key: cord-103208-krann2ir
authors: Barber-Axthelm, Isaac M; Kelly, Hannah G; Esterbauer, Robyn; Wragg, Kathleen; Gibbon, Anne; Lee, Wen Shi; Wheatley, Adam K; Kent, Stephen J; Tan, Hyon-Xhi; Juno, Jennifer A
title: Coformulation with tattoo ink for immunological assessment of vaccine immunogenicity in the draining lymph node
date: 2020-08-29
journal: bioRxiv
DOI: 10.1101/2020.08.27.270975
sha: 
doc_id: 103208
cord_uid: krann2ir

Characterisation of germinal centre B and T cell responses yields critical insights into vaccine immunogenicity. Non-human primates are a key pre-clinical animal model for human vaccine development, allowing both lymph node and circulating immune responses to be longitudinally sampled for correlates of vaccine efficacy. However, patterns of vaccine antigen drainage via the lymphatics after intramuscular immunisation can be stochastic, driving uneven deposition between lymphoid sites, and between individual lymph nodes within larger clusters. In order to improve the accurate isolation of antigen-exposed lymph nodes during biopsies and necropsies, we developed and validated a method for co-formulating candidate vaccines with tattoo ink, which allows for direct visual identification of vaccine-draining lymph nodes and evaluation of relevant antigen-specific B and T cell responses by flow cytometry. This approach improves the assessment of vaccine-induced immunity in highly relevant non-human primate models.

Peripheral lymphoid tissues, including lymph nodes (LN), tonsils and mucosal associated lymphoid tissues are critical sites for the generation of adaptive immunity and immunological memory. After intramuscular (IM) administration, vaccine antigens drain via the lymphatics to be concentrated and retained within the LN, where they are subject to immune surveillance.

Antibodies are a key protective correlate for most human vaccines, with high-affinity variants generated within germinal centres (GC) via tightly regulated interactions of antigen, GC B (B GC ) cells, and T follicular helper (T fh ) cells ( 1 ) . Efficient generation of GC by immunisation is therefore a key determinant of vaccine success or failure, controlling the kinetics, magnitude and quality of the resultant serological response ( 2 , 3 ) .Direct characterisation of antigen-specific B GC or T fh cells can provide important insights into vaccine immunogenicity and the biogenesis of protective immune responses. However, interrogation of LN B GC and T fh cells in humans is challenging, requiring invasive surgical excision or the collection of a small number of cells by fine needle aspirates (FNA) ( 4 , 5 ) . In contrast, pre-clinical animal models such as non-human primates (NHPs) offer the opportunity to collect longitudinal LN and peripheral blood samples during vaccine studies ( 6 , 7 ) . A factor critical to the detection of these immune responses is the accurate sampling of LNs that drain the injection site, which can be technically challenging due to the sporadic route of antigen trafficking in vivo.

There are multiple factors that can confound accurate sampling of vaccine draining LNs. In humans, IM vaccination into the deltoids sees antigen drain predominately to the axillary LN, with vaccine responses largely restricted to draining LNs in anatomic proximity to the injection site ( 8 -1 2 ) . Vaccination in the quadriceps is expected to drain predominately to the deep inguinal lymph nodes, which subsequently drains to the external iliac LN.

. In some individuals, lymphatic drainage from the proximal pelvic limb musculature may bypass the ipsilateral inguinal LNs, and drain directly into the iliac LNs

. While lymphatic drainage patterns of the thoracic limb are conserved in rodents and NHPs

, pelvic limb lymphatics predominately drain to the iliac LN, with inconsistent drainage to the ipsilateral inguinal LN . We and others observed substantial variability in vaccine induced responses when random LN in the draining region are sampled in NHPs ( 1 9 ) and humans, likely in part due to not sampling the responding LN. The ability to directly track antigen drainage following vaccination would substantially improve the accuracy of LN biopsies and assessment of GC immune responses, particularly in large animals such as NHPs. While this can be partially mitigated by substituting subcutaneous (SC) for IM vaccine administration ( 2 ) , the majority of human vaccines are given IM and it is desirable to maintain comparable delivery in pre-clinical animal models.

Previous studies have used tracking dyes to broadly identify LN drainage patterns in rodent and NHP animal models

.Tracking dyes have also been utilised clinically to identify sentinel LNs in cancer patients for biopsy or surgical resection

. However, the potential for mixing vaccine antigens with tracking dyes for long-term demarcation of draining LNs in vivo is untested. Here we show that that co-formulating influenza haemagglutinin (HA) or severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) spike immunogens with adjuvant and tattoo ink allows for the ready visual identification of draining LN without compromising downstream analyses of cellular and humoral immunity. We propose tattoo ink-based vaccine tracking is an effective method for the differentiation of vaccine-draining LNs during extended periods of time post-vaccination, and facilitates a more accurate quantification and phenotypic characterisation of vaccine-specific B and T fh cells in draining LNs, in both murine and NHP animal models.

Tracking dyes used in vaccinations should (i) visibly stain the draining LN with co-deposition of antigen, and (ii) not affect vaccine immunogenicity or downstream analyses, such as flow cytometric quantification of antigen-specific B GC and T fh populations. We first tested the impact of 3 candidate tracking dyes (Evans blue dye, India ink, and tattoo ink) on cell viability and autofluorescence in vitro. Human PBMC cultured in media with 0.05% Evans blue dye for 1 hour resulted in substantial cytotoxicity and loss of lymphocytes (Fig 1A) . In contrast, culture with 1% India ink or 1% tattoo ink did not affect lymphocyte viability (Fig 1B) . Despite the short duration of co-culture (1hr), incubation of PBMC with 1% India ink demonstrated alterations in cellular autofluorescence as measured by flow cytometry on channels off the blue, violet and UV lasers (Fig 1C) . Given tattoo ink demonstrated less autofluorescence (Fig 1C) , we proceeded to test the utility of tattoo ink co-formulated with vaccine antigens in mice.

C57Bl/6J mice were immunised IM in the right gastrocnemius and left quadriceps with A/Puerto Rico/8/1934 haemagglutinin (PR8-HA; 5μg) formulated with Addavax adjuvant and tattoo ink (0.5%). Based on previous studies

, antigens delivered to the right gastrocnemius will predominately drain to the right popliteal LN and the right iliac LN, with variable drainage to the right inguinal LN (Fig 1D) . Antigens delivered in the left quadriceps will predominately drain to the ipsilateral iliac LN, with variable drainage to the ipsilateral inguinal LN (Fig 1D) . Lymph nodes were harvested and assessed visually for ink staining 14 days post-vaccination. On the left quadricep side, non-draining LN (left popliteal and axillary LN) showed no evidence of ink uptake, with the left iliac LN consistently exhibiting ink uptake (Fig 1E, Table 1 ). In contrast, right popliteal and right iliac LNs exhibited obvious ink uptake by eye following right gastrocnemius injection, while absent in the right axillary LN (Fig 1E, Table 1 ). These observations are consistent with previous reports that lymphatics from the pelvic limbs drain into the iliac LNs in mice

. Dye labelling of inguinal LN was variable, with approximately 20% of the left inguinal LNs, and 60% of the right inguinal LNs labelled (Fig 1E, Table 1 ). This may reflect differences in lymphatic drainage patterns between proximal and distal muscle groups of the pelvic limb. Overall, our results indicate that tattoo ink can label draining LNs when administered in combination with antigen without substantially impacting lymphocyte viability and autofluorescence.

To assess the extent to which ink staining tracks with vaccine-induced GC activity in mice, we quantified both total B GC (B220 + IgD -GL7 + CD38 dim ) or HA-specific B GC cell frequency ( 2 8 ) in LNs with and without gross ink uptake (Fig 2A) . (Fig 2A, B) . These data indicate that the tattoo ink vaccine formulation does not hinder the identification of total or antigen-specific B GC by flow cytometry, and that ink-dyed draining LN are more likely to contain vaccine responses than suggesting that the degree of ink accumulation mirrors antigen load in the LN following vaccination.

While mice generally have only 1-2 LNs at each lymphoid site

, primates and humans commonly exhibit chains or clusters of 2-14 LNs

. This creates additional complexities when evaluating the adaptive immune response, as vaccine antigen may drain to a small subset of the LNs at a given site. We tested if sampling accuracy, and the characterisation of vaccine-elicited immune responses ex vivo, could be improved using tattoo ink to label vaccine-draining LNs in NHPs.

Pigtail macaques (Macaca nemastrina) were immunised in the right quadriceps with SARS-CoV-2 spike (100μg) formulated with monophosphoryl lipid A (MPLA) liposomal adjuvant ( 2 9 ) , and were boosted IM in the right and left quadriceps with SARS-CoV-2 spike (100μg) formulated with MPLA and tattoo ink (1.0%). Administration at these sites was expected to drain primarily to the left and right iliac LNs, with inconsistent drainage to the left and right inguinal LNs (Fig 3A) (

. Animals were additionally immunised in the right and left deltoids with human immunodeficiency virus-1 (HIV-1) fixed trimeric envelope protein (SOSIP) vaccines (100μg) formulated with MPLA and 1.0% tattoo ink (ink in the right deltoid only), with expected drainage to the axillary LNs (Fig 3A) (

. Animals were humanely euthanized 13-14 days after the second immunisation, and necropsies were performed to evaluate draining and nondraining LNs for the presence of tattoo ink.

Among the draining LNs, tattoo ink was visible in at least one LN within the right and left iliac chains in 7 of 8 animals. ( Table 2 ). Dye labelling of inguinal LNs was variable, with LNs containing tattoo ink being identified in the left and right inguinal lymphoid sites found in 1 of 8 and 4 of 8 animals, respectively (Fig 3 B and C, Table 2 ). Dye labelling was observed in the right axillary LNs in 8 of 8 animals, while dye labelling in the left axillary LNs was observed in 1 of 8 animals (Fig 3D-F, Table 2 ). Stochastic drainage to the ipsilateral inguinal LN aligns with previous reports ( 2 ) , and is important to consider as the inguinal LN is a common and readily accessible site for LN sampling via FNA or biopsy. No tattoo ink was grossly visible in any nondraining LN clusters; including the popliteal, para-aortic, mesenteric, mediastinal, tracheobronchial and submandibular LNs (data not shown). Tattoo ink was generally observed in only a single, or limited number of LN recovered from a given lymphoid site (Fig 3B-F) ,

suggesting the widely used practice of pooling all LN for immunological analysis could result in significant dilution of vaccine-specific responses and unpredictable effects on reported frequencies of antigen-specific B and T cell responses.

Longitudinal studies involving LN biopsies in macaques commonly sample the more readily accessible inguinal LN, to which antigen drainage is highly stochastic ( 2 ) . Among the 8 animals . However, long term stability in vivo, and the influence of dye on downstream immunological analysis is unclear. Prolonged labelling of vaccine draining LNs is a relevant consideration, as it allows serial sampling to evaluate changes in the immune response over weeks to months after initial vaccination. Visible dye staining in rats was reported 10-14 days after intraperitoneal administration of pontamine sky blue dye ( 2 2 ) . Tattoo ink, by design, is a stable, relatively inert compound that can persist in LNs for extended periods time. In humans, several case reports have noted tattoo ink being incidentally identified in draining LNs up to 30 years after the tattoo was originally placed

. The longevity of tattoo ink in draining LNs when formulated with vaccines still needs to be determined. However, our data demonstrates that draining LNs containing tattoo ink can be readily identified 2 weeks after administration in both mice and NHPs, with the likelihood of it persisting for considerably longer.

Published methods for identifying vaccine-draining LNs include administration of fluorescently labelled immunogens that can be identified by In Vivo Imaging Systems (IVIS)

, and administration of Tc 99 sulfur colloid that can be identified with a gamma probe ( 3 4 ) . Vaccines may also be administered in specific anatomical locations, or specific routes, to increase the likelihood of antigen drainage to a specific LNs. For example, injection in the SC flank in mice for selective drainage to the ipsilateral inguinal LN

, or SC immunisation in the anterior thigh for selective drainage to the inguinal and iliac LNs in NHPs ( 2 ) . The route of vaccine can impact the associated immune response; for example, reports of SC immunisation eliciting stronger neutralising antibody response compared to IM immunisation in NHPs)

.

Identification of vaccine draining LNs with tattoo ink provides a simple approach for active LN identification without requiring specialised equipment, while permitting IM vaccination routes of greatest clinical relevance for human vaccines.

One consideration with our proposed method is the potential for both endogenous and exogenous pigments to confound the identification of ink containing lymph nodes. Hemosiderin, and ironcontaining haemoglobin breakdown product, can accumulate in draining lymph nodes from congested, haemorrhagic, or inflamed areas . Similarly, carbon-containing, particulate debris can accumulate in tracheobronchial and mediastinal lymph nodes, following inhalation and drainage from the pulmonary tree 

Mouse studies and related experimental procedures were approved by the University of 

Full length influenza H1N1 A/Puerto Rico/8/34 hemagglutinin (PR8-HA)

, SARS-CoV-2 spike protein (S) ( 4 5 ) , and the receptor binding domain of the SARS-CoV-2 spike protein (RBD) F  r  e  q  u  e  n  c  y  o  f  t  a  t  t  o  o  i  n  k  l  a  b  e  l  l  i  n  g  i  n  m  u  r  i  n  e  L  N  s . 

The lymph node at a glance -how spatial organization optimizes the immune response

Rapid germinal center and antibody responses in non-human primates after a single nanoparticle vaccine immunization

Direct probing of germinal center responses reveals immunological features and bottlenecks for neutralizing antibody responses to HIV Env trimer

Quantification of residual germinal center activity and HIV-1 DNA and RNA levels using fine needle biopsies of lymph nodes during antiretroviral therapy

Normal human lymph node T follicular helper cells and germinal center B cells accessed via fine needle aspirations

Rhesus macaque B-cell responses to an HIV-1 trimer vaccine revealed by unbiased longitudinal repertoire analysis

Serial study of lymph node cell subsets using fine needle aspiration in pigtail macaques

Vaccine priming is restricted to draining lymph nodes and controlled by adjuvant-mediated antigen uptake

Cancer treatment and research

Deep lymphatic anatomy of the upper limb: an anatomical study and clinical implications

Axillary lymph node accumulation on FDG-PET/CT after influenza vaccination

Lymph node activation by PET/CT following vaccination with licensed vaccines for human papillomaviruses

The normal anatomy of the lymphatic system in the human leg

Three dimensional demonstration of the lymphatic system in the lower extremities with multi detector row computed tomography: A study in a cadaver model

Perforating and deep lymphatic vessels in the knee region: an anatomical study and clinical implications

The lymph system in mice

The lymphatics of Japanese macaque

Lymph node mapping in the mouse

Selfassembling influenza nanoparticle vaccines drive extended germinal center activity and memory B cell maturation

The lymphatic system in rhesus monkeys (Macaca mulatta) outlined by lower limb lymphography

Tracking the luminal exposure and lymphatic drainage pathways of intravaginal and intrarectal inocula used in nonhuman primate models of HIV transmission

Patterns of lymphatic drainage in the adult laboratory rat

Triple injection" lymphatic mapping technique to determine if parametrial nodes are the true sentinel lymph nodes in women with cervical cancer

How long will I be blue? Prolonged skin staining following sentinel lymph node biopsy using intradermal patent blue dye

Efficacy of methylene blue dye in localization of sentinel lymph node in breast cancer patients

Safety and technical success of methylene blue dye for lymphatic mapping in breast cancer

The predictive value of methylene blue dye as a single technique in breast cancer sentinel node biopsy: a study from Dharmais Cancer Hospital

Subdominance and poor intrinsic immunogenicity limit humoral immunity targeting influenza HA-stem

Safety evaluation of monophosphoryl lipid A (MPL): An immunostimulatory adjuvant

Tattoo pigment in an axillary lymph node simulating metastatic malignant melanoma

Tattoo pigment mimicking axillary lymph node calcifications on mammography

Tattoo pigment in axillary lymph node mimicking calcification of breast cancer

Targeting HIV Env immunogens to B cell follicles

Vaccine-draining lymph nodes of cancer patients for generating anti-cancer antibodies

Vaccine draining lymph nodes are a source of antigenspecific B cells

Immunization with tumor neoantigens displayed on T7 phage nanoparticles elicits plasma antibody and vaccinedraining lymph node B cell responses

Elicitation of robust tier 2 neutralizing antibody responses in nonhuman primates by HIV envelope trimer immunization using optimized approaches

Histopathology of the lymph nodes

Pathologic basis of veterinary disease

Common reactive erythrophagocytosis in axillary lymph nodes

Primary nodal anthracosis identified by EBUS-TBNA as a cause of FDG PET/CT positive mediastinal lymphadenopathy

Cleavage strongly influences whether soluble HIV-1 envelope glycoprotein trimers adopt a native-like conformation

Structure and immunogenicity of a stabilized HIV-1 envelope trimer based on a group-M consensus sequence

Flow cytometry reveals that H5N1 vaccination elicits cross-reactive stem-directed antibodies from multiple Ig heavy-chain lineages

Humoral and circulating follicular helper T cell responses in recovered patients with COVID-19

Fiji: an open-source platform for biological-image analysis

NIH Image to ImageJ: 25 years of image analysis

The authors would like to thank the staff at the Monash Animal Research Platform (MARP) andGippsland Field Station, including Irwin Ryan, for their assistance with the non-human primate study. We thank Robin Shattock (