A vaccine strategy that protects against genital herpes by establishing local memory T cells

Abstract

Most successful existing vaccines rely on neutralizing antibodies, which may not require specific anatomical localization of B cells. However, efficacious vaccines that rely on T cells for protection have been difficult to develop, as robust systemic memory T-cell responses do not necessarily correlate with host protection. In peripheral sites, tissue-resident memory T cells provide superior protection compared to circulating memory T cells. Here we describe a simple and non-inflammatory vaccine strategy that enables the establishment of a protective memory T-cell pool within peripheral tissue. The female genital tract, which is a portal of entry for sexually transmitted infections, is an immunologically restrictive tissue that prevents entry of activated T cells in the absence of inflammation or infection. To overcome this obstacle, we developed a vaccine strategy that we term 'prime and pull' to establish local tissue-resident memory T cells at a site of potential viral exposure. This approach relies on two steps: conventional parenteral vaccination to elicit systemic T-cell responses (prime), followed by recruitment of activated T cells by means of topical chemokine application to the restrictive genital tract (pull), where such T cells establish a long-term niche and mediate protective immunity. In mice, prime and pull protocol reduces the spread of infectious herpes simplex virus 2 into the sensory neurons and prevents development of clinical disease. These results reveal a promising vaccination strategy against herpes simplex virus 2, and potentially against other sexually transmitted infections such as human immunodeficiency virus.

Figure 1. Effector T cells are recruited to the vagina by topical chemokine treatment

a, Experimental schematic. Donor gBT-I CD8 T cell recipients were unimmunized (naïve), or immunized either ivag or s.c. with TK- HSV-2. Five days post-infection, s.c. immunized mice was treated vaginally with either the chemokines (pull) or PBS. One day post-pull, the frequency (b) and number (c, d) of donor gBT-I CD8 T cells (b, c) and CD4 T cells (d) were determined in the indicated tissues. Plots are gated on total CD8 T cells and numbers indicate % gBT-I (CD45.1+) (b). Numbers in graphs indicate fold difference in T cell number between s.c. vs. s.c. + pull. Statistical significance was determined by two-tailed unpaired Student’s t-test. Data are pooled from 2–7 independent experiments; n = 6–21 per group. All error bars show standard error (SEM), and *; p = 0.05–0.01, **; p = 0.01 – 0.001, ***; p < 0.001 throughout the manuscript (individual p values are found in Supplementary Information).

Figure 2. Chemokine pull is specific for highly activated effector T cells

Mice were s.c. immunized and given chemokines or PBS at day 5, 15 or 28 post-infection and analyzed one day post-pull. a, CXCR3 expression on donor gBT-I or CD44+ CD4 T cells from the spleen one day post pull from s.c. immunized mice (open black) and s.c. + pull. (open red). Shaded histograms are CD44^lo CD8 or total CD4 T cells. b, The gBT-I cell number in the vagina (left) or spleen (middle) and frequency in the spleen (right) were examined one day post-pull. c, d, The number of CD44+ CD4 T cells (c) and endogenous CD44+ CD8 T cells (d) was determined in the vagina one day post-pull on the indicated days. Statistical significance was determined by two-tailed unpaired Student’s t-test. Data are pooled from 2–3 independent experiments; n = 6–18 per group.

Figure 3. Virus-specific T cells recruited by chemokine pull are retained in the vagina long-term

Mice were immunized and treated as shown in Figure 1a. Four (a-c) or twelve (d, e) weeks post-pull, T cell numbers and frequency were assessed in various tissues., At 4 weeks post-pull, numbers of gBT-I cells (a) and endogenous CD4 T cells (c) were determined in the indicated tissues. Number in graph shows fold difference in gBT-I number between s.c. vs. s.c. + pull groups. b, Frequency of gBT-I cells was measured in the vagina. Plots are gated on total CD8 T cells. Numbers show % gBT-I (CD45.1+). At 12 weeks post-pull, the numbers of gBT-I cells (d) and endogenous CD4 T cell (e) were determined in the vagina (left) and compared to numbers at 4 weeks post-pull (right). # × N.D. = number of animals with no cells detected in the tissue. Statistical significance was determined by two-tailed unpaired Student’s t-test; ns = not significant. Data are pooled from 2–3 independent experiments; n = 4–15 per group.

Figure 4. Prime and pull protects mice from lethal genital HSV-2 challenge

Mice were immunized as shown in Figure 1a and challenged vaginally with a lethal dose of HSV-2 four weeks post-pull. Weight loss (a), disease severity (b), and survival (c) were assessed. d, HSV-2 was titered from vaginal washes collected at the indicated time points. Dashed line indicates limit of detection. ND = none detected. n = 11 (unimmunized), n = 9 (ivag), n = 12 (s.c., s.c. + pull). e, Viral titers were measured in the dorsal root ganglia 6–7 days post-challenge. # × N.D. = number of mice in which no virus was detected. n = 6–11 per group. Statistical significance was measured by two-way ANOVA (a-b, d), log-rank (Mantel-Cox) test (c) or two-tailed unpaired Student’s t-test (e). Data are pooled from 3–5 independent experiments.