter gene plasmids. For the neutralization assay, 5,000 TCID50 of virus was incubated with multiple dilutions of the test sample in triplicate for 1 h at 37°C in a total volume of 150 μl in 96-well flat-bottom culture plates. A 100-μl suspension of cells (5 × 105 cells/ml of growth medium containing 25 μg DEAE dextran/ml but lacking puromycin, G418, and hygromycin) was added to each well. One set of control wells received cells plus virus (virus control), while another set received cells only (background control). Plates were incubated until approximately 10% of the cells in virus control wells were positive for green fluorescent protein expression by fluorescence microscopy (approximately 3 days). At this time, 100 μl of the cell suspension was transferred to a 96-well white solid plate (Costar) for measurement of luminescence using the Britelite luminescence reporter gene assay system (Perkin-Elmer Life Sciences). Luminescence versus sample dilution curves were plotted, and the neutralization titer is the interpolated dilution at which luminescence, expressed in relative luminescence units (RLU), was reduced by 50% from that in virus control wells after subtraction of background RLU. Cell-free stocks of TCLA-SIVmac251 and SIVmac251-UCD were generated in H9 cells and in rhesus macaque PBMC, respectively.
Kaplan-Meier survival analysis was performed to compare the infection rates in macaques after inoculation with serial 10-fold dilutions of the ramp-up-stage and set-point-stage plasma pools. GraphPad Prism, version 4.a for Apple OSX10.4 (GraphPad Software, San Diego, CA), and Macintosh G5 computers (Apple Inc., Cupertino CA) were used for all analyses.
The rhesus macaques that contributed individual plasma samples to the plasma pools were exposed to SIVmac251 during a series of weekly low-dose vaginal SIVmac251 inoculations as described elsewhere (24). The details of the pre-ramp-up-stage plasma pools produced for these studies are summarized in Table Table11 and Fig. Fig.1.1. Pool A was made by combining 16 individual vRNA-negative (vRNA−) (<125 copies/ml) plasma samples from two donor macaques that had no detectable SIV RNA or DNA in tissues on necropsy (Fig. (Fig.1).1). Plasma pool B was made by combining 21 individual vRNA− (<125 copies/ml) plasma samples collected from three donor macaques that similarly lacked detectable SIV vRNA in plasma but had low levels of SIV RNA or DNA in tissues at necropsy (Fig. (Fig.1).1). Pool C was made by combining 60 individual vRNA− (<125 copies/ml), pre-ramp-up-stage plasma samples from six donor macaques that eventually became viremic after additional inoculations (Fig. (Fig.1).1). Samples included in pool C were collected at least 7 days prior to the collection of the first plasma vRNA-positive (vRNA+) samples (Fig. (Fig.1).1). After the individual samples were mixed to make the pools, all three of these plasma pools had <3 vRNA copies/ml by RT-PCR. Finally, the plasma sample from monkey 26513 with the transient “blip” (731 vRNA copies/ml) during the vaginal inoculation series was designated “pool D.” All the animals contributing to the pre-ramp-up plasma pools were negative for anti-SIV antibodies in plasma at the time the samples were collected.
To determine if infectious virus was present in the pre-ramp-up-stage plasma of macaques exposed vaginally to SIVmac251, almost the entire volume of each pool (Table (Table1)1) was used in a series of plasma transfer experiments. None of the rhesus macaques inoculated intravenously (i.v.) with plasma pool A (n = 2) or B (n = 2) became infected, as judged by the lack of detectable vRNA in plasma, vDNA in PBMC, anti-SIV antibody responses in plasma, or anti-SIV cellular immune responses in PBMC over the 10- to 12-week follow-up period (Fig. (Fig.22 and data not shown). The single animal inoculated with 2 ml of “pool D,” which consisted of the single plasma sample with the blip of vRNA (731 copies/ml), also failed to show any evidence of infection (Fig. (Fig.2).2). In contrast, both animals inoculated with pool C became plasma vRNA+ 1 week after inoculation (Fig. (Fig.2).2). In fact, the plasma vRNA levels at 7 days postinoculation (p.i.) exceeded 106 copies/ml in these animals, peaked at even higher levels at day 14 p.i., and remained well above 106 copies/ml in both animals through 13 weeks p.i. Both of these animals developed strong anti-SIV antibody responses and SIV-specific cellular immune responses (data not shown), but these responses were unable to control viral replication. Thus, infectious virus with a very high in vivo replication capacity was present in the pre-ramp-up-stage plasma samples used to produce plasma pool C, although the pool tested vRNA negative by the RT-PCR assay (<3 copies/ml) and each contributing sample also tested vRNA negative by the branched-DNA assay (<125 copies/ml). This result provides strong evidence that infectious virus is present during the pre-ramp-up stage of SIV infection that follows mucosal SIV inoculation and can result in productive viral dissemination.
The rhesus macaques that served as plasma donors for these studies had been exposed to SIVmac251 by vaginal inoculation and developed disseminated infections characterized by acute high-titer viremia, seroconversion, and progression to stable viral set points (Fig. (Fig.3).3). The ramp-up-stage plasma pool was made by combining the first vRNA+ plasma samples collected from seven inoculated macaques; these samples were collected about 1 week before peak plasma vRNA levels were achieved (Fig. (Fig.3).3). The ramp-up-stage plasma pool contained 1.2 × 105 vRNA copies/ml; however, attempts to isolate SIV using CEMx174 cells and SEA-stimulated rhesus macaque PBMC were unsuccessful (data not shown). The set-point-stage plasma pool was made by combining SIV vRNA+ plasma samples collected from seven vaginally inoculated macaques between 10 and 16 weeks after an individual's peak plasma vRNA level was detected. The set-point-stage plasma contained 1.5 × 108 vRNA copies/ml and 103 TCID50 using CEMx174 cells as the indicator cells (data not shown). Details of the ramp-up- and set-point-stage plasma pools are summarized in Table Table22.
To determine the number of SIV RNA copies in the ramp-up-stage plasma pool that is needed to establish SIV infection in naïve monkeys, a serial titration/i.v. inoculation approach was used, challenging animals with inocula containing increasing amounts of vRNA until infection was established. Initially a 100-μl aliquot of a ramp-up-stage plasma pool was diluted to 2,000 copies of vRNA/ml in 5.9 ml of 0.9% saline for injection, and then 10-fold serial dilutions of this aliquot were used to produce the inocula with 2 and 20 copies of vRNA. One SIV-naïve rhesus monkey was inoculated i.v. with 1 ml of saline containing 2 vRNA copies (Fig. (Fig.3),3), but this animal (animal 32970) did not develop a typical disseminated systemic infection. Although vRNA (100 copies/ml) was detected in this animal's plasma at 9 days p.i., all other plasma samples were negative, and vDNA was not detected in the PBMC. Further, no anti-SIV antibody response developed during the 8-week p.i. waiting period (Fig. (Fig.33 and data not shown). Two additional animals were inoculated with 1 ml of saline containing 20 copies of vRNA from the ramp-up-stage plasma pool; both (animals 33815 and 35036) became systemically infected and developed typical SIV infections. In both of these animals, plasma vRNA levels peaked at approximately 2 weeks p.i. and gradually declined to set point levels above 105 vRNA copies/ml plasma (Fig. (Fig.3).3). vDNA was present in PBMC from week 1 p.i. until necropsy, and anti-SIV antibodies were readily detectable in plasma by 4 weeks p.i. (data not shown). It is worth noting that donor macaque 29271 contributed 75% of the total vRNA, corresponding to 15 of the 20 vRNA copies, in the infectious ramp-up-stage pool inoculum.
To determine the number of SIV RNA copies from the set-point-stage plasma pool needed to establish SIV infection in naïve macaques, an aliquot of the set-point-stage plasma pool was diluted to approximately 1.5 copies of vRNA/ml of saline and inoculated i.v. into two naïve rhesus macaques (Fig. (Fig.4).4). The production of this inoculum involved making eight serial 10-fold dilutions to produce a 10-ml aliquot with 1.5 vRNA copies/ml. Neither of the animals (animals 33952 and 34846) inoculated with 1.5 vRNA copies of the set-point-stage plasma pool became infected: there was no detectable plasma vRNA, PBMC vDNA, or anti-SIV antibody response for 10 weeks p.i. At 10 weeks p.i., 1 ml of a dilution of the set-point-stage plasma pool containing 15 copies of vRNA/ml saline was inoculated i.v. into the same two rhesus macaques. Neither of the animals inoculated with 15 vRNA copies of the set-point-stage plasma pool showed any evidence of infection during a 10-week observation period (Fig. (Fig.4).4). At 20 weeks p.i. (10 weeks after the second round of inoculations), 1 ml of a dilution of the set-point-stage plasma pool containing 150 copies of vRNA/ml saline was inoculated i.v. into the same two rhesus macaques and an additional SIV-naïve monkey. None of the three animals (animals 33952, 34846, and 34373) inoculated with 150 vRNA copies of the set-point-stage plasma pool had evidence of infection during a 10-week observation period (Fig. (Fig.4).4). At 30 weeks p.i. (10 weeks after the third round of inoculations), 1 ml of a dilution of the set-point-stage plasma pool containing 1,500 copies of vRNA/ml saline was inoculated i.v. into the same three rhesus macaques. All three animals (animals 33952, 34846, and 34373) developed typical systemic SIV infections; two of these animals had a rapid onset of viremia with a peak at about 14 days p.i., while the third animal had delayed viremia, with the peak at about 28 days p.i. (Fig. (Fig.4).4). While two of the set-point-stage plasma pool recipient animals had lower set point plasma vRNA levels, the plasma vRNA levels remained high in one animal (animal 34373). Macaques 25479 and 29459 contributed 135 and 1,350 vRNA copies, respectively, to the infectious inoculum with 1,500 copies of vRNA.
Kaplan-Meier survival analysis was performed to compare the numbers of SIV RNA copies in the ramp-up-stage plasma pool and the set-point-stage plasma pool that were needed to infect naïve recipients by i.v. inoculation. As shown in Fig. Fig.4C,4C, the number of vRNA copies in the ramp-up-stage plasma pool needed to establish infection after i.v. inoculation of a naïve host was significantly lower (P < 0.0027) than that for the set-point-stage plasma pool.
To evaluate whether plasma virions obtained at the set point stage were intrinsically less infectious than those from the pre-ramp-up and ramp-up stages of infection, or whether factors in the set-point-stage plasma might interfere with infectivity, an aliquot of the ramp-up-stage plasma pool was diluted to 20 vRNA copies/0.5 ml of saline, and 0.5 ml of the heat-inactivated set-point-stage plasma pool was added to make a final concentration of 20 vRNA copies/ml. One milliliter of this mixture was inoculated i.v. into two naïve rhesus macaques (Fig. (Fig.5;5; Table Table3).3). Neither of the animals inoculated with 20 vRNA copies of the ramp-up-stage plasma pool mixed with 0.5 ml of the heat-inactivated set-point-stage plasma pool developed typical systemic infections; only the day 2 p.i. plasma from each animal had <250 vRNA copies/ml, and all other plasma samples were negative. Further, no PBMC vDNA or anti-SIV antibodies were detected in these animals for the 10-week p.i. observation period (Fig. (Fig.55 and Table Table3;3; also data not shown). In contrast, one animal inoculated i.v. with an identical aliquot of the ramp-up-stage plasma pool diluted in 0.5 ml of heat-inactivated plasma from a SIV-naïve control monkey to contain 20 vRNA copies developed a typical pattern of viremia and infection. One animal inoculated i.v. with 0.5 ml of the heat-inactivated set-point-stage plasma pool remained uninfected (Fig. (Fig.5).5). The set-point-stage plasma pool had significant levels of antibodies capable of in vitro neutralization of the SIVmac251 challenge stock; a 1:2,337 dilution of the set-point-stage plasma pool reduced the RLU by 50% from that of the virus-only control in the neutralizing antibody assay.
Many of the critical events associated with HIV pathogenesis seem to occur in the first few days after infection (21, 30). The development of effective intervention strategies to prevent HIV transmission and AIDS requires a detailed understanding of this acute stage of infection. Among regular blood donors, periods of transient viremia in the pre-ramp-up stage of HIV infection have been reported (12). Intermittent detection of very low levels of vDNA in the PBMC of highly exposed seronegative subjects (so-called occult infections) has been reported (45). Similarly, occult systemic SIV infections with periods of transient viremia occur in some rhesus macaques mucosally inoculated with low doses of pathogenic SIV (24, 27, 31). In the present study, we did not observe SIV transmission following infusion of 2 ml of plasma from a macaque with “blip” viremia (731 vRNA copies/ml). We also failed to transmit SIV following infusions of 40 ml of plasma collected from three macaques that had been exposed to serial intravaginal SIV challenges and had evidence of low levels of SIV DNA and RNA in tissues but failed to develop systemic plasma viremia. Although these results suggest that many occult infections, evidenced by transient blips of vRNA in plasma or viral DNA in tissue, may not be a concern with respect to blood transfusion safety, definitive conclusions in this regard cannot be made due to the limited size and scope of the present study.
Our study clearly demonstrated that a pre-ramp-up-stage plasma pool testing negative for vRNA (<3 copies/ml) and composed of plasma samples collected from six animals at least 1 week prior to the presence of measurable plasma vRNA contains infectious virus that can be transmitted to naïve macaques by i.v. inoculation. The six animals from whom this infectious pre-ramp-up-stage plasma was collected developed productive systemic infections. Moreover, it appears that the pre-ramp-up-stage virus is well adapted to replicate in the SIV-naïve host. Each recipient received 85 ml of a plasma pool containing <3 vRNA copies/ml, or <255 vRNA copies total. Nevertheless, infection was efficiently and effectively established; the animals infected with the pre-ramp-up-stage plasma had very high levels of viral replication 1 week after inoculation, and plasma vRNA levels remained well above 106 copies/ml during the 14-week observation period. These data clearly demonstrate the infectious and pathogenic potential of pre-ramp-up-stage virus and underscore the point that depending on the volume of the inoculum, even samples that test below stringent vRNA copy-per-milliliter thresholds may still transmit infection.
The plasma transfer experiments using dilutions of pools of ramp-up- and set-point-stage plasma collected from macaques after vaginal SIV inoculation demonstrated that the number of infectious virions per vRNA copy is significantly lower in set-point-stage plasma than in ramp-up-stage plasma. In fact, since each virion has 2 RNA copies, the nominal particle infectivity ratio in the ramp-up-stage plasma pool was 1 to 9 infectious units/10 virions. In marked contrast, the particle infectivity ratio in the set-point-stage plasma pool was significantly lower, at 1 to 9 infectious units/750 virions. The relatively low (1:75 to 1:750) ratio of infectious virions to total virions in set-point-stage plasma could be consistent either with the generation of less fit mutant viruses in vivo, due to the nucleotide substitution errors introduced by the SIV reverse transcriptase and host polymerases, or with the presence of antibodies that coat and neutralize a large proportion of the virions in set-point-stage plasma, or both. The hypothesis that set-point-stage virions are coated with antibodies that interfere with infectivity is consistent with the observed inactivation of the ramp-up-stage virions after they were mixed with the heat-inactivated set-point-stage plasma pool. However, we are conducting additional evaluations to determine what component(s) of set-point-stage plasma confers this activity. In this context, it is worth noting that in an early report, HIV immune globulin failed to protect chimpanzees against experimental challenge with HIV (38) but then was clearly protective when used at a higher dose (39). HIV immune globulin obtained from HIV-infected chimpanzees can reduce the infectivity of HIV in rhesus macaques (17) as well as blocking simian/human immunodeficiency virus infection (41).
While it may be relatively easy to explain the presence of 1 to 2 log units of noninfectious virions in set-point-phase plasma, the paucity of noninfectious virions in ramp-up-stage plasma is more difficult to explain. It is possible that although mutant genomes arise by reverse transcriptase errors during the ramp-up stage, insufficient mutations accumulate in a very fit founder virus genome during the rounds of replication between infection and ramp-up viremia to have a distinguishable effect on the fitness of the mutant virions.
A less likely hypothesis is that efficient purifying selection working through substrate competition eliminates less prolific genomes before they produce virions that reach the plasma in the short time between infection and the ramp-up stage of infection.
Differential infectivity of a virus in plasma during the acute versus the chronic stage is not unique to SIV and presumably HIV; a similar phenomenon has been reported for hepatitis C virus (HCV) and hepatitis B virus (HBV) infections. The HCV strain H inoculum consists of serum collected 7 weeks posttransfusion from a patient in the acute stage of HCV infection, while the HCV strain F inoculum is derived from sera collected 1 year posttransfusion from a patient in the chronic stage of HCV infection (2, 10, 15). The ratio of vRNA copies to infectious units is approximately 1:1 for the acute-stage strain H inoculum and >1:103 for the chronic-stage strain F inoculum (2, 15). In fact, as few as 20 HCV RNA copies in acute-stage serum can transmit HCV infection to a naïve chimpanzee, while HCV transmission with plasma collected after seroconversion requires 1,000-fold higher levels of HCV (18). Similar findings of very high infectivity of ramp-up- versus set-point-stage plasma have been established for HBV, both in chimpanzees and in chimeric mice with humanized livers (20, 43); in the latter system, ramp-up-stage HBV serum is about 100 times more infectious than later-stage serum. As with SIV and HIV, the potential explanations for the relatively low infectivity of chronic-stage plasma in HCV infection include the presence of large numbers of defective virions or noninfectious antibody-virion immune complexes. The latter explanation is generally favored, since the results of one study suggest that a significant proportion of the HCV virions in the chronic-stage strain F inoculum exist as immune complexes (15).
In human blood banking, NAT is currently used to screen blood donations for HIV, HCV, and HBV in order to prevent transfusion of blood collected during the window period between the development of infectious viremia and seroconversion. This window period includes the pre-ramp-up and ramp-up stages of infection. NAT screening was implemented in the United States and other countries in 1999 using MP screening in which 16 or more donor specimens are pooled prior to testing (8, 42). Prior to the use of MP-NAT, HIV was transmitted via window-stage donations at significant rates (5, 22). Although very rarely, HIV has been transmitted by window period blood donations that were determined to have ≤150 vRNA copies/ml even after the adoption of MP-NAT (9, 11, 35, 36). The occurrence of rare HIV transmission events by donations with no evidence of anti-HIV antibodies and very low vRNA levels is consistent with our findings. Further, our study documents transmission of infection by pre-ramp-up-stage and diluted ramp-up-stage plasma with vRNA levels below even the limit of detection of individual-donation NAT, suggesting that even individual-donation NAT, which was recently implemented in the Republic of South Africa (14), may not be sensitive enough to interdict all HIV-infected donations. Thus, implementation of pathogen inactivation methods to sterilize blood transfusions (1) may be required to achieve the next level of safety.
Understanding the relative infectiousness of plasma virions at different stages of HIV infection/exposure not only provides important information for assessing the safety of blood donations and donor-screening policies but may also yield significant insights into critical biological differences between transmitted virus and the virus variants that emerge during infection of the host (19). Further studies aimed at understanding the viral phylogenetics in the plasma pool donors and recipients are under way.
The authors thank B. Hahn and G. Shaw for helpful discussions, the Primate Services Unit at the CNPRC, and Ding Lu, Lili Guo, and Kristen Bost for excellent technical assistance.
This work was supported by Public Health Service grants U51RR00169, from the National Center for Research Resources, and P01 AI066314, from the National Institute of Allergy and Infectious Diseases, by a gift from the James B. Pendleton Charitable Trust, and in part by federal funds from the National Cancer Institute, National Institutes of Health, under contracts NO1-CO-124000 and HHSN266200400088C.
Published ahead of print on 7 January 2009.