|Home | About | Journals | Submit | Contact Us | Français|
Patients infected with HIV-1 typically seroconvert within weeks of primary infection. In rare cases, patients do not develop antibodies against HIV-1 despite demonstrable infection. We describe an HLA-B*5802 positive individual who presented with AIDS despite repeatedly negative HIV-1 antibody screening tests. Phylogenetic analysis of env clones revealed little sequence diversity, and weak HIV-1 specific CD8+ T cell responses were present to Gag epitopes. The patient seroconverted after immune reconstitution on HAART. Lack of an antibody response to HIV-1 is rare and appears to be due to a defect in HIV-1-specific immunity rather than infection with attenuated virus.
Acute infection with HIV-1 is characterized by high viral loads and a rapid decrease in CD4+ T cells . A humoral response with antibodies that target multiple viral epitopes classically follows acute infection by three weeks , and is the mainstay of clinical diagnosis . In rare cases, patients do not develop humoral immunity to HIV-1 [4-6]. We describe the clinical, immunologic and virologic features of one such patient.
In January 2007, a 59-year-old man presented to Johns Hopkins Hospital with complaints of chest pain, shortness of breath and cough, all of which had worsened over a three week period. He reported a negative HIV test three months earlier. The patient was admitted to the hospital with a diagnosis of community-acquired pneumonia. A screening HIV-1 ELISA was negative. He was found to be pancytopenic, which prompted concern for malignancy or occult infection. Serum and urine protein electrophoresis testing was normal. A peripheral blood smear revealed a normochromic, normocytic anemia, and leukocyte flow cytometry was negative for leukemia or lymphoma markers. Quantitative plasma immunoglobulin levels revealed IgG and IgA levels that were slightly above the reference range; IgM levels were normal. Testing for human T-cell leukemia virus 1 and 2 (HTLV-1 and 2) was negative.
The patient remained persistently hypoxic despite antibiotic therapy and underwent bronchoscopy and broncho-alveolar lavage (BAL), which revealed Pneumocystis jiroveci. CD4+ T cell count was 90 cells/mm3, and an HIV-1 RNA was greater than 100,000 copies/mL. Although repeat HIV-1 ELISA and Western blot were again negative [Figure 1], the patient was diagnosed with HIV-1 infection and AIDS. An HIV-1 genotype revealed clade B pan-sensitive virus, with minor mutations in reverse transcriptase and protease coding regions. The patient denied illicit drug use but had engaged in unprotected sex with a commercial sex worker two months prior to hospital admission. He improved clinically, and was started on combination antiretroviral therapy with tenofovir / emtricitabine / efavirenz. He has experienced a steady rise of his CD4+ T cell count to greater than 500 cells/mm3, and his viral load has fallen to 57 copies/mL. Four months after the initiation of ART, HIV serology was repeated and was strongly positive [Figure 1].
A portion of the data presented in the Results section was obtained as part of the diagnostic work up, including the patient's history, initial HIV-1 antibody results, HIV-1 genotypic resistance profiling and clade determination, serum immunoglobulin assays including IgG studies to common antigens, CD4+ T cell counts and viral loads. The patient's initial negative western blot assay was confirmed with a commercial kit. HIPAA confidentiality regulations were followed in compiling this information from the patient's medical records.
After informed consent was obtained, the patient donated blood for laboratory evaluation on two separate occasions. Phlebotomy was performed in accordance with a protocol approved by the Johns Hopkins Institutional Review Board. Plasma and peripheral blood mononuclear cells (PBMCs) were obtained via Ficoll gradient centrifugation. The patient's PBMCs were tested in an interferon-γ ELISPOT using antibodies obtained from Mabtech. Duplicate wells containing105 PBMC were stimulated with overlapping peptides representing the entire Gag proteome obtained from the AIDS Research and Reference Reagent Program. All plates were evaluated with an automated ELISPOT reader system (Carl Zeiss MicroImaging, Inc.) with KS4.8 software by an independent scientist in a blinded fashion (Zellnet Consulting). A positive response was defined as more than a mean of 5 positive cells per well. Negative controls routinely had fewer than a mean of 1.5 positive cells/well. HLA genotyping was performed by evaluation of the patient's genomic DNA obtained from PBMC .
Replication-competent virus was cultured from the patient's resting CD4+ T cells, and gag and pol sequencing was performed . The C2-V4 region from env was amplified, cloned, and sequenced from plasma from the first time point, and phylogenetic analysis was carried out on these sequences as previously described . Sequences have been submitted to Genbank (accession numbers GQ465542-GQ465590). Virus from patient plasma was grown in MT-2 cells [obtained from the AIDS Research and Reference Reagent Program] and in healthy donor CD4+ lymphoblasts for five days, and p24 quantitation (Perkin Elmer) was performed according to manufacturer's instructions at day 0, 3 and 5 to determine viral tropism.
We performed quantitative immunoglobulin testing in order to rule out a global defect in antibody production. Our patient had slightly elevated levels of IgG and IgA and normal levels of IgM. In addition, IgG antibodies to CMV, hepatitis A, HSV-1, and HSV-2 were present. He had no evidence of other known chronic viral infection, including HCV, HIV-2, HTLV-1 or 2. In order to estimate how long the patient had been infected, we performed phylogenetic analysis of plasma env C2-V4 sequences. The analysis of 52 independent clones revealed minimal sequence diversity (Figure 2A). Genetic homogeneity among HIV-1 quasispecies is highly suggestive of recent infection, as the virus has not had time to diversify . However, we cannot rule out the possibility that this observed genetic homogeneity is a reflection of the absence of selective pressure mediated by neutralizing antibodies. We measured viral co-receptor tropism by a viral culture assay in which the patient's virus, isolated from plasma, was grown in both primary CD4+ lymphoblast and MT-2 cell line cultures. CD4+ lymphoblasts express CCR5 and CXCR4, and MT-2 cells express CXCR4 only. The patient's virus grew in CD4+ lymphoblasts but not in MT-2 cells as measured by p24 quantification (Perkin-Elmer). The cell cultures were also infected with two viral strains with known co-receptor tropism, IIIb (X4) and Ba-L (R5), to serve as controls. Ba-L virus grew in CD4+ lymphoblast cultures, while IIIb virus grew in MT-2 cells. The patient's virus is CCR5-tropic by this assay [Figure 2B].
ELISPOT assays evaluating interferon-γ responses to Gag epitopes were performed at two different time points to assess the HIV-1-specific cellular immune response of this patient; the first during the patient's seronegative period and the second after immune reconstitution and seroconversion. At both early and late time points, he had weak responses to two Gag epitopes. In response to the EKAFSPEVIPMFSAL epitope (Gag 161-175), 50 SFC / 106 cells were detected at the early time point and 95 SFC / 106 cells at the late time point. In response to the SPEVIPMFSALSEGA epitope (Gag 165-179), 50 SFC / 106 cells were detected at the early time point and 75 SFC / 106 cells at the late time point. While the first Gag epitope contains the HLA-B*57/5801 restricted epitope KF11 (162-172), no response was seen to this optimal epitope (data not shown). His cellular immune response to HIV-1 Gag appears to have been present throughout infection, but did not expand with immune reconstitution despite the emergence of a humoral response to HIV-1 antigens [Figure 1].
To further characterize the patient's immunologic response to infection, we isolated genomic DNA obtained from his PBMCs and carried out HLA genotyping, which identified the following alleles and allele groups: HLA-A*02, HLA-A*26, HLA-B*35 and HLA-B*5802. HLA-B*5802 has been associated with rapid progression . The presence of this allele, along with his clinical history of a recent high-risk sexual exposure and the evidence of minimal quasispecies diversity by phylogenetic analysis, suggests that our patient experienced an accelerated disease course.
The patient had clinical and laboratory evidence of AIDS despite HIV-1 antibody tests that remained persistently negative until after he had achieved immune reconstitution with highly active antiretroviral therapy (HAART). False negative HIV-1 screening tests are rare and are usually attributed to testing within several weeks of primary infection, before a detectable humoral response develops . Little evidence exists to support the phenomenon of “seroreversion,” in which patients with HIV-1 infection lose their antibody response over time . The patient described here, who presented with symptomatic AIDS and was persistently seronegative over a five month period, does not fall into either of these categories, and was diagnosed with seronegative HIV-1 infection. Making this diagnosis often involves a combination of repeatedly negative serum HIV-1 ELISA and Western blot tests without prior positive testing, clinical evidence of AIDS or HIV-related symptoms, and evidence of HIV infection based on a positive p24 antigen, positive HIV-1 DNA or RNA polymerase chain reaction or HIV-1 viral culture. Previously described cases of seronegative HIV-1 infection have not documented spontaneous seroconversion; patients who were not treated for HIV-1 with antiretroviral therapy died or were lost to follow up while several of those treated with HAART, including our patient, developed HIV-1 antibodies after immune reconstitution [4-6].
In sequencing gag, pol and env genes from this patient's virus, we were unable to identify any large deletions that could have explained the lack of an HIV-specific antibody response. Both clinical evidence and viral culture assays suggested that he was infected with a replication competent virus. This is in keeping with prior descriptions of seronegative HIV-1 infection [4-6], in which a host immune defect appears to be responsible for seronegativity rather than infection with an attenuated or mutated virus.
The role of antibodies in the control of untreated HIV-1 infection is not well understood. Several lines of evidence suggest that neutralizing antibodies do not have a major role in controlling viral replication. In elite suppressors, HIV-1 infected patients with stable CD4 counts who maintain viral loads below 50 copies/mL without antiretroviral therapy, a recent report demonstrated minimal neutralizing antibody titers to autologous virus despite their lack of disease progression. Conversely, untreated patients with progressive disease were found to have the highest antibody titers to autologous virus . Given a lack of evidence that HIV-1 antibodies can slow progression of disease, it seems unlikely that seronegativity alone is responsible for what appears to be the rapid disease progression seen in the patient described here.
In contrast to neutralizing antibodies, cytotoxic T cells (CTLs) appear to be central to host defense against HIV-1 infection. This is evident in SIV models in which CD8+ cell depletion leads to increased viral loads that are suppressed with recovery of these cells . It is intriguing that one of the HLA alleles present in this patient, HLA-B*5802, is associated with rapid progression. This allele differs by only three amino acids from HLA-B*5801, an allele associated with low viral loads and good disease prognosis .
This patient did have weak but reproducible T cell response to two Gag epitopes while he was seronegative, something that has not been demonstrated previously in these patients. The only other case report to describe HIV-1- specific T cell immunity in seronegative patients found no T cell responses to the entire HIV-1 proteome in two patients . Interestingly, once he achieved immune reconstitution, his antibody response expanded dramatically (Figure 1), but his T cell responses remained unchanged.
We cannot draw definitive conclusions regarding the pathophysiology of seronegative HIV-1 infection from a single case report; however some intriguing clues are present. It has been previously demonstrated that CTLs usually appear within weeks of primary infection, and are largely responsible for controlling both viremia and the tempo of disease progression. One explanation of the pathophysiology of seronegative HIV-1 infection is that transmitted virus, replicating without significant immune pressure due to an absent  or narrow CTL response such as the one seen in the patient described here, leads to overwhelming immune suppression and symptomatic AIDS. Thus the hallmark of seronegative HIV-1 infection, lack of detectable HIV-1 antibodies, may represent a marker for deficient HIV-specific cellular immunity in these patients, which in turn may be the underlying etiology of their unique and dramatic clinical course. This hypothesis needs to be tested in a larger number of patients in order to better understand the interaction between cellular and humoral immunity in the control of HIV-1 infection, which could have major implications for HIV-1 vaccine design.
The authors thak Jordyn Gamiel and Oliver Laeyendecker for performing the Western blot assay shown in Figure 1.
This study was supported by NIH grants R56 AI73185-01A1 and R01 AI080328 (J.N.B.) as well as the Howard Hughes Medical Institute (R.F.S.)
The authors report no conflict of interest.
The case report described here was presented at the 2009 Conference on Retroviruses and Opportunistic Infections in Montreal, Canada.