IMMUNOLOGY ASSAYS AND METHODS

INTRODUCTION

All AVEG vaccine trials include assessments of the immunological response to HIV induced by vaccination. These assessments include antibody binding assays such as Western Blot and ELISA, functional antibody assays such as lymphocyte fusion inhibition and virus neutralization, and assays of cellular immunity such as HIV-specific cytotoxic T-lymphocyte (CTL) activity and lymphocyte proliferation. These assays are performed on-site at the AVEUs, at the Central Immunology Laboratory (CIL) at Duke University, and at the Central Mucosal Immunology Laboratory (MIL) at the University of Alabama at Birmingham.

The section below is a description of assays and assay methods used for AVEG studies.

METHODS

AVEU (Local) Assays

Assays to screen for infection with HIV-1

Western Blot (WB) assays. Commercially available FDA-approved WB kits are used for screening volunteers. The Cambridge-Biotech Corporation kit distributed by Ortho Diagnostic Systems and the Organon-Teknika kit are used for the WB assay for immunogenicity evaluations in most AVEG studies. Each band is scored as negative (designated as 1), +/ if observed and less intense than that of the p24 band of the weakly reactive control strip (designated as 2), + if less reactive than the p24 band of the strongly reactive control strip (designated as 3), and ++ if the intensity is the intensity of the p24 band on the strongly reactive strip (designated as 4). Bands with +/ intensity or greater are termed reactive. The Western Blot is considered positive if there is (+) intensity of any two of the following three bands: p24, gp41, gp120/160. Band reactivity for any specimen that does not fulfill the criteria for being positive is an indeterminate result.

ELISA (EIA) assays. The Abbott whole virus lysate EIA kit was uniformly used at each AVEU for immunogenicity evaluations in early AVEG studies. The Abbott HIV 1,2 EIA kit has replaced the Abbott whole virus lysate EIA kit for routine diagnostic use. For both assays, the ratio of the specimen optical density versus the cut-off OD is calculated. Ratios of the experimental OD to the control OD 1 are considered positive. Specialized FDA or AVEG-approved ELISA kits may be performed on a protocol-specific basis to distinguish vaccine responses from HIV infection (see Exhibit 5).

Lymphocyte proliferation assays. The consensus protocol for lymphoproliferation assays is as follows (1-2). For each assay, volunteer peripheral blood mononuclear cells (PBMC) are stimulated with one or more HIV antigens corresponding to the vaccine product (antigen-specific response), Phytohemagglutinin (T cell mitogen response), Candida albicans, or Tetanus toxoid (recall antigen response). 100 µl of 1x10⁶ PBMC/ml in RPMI-1640 medium supplemented with L-glutamine, Penicillin/Streptomycin, Gentamicin and 10% human AB serum are distributed into wells of a 96-well microculture plate. A second aliquot of 100 µl of medium alone (control) or medium containing HIV antigen (1-25 µg/ml final concentration is the usual range), Phytohemagglutinin (Sigma Cat. #L-9132, 5 µg/ml final concentration), Candida antigen (Greer Laboratories, 20 µg/ml final concentration), or Tetanus toxoid (Connaught Laboratories) is then added. Triplicates or quadruplicates of the control and experimental cultures are routinely set up.

The plates are incubated at 37C in a humidified atmosphere containing 5% CO₂. Cultures containing PHA are incubated for 96 hours before harvesting and the cultures containing HIV antigen, Candida or Tetanus toxoid are incubated for 144 hours. One mci/well (50 µl) of a fresh stock of ³H-thymidine is added to the cultures 16-18 hours before harvesting. The geometric mean cpm of ³H-thymidine incorporation for each set of replicates is then ascertained. A delta cpm (i.e., the difference between the stimulated geometric mean and the medium control geometric mean), and a stimulation index (i.e., the ratio of stimulated to medium control values) are calculated. The PBMC of a normal, non-vaccinated individual are included in each assay as a control. For some protocols, additional mitogens or antigens are included. Exhibit 1 is a complete listing of the antigens and mitogens which have been utilized to date for lymphoproliferation assays.

Cytotoxic T lymphocyte (CTL) assays. CTL assays are performed at several of the AVEUs in support of the trials. For the detection of resting memory HIV-specific CTL activity induced by immunization with a candidate AIDS vaccine, an in vitro stimulation step is required prior to assaying for cytolytic activity. A number of different stimulation strategies have been developed and successfully employed at the AVEUs to increase the number of functionally mature volunteer CTL effector cells (3-10). These include: a) recombinant Epstein-Barr virus--HIV-gp160 transformed autologous B cells (3), b) whole inactivated HIV-1 (4), c) autologous blood-derived macrophages infected with HIV-1 (5), d) autologous lymphocytes or Epstein-Barr virus (EBV) transformed B cells (± transfection of CD4 receptor or enrichment for CD4+ lymphocytes) infected with recombinant poxvirus--HIV-1 vectors (± inactivation with psoralen and ultraviolet light) (6-10), and e) HIV-1 peptide pulsed autologous EBV-transformed B cells. Many of the stimulation strategies are also employed as targeting strategies. A typical standard 4-6 hours ⁵¹Cr-release assay is used to measure lytic activity.

Skin testing for DTH reactivity. Delayed Type Hypersensitivity (DTH) induced by immunization with candidate AIDS vaccine is assessed by skin testing. Skin testing in AVEG protocols uses a small amount of protein or peptide corresponding to the HIV-1 envelope, gag, etc., gene products present in the candidate vaccine as the HIV-specific skin test antigen. All HIV-specific skin testing is done with HIV-1 proteins covered by an IND specific for their use as DTH reagents. If a volunteer has been induced to make a DTH response by immunization, a localized inflammation will be activated at the site of injection of the HIV skin test reagent 48-72 hours after administration. A skin test is considered positive if the diameter of the inflamed area measures >5 mm. Measurements are taken of the greatest diameter and the perpendicular diameter. Three commonly used non-HIV skin test reagents are employed as positive controls: Tetanus, Candida, and mumps. Most people have been exposed to or vaccinated with these and will have a positive reaction to these recall control antigens. Saline or another suitable product, such as Mock prep, are employed as negative controls.

HIV-1 in vitro resistance assay. Fresh or cryopreserved PBMC (106/mL) are stimulated with 5 ng/mL a-CD3 monoclonal antibody OKT3 (Ortho Diagnostics, Schenectady, NY) and 2 U/mL interleukin (IL)-2 (Boehringer Mannheim, Indianapolis). Lymphoblast stimulation within 2-4 days is confirmed by microscopic examination. Cells are washed, concentrated, and split into aliquots of 5 X 106 PBMC. PBMC from HIV-positive patients are challenged with HIV-1MN or HIV-1BaL in 200mL for 1 h. Depending on the number of available cells, doses range from 3.5 to 20 PBMC TCID50. One group is left unchallenged (but otherwise treated identically) to evaluate endogenous viral growth. PBMC from uninfected volunteers are challenged with similar doses of HIV-1MN and HIV-1BaL. PBMC are then washed free of unbound virus and plated at 5 X 106/well in 2 mL that is half complete medium plus IL-2 and half conditioned medium (complete medium from the previous 48 h of incubation). Supernatants are harvested for batched EIA of viral p24 (Organon Teknika, Durham, NC) and cultures refed with complete medium plus IL-2 on days 3, 7, 10, and 14. Cultures producing endogenous HIV are not readily evaluated for susceptibility to exogenous challenge, as this would require quantitative genotyping of the virus strains produced in culture.

CENTRAL IMMUNOLOGY LABORATORY ASSAYS

Antigen/peptide ELISA assays. Custom ELISA assays tailored to a particular protocol are performed with the antigens/peptides listed in Exhibit 3.

Serological tests for binding antibodies are performed using a number of different antigens, including various synthetic peptides and recombinant proteins. Optimal plates and plating conditions are determined for each antigen. Serum samples are always tested in duplicate and two negative and two HIV+ control sera are included in each assay. Depending on the protocol, serum samples are tested at either a single solution (1/50 for synthetic peptide and 1/100 for recombinant protein antigens) or eight serial two-fold dilutions beginning at 1/50. The positive control sera are titrated in the same way, regardless of whether the test samples are assayed at a single fixed dilution or are titrated. For each ELISA assay, duplicate antigen-containing and non-antigen-containing wells are set up for each serum. After incubation of the serum and antigen, the wells are washed and anti-human Ig antibody conjugated to alkaline phosphatase is added. The wells are again washed and the substrate p-nitro-phenyl phosphate is added to each well. After a short incubation to allow color development, optical density readings are made of each well and an average OD reading for each pair of replicates is calculated. A Delta score (i.e., OD antigen - OD non-antigen) is regarded as positive if it exceeds 0.2. For the fixed dilution samples, the optical densities in duplicate wells are averaged. The average and range values are transmitted to the Data Coordinating and Analysis Center (DCAC), The Statistical Center for HIV/AIDS Research and Prevention (SCHARP). The serum titration optical density values are averaged at each serum dilution and the average is plotted as a function of serum dilution. The values reported to the DCAC are the reciprocal serum dilutions which yield 50% maximum binding.

Anti-HIV-1 ADCC assays. The antibody-dependent cellular cytotoxicity (ADCC) assay (11) is a serologic assessment of the presence and relative titer of serum antibodies capable of directing the lysis of HIV-1 expressing targets by normal donor PBMC. The percent-specific lysis is measured relative to the release of ⁵¹Cr from labeled target cells in comparison to maximum and spontaneous isotope release values. The standard assay employs 5x10³ labeled targets, 1.65x10⁵ normal donor PBMC, and serial 10-fold dilutions of test serum ranging from 1:20 to 1:20,000 (final dilution). Each serum dilution is tested in triplicate wells. The incubation period is 6 hours. The percent-specific lysis is calculated according to the formula: [(experimental release-spontaneous release)/(maximum release-spontaneous release)]x100. The routine assay utilizes the CEM-NKR cell line, which has been immunoselected for resistance to spontaneous killing by NK cells in the donor PBMC. Each assay includes a determination of background lysis of HIV-1 expressing CEM-NKR cells incubated with normal donor PBMC in absence of any serum. One serum specimen obtained from an HIV seropositive subject and one serum specimen obtained from an HIV seronegative donor are also used in each assay as positive and negative controls, respectively.

Anti-HIV-1 ADCC-directing activity is measured on CEM-NKR target cells which are either infected with HIV-1 (usually the HTLV-IIIB or MN strains) or saturated with purified recombinant gp120 passively adsorbed onto cells through the high affinity gp120/CD4 interaction (usually 5 µg of recombinant gp120 are used to coat 1x10⁶ target cells for 2 hours at 37C). The recombinant gp120 IIIB and MN are supplied by VaxGen and the rpg120 SF-2 is supplied by Chiron Vaccines. The HIV-1 specific ADCC value is obtained by subtracting the background lysis (percent-specific lysis against HIV-1 expressing CEM-NKR targets in the absence of serum) from the percent-specific lysis against HIV expressing CEM-NKR targets incubated with the test sera. A value of 15% or greater is considered positive for ADCC-directing antibody reactivity.

Cell fusion inhibition assays. The cell fusion inhibition assay infected cell partners are produced in the CD4+ cell line, CEM. Cells are infected with a viral isolate, cultured for approximately 10-14 days, during which time the culture reaches maximum CPE and the majority of the cells are killed. The stable surviving fraction of cells are infected and identified as "Grow throughs," and are used as the "infected cells" in the assay. The uninfected partner may be either CEM or MOLT 4, the latter giving slightly larger syncytia produced on a shorter time scale.

To assess fusion inhibition, 5,000-10,000 chronically infected CEM cells are mixed with 70,000 uninfected MOLT 4 cells (or CEM cells) in the presence of a 1/10 dilution of test and negative and positive control sera in a 96-well, ¸ area plate (12). About 24 hours later, the wells are visualized at low magnification (40x) for presence of multinucleated syncytia. In the fusion screen assay, only a 1/10 serum concentration is tested in duplicate. If strong inhibition (0-15 syncytia) is noted for a particular serum in the screen, it is considered positive and retested at multiple dilutions to determine its titer. Titer is given as the reciprocal of the serum dilution required to inhibit syncytia number relative to control wells by 90%. A titer of 10 is considered positive.

Cell fusion inhibition-peptide competition assays. Fusion inhibition-peptide competition assays are performed by executing additional experiments where increasing concentrations of a peptide are added to the assay wells. In a given assay, if <5 syncytia are observed when no peptide is added, the fusion inhibition assay response is said to be depleted at the first concentration where 3⁺ (i.e., 26-35) or more syncytia are observed.

Neutralization assays. Three neutralization assays have been utilized in the project.

The multiple viral dose neutralization assay is performed in Dr. Matthews' laboratory with both lab-adapted and primary HIV isolates (13-15). Each serum dilution is tested against six concentrations of virus inocula in duplicate that are adjusted by 4- or 5-fold dilutions to include 0.1-100 TCID₅₀ of virus. After 30 minutes of incubation together at room temperature, the virus/antibody mix is then added to AA5 cells and the cells are cultured for 6-10 days (depending on virus isolate under study) to test for virus proliferation. After that period of time, 10 µl of supernatant is tested for reverse transcriptase (RT) activity. The TCID₅₀ for each control and test sera are determined by the Reed and Munch formula, and the neutralizing titer is defined as the reciprocal of the serum dilution resulting in a one-log reduction in TCID₅₀. Titers of 10 are considered positive. The interpolation of titer is performed on the log scale.

Neutralization of primary HIV isolates is accomplished in a similar experimental design as for neutralization of the prototypic lab-adapted isolates on cell lines. That is, the endpoint for both assays is a log reduction in infectious titer of the virus in question as estimated by TCID₅₀ measurements. To accomplish this, multiple virus dilutions covering infectious titers of 100 to 0.05 TCID₅₀ are treated with test sera. After appropriate incubation and culture to allow virus expansion, virus infection is monitored by the RT assay. A serum is scored as positive for neutralization if the estimated TCID₅₀ in the test serum is at least one log lower than that in treatment groups containing normal human sera at the same dilutions.

The difference between primary isolate versus prototypic isolate assays is in the target cells used and time of the assay. The prototypic assay uses cell lines (usually AA5), whereas the primary isolate studies utilize PBMC activated for 4 days with a combination of anti-CD3 (OKT3, 0.05 µg/ml) and anti-CD28 (0.1 µg/ml) antibodies. The source of PBMC is from ten healthy HIV-uninfected human donors (buffy coats from Interstate Blood Bank). The cells are banded on LSM, washed, and frozen at 50x10⁶/ml in fetal calf serum and 10% DMSO. The cells are thawed and activated as indicated for 4 days before the infection assay is started. Experiments are set up in the same format as for prototypic isolates in a 96-well microtiter U-bottom plate. Cells are at 2x10⁶/ml in 5% IL2 containing media and a final volume of 100 µl. At 5-7 days post infection, 80 µl of fresh media is added to each well and incubation is continued until Day 10-12, at which time the experiment is terminated. Supernatants are then tested for RT activity and the data is treated in the same manner as for the prototypic/cell line experiments.

A single virus dose assay, similar to the multiple virus dose neutralization assay, except that only one concentration of HIV was employed, was used in some of the earlier protocols; e.g., Protocol 002. The virus inocula for the single viral dose assay (15), performed in Dr. Langlois' laboratory, is fixed at about 10 infectious units TCID₁₀₀ for each serum dilution. That amount of virus is then incubated with serial dilutions of test sera in duplicate for 30 minutes at room temperature. The neutralizing titer of the sera is interpolated as the reciprocal of that serum dilution that inhibits RT activity (in terms of cpm incorporated/10 µl supernatant assayed) by 90%. Titers of 20 are not considered to be significant.

The vital dye neutralization assay is performed in Dr. Montefiori's laboratory. It is a single virus dose assay fixed at about 100 TCID₅₀ for each serum dilution made in 96-well plates (16-18). Virus is incubated with serial dilutions of test sera in triplicate at 37C for 30 minutes after which time MT-2 or CEMx174 cells are added. Infection of these cells leads to massive syncytium formation and virus-induced cell killing. Viable cells are quantitated by vital dye (neutral red) uptake after about 3 days of culture. Neutralization is seen as protection from virus-induced cell killing, where percent protection is defined as the difference in absorption (A₅₄₀) between test wells (cells + sample + virus) and virus control wells (cells + virus) divided by the difference in absorption between cell control wells (cells only) and virus control wells. Plates are harvested when cytopathic effect in virus control wells 70% but 100%. Neutralizing titers are defined as the reciprocal of the serum dilution to produce 50% protection.

Primary isolate neutralization assays in PBMC are performed in Dr. Montefiori's laboratory (18, 28-29). A fixed dose of cell-free virus (500 TCID₅₀) is incubated with an equal volume of each serum dilution (50 µl total) for 1 hour at 37 C in 96-well U-bottom plates. PHA-PBMC (overnight stimulation) are added to each well (4 x 10⁵ cells in 150 µl of IL-2 growth medium) and incubated for 24 hr at 37 C. The PBMC are then washed 3x with 200 l of growth medium to remove the virus inoculum and antibodies. Washed cells are resuspended in fresh IL-2 growth medium and incubated for approximately 6 days. Culture supernatants (25 l) are collected on a daily basis and mixed with 225 l of 0.5% Triton X-100 for later p24 quantification; the volume of IL-2 growth medium removed each day is replaced with fresh medium. Viral p24 is quantified in the DuPont antigen ELISA. Viral p24 is first measured in supernatants from virus control wells (no serum sample) for each harvest day. These values are used to construct the virus replication curve. Viral p24 is measured in all remaining supernatants for the harvest day that corresponds in time to when virus replication in control wells was in a linear phase of increase, usually exceeding 5 ng p24/ml. This assures optimum sensitivity in the assay (28). Neutralization is determined as a function of a reduction in p24 synthesis relative to corresponding prebleed serum samples or other serum samples from healthy seronegative individuals. Neutralization titers are reported as the serum dilution where p24 synthesis was reduced 80%. Neutralization potency is reported as the % reduction in p24 synthesis at the lowest serum dilution tested (usually 1:2 dilution).

The cMAGI/MAGI neutralization assays for primary and laboratory Isolates score for reduction of infectious virus titer employing the indicator cell lines MAGI or the CCR5 expressing derivative cMAGI. The MAGI (Multinuclear Activation of a Galactosidase Indicator) cell line was derived from parental HeLa cells by introducing genes for CD4 and an HIV-1 LTR-driven -gal reporter with an amphotropic retrovirus vector (30). The cMAGI cell line was derived from the MAGI cell line by introduction of the CCR5 gene using the amphotropic retroviral vector, PA317 (31). The cMAGI cells support replication of primary NSI (R5) isolates and laboratory adapted X4 viruses, while the MAGI cells support replication of only X4 viruses. Both cell lines exploit the ability of HIV-1 tat to transactivate the expression of a -galactosidase reporter gene driven by the HIV-LTR. The -gal reporter has been modified to localize in the nucleus and can be detected with the X-gal substrate as intense nuclear staining within a few days of infection. The number of stained nuclei can thus be interpreted as equal to the number of infectious virions in the challenge inoculum if there is only one round of infection prior to staining. In practice, we add an inhibitor of infection and cell-cell fusion (the DP178 peptide, (32) 24 hours post-infection in order to permit a readout that confidently represents a single round of infection. Infected cells are enumerated using a CCD-imager and both primary and laboratory adapted isolates show a linear relationship between virus input and the number of infected cells visualized by the imager. Precision of the infected cell counts is excellent in this format and reproducibly yields estimates of virus titers with a standard deviation of 15%. In the MAGI and cMAGI assays a 50% reduction in infectious titer (Vn/Vo = 0.5) is significant and provides the primary cutoff value for assessing virus neutralization. A secondary cutoff of Vn/Vo = 0.1, corresponding to a 90% reduction in infectious titer is also assessed.

Dr. Thomas Matthews and members of his laboratory developed the MAGI and cMAGI-based neutralization assays at the CIL. Over the past two years, Dr. Matthews' laboratory has used and tested this assay format extensively with excellent results. The cMAGI and MAGI assays are currently performed at the CIL in Dr. Greenberg's laboratory. Each serum dilution is tested in triplicate against a single virus inocula adjusted to yield approximately 1200 infected cells/well. After 30-60 minutes of incubation together at room temperature, the virus/antibody mix is added to the cMAGI or MAGI cells and 24 hours later, the DP178 peptide is added to inhibit multiple rounds of infection and cell-cell virus spread. The cells are cultured for 2 more days, fixed and stained with the X-gal substrate to detect infected cells. The number of infected cells for each control and test sera are determined with the CCD-imager and the neutralizing titer is defined as the reciprocal of the serum dilution resulting in a 50% reduction in infectious virus titer. Titers of 10 are considered positive.

C'-ADE. Complement-mediated antibody-dependent enhancement (C'-ADE) of HIV-1 infection is measured in MT-2 cells at a single virus dose of about 10 TCID₅₀ (16,19). The virus is incubated with serial dilutions of test sera and a 1:40 dilution of fresh normal human serum as a source of complement in duplicate at 37C for 30 minutes. MT-2 cells are added; incubation is continued for 3 hours. Virus and antibodies are then removed by a series of washes. The washed cells are incubated in fresh growth medium for 3 days, after which time 25 µl of cell suspension is assayed for p24 content. An increase in p24 production that is more than two times greater than background (no test serum) is considered positive for infection-enhancement. C'-ADE activity is given as titer, peak and power values, where titer is the reciprocal of the last serum dilution to show enhancement, peak is the reciprocal serum dilution producing the greatest enhancement, and power is the magnitude of enhancement at the peak dilution, which is given as the fold-increase in p24 production. Control sera from healthy, HIV-1-negative individuals never produces more than a 2-fold increase in p24 production at any dilution tested. Alternatively, viable cells may be quantitated by neutral red uptake at the end of the three-day incubation. C'-ADE titers are defined as the reciprocal serum dilution to produce a 15% reduction in viable cells.

gp120/CD4 blocking [titration of antisera for blockade of gp120 binding to CD4 positive cells or soluble CD4]. Data from two different gp120/CD4 blocking assays are available. The first assay is essentially the same as described in Skinner et al (20). Briefly, purified gp120 (baculovirus recombinant IIIB gp120 from Repligen) is radiolabeled with I-125 using Bolton and Hunter reagent. About 3x10⁵ cpm of this solution is incubated for 60 minutes with 1.2x10⁶ CD4 positive CEM cells and serial dilutions of serum in a total volume of 120 µl. The cells are washed by centrifugation through a mixture of 60% dibutyl phthalate and 40% bis(2-ethylhexyl) phthalate. The amount of gp120 bound to cells is estimated by counting the cell pellet in a gamma counter. Each point is measured in duplicate and OKT4A antibody is used as a positive control for complete blockade of gp120 binding to CD4. The 50% inhibitory titer is defined as the inverse of the serum dilution which reduces the bound counts per minute to one-half the counts per minute bound in the presence of normal serum.

A second assay, which has supplanted the one above in recent studies, is performed using reagents and methodology provided by Michael Peterson formerly of Genentech. Briefly, the assay measures levels of antibody in test sera which block binding of soluble CD4 to rgp120 coated onto ELISA 96-well plates. Binding of sCD4 to gp120 is then monitored using a horseradish peroxidase-conjugated monoclonal antibody to CD4. The estimation of blocking titer is accomplished by testing serial dilutions of sera and interpolating the results to identify a serum dilution which inhibits CD4 binding to gp120 by 50%. Positive and negative control sera are included in each assay to establish background and uninhibited levels of binding. The gp120/CD4 blocking assays are considered positive if titers >20.

Cytotoxic T-lymphocyte (CTL) assays. Determinations of vaccine-induced CTL responses are made in association with most AVEG protocols, with specific emphasis given to those studies involving live vectors and/or novel adjuvants. The overall strategy for measurement of MHC-restricted CTL activities differs slightly for infected versus non-infected vaccinees. For protocols involving HIV-1-infected individuals, fresh PBMC are utilized as effector cells for direct CTL assays. These include an effector:target cell titration, as well as a limited dilution analysis of CTLp on a subset of individuals.

For CTL determinations on non-infected vaccinees, no direct CTL assays are performed, but rather a standard in vitro stimulation (IVS) strategy is employed. Briefly, this first involves acute infection of a portion of the vaccine PBMC sample with either a single recombinant vaccinia/HIV vector or pool of vectors (depending on the specific vaccine protocol) that most closely approximates the vaccine immunogen. After 90 minutes, the infected PBMC are washed and utilized as stimulators in co-cultures with the remaining uninfected PBMC at a responder:stimulator (R:S) ratio of 4:1 and a total cell density of 1x10⁶ cells/ml in RPMI 1640 medium containing 5% pooled human AB serum, 50 µg/ml gentamicin sulfate, and 330 U/ml of IL-7. IL-2 (20 U/ml) is added to the cultures on Day 7 and CTL activity is assessed on Day 14 (±2 days).

The actual CTL assays are performed in an identical fashion for both fresh PBMC and in vitro stimulated effectors. Effector cell populations are subjected to individual phenotypic depletions with magnetic microspheres conjugated to anti-mouse IgG (control), and anti-CD4 and anti-CD8 monoclonal antibodies. Following individual depletions, effectors are titrated against a standard number of ⁵¹Cr labeled, autologous, EBV-transformed B-lymphocyte cell lines (BLCL) infected with either control vaccinia (vSC8, vCP1170 or NYCBH) or vaccinia/HIV-1 constructs (listed in Exhibit 2). Generally, two effector:target (E.T.) ratios are utilized (either 100:1 and 50:1 or 50:1 and 25:1) against all targets. Unlabeled autologous BLCL infected with control vaccinia vector are added to each well as "cold target competitors" to yield a cold:hot target ratio of 40:1. These competitors reduce the anti-vaccinia background generated as a byproduct of the in vitro stimulation. For protocols involving defined peptide immunogens, peptides (5-20 µg/ml) are substituted for vaccinia vector-infected PBMC in the stimulation phase and peptide-coated, autologous, ⁵¹Cr labeled BLCL are utilized as targets. The recombinant poxvirus vectors, as well as the synthetic peptides utilized in the assessment of HIV-specific CTL reactivities elicited in conjunction with Phase I/II protocols, are shown in Exhibit 2.

A BLCL target line mismatched at both MHC Class I or II and infected with identical poxvirus control or HIV constructs is included as an MHC restriction control whenever possible.

HIV-1-specific CTL reactivity is defined by: 1) a significant level of specific lysis of autologous HIV-1 antigen expressing targets compared to controls, 2) depletion of lytic activity by removal of CD4+ and/or CD8+ effector lymphocytes, and 3) absence of specific lytic activity against HLA-mismatched targets.

Bulk CTL is currently defined to be positive if the HIV-1 antigen expressing targets relative to control have a specific lysis >10%. CD8 + CTL occurs when lytic activity decreases by 50% after removal of CD8 cells, while removal of CD4 cells maintains specific lysis of at least 5%. An analogous rule is applied for CD4 + CTL.

For some analyses, data are summarized for the 50:1 and 100:1 E:T ratios. The CTL assay is negative if the 100:1 E:T ratio results do not support the findings of elevated cytolytic activity at the 50:1 E:T ratio. Percent-specific lysis results at 50:1 E:T are used when available, unless 100:1 E:T indicates cytolytic activity and the 50:1 E:T results are either negative or not performed.

ELISA Screening Algorithm

In older AVEG studies, the evaluation of vaccinee sera for functional antibody activity (e.g., fusion inhibition, ADCC, virus neutralization, gp120/CD4 blocking) was limited to samples that had already been shown to have anti-HIV activity by the antigen ELISA assay discussed above. Under this scheme, fusion inhibition, neutralization, and gp120/CD4 blocking assays were only performed on samples which were positive on the basis of the ELISA screening algorithm. In this case, a special computation of percent responders and other quantities is used. The algorithm and these computations are described in a technical report. In study reports and summary tables, the only computations presented which are based on ELISA screen are those for protocols that were activated prior to Protocol 006.

MUCOSAL IMMUNOLOGY LABORATORY ASSAYS

Enzyme-Linked Immunosorbent Assays (ELISAs)

ELISA assays are performed with the antigens/peptides listed in Exhibit 3.

Antigen-specific class and subclass ELISAs. ELISA plates are coated with relevant antigen. After blocking and washing, 2-fold dilutions of test and control samples are added to triplicate wells and incubated. An appropriate dilution of biotin-conjugated affinity-purified goat anti-human IgM, IgG, IgA, or biotin-conjugated mouse monoclonal anti-human IgG1, IgG2, IgG3, or IgG4 antibody is added. The polyclonal anti-human IgM, IgG, and IgA reagents have been affinity-purified and cross-adsorbed to specificity on columns of human myeloma proteins. Horseradish peroxidase-conjugated avidin is added and color is developed with enzyme substrate. Due to the extremely wide range of immunoglobulin concentrations in external secretions (up to 100-fold differences), expression of data in O.D. units is not meaningful. Therefore, a standard curve is constructed using anti-IgA coated plates which are subsequently incubated with consecutive dilutions of a solution containing known amounts of secretory IgA. The optical densities of unknown samples are plotted against the standard curve of optical density versus µg/ml of diluted secretory IgA. Data acquisition, the generation of standard curves and interpolation of unknowns, is performed by computer using DeltaSoft software [BioMetallics Inc., Princeton, NJ (21)]. The assay endpoint is the mass of antigen-specific binding protein per volume of sample (ng/ml).

ELISAs for levels of total immunoglobulins. Assays to determine the total levels of immunoglobulins of all previously stated isotypes and subclasses are performed essentially as described above for the antigen-specific assays, except that the ELISA plates are coated with polyclonal anti-human immunoglobulin class-specific antibodies or monoclonal anti-human IgG or -IgA subclass-specific antibodies. Detection of the immunoglobulins trapped by the coating antibodies are performed with biotinylated antibodies as above--detection of the trapped IgG or IgA subclasses uses biotinylated polyclonal anti-IgG or anti-IgA specifically adsorbed against mouse Ig. Standards for these assays consist of a pool of normal serum with known immunoglobulin concentrations (Immunoglobulin Reference Standard, Binding Site, Inc.) or polyclonal secretory IgA purified from colostrum or myeloma proteins previously purified and characterized in the MIL. The standard curves are constructed using the optical densities of solutions containing known amounts of immunoglobulins; the optical density of the unknown sample is read against the standard curves.

ELISA for levels of human serum albumin (HSA) in external secretions. ELISA plates are coated with polyclonal anti-HSA (Sigma), blocked with PBS/Tween 20 and incubated with solutions containing a known amount of HSA (Sigma) (for the construction of a standard curve) or appropriately diluted samples of external secretions. Trapped HSA is detected by biotinylated anti-HSA. The results are expressed in µg/ml of HSA using methodology described above.

Development of Western Blot (WB) Strips Containing HIV Antigens

Western Blot is performed for the detection of total (not isotype-specific) and IgG antibodies to HIV antigens. In order to increase sensitivity so that we are able to detect low titers of IgA antibodies in diluted samples of external secretions, a chemiluminescence kit as described by Mohamed et al (22) is used. Positive bands are scored subjectively on a scale of 0-3+.

Enumeration of Ig-Secreting Cells of Particular Isotype or Subclass

The enumeration of IgM-, IgA-, and IgG-secreting cells of a given subclass is performed by the ELISPOT assay (23-24). Nitrocellulose-bottomed 96-well plates [Milliter, Millipore Corp., Bedford, CA] are coated with relevant HIV antigens or anti-Ig. Suspensions of cells are incubated on the plates to allow secreted Ig or antibodies to bind locally to the coating. Samples are tested in duplicate using 2-4 different cell concentrations. After washing out the cells, the plates are developed with enzyme-labeled antibodies of desired isotype specificity, followed by the corresponding chromogenic substrate. The spots indicating Ig or specific antibodies are counted under a low-power dissection microscope and the results expressed in relation to the number of plated cells. For enumeration of cells secreting IgA and IgG subclasses, monoclonal reagents are used by biotin- or enzyme-labeled anti-mouse IgG. The number of cells secreting HIV-antigen-specific antibodies of a certain isotype are then related to the total number of cells secreting that specific isotype.

Enumeration of Cytokine-Secreting Cells

The above described ELISPOT technique is also employed for enumeration of cells secreting cytokines (e.g., IFN- and IL-6). For this purpose, mouse monoclonal antibodies against IFN- [Chemicon, Tenecula, CA] and IL-6 [Collaborative Medical Products, Bedford, MA] will be used for coating the nitrocellulose plates. At the termination of the cell incubation period (18-20 hours for optimal release of the assayed cytokines), the wells will be developed with polyclonal, enzyme- or biotin-labeled reagent rabbit IgG anti-IFN- [Endogen, Boston, MA] or goat IgG anti-IL-6 [Collaborative Medical Products, Bedford, MA] and developed as described above.

To demonstrate that the spots obtained are due to the de novo synthesis of Ig or cytokines, cycloheximide treatment is applied as follows (23). Isolated cells are cultured for 4 hours at 37C with or without 50 µg/ml cycloheximide, then washed and plated on nitrocellulose wells. The cells previously treated with cycloheximide are incubated again with the same amount of cycloheximide during plating. A considerable reduction (85%) in the number of SFC is observed if Ig are actively synthesized, but a similar number of spots is observed if the cells only ingest and release exogenous Ig or cytokine.

Exhibit 4 provides a listing of specimen types for mucosal assays.

Site-Specific Assays Performed in Support of Some AVEG Trials

1. Flow cytometric indirect immunofluorescence assay (FIFA) to detect antibody to native HIV-1 envelope glycoprotein of laboratory-adapted HIV-1 strains. This assay is performed in Dr. Geoffrey Gorse's laboratory at the Saint Louis University AVEU.

The immunofluorescence assay (FIFA) is performed using target cells (H9 cells) which are a single cell clone derived from HUT78 cells, human cutaneous T-cell lymphoma cell line. The target cells are either uninfected or chronically infected with HIV-1 and washed and incubated in test serum. Labeling of uninfected target cells with envelope glycoprotein is done by washing and incubation with recombinant envelope glycoprotein. After incubation with the envelope glycoprotein, the target cells are washed and resuspended in RPMI 1640 medium. Human test serum at a dilution of 1:90 or higher is added. After incubation with serum at 4C, the cells are washed, FITC conjugated F(ab)₂ goat anti-human immunoglobulin G is added with bovine serum albumin. After washing, the cells are fixed overnight in 1-2% (vol/vol) formaldehyde. After a final wash in PBS, the logarithm of the green-fluorescence intensity is analyzed on a FACScan flow cytometer. The data are collected first in a forward-scatter versus light-scatter cytogram which displays size and the degree of granularity. This is gated to eliminate particles which are not of sufficient size to be intact cells. Cells passing through this gate are then displayed on a histogram of the logarithm of the green-fluorescence intensity versus relative cell number. For each assay of HIV-1 infected or envelope glycoprotein-coated H9 cells, cells stained with control serum from uninfected, unvaccinated individuals are used to set an appropriate boundary with which to determine the fraction of cells stained with vaccinee serum that were shifted into the positive fluorescence intensity region. The fractions of cells with intensity shift above the boundary (termed percent positive fluorescence) is then used as a measure of antibody content. Raw data derived from he flow cytometric histograms are used to determine the net percent positive fluorescence for each serum specimen and HIV-1 envelope glycoprotein-expressing target cell population, which is defined by the following equation: net percent positive fluorescence = [(percent positive fluorescence with test serum used against HIV-1 antigen-expressing target cells) - (percent positive fluorescence of the test serum used against control target cells)] - [(percent positive fluorescence with the negative control serum used against HIV-1 antigen-expressing target cells) - (percent positive fluorescence with the negative control serum used against control target cells)]. The specimen is considered positive if net percent positive fluorescence is >4.0% (25-27).

2. Antibody to native primary HIV-1 isolate envelope glycoproteins by flow cytometric indirect immunofluorescence assay (FIFA). This assay is performed in Dr. Geoffrey Gorse's laboratory at the Saint Louis University AVEU.

Primary HIV-1 isolates are grown in phytohemagglutinin-stimulated peripheral blood mononuclear cells (PBL) obtained from normal, healthy, unvaccinated donors and the viral titer of each pool is determined in PBL using the Reed and Muench method. Target cells in the assay are PBL from normal, healthy, unvaccinated, uninfected adults. PBL are stimulated for 3-4 days with phytohemagglutinin, washed and depleted of CD8+ T-cells by anti-CD8 monoclonal antibody immunomagnetic bead depletion. PBL are infected with a primary HIV-1 isolate by exposure of the cell button to virus in the presence of polybrene for 21/2 hours and then washed. The PBL are then incubated in medium containing IL-2 for 5-7 days and washed. The control PBL from the same donor are stimulated with phytohemagglutinin and left uninfected 3-4 days in advance of the day of staining. Cells on the day of staining are depleted of B-cells by anti-CD19 monoclonal antibody immunomagnetic bead depletion. On the day of staining 10⁶ infected and uninfected PBL are aliquoted into separate tubes, the cell buttons are incubated at 4C with a 1:90 final dilution (or higher) of test serum for 30 minutes and washed. PBL are stained with goat anti-human immunoglobulin (ab)₂ conjugated with FITC, and mouse anti-CD3 conjugated with phycoerythrin (PE) and FBS. Cells are washed, fixed in 1% formalin overnight and washed and re-suspended in PBS. Cells are then gated based on size and light scatter for degree of granularity using a FACscan flow cytometer. Boundaries are set based on staining of uninfected and infected cells with positive and negative control sera. The proportion of cell staining positively by both orange (PE) and green (FITC) log fluorescence is determined. The percent positive green fluorescence of each serum is calculated by subtracting the proportion of gated, uninfected cells in quadrant 2 from the proportion of infected cells in the same staining regions. This quadrant 2 region represents cells that have positive orange and green fluorescence staining with PE and FITC. The net percent positive fluorescence is the result of subtracting the percent positive fluorescence of the negative control serum from that of the test serum. The net percent positive fluorescence is considered positive for a serum specimen if it is >4.0% (25-27).

3. Vaccinia neutralizing antibodies. This assay is performed in Dr. Barney Graham's and Dr. Paul Spearman's laboratories at the University of Vanderbilt AVEU.

REFERENCES

1. Keefer MC, Bonnez W, Roberts NJ Jr, Dolin R, Reichman RC. Human immunodeficiency virus (HIV-1) gp160-specific lymphocyte proliferative responses of mononuclear leukocytes from HIV-1 recombinant gp160 vaccine recipients. J Infect Dis. 1991;163:448-453.

2. Gorse GJ, Belshe RB, Newman FK, Frey SE, NIAID AIDS Vaccine Clinical Trials Network. Lymphocyte proliferative responses following immunization with human immunodeficiency virus recombinant gp160. Vaccine. 1992;10:383-388.

3. Hammond SA, Bollinger RC, Stanhope PE, Quinn TC, Schwartz D, Clements ML, Siliciano RF. Comparative clonal analysis of human immunodeficiency virus type 1 (HIV-1)-specific CD4+ and CD8+ cytolytic T lymphocytes isolated from seronegative humans immunized with candidate HIV-1 vaccines. J Exp Med. 1992;176:1531-1542.

4. El-Daher N, Keefer MC, Reichman RC, Dolin R, Roberts NJ Jr. Persisting human immunodeficiency virus type 1 gp160-specific human T lymphocyte responses including CD8+ cytotoxic activity after receipt of envelope vaccines. J Infect Dis. 1993;168:306-313.

5. McElrath MJ, Hoffman M, Klucking S, Corey L, Greenberg PD. HIV-infected macrophages as efficient stimulator cells for detection of cytotoxic T cell responses to HIV in seronegative and seropositive vaccine recipients. AIDS Res Hum Retroviruses. 1994;10:541-549.

6. McElrath MJ, Rabin M, Hoffman M, Klucking S, Garcia JV, Greenberg PD. Evaluation of human immunodeficiency virus type 1 (HIV-1)-specific cytotoxic T-lymphocyte responses utilizing B-lymphoblastoid cell lines transduced with the CD4 gene and infected with HIV-1. J Virol. 1994;68:5074-5083.

7. Lubaki MN, Egan MA, Siliciano RF, Weinhold KJ, Bollinger RC. A novel method for detection and ex vivo expansion of HIV type 1-specific cytolytic T lymphocytes. AIDS Res Hum Retroviruses. 1994;10:1427-1431.

8. McElrath MJ, Corey L, Greenberg PD. Cytotoxic T cells in adults and children. Evaluation of cytotoxic T cell responses to candidate HIV-1 vaccines in HIV-1-uninfected individuals. AIDS Res Hum Retroviruses. 1994;10:S69-S72.

9. Egan MA, Pavlat WA, Tartaglia J, Paoletti E, Weinhold KJ, Clements ML, Siliciano RF. Induction of human immunodeficiency virus type 1 (HIV-1)-specific cytolytic T lymphocyte responses in seronegative adults by a nonreplicating, host-range-restricted canarypox vector (ALVAC) carrying the HIV-1_MN env gene. J Infect Dis. 1995;171:1623-1627.

10. Gorse GJ, Patel GB, Newman FK, Mandava M, Belshe RB (NIAID AIDS Vaccine Clinical Trials Network). Recombinant gp160 vaccination schedule and MHC HLA type as factors influencing cellular responses to HIV-1 envelope glycoprotein. Vaccine. 1995;13:1170-1179.

11. Weinhold KJ. Nonrestricted forms of anti-HIV-1 cytotoxicity. In: Aldovini, Walker, eds. Techniques in HIV Research. Stockton Press; 1990:187-199.

12. Putney, et al. HTLV/LAV-neutralizing antibodies to an E. coli-produced fragment of the virus envelope. Science. 1986;234:1392-1395.

13. Goff, Stephen, et al. Isolation and properties of moleney murine leukemia virus mutants: use of a rapid assay for release of virion reverse transcriptase. J Virol. 1981;38:239-248.

14. Willey, Ronald, et al. In vitro mutagenesis identifies a region within the envelope gene of the human immunodeficiency virus that is critical for infecticity. J Virol. 1988;62:139-147.

15. Langlois AJ, Matthews TJ, Weinhold KJ, et al. Detection of HIV-1 neutralizing antibodies by a simple, rapid, calorimetric assay. AIDS Res Hum Retroviruses. 1988;4:63-69.

16. Montefiori DC, Graham BS, Kliks S, Wright PF, NIAID AIDS Vaccine Clinical Trials Network. Serum antibodies to HIV-1 in recombinant vaccinia virus recipients boosted with purified recombinant gp160. J Clin Immunol. 1992;12:429-439.

17. Montefiori DC, Robinson WE Jr, Schuffman SS, Mitchell WM. Evaluation of antiviral drugs and neutralizing antibodies against HIV by a rapid and sensitive microtiter infection assay. J Clin Microbiol. 1988;26:231-235.

18. Pilgrim AK, Pantaleo G, Cohen OJ, Fink LM, Zhou JY, Zhou JT, Bolognesi DP, Fauci AS, Montefiori DC. Neutralizing antibody responses to human immunodeficiency virus type 1 in primary infection and long-term nonprogressive infection. J Infect Dis. 1997;176:924-932.

19. Montefiori DC, Pantaleo G, Fink LM, Zhou JT, Zhou JY, Bilska M, Miralles GD, Fauci AS. Neutralizing and infection-enhancing antibody responses to human immunodeficiency virus type 1 in long-term nonprogressors. J Infect Dis. 1996;173-60-67.

20. Skinner MA, Langlois AJ, McDanal CB, et al. Neutralizing antibodies to an immunodominant envelope sequence do not prevent gp120 binding to CD4. J Virol. 1988;62:4195-4200.

21. Rodbard D, Munson PJ, de Lean A. Improved curve fitting, parallelism testing, characterization of sensitivity and specificity, validation, and optimization for radioligand assays. In: Radioimmunoassay and Related Procedures in Medicine. Int. Atomic Energy Agency, Vienna; 1977:469.

22. Mohamed OA, Ashley R, Goldstein A, et al. Detection of rectal antibodies to HIV-1 by a sensitive chemiluminescent Western Blot immunodetection method. J Acquir Immune Defic Syndr. 1994;7:375-380.

23. Czerkinsky C, Nilsson LA, Tarkowski A, Koopman WJ, Mestecky A, Ouchterlong O. The solid-phase immunospot assay (ELISPOT) for enumerating antibody-secreting cells: methodology and applications. In: Kemeny DM, Challacombe SJ, eds. ELISA and Other Solid-Phase Assays: Theoretical and Practical Aspects. Chichester, England: Wiley; 1988:217-239.

24. Czerkinsky C, Moldoveanu Z, Mestecky J, Nilson LA, and Ouchterlony Ö. A novel two colour ELISPOT assay. I. Simultaneous detection of distinct types of antibody-secreting cells. J Immunol Methods. 1998;115:31-37.

25. Gorse GJ, Frey SE, Newman FK, Belshe RB, AIDS Vaccine Clinical Trials Network. Detection of binding antibodies to native and recombinant human immunodeficiency virus type 1 envelope glycoproteins following recombinant gp160 immunization measured by flow cytometry and enzyme immunoassays. J Clin Microbiol. 1992;30:2606-2612.

26. Gorse GJ, Frey SE, Patel G, Newman FK, Belshe RB, NIAID AIDS Vaccine Clinical Trials Network. Vaccine-induced antibodies to native and recombinant human immunodeficiency virus type 1 envelope glycoproteins. Vaccine. 1994;12:912-918.

27. Gorse GJ, Patel GB, Newman FK, Belshe RB, Berman PW, Gregory TJ, Matthews TJ (NIAID AIDS Vaccine Evaluation Group). Antibody to native human immunodeficiency virus type 1 envelope glycoproteins induced by IIIB and MN recombinant gp120 vaccines. Clin Diagn Lab Immun. 1996;3:378-386.

28. Zhou JY, Montefiori DC. Antibody-mediated neutralization of primary isolates of human immunodeficiency virus type 1 in peripheral blood mononuclear cells is not affected by the initial activation state of the cells. J Virol. 1997;71:2512-2517.

29. Montefiori DC, Collman RG, Fouts TR, Zhou JY, Bilska M, Hoxie JA, Moore JP, Bolognesi DP. Evidence that antibody-mediated neutralization of human immunodeficiency virus type 1 is independent of coreceptor usage. J Virol. 1998;72:1886-1893.

30. Kimpton J, Emerman M. Detection of replication-competence and pseudotype human immunodeficiency virus with a sensitive cell line on the basis of an activation of an integrated beta-galactosidase gene. J Virol. 1992;66:2232-2239.

31. Chackerian B, Long EM, Liciw PA, Overbaugh J. Human immunodeficiency virus type 1 co-receptors participate in post-entry stages in the virus replication cycle and function in simian immunodeficiency virus infection. J Virol. 1997;71:3932-3939.

32. Wild C, Greenwell T, Matthews T. Evidence that antibody-mediated neutralization of human immunodeficiency virus type 1 is independent of coreceptor usage. AIDS Res Hum Retroviruses. 1993;9:1051-1053.