Assessment of Anti–HIV-1 Antibodies in Oral and Nasal Compartments of Volunteers From 3 Different Populations : JAIDS Journal of Acquired Immune Deficiency Syndromes

Secondary Logo

Journal Logo

Basic and Translational Science

Assessment of Anti–HIV-1 Antibodies in Oral and Nasal Compartments of Volunteers From 3 Different Populations

Bergin, Philip J. PhD*; Langat, Robert BSc; Omosa-Manyonyi, Gloria MD; Farah, Bashir BSc; Ouattara, Gina MD; Park, Harriet MPH; Coutinho, Helen BSc*; Laufer, Dagna MD; Fast, Pat MD; Verlinde, Carl MSc; Bizimana, Jean MSc§; Umviligihozo, Gisele MSc§; Nyombayire, Julien MD§; Ingabire, Rosine MD§; Kuldanek, Kristin BSc; Cox, Josephine PhD; McMorrow, Martin MBA*; Fidler, Sarah MBBS, PhD; Karita, Etienne MD§; Gilmour, Jill PhD*,‡; Anzala, Omu MD

Author Information
JAIDS Journal of Acquired Immune Deficiency Syndromes: October 1, 2016 - Volume 73 - Issue 2 - p 130-137
doi: 10.1097/QAI.0000000000001094



HIV-1 is often transmitted across the genital mucosa during sexual intercourse. HIV-1 crosses the mucosal epithelial barrier infecting cells in the underlying lamina propria where it replicates and then disseminates through the draining lymph nodes into the periphery.1 Of the 4 potential HIV vaccine candidates tested for efficacy,2–5 only the Thai RV144 vaccine trial has demonstrated a moderate reduction on the acquisition of HIV-1, linked to the induction of HIV-1 Env-specific IgG antibodies.6–11 Data from a secondary analysis of the CAPRISA 004 tenofovir microbicide trial12 suggested that successful HIV preventative approaches will require induction of an effect at the mucosal surface, likely to include anti-HIV immune responses that can either prevent the initial mucosal infection, or eradicate the virus before it disseminates into the periphery. Systemic vaccination is known to be able to induce a long-lasting mucosal response.13 However, methods have been developed to bias the response to a vaccine regimen to target toward a mucosal surface responses; vaccines may be administered through mucosal routes14 or the use of a replicating vector may extend the time the immune systems is exposed to vectors and inserts,15 potentially allowing time for a more efficacious response. Recently, a replicating Sendai vector encoding HIV-1 Gag has been tested in a phase 1 trial ( Identifier NCT01705990). Administered intranasally, this replicating vector, and others like it,15,16 may preferentially direct vaccine-induced immune responses to the mucosa and/or linked mucosal compartments such as gastric-associated lymphoid tissue. To support the assessment of future clinical trials intended to induce a mucosa response, it is therefore necessary to standardize clinical collection methods that are acceptable to volunteers and will help ensure compliance for multiple sampling visits. Although genital (both cervico-vaginal and rectal) sampling is increasingly being applied in HIV vaccine trials, it is considered to be invasive and may be less acceptable in several cultures and regions.17,18 Collection of nasal19 and/or salivary samples may be a less intrusive, more acceptable option,20 and may indicate what is happening at other mucosal surfaces.

Here, we present a pilot study demonstrating that anti-HIV antibodies were detected in nasal secretions collected from the turbinate region of the nasal cavity (TB), deeper in the nasopharyngeal tract (NP), and oral secretions of HIV-1–infected individuals. Antibody levels at the collection sites were compared. Two devices were also compared in the collection of nasal cavity secretions.


Study Population and Ethics

Volunteers were recruited from Nairobi, Kenya (Kenya AIDS Vaccine Initiative-Institute of Clinical Research; KAVI-ICR); Kigali, Rwanda (Project San Francisco; PSF); and London, United Kingdom (St. Mary's Clinic) to allow for population-specific differences and prepare for future clinical studies. Approval was granted by the Kenyatta National Hospital/University of Nairobi Ethics Committee (KAVI), Rwanda National Ethics Committee and Emory University Institutional Review Board (PSF), and the National Research Ethics Service Committee London—Brent, London (St. Mary's). All volunteers provided written informed consent. Known HIV-1–infected and uninfected volunteers (18–60 years) were enrolled (Table 1). Seronegative status was confirmed using standard diagnostic methods according to the national HIV testing guidelines. The study excluded volunteers with severe abnormalities in the mouth or nasal cavity.

Volunteer Demographics

Study Design

This was a descriptive study assessing anti-HIV antibodies in a cohort of known HIV-1–infected volunteers compared against confirmed HIV seronegative (by point-of-care methods). At St. Mary's, all samples were collected at a single visit. KAVI and PSF incorporated a longitudinal component with samples collected at 2 distinct time points. At KAVI, samples were collected over 2 visits within 2 weeks, comparing the nasal cavity and nasopharyngeal tract with flocked swab (FS) (Copan Diagnostics Inc., Murrieta, CA) samples at the first visit, and comparing FS and synthetic absorptive matrix (SAM) (SAM; Hunt Developments (UK) Ltd, Midhurst, United Kingdom) strip collection from alternating nasal cavities at the second visit. PSF collected identical samples over 2 visits within 1 month, to assess parity of sampling over time. The study was designed to confirm that all 3 centers could collect comparative samples from identical regions of the nasal and oral compartments, and to determine whether the samples collected from HIV-1–infected volunteers had detectable anti–HIV-1 antibodies. Previous optimization studies (data not shown) suggested that groups of 15 HIV-1–infected and 15 seronegative volunteers would be adequate to demonstrate anti–HIV-1 antibodies in a consistent manner (should they be present) but also to demonstrate any significant differences between the 3 centers. In addition, PSF recruited 5 additional volunteers for training purposes.

Sample Collection and Processing

Samples from the nasal cavity (TB) and the nasopharyngeal tract (NP) were collected using FSs (TB; inserted until the swab head was covered by the cavity, NP; inserted 8 cm into the nasopharyngeal tract or until resistance was met). The swab was rotated 360 degrees around the cavity, to maximize collection, and secretions eluted from the swab by centrifugation in phosphate-buffered saline containing 1% bovine serum albumin, 0.5% Tween-20%, and 0.05% sodium azide (elution buffer) at 4°C. In addition, volunteers at KAVI and St. Mary's provided turbinate secretions collected using a SAM strip. Samples were eluted from the SAM strip by centrifugation at 4°C, using elution buffer (as above), in a Costar spin-x centrifuge tube (0.22 µm pore size, Sigma, Gillingham, United Kingdom). Whole oral fluid (OT) was collected into Falcon tubes as described previously.21 Briefly; volunteers allowed whole OTs to pool in the mouth, then transferred by drooling into a 50 mL Falcon tube. The sample was centrifuged to remove pellet particulate matter and separate OT from any contaminating mucus. Whole OT was collected from the tube, ensuring that any contaminating mucus was left in the tube. Parotid saliva was collected from the parotid duct (Stensen duct) of the parotid glands using Salimetrics oral swabs (SOS; Salimetrics LLC, State College, PA) placed over the opening of the Stensen duct and held in place for 5 minutes. Parotid saliva was eluted from the SOS by centrifugation at 4°C, transudate was centrifuged at 4°C for 15 minutes to pellet any contaminants and separate the mucus component, and the supernatant collected. All samples were stored on ice and processed within 2 hours of collection, eluted samples were aliquoted and stored at −80°C. All samples were collected with the assistance of study staff with Standard Operating Procedures (SOPs) clearly defining the collection location and methods.

Peripheral blood was collected into BD vacutainer SST tubes (Fisher Scientific, Loughborough, United Kingdom) by venipuncture as previously described22 and centrifuged. Aliquoted serum was stored at −80°C.

Binding Antibody ELISAs

Samples were tested for gp140 Env (clade A UG37) and Gag p24 (clade B BH10) IgG and IgA antibodies using an Enzyme-linked immunosorbent assay (ELISA) modified from previously described methods.23,24

ELISA plates (Medium binding; Greiner Bio-One Inc., Sigma, United Kingdom) were coated with recombinant HIV-Gag p24 antigen (2.5 μg/mL; Aalto Bio Reagents, Ltd., Dublin, Ireland) or recombinant HIV-Env subtype A UG37 gp140 antigen (5.0 μg/mL; Polymun, Vienna, Austria) for 1 hour. After washing in PBS Tween (PBS-T; Sigma), the plates were blocked using 10% Heat Inactivated Fetal Bovine Serum (FBS) (Sigma) in PBS-T. Plates were frozen at −80°C and stored for up to 6 months. Before sample analysis, plates were thawed and washed in PBS-T. Mucosal sample aliquots were diluted 1:20, and plasma samples were diluted 1:100. Samples were incubated on the ELISA plates for 1 hour at 37°C then washed in PBS-T. IgG-binding antibodies were detected using peroxidase-labeled goat anti-human IgG (Fc specific) (Sigma) and signal developed using Sureblue TMB Microwell Peroxidase Substrate (Insight Biotechnology Ltd., Wembley, United Kingdom) and TMB Stop solution (Insight Biotechnology Ltd.). IgA antibodies were detected using Biotin-labeled Goat anti-Human IgA (alpha chain) (Insight Biotechnology Ltd.) followed by an avidin peroxidase (Sigma) multiplication step and signal developed as above. After development, plates were read at 450 nm using a Tecan Sunrise plate reader at the Human Immunology Laboratory (HIL) or Tecan Infinite M200 plate reader (KAVI). All samples collected at KAVI were assessed at KAVI, whereas PSF and St. Mary's samples were assessed at the HIL. Positive responses were defined as optical density (OD) readings above the following cut-offs: ≥0.2 for IgG and ≥0.3 IgA; all positive samples were subsequently titrated for limiting dilution assessment. A panel of HIV-1 seropositive and seronegative plasma samples (Blood bank, Cape Town, South Africa) was used to ensure parity between the 2 testing laboratories.


Figures were made, and statistical analysis was performed using GraphPad Prism version 6.0. Statistically significant differences were determined using a Mann–Whitney nonparametric analysis, 2-tailed with 95% confidence, unless otherwise stated.



Volunteers were equally split between HIV-infected and uninfected (Table 1). The study at PSF preferentially enrolled female volunteers.

Demonstration of Concordance Between ELISA Methods

To ensure parity of data between HIL and KAVI, a blinded panel of 10 HIV-infected samples and 2 HIV seronegative samples was screened for IgG and IgA antibodies to Gag (p24) and envelope (gp140). Positive samples were titered. Staff at HIL and KAVI correctly identified all positive and negative samples and determined endpoint titer to within 1 dilution step of the predetermined value (data not shown).

Salivary and Nasal Sampling Was Well Accepted

All volunteers agreed to provide all nasal and salivary samples. This was repeated where samples were taken at 2 time points, with the exception of 2 volunteers who did not return for the second visit. Although volunteers did provide all nasal samples, they reported more discomfort from the FS samples collected deeper in the nasopharyngeal tract (NP).

Anti–HIV-1 Antibodies Are Present at Comparable Levels in the TB and NP of HIV-1–Infected Individuals

Both IgG- and IgA-binding antibodies against HIV-Env gp140 and HIV-Gag p24 IgG were detected in nasal secretions from HIV-infected volunteers (Fig. 1A and Table 2). Anti–HIV-1 IgG was found at consistently higher amounts than IgA in all samples tested, ranging from 1.1 to 6.1-fold in OD, depending on the sample type (Table 3). This difference was more pronounced in the serum, which had more IgG present (by OD reading and absolute titer), with an IgG to IgA fold difference, ranging from 48 to 1195 and 47 to 262 in mean fold difference in absolute titers (IgG/IgA for anti-gp140 and p24, respectively; Table 3). Although this trend was seen at all centers, the samples from East Africa (KAVI and PSF) exhibited higher titers of antibodies than were seen in the samples from London (St. Mary's). This may be due to the treatment status of the volunteers as anti-retroviral therapy (ART)–treated volunteers are likely to have fewer viruses inducing an immune response. Although information was available as to ART status, length of time on ART and date of initiation of ART were not collected for this study. However, CD4 counts were collected, and as expected, there was a trend toward a higher CD4 count in volunteers on therapy (see Supplemental Digital Content,

Anti–HIV-1 antibodies are detectable in both nasal and oral secretions. Samples were collected with a FS from both the turbinate region (TB) of the nasal cavity and from the nasopharyngeal tract (NP) of (A) HIV-1–infected volunteers and (B) HIV-1–uninfected volunteers. Samples of the whole OT were collected into Falcon tubes, and saliva (SA) was collected by placing a SAM strip over the duct opening of the parotid gland duct opening of (C) HIV-1–infected volunteers and (D) HIV-1–uninfected volunteers. Graphs show the presence of anti–HIV-1 Env (gp140; black points) or anti–HIV-1 p24 (Gag; blue points) IgG (filled circles) and IgA (filled diamonds). Graph shows individual volunteers and mean ± SD for each group, dotted line (…….) represents the IgG positivity cut-off of 0.2 OD units, whereas the dashed line (. _ . - .) represents the IgA positivity cut-off of 0.3 OD units. Any significant differences are shown with their exact P value.
Quantity of IgG Antibodies Higher Than IgA Antibodies in Mucosal Samples Collected From HIV-1–Infected Volunteers*
Mucosal IgG and IgA Titers From HIV-1–Infected Samples Collected at 3 Clinical Centers

In all cases where antibody was present in the NP compartment, it was also present in the TB. There were no significant differences between Env and Gag antibody levels collected from the TB and NP (Fig. 1A) of HIV-1–infected individuals. The levels of cross-reactive HIV-1 binding from sample eluate were higher in NP than TB from HIV-1 seronegative volunteers for anti-Env IgA and anti-p24 IgG (Fig. 1B; P < 0.001 and P = 0.0005, respectively).

Anti–HIV-1 Antibodies Are Present in Saliva

Saliva was collected from the parotid duct using an SOS, and whole OT (saliva mixed with transudate from serum) was collected in a tube. During processing in the laboratory, it was noted that the whole saliva sample was often contaminated with particulate matter and mucus. On several occasions, it was not possible to harvest any supernatant from the sample (6/30 volunteers at St. Mary's), with either little sample being present or the bulk of the sample being thick mucus. Parotid saliva was much cleaner with little or no particulate contamination or mucus.

Anti–HIV-1 IgG was present in either OT or parotid saliva in 47/50 HIV-1–infected volunteers, but anti–HIV-1 IgA was present in only 24/50 HIV-1–infected volunteers (Fig. 1C and Table 2). Both anti-Env gp140 and anti-p24 antibody were found in the samples (Fig. 1C) at very similar levels, but the yield of anti-p24 IgA in the whole OT samples was higher than the samples collected from the parotid gland (P= 0.0323; Fig. 1C). There was a trend toward higher yields of IgG in the parotid saliva compared with the transudate (Fig. 1C). Variation was detected in background levels in saliva from HIV-1–uninfected volunteers.

FS Is Superior to SAM Strip for Collecting Samples With Detectable Anti–HIV-1 Antibodies

At KAVI and St. Mary's, a SAM strip used to collect turbinate nasal secretions was well tolerated, but yielded lower levels of anti–HIV-1 IgG antibodies than the FS (Fig. 2A; P < 0.0001 and P = 0.0004, respectively) with a similar trend seen for IgA. There was variation in background readings from HIV-1–uninfected samples (Fig. 2B). Two of the twenty (2/20) HIV-1–uninfected volunteers from PSF had detectable levels of anti-p24 IgG and/or IgA detectable in all samples (Figs. 1B, D, 2B). Anti-p24 IgA was also present in serum, confirming the mucosal results (data not shown). HIV-1–infected volunteers were previously identified as seropositive, and all uninfected volunteers were demonstrated as seronegative using point-of-care diagnosis, suggesting that the ELISA used here may pick up low levels of antibodies in cross-reactive sera and mucosal secretions.

FSs collect nasal secretions with a higher yield of anti–HIV antibody than SAM strips. Samples were collected from the turbinate region (TB) of the nasal cavity using a SAM strip and an FS of (A) HIV-1–infected volunteers and (B) HIV-1–uninfected volunteers. Graphs show the presence of anti–HIV-1 Env (gp140; black points) or anti–HIV-1 p24 (Gag; blue points) IgG (filled circles) and IgA (filled diamonds). Graph shows individual volunteers and group mean ± SD, dotted line (…….) represents the IgG positivity cut-off of 0.2 OD units, whereas the dashed (. _ . - .) line represents the IgA positivity cut-off of 0.3 OD units. Significant differences are shown with their exact P value.

HIV-1–Infected Volunteers on ART Demonstrated Less Anti–HIV-1 Antibody Than Untreated Volunteers

There was either a significant difference or trend toward lower levels of anti–HIV-1 antibody in all samples for volunteers on ART compared with naive/untreated volunteers (Fig. 3).

Effect of ART treatment on levels of circulating anti-HIV antibodies in mucosal secretions and serum. Graph shows individual volunteers from PSF with their ART treatment status. Circles represent IgG antibody responses, whereas diamonds indicate IgA response. Black indicates an anti-Env antibody, whereas blue represents responses against p24. Graph shows responses together with mean ± SD for NA (naive/untreated) and ART (ART treated) volunteers for (A) OT, (B) parotid saliva, (C) turbinate nasal secretions, (D) nasopharyngeal secretions, and (E) serum. Dotted line (…….) represents the IgG positivity cut-off of 0.2 OD units, whereas the dashed (. _ . - .) line represents the IgA positivity cut-off of 0.3 OD units.


Of the 4 HIV vaccine efficacy studies to date, only 1 has shown some protective efficacy.2,11 The RV144 trial suggested that vaccines eliciting IgG antibodies against a specific part of Env in the V1/V2 region may prevent infection and that serum IgA antibodies may correlate with the risk of infection.6,10,11,25 At the very least, antibodies will form a component of an effective HIV-1 vaccine. Because the transmission of HIV-1 occurs primarily across genital mucosa, it is likely that antibodies elicited by vaccination should be present at the mucosa, ideally both vaginally and rectally.

To better target mucosal surfaces, several groups, including ours, are working on vaccine candidates that may be administered mucosally using needle-free methods readily applicable to resource-limited settings, such as intranasally.16 Here, we assessed different locations and collection methods for nasal sampling to determine a method that detects humoral responses and may be more acceptable to volunteers than more invasive genital or rectal sampling,17,18,20 thus increasing the likelihood of compliance with multiple collections throughout a clinical trial.

More HIV-1–infected volunteers had detectable levels of anti–HIV-1 IgG than IgA in nasal secretions. In the majority of uninfected volunteers, no anti–HIV-1 antibodies were present in any sample, except for 3/20 volunteers with low levels of anti-p24, and several volunteers with high levels of nasal anti–Env IgG, antibodies were detectable. These anti-p24 antibodies were present in multiple samples suggesting that these 3 volunteers, while seronegative by clinical diagnostic antibody tests, had persistent systemic cross-reactive antibodies against HIV-1 p24. Several seronegative volunteers expressed high levels of anti-Env IgG in their nasopharyngeal tract, and this was more pronounced in the NP rather than the TB samples, suggesting the presence of cross-reactive antibodies or contamination in the respiratory tract.

The SAM strip has been used in several studies to take repeated samples directly from the nasal mid-turbinate over very short periods of time.26 Although these samples, used to measure RNA and cytokines, generated very reproducible data, we found that eluates from the SAM strip samples tended to have lower antibody titers than eluates from FSs. Less than 50% (13/30) of the HIV-infected volunteers had detectable antibody in the SAM strip samples compared with two-thirds (20/30) of those collected using the FS. Because of their much smaller surface area, the SAM strips collect a comparatively small amount of nasal secretion compared with the FSs that are rapidly replenished by the local immune tissue. In comparison, the FSs have a much larger surface area and collect secretions from either the entire nasal cavity comprising all 3 turbinate areas or from a much larger area in the nasopharyngeal tract.

Other groups have detected antibodies in the nasal cavity when assessing other vaccines, such as influenza vaccines,27 and diseases such as rhinitis28 and respiratory syncytial virus19,29 infection. Deep in the nasopharyngeal tract is the standard location for collection of nasal samples, and although safe to the volunteer, it can lead to discomfort. Previous studies in East Africa have established that to collect repeated samples, sampling needs to be simple, fast, and not too uncomfortable.20 Feedback from volunteers and study clinicians indicated that the deeper NP collections were less tolerated than the TB collections (personal communication). The relative ease and comfort of TB sampling, combined with equivalent antibody detection compared with NP sampling, makes the former more attractive for repeated collections.

Anti–HIV-1 antibodies were detectable in saliva and were only present when antibodies were also present in the nasal sample. IgG was present at much higher levels (OD and absolute titers) and in more samples than IgA, but often there was mucus present in the saliva samples, with more present in the whole saliva than in the parotid saliva. Several studies have shown that IgA, and to some extent IgG, can bind to genital mucus,30,31 and it is likely that IgA may be bound in a complex with the oral mucus, and is lost during processing. We detected no significant differences between antibody levels in parotid saliva and whole OT with the ratios between IgG and IgA remaining constant between the 2 sample types. There was significantly more antibody in the serum than in the salivary samples. Although it is likely that much of the salivary antibody is transudated across the oral mucosa from the peripheral blood, it is possible that we are detecting antibodies produced and secreted into the salivary compartment. In this study, we did not look for the presence of dimeric IgA, secretory component, or the J-chain and are unable to confirm whether the IgA being detected in the salivary samples is secreted or monomeric serum-derived IgA. Mucosal antibodies were only detected in volunteers who also had serum antibodies. High titers of mucosal antibodies corresponded to volunteers with high titers of serum antibodies.

HIV-1–infected volunteers on therapy had lower levels of anti–HIV-1 antibody at all mucosal surfaces. There was no significant difference in the levels of HIV-1 Env IgG antibody in serum, but there was a difference in levels of anti-p24 IgG antibody. It is likely that anti–HIV-1 antibodies will decrease as ART suppresses viremia and the associated immune stimulation. Because the levels of viremia decrease, it is likely that the B-cell response will switch from actively antibody-secreting plasmablasts toward maintaining a memory B-cell pool, resulting in a decrease in the level of circulating and compartmentalized antibodies against HIV.

We, and others, have previously collected samples from the genital mucosa of volunteers enrolled into observational studies and phase 1 clinical trials in Kenya.20,32–35 Although mucosal sampling was feasible with some volunteers, there were issues with compliance over repeated sampling of the genital mucosa, particularly with rectal sampling.20 Here, we present an assessment of nasal and salivary sampling in HIV-1–infected versus HIV-1–uninfected volunteers. Sampling from both areas was acceptable to the volunteers, with almost all volunteers providing all samples requested (except for 2 volunteers who did not return for the second visit), as opposed to the much lower compliance seen in our previous studies collecting genital secretions in Kenya.20 In these previous studies, almost two-third of female volunteers was willing to provide cervico-vaginal samples (by Merocel sponge and/or vaginal Softcup), whereas only 1/4 (24%) of the volunteers (male and female) were willing to provide rectal samples. For the purposes of assessing humoral responses, collection from the turbinate region of the nasal cavity seemed to be as effective as samples collected deeper in the nasopharyngeal tract, and was more acceptable to volunteers who reported discomfort from the “deeper” sample collection in the nasopharyngeal tract (personal communication with study physicians). Nasal samples also seemed to provide a more concentrated antibody than a sample collected from the oral cavity. Nasal collection from turbinate region is a useful option for assessing anti-HIV antibodies in the context of a clinical trial, or observational study, although it remains to be clarified if levels of antibody in the nasal cavity and salivary compartment correspond to levels at the genital mucosa; the site of transmission of HIV-1 during sexual contact.

In conclusion, this study clearly demonstrates that FS collection from the whole turbinate region (the nasal cavity) is sufficient for detecting antibody responses, and is more acceptable to the volunteer than a more invasive collection procedure penetrating up to 10 cm into the nasopharyngeal tract. Although repeated genital sampling has previously shown poor compliancy,20 in this study African and UK participants alike accepted nasal sampling. This sampling method is likely to be useful for assessing antibody responses to a nasal administration during a vaccine clinical trial, and in other observational or epidemiological studies.


The authors thank all volunteers who took part in the study and clinic staff at KAVI, PSF, and St. Mary's. They thank Mr Kenneth Legg for his assistance in performing the London study and Ms. Emmy McGarry for excellent assistance in performing assays; Kristina Broliden and Taha Hirbod, Mr Duncan Hunt, Prof. Peter Openshaw, Trevor Hansel, Prof Robin Shattock, and Alethea Cope for technical assistance and advice. They also thank Ms. Alpa Mulji for her transfer of the standardized ELISA method to KAVI with support of the GSK Pulse program.


1. Haase AT. Targeting early infection to prevent HIV-1 mucosal transmission. Nature. 2010;464:217–223.
2. Hammer SM, Sobieszczyk ME, Janes H, et al. Efficacy trial of a DNA/rAd5 HIV-1 preventive vaccine. N Engl J Med. 2013;369:2083–2092.
3. Gray G, Buchbinder S, Duerr A. Overview of STEP and Phambili trial results: two phase IIb test-of-concept studies investigating the efficacy of MRK adenovirus type 5 gag/pol/nef subtype B HIV vaccine. Curr Opin HIV AIDS. 2010;5:357–361.
4. Gray GE, Allen M, Moodie Z, et al. Safety and efficacy of the HVTN 503/Phambili study of a clade-B-based HIV-1 vaccine in South Africa: a double-blind, randomised, placebo-controlled test-of-concept phase 2b study. Lancet Infec Dis. 2011;11:507–515.
5. Buchbinder SP, Mehrotra DV, Duerr A, et al. Efficacy assessment of a cell-mediated immunity HIV-1 vaccine (the Step Study): a double-blind, randomised, placebo-controlled, test-of-concept trial. Lancet. 2008;372:1881–1893.
6. Karasavvas N, Billings E, Rao M, et al. The Thai Phase III HIV Type 1 Vaccine trial (RV144) regimen induces antibodies that target conserved regions within the V2 loop of gp120. AIDS Res Hum Retroviruses. 2012;28:1444–1457.
7. Zolla-Pazner S, deCamp A, Gilbert PB, et al. Vaccine-induced IgG antibodies to V1V2 regions of multiple HIV-1 subtypes correlate with decreased risk of HIV-1 infection. PloS One. 2014;9:e87572.
8. Liao HX, Bonsignori M, Alam SM, et al. Vaccine induction of antibodies against a structurally heterogeneous site of immune pressure within HIV-1 envelope protein variable regions 1 and 2. Immunity. 2013;38:176–186.
9. Yates NL, Liao HX, Fong Y, et al. Vaccine-induced Env V1-V2 IgG3 correlates with lower HIV-1 infection risk and declines soon after vaccination. Sci Transl Med. 2014;6:228–239.
10. Haynes BF, Gilbert PB, McElrath MJ, et al. Immune-correlates analysis of an HIV-1 vaccine efficacy trial. N Engl J Med. 2012;366:1275–1286.
11. Rerks-Ngarm S, Pitisuttithum P, Nitayaphan S, et al. Vaccination with ALVAC and AIDSVAX to prevent HIV-1 infection in Thailand. N Engl J Med. 2009;361:2209–2220.
12. Abdool Karim Q, Abdool Karim SS, Frohlich JA, et al. Effectiveness and safety of tenofovir gel, an antiretroviral microbicide, for the prevention of HIV infection in women. Science. 2010;329:1168–1174.
13. Avdicova M, Crasta PD, Hardt K, et al. Lasting immune memory against hepatitis B following challenge 10-11 years after primary vaccination with either three doses of hexavalent DTPa-HBV-IPV/Hib or monovalent hepatitis B vaccine at 3, 5 and 11-12 months of age. Vaccine. 2015;33:2727–2733.
14. Kaneko H, Bednarek I, Wierzbicki A, et al. Oral DNA vaccination promotes mucosal and systemic immune responses to HIV envelope glycoprotein. Virology. 2000;267:8–16.
15. Thomas MA, Song R, Demberg T, et al. Effects of the deletion of early region 4 (E4) open reading frame 1 (orf1), orf1-2, orf1-3 and orf1-4 on virus-host cell interaction, transgene expression, and immunogenicity of replicating adenovirus HIV vaccine vectors. PloS One. 2013;8:e76344.
16. Excler JL, Parks C, Ackland J, et al. Replicating viral vectors as HIV vaccines: summary report from the IAVI-sponsored satellite symposium at the AIDS vaccine 2009 conference. Biological. 2010;38:511–521.
17. Lazarus EM, Otwombe K, Adonis T, et al. Uptake of genital mucosal sampling in HVTN 097, a phase 1b HIV vaccine trial in South Africa. PloS One. 2014;9:e112303.
18. Seaton KE, Ballweber L, Lan A, et al. HIV-1 specific IgA detected in vaginal secretions of HIV uninfected women participating in a microbicide trial in Southern Africa are primarily directed toward gp120 and gp140 specificities. PloS One. 2014;9:e101863.
19. Esbenshade JC, Edwards KM, Esbenshade AJ, et al. Respiratory virus shedding in a cohort of on-duty healthcare workers undergoing prospective surveillance. Infect Control Hosp Epidemiol. 2013;34:373–378.
20. Omosa-Manyonyi G, Park H, Mutua G, et al. Acceptability and feasibility of repeated mucosal specimen collection in clinical trial participants in Kenya. PloS One. 2014;9:e110228.
21. Hirbod T, Reichard C, Hasselrot K, et al. HIV-1 neutralizing activity is correlated with increased levels of chemokines in saliva of HIV-1–exposed uninfected individuals. Curr HIV Res. 2008;6:28–33.
22. Jaoko W, Karita E, Kayitenkore G, et al. Safety and immunogenicity study of Multiclade HIV-1 adenoviral vector vaccine alone or as boost following a multiclade HIV-1 DNA vaccine in Africa. PloS One. 2010;5:e12873.
23. Keefer MC, Gilmour J, Hayes P, et al. A phase I double blind, placebo-controlled, randomized study of a multigenic HIV-1 adenovirus subtype 35 vector vaccine in healthy uninfected adults. PloS One. 2012;7:e41936.
24. Cranage MP, Fraser CA, Cope A, et al. Antibody responses after intravaginal immunisation with trimeric HIV-1 CN54 clade C gp140 in Carbopol gel are augmented by systemic priming or boosting with an adjuvanted formulation. Vaccine. 2011;29:1421–1430.
25. Kim JH, Excler JL, Michael NL. Lessons from the RV144 Thai phase III HIV-1 vaccine trial and the search for correlates of protection. Annu Rev Med. 2015;66:423–437.
26. Scadding GW, Calderon MA, Bellido V, et al. Optimisation of grass pollen nasal allergen challenge for assessment of clinical and immunological outcomes. J Immunol Methods. 2012;384:25–32.
27. Ainai A, Tamura S, Suzuki T, et al. Intranasal vaccination with an inactivated whole influenza virus vaccine induces strong antibody responses in serum and nasal mucus of healthy adults. Hum Vaccines Immunotherapeutics. 2013;9:1962–1970.
28. Rondon C, Romero JJ, Lopez S, et al. Local IgE production and positive nasal provocation test in patients with persistent nonallergic rhinitis. J Allergy Clinical Immunology. 2007;119:899–905.
29. Slobod KS, Shenep JL, Lujan-Zilbermann J, et al. Safety and immunogenicity of intranasal murine parainfluenza virus type 1 (Sendai virus) in healthy human adults. Vaccine. 2004;22:3182–3186.
30. Fahrbach KM, Malykhina O, Stieh DJ, et al. Differential binding of IgG and IgA to mucus of the female reproductive tract. PloS One. 2013;8:e76176.
31. Domagala A, Kasprzak M, Kurpisz M. Immunological characteristics of cervical mucus in infertile women. Zentralbl Gynakol. 1997;119:616–620.
32. Lajoie J, Kimani M, Plummer FA, et al. Association of sex work with reduced activation of the mucosal immune system. J Infec Dis. 2014;210:319–329.
33. Arnold KB, Burgener A, Birse K, et al. Increased levels of inflammatory cytokines in the female reproductive tract are associated with altered expression of proteases, mucosal barrier proteins, and an influx of HIV-susceptible target cells. Mucosal Immunol. 2016;9:194–205.
34. Birse KD, Cole AL, Hirbod T, et al. Non-cationic proteins are associated with HIV neutralizing activity in genital secretions of female sex workers. PloS One. 2015;10:e0130404.
35. Hasselrot K, Cheruiyot J, Kimani J, et al. Feasibility and safety of cervical biopsy sampling for mucosal immune studies in female sex workers from Nairobi, Kenya. PloS One. 2012;7:e47570.

mucosal; HIV; vaccine; nasal; salivary; antibody

Supplemental Digital Content

Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.