Facebook link Linked In link You Tube link

FOCIS ePub Header

June 15, 2015

Editor: Andrew H. Lichtman, MD, PhD, Brigham & Women's Hospital
Editorial Board: Abul K. Abbas, MD, University of California, San Francisco | Carla J. Greenbaum, MD, Benaroya Research Institute | Andrew H. Lichtman, MD, PhD, Brigham & Women's Hospital

Highlights in Recent Literature | Clinical Immunology Highlights | Immunphenotyping | Clinical TrialsPDF Version Previous Issues

Highlights from Recent Literature

Regulating the Regulators: Shutting Down Regulatory T Cells with GARP-TGFβ1 Antibodies

A review of Julia Cuende et al. Monoclonal antibodies against GARP/TGFβ-1 complexes inhibit the immunosuppressive activity of human regulatory T cells in vivo. Science Translational Medicine (2015) 7, 274ra18. PMID: 25904740

Regulatory T cells (Treg) play critical roles in initiating and maintaining self tolerance but they can also interfere with tumor immunity and permit the persistence of chronic infections. One of the mechanisms by which Treg suppress other T cells is by presenting inactive TGFβ1 on the cell surface that becomes activated to immunosuppressive active TGFβ1 by unknown mechanisms. The authors demonstrate that the protein GARP plays a key role in TGFβ1 activation and that a monoclonal antibody that blocks this conversion inhibits Treg function.

  • GARP is a transmembrane protein expressed by Treg but not by other T cell subsets and it binds to latent TGFβ1 on the Treg cell surface.
  • The authors generated 31 new mouse and llama-anti human GARP antibodies and cloned these onto the human IgG1 backbone, creating humanized anti-GARP antibodies.
  • Two of these antibodies inhibited the production of active TGFβ1 by Treg, suggesting that GARP is directly involved in production of active TGFβ1 by human Treg.
  • These two inhibitory antibodies both mapped to a similar domain of the molecule, hGARP101-141, a part that is involved with TGFβ1 association on the cell surface.
  • The two inhibitory antibodies blocked Treg mediated suppression in in vitro assays and in a model of human anti-mouse xenogenic graft vs. host disease (GvHD).

By creating and studying antibodies against GARP, a protein that associates with latent TGFβ1 on the surface of Treg, the authors demonstrated that GARP is involved in the activation of TGFβ1 by Treg and that inhibition of this activity abrogates Treg mediated suppression in vitro and in an in vivo human anti-mouse GvHD model. Neutralizing anti-GARP antibodies are therefore a novel new class of Treg inhibitory therapeutics that may be valuable in the treatment of cancers and chronic infections.

Reviewed by Rachael A. Clark, MD, PhD, Brigham and Women's Hospital

Stem-like CD8T Cells Mediate Long-term Protection After Yellow Fever Vaccination

A review of Silvia A. Fuertes Marraco, et al. Long-lasting stem cell–like memory CD8+ T cells with a naïve-like profile upon yellow fever vaccination. Science Translational Medicine (2015) 7, 282ra48. PMID:25855494

The yellow fever (YF) vaccine is a live attenuated virus that confers lifelong protection in humans. Protection is thought to be mediated by CD8+ T cells but there are no studies on how immunity is maintained long term. The authors studied long term immune responses in vaccinated patients and found that a stem cell-like population of CD8+ memory T cells persisted and mediated immune protection for decades after vaccination.

  • The authors studied a cohort of 41 patients who had received the vaccine between .27 years and 35 years before the current study.
  • T cells specific for yellow fever were detected in 38 of 41 patients. Early after vaccination, effector cells dominated the response, with the frequency of these cells tending to decrease over time.
  • However, a separate population of yellow fever specific naïve-like CD8+ T cells were found to be stably maintained for more than 25 years and were capable of self renewal in vitro.
  • Phenotypic and gene transcription studies demonstrated that these YF-specific long-lasting CD8+ T cells were similar to but distinct from genuine naïve cells from unvaccinated donors and also that they resembled a recently described stem cell-like population of memory T cells (Tscm)

The YF vaccine is one of the few vaccines that induces lifelong immunity in humans and the mechanism behind this long term protection was previously unknown. The authors demonstrate that long-term protection against YF in vaccinated patients was mediated by a unique population of T cells that had some features in common with naïve T cells, although they were clearly distinct, and also in common with a recently described population of stem cell-like memory T cells. These studies are the first to identify these cells in humans, implicate them in long-lived immune responses and demonstrate that they can persist for decades. Further studies of this cell type may lead to better vaccination strategies that confer life-long immune protection.

Reviewed by Rachel A. Clark, MD, PhD, Brigham and Women's Hospital

It's Only Skin Deep?

A review of Watanabe et al. Human skin is protected by four functionally and phenotypically discrete populations of resident and recirculating memory T cells. Science Translational Medicine (2015) 279: 279ra39. PMID: 25787765

To better understand the components of the human cutaneous immune response, this study from the Clark lab extends earlier work on murine skin T cells and human cutaneous T cell lymphoma (CTCL) patients to characterize the populations of skin tropic non-recirculating (resident) memory T cells and recirculating T cells.


  • To characterize the human cutaneous T cell immune response, they began by optimizing a mouse model of human skin. They began by grafting human foreskin to a healthy mouse, followed by an infusion of human PBMCs. Of note, human foreskin was chosen as a model of T cell depleted skin, based on prior work. Allogeneic T cells were injected and they specifically infiltrated the grafted skin and created a dermatitis. Over time, populations of resident T memory cells (Trm) developed in the skin graft
  • Anti-CD52 (alemtuzumab) administration specifically depleted circulating T cells (in blood) and over time the recirculating population of skin T cells was also depleted, while sparing a subgroup of T cells.
  • They focused on the non-recirculating Trm and compared the T cells remaining after alemtuzumab therapy in human skin grafts in mice and in patients with cutaneous T cell lymphoma. Retained skin Trm in those settings were concordant, with elevated CD69 and a subset coexpressing CD103.
  • Trm were divided into two groups, CD103 positive and negative. To study these carefully, human adult skin was split using enzyme treatment into epidermal and dermal layers and studied the T cell populations in each layer of skin. The number and function of these T cell subsets were then compared. Interestingly, Trm made more cytokine than circulating T cells. In addition, CD103+ Trm had limited proliferation potential though increased cytokine production.
  • While keratinocyte contact was shown to be needed for CD103 expression in T cells, it was shown to be mediated, at least in part, by TGFb release.
  • Circulating skin homing T cells were also divided into two groups: skin tropic central memory T cells (CCR4+ CCR7+ and L-selectin+, or Tcm) and CCR7+/L-selectin negative (or migratory memory cells, Tmm).
    • These cells (Tcm and Tmm) were compared to Tem by CyTOF mass cytometry. Interestingly, cytokine profile of Tmm was between Tem and Tcm. In Tem, Th1, Th17 cytokines were highest and in Tcm, Th2 cytokines were highest.
  • To characterize Tmm in vivo, patients with CTCL were studied. Interestingly, atypical lesions correlated with presence of Tmm.

This group was able to delve deeper than previous studies into the nature and migration patterns of human cutaneous CD4+ T cell responses. They confirmed the presence of four groups of skin tropic human T cells (two recirculating populations and two resident populations). They employed clever tools, including using infant foreskin engrafted onto mice with subsequent PBMC infusion to create inflammation and thus foster the development of the appropriate T cell infiltrate in the graft. In order to fully treat skin diseases including CTCL and psoriasis, we need an increased understanding of the baseline pathways of cutaneous immunity to allow optimization of our therapies. In addition, this knowledge improves our understanding of a fundamental component of our immune system.

Reviewed by Sarah Henrickson, MD, PhD, Children's Hospital of Philadelphia

Interrogating the Pathogenicity of Anti-Myelin T Cells in MS

 A review of Cao et al. “Functional inflammatory profiles distinguish myelin-reactive T cells from patients with multiple sclerosis.” Science Translational Medicine (2015) 287: 287 PMID: 25972006.

While there are anti-myelin T cells in patients with MS, it has been hard to implicate these cells directly to disease pathogenesis as similar numbers of anti-myelin T cells have been found in healthy controls. What has not been possible to examine previously is whether there is a difference between the function of those cell populations in the two different patient groups. In this study, anti-myelin T cells are isolated from patients with MS and healthy controls and compared functionally (both by cytokine production and proliferation) and transcriptionally.


  • In order to better characterize the autoreactive T cells in MS patients and healthy controls, T cells were sorted into appropriate subsets (i.e. +/- CCR6+, using the CCR6+ cell population as enriched in Th17 and Tr1 cells) and then cultured in vivo with activating stimuli for two weeks. This yielded libraries of polyclonally stimulated cells. In this study, 13,324 T cell libraries from 23 patients and 22 controls were created.
    • T cells from each library were characterized for their antigenic specificity using myelin peptides and Candida. 
    • T cells from MS patients had greater proliferation and more inflammatory cytokines (IFNg, IL-17 and GM-CSF production) in response to myelin antigens when compared to T cells from healthy controls 
    • Principal components analysis was performed on T cell libraries from 13 MS patients and 13 controls. 
      • CCR6+ memory cells in MS patients had two signatures, one that had higher production of GM-CSF and one with higher production of IL-17 or IFNγ.
      • Interestingly, healthy controls had a population of IL-10 secreting (inhibitory) T cell libraries stimulated by myelin antigen
  • In order to investigate whether complex and varied functional responses were based on assays that polyclonally activated cells (or whether individual cells had various functional profiles), single cell cloning was performed on libraries with the most proliferation.
    • 144 clones were generated from 2 MS patients and 2 healthy controls (all HLA typed and DR4+, which has been shown to be an MS susceptibility locus) by sorting cells using HLA-DR4 tetramers loaded with peptides from MOG and PLP.
    • The clones were stimulated with autologous monocytes with DR4 peptides from MOG and PLP. 
    • Myelin responsive clones from MS patients had higher production of IL-17, GM-CSF and/or IFNγ.
    • Interestingly, 22 of 23 total IL-10 secreting clones isolated came from healthy controls.
  • Finally, RNA sequencing was performed on sorted populations of myelin peptide tetramer positive and negative CCR6+ CD4+ memory T cells from T cell libraries (from 5 MS patients and 4 healthy controls) which showed the greatest amount of proliferation.
    • GO enrichment analysis and GSEA were performed
      • 197 genes were found only in the MS patient tetramer positive samples
      • GSEA analysis revealed 305 gene sets enriched in expression in MS patient tetramer positive cells (versus tetramer negative cells), 
        • 19 were shared with tetramer positive healthy controls
        • 112 were unique to MS patient tetramer positive samples
      • The pathways in the mouse model of MS, experimental autoimmune encephalitis, were compared to the MS patient CCR6+ tetramer positive memory cells, with evidence of overlap with pathways previously characterized in pathogenic murine Th17 cells
      • They also compared to signatures from rat EAE, murine signatures from specific cell subsets (Th1, Th17, Th2) and human Th17 effector memory signatures
        • They found human Th17 effector memory, human Th2 and mouse Th17 enrichment in the patient samples
    • Leading edge analysis connected IL-8, IFNγ, CCL20 and CCL5 in MS patient tetramer positive cells.
      • When samples were clustered, it was noted that the resulting groups were based on disease status or tetramer status. 
    • Pathway analysis was performed and defined possible mechanistic networks that can be used to differentiate MS patient autoreactive T cells, with roles shown for Th17 and T follicular helper cell associated cytokines among others.

This study demonstrates that while autoreactive cells can be found in both affected patients with MS and healthy controls, there are functional differences at the levels of proliferation and cytokine production as well as transcriptional differences, in the autoreactive cells from patients versus healthy controls. The functional differences and transcriptional differences link mechanisms proposed in the major murine model of MS (EAE) as well as previous human studies at least in part to the autoreactive cells identified in the MS patients. In addition, protective signatures are seen in the autoreactive cells in healthy controls, which may help explain the presence of the cells without disease in healthy controls. There are many more studies that will need to be performed to further characterize these rare cells in their native state, but this study furthers our understanding of a longstanding paradox in the field.

Reviewed by Sarah Henrickson, MD, PHD, Children's Hospital of Philadelphia

Immune Effects of a Personalized Melanoma Vaccine 

A review of Carreno, B. M. et al. “A dendritic cell vaccine increases the breadth and diversity of melanoma neoantigen-specific T cells”. Science  348, 803–808 (2015). PMID:25837513

Melanoma cells harbor up to hundreds of somatic mutations that provide a putative source of tumor-specific neoantigens for personalized tumor vaccines. Although the induction of T cell immunity against tumor mutation-specific neoantigens has been reported, it remains unknown whether vaccination can augment neoantigen-specific T cell responses. Carreno et al. explore the effect of a personalized vaccine on the repertoire of neoantigen specific T cells. The authors sequenced the exomes of three patients with stage 3C cutaneous melanoma, then created and administered a personalized vaccine composed of seven neoepitope peptides pulsed on to autologous dendritic cells.

  • In all patients, a CD8 T cell-mediated response to at least one neoepitope was detected in pre-vaccination PBMCs.
  • Vaccination augmented pre-existing T cell responses, and also induced new responses in all patients.
  • Isolated ex vivo T cells specific for these neoepitopes were functional, and all but one neoepitope-specific T cell line could distinguish the mutant epitope from the wild-type.
  • Vaccination increased the frequency of pre-existing neoepitope specific T cells, and induced many new clones, increasing the diversity and size of the anti-tumor T cell repertoire.

Carreno et al. provide a valuable paradigm for the discussion of cancer neoepitopes; they classify pre-existing neoantigens as “dominant”, and vaccine-induced neoantigens as “subdominant”. The abundance of these subdominant neoantigens highlights the importance of vaccination in an immunotherapy regimen, as T cells targeting these epitopes are unlikely to arise from immunomodulatory treatment (anti-CTLA4, anti-PD1, etc.) alone. The authors have shown that personalized cancer vaccines are able to induce a strong anti-tumor immune response. The pipeline for the creation of these vaccines is currently too expensive and time consuming for widespread use; however falling sequencing costs and improved in silico methods for antigen identification are bringing personalized vaccine therapy closer to fruition.

Reviewed by Alexander Hopkins, Johns Hopkins University, Cellular and Molecular Medicine Program and Eric Lutz, PhD, Johns Hopkins University, Sidney Kimmel Cancer Center.

Identification of Human T Cell Receptors with Optimal Affinity to Cancer Antigens Using Antigen-negative Humanized Mice

 A review of  Obenaus, M. et al. “Identification of human T cell receptors with optimal affinity to cancer antigens using antigen-negative humanized mice.” Nature Biotechnology 33, 402–407 (2015). PMID: 25774714

High affinity tumor-reactive T cells that bind to unmutated tumor-associated antigens (TAAs) are subject to deletion via central tolerance in the thymus and thus are rarely found in the periphery. Those that do escape deletion are often of low affinity and are susceptible to peripheral tolerance mechanisms. Obenaus et al. use antigen negative, HLA-A2 and human T cell receptor (TCR) transgenic mice to isolate CD8+ T cells specific for the unmutated tumor antigens MAGE-A1 and NY-ESO. These T cells have improved cytolytic activity and interferon (IFN) γ production when compared to MAGE-A1- and NY-ESO-reactive CD8+ T cells isolated from human donors. Furthermore, these TCRs derived from mice exhibit no alloreactivity or functional antigen cross-reactivity, contributing to their potential utility in future adoptive cell therapies.

  • ABabDII mice, transgenic for human TCR-αβ gene loci and chimeric HLA-A2, were immunized with peptide corresponding to the HLA-A2 restricted epitope MAGE-A1278, which differs from the murine homolog by 6 out of 9 amino acids. CD8+ T cells reactive to MAGE-A1278 HLA-A2 multimers were detected after a boost 300 days later, indicating a functional memory response. RACE-PCR amplification identified 3 TCR clones, reflecting a single expanded TCR clone from each of 3 mice. Transduction of human T cells with these TCRs (T1367, T1405, and T1705) showed cytolytic activity against and IFNγ production in response only to HLA-A2+, MAGE-A1+ cells.
  • T1367 was tested against 114 known HLA-A2 restricted peptides and showed no off-target cross-reactivity. Alanine substitution identified the MAGE-A1 epitope recognized by T1367 and other peptides with similar motifs were identified in the human proteome and tested for T1367 recognition. Although a peptide derived from the SAMD9 gene could be recognized by T1367, it was only recognized at very high peptide concentrations and not on T cells, which naturally express SMAD9, demonstrating a lack of off-target toxicity.
  • When CD8+ T cells were transduced with MAGE-A1-specific TCRs isolated from human donors (hT27 and hT89) compared to the murine-derived TCRs, the CD8+ T cells expressing murine-derived TCRs showed greater cytolytic activity and IFNγ production. The human-derived TCRs were only stimulated by high concentrations of peptide loaded onto the T2 target cell line but only hT27 could recognize MAGE-A1+ cancer cells, whereas all murine-derived TCRs were stimulated by both the MAGE-A1+ cancer cells and T2 cells loaded with substantially lower concentrations of MAGE-A1278. When mice bearing large MAGE-A1+ tumors were treated with either ht27- or T1367-transduced T cells, only the T1367-transduced T cells were capable of inducing tumor regression.
  • Similar results were seen when ABabDII mice were immunized against NY-ESO: the murine-derived TCRs exhibited improved functionality when compared to TCRs isolated from human donors. The murine-derived TCRs produced more IFNγ and were capable of responding to lower antigen concentrations than human-derived TCRs.

Isolation of tumor antigen-specific T cells from patients has yielded low avidity T cells due to mechanisms of central and peripheral tolerance, which limits the available repertoire of tumor-reactive T cells. Obenaus et al. demonstrate that antigen-negative hosts expressing cognate human TCR-αβ gene loci and chimeric HLA molecules provide a system for identifying and expanding tumor-reactive T cells with increased functional avidities capable of inducing an anti-tumor response.

Reviewed by Heather Kinkead, Johns Hopkins University, Cellular and Molecular Medicine Program and Eric Lutz, PhD, Johns Hopkins University, Sidney Kimmel Cancer Center

Back to top

Highlights From Clinical Immunology, the Official Journal of FOCIS

B Cells and GCs in EAE

A review of Batoulis H. et al. “Central nervous system infiltrates are characterized by features of ongoing B cell-related immune activity in MP4-induced experimental autoimmune encephalomyelitis.” Clinical Immunology. (2015) 158, 47–58 PMID: 25796192

B cell aggregates and perhaps tertiary lymphoid organs (TLOs) form in the meninges of multiple sclerosis (MS) patients, and these have been correlated with more aggressive disease. Nonetheless, the contribution of B cells, antibodies, and TLOs to MS is poorly understood. Experimental autoimmune encephalomyelitis (EAE) is a demyelinating disease of mice that is widely used as model of human MS. EAE is induced by immunization with myelin proteins and strong adjuvant, but the disease is usually helper T cell dependent. The authors have previously described a B cell-dependent variation of the EAE model, induced by immunization of C57BL/6 mice with MP4, a fusion protein of myelin basic protein (MBP) and parts of proteolipid protein (PLP). In this report, the authors goal was to determine the function and disease impact of B ell infiltrates in the CNS in MP4-EAE. There main findings were:

  • Using immunohistochemistry, cells in cerebellar B cell aggregates were CXCL13 and CD10 positive, but negative for makers of B1 or regulatory B cells.
  • Cerebellar germline transcripts indicative of IgG2b and IgG3 class switching ere found in the majority of mice with MP4-EAE at 50 and 70 days after onset.
  • NexGen sequencing of cNDA from cerebellum and spleen of MP4-EAE mice indicated restricted VH usage and abnormally long, positively charged CDR3 loops, which have been associated with autoimmune disease.
  • Elispot analyses indicated antigen-specific B cell responses spread from the immunizing antigens to other myelin and neuronal antigens.

The authors state that these data are consistent with the development of tertiary lymphoid organs with active germinal centers that contribute to chronic B cell-dependent EAE. Further direct analysis of B cells within aggregates in their model, using laser capture, will be required to better characterize the TLO-like structures. The relationship between their MP4-EAE model, and T cell dependent EAE or human MS, where TLOs are in meninges and not in the cerebellum, remains to be clarified.

Reviewed by Andrew H. Lichtman, MD, PhD, Brigham and Women’s Hospital

TIM3 on NKs in HCV Infection

A review of Golden-Mason L. et al. “Hepatitis C viral infection is associated with activated cytolytic natural killer cells expressing high levels of T cell immunoglobulin- and mucin-domain-containing molecule-3.”Clinical Immunology (2015) 156, 1–8. PMID 25797693

Chronic viral infections, including hepatitis C virus (HCV) infection, are associated with exhausted T cells which express high levels of various inhibitory molecules, and are unresponsive to viral antigens. Among these inhibitory molecules is T cell immunoglobulin- and mucin-domain-containing molecule-3 (Tim-3). Previous studies have shown highTim-3 expression on human natural killer (NK) cells, but the expression and function of NK Tim-3 in chronic viral infections is not well characterized. In this study, the authors examined Tim-3 expression on NK cells in chronic hepatitis-C virus (HCV)-infected patients. They examined blood NK cells from 37 chronically HCV-infected subjects and 20 uninfected controls, using FACS, qRT-PCR, and functional assays. The findings include the following:

  • Tim-3 expression was increased on activated NK cells in patients with chronic HCV.
  • The high NK levels of Tim-3 was not reversed by effective antiviral therapy with IFN alpha and ribavirin.
  • IFN-alpha treatment causes a greater increase in NK TRAIL expression in Tim-3 high cells than Tim-3 low cells.
  • Tim-3 high cells in the INF-alpha treated patients showed greater lymphokine-activated killing activity, more viral control, and more degranulation compared toTim-3 low NK cells, although cytokine production was the same.

These results show that high Tim-3 expression on NKs in the setting of chronic HCV infection is not a marker of functional exhaustion, but rather of activation and enhanced cytotoxic activity. This is consistent with other published data, but in contrast to studies showing that NK Tim-3 marks exhausted NK cells in HBV infection. The authors advise that more studies on Tim-3 expression and function on different cell types and in different infections are needed to better understand how stimulating and blocking Tim-3 reagents might be considered therapeutically.

Reviewed by: Andrew Lichtman, MD, PhD, Brigham and Women’s Hospital

Back to Top

Human Immunophenotyping Update

Immunophenotyping of Human NK Cells 

Catherine A. Blish, MD, PhD, Stanford University School of Medicine 

Natural killer (NK) cells are innate lymphocytes that can rapidly respond to tumor or infected cells by killing (cytolysis) or by secreting cytokines. Unlike adaptive T and B lymphocytes that somatically rearrange antigen-specific receptors, NK cells express a variety of germline-encoded activating and inhibitory receptors on their cell surface. These receptors include the killer immunoglobulin-like receptors (KIR), C-type-lectin-like receptors (for example, NKG2A, NKG2C, and NKG2D), leukocyte immunoglobulin-like receptor subfamily b member 1 (LILRB1, also known as ILT-2 and CD85j), natural cytotoxicity receptors (NCRs, including NKp30, NKp44, NKp46, and NKp80), and signaling lymphocyte activation molecule (SLAM) family receptors (for example, 2B4, NTB-A). NK cells also express a variety of adhesion molecules such as CD2 and DNAM-1 that influence their function. These receptors allow NK cells to recognize ‘altered self’ on virus-infected, malignant or stressed cells.

NK cells rely on combinatorial signaling from this diverse array of receptors (1). The inhibitory receptors, which include inhibitory KIR (e.g., KIR3DL1, KIR2DL1, KIR2DL2, KIR2DL3, KIR2DL5), LILRB1, and NKG2A, are specific for self-HLA class I molecules, and provide an “all clear” or “back off” signal to the NK cell. Engagement of these receptors by HLA dampens the NK cell response. However, engagement of activating receptors, which recognize a variety of stress-related molecules and pathogen- or tumor-derived ligands, activates the NK cells to kill or secrete cytokines by indicating that the cell is a threat. Since the ultimate outcome of the interaction between an NK cell and a target cell is determined by the combinatorial effects of these inhibitory and activating signals, NK cell phenotype and function are closely and uniquely linked.

This vast array of receptors has presented a challenge for the immunophenotyping of NK cells. There has been increasing recognition that the NK cell receptor repertoire, as determined by the expression patterns of this array of receptors, is incredibly complex (2,3). Further complicating this matter is the challenge of determining which cells are actually NK cells, because unlike a B or a T cell, there is no single receptor that defines the subset. In the classic definition, NK cells are broken down into two subsets based on expression of CD56 and CD16 (the FcγRIII). In the peripheral blood, the vast majority (~90%) of NK cells express CD56 at low levels and also express CD16. These CD56dimCD16+ NK cells are considered mature NK cells that are highly effective in cytolysis. The more rare CD16brightCD16- NK cells are thought to be relatively immature NK cells that are specialized for cytokine secretion (4). However, this classification ignores many subsets, including intermediate populations such as rare CD56brightCD16+ NK cells, or the CD56-CD16+ NK cells that are particularly prominent during chronic viral infections (5). Thus, NK cells are best defined as much as what they are not (not B cells, T cells, monocytes, or dendritic cells) as by what they are (lymphocytes with cytolytic activity or cytokine secretion that express CD56 and/or CD16). Finally, within these populations there are a vast array of different subsets based on expression of other NK cell receptors, whose functional significance we are only beginning to unravel. In fact, based on combinatorial expression of 28 NK cell receptors, Horowitz et al. estimated that each individual has between 6,000-30,000 unique NK cell phenotypes, and more than 100,000 subsets in a small population of 22 individuals (3).

With this framework in mind, the first step in immunophenotyping is to carefully identify the NK cell subset. While the most common approach to do this is to identify CD3-CD56+ NK cells, as discussed above, this approach fails to identify CD56- NK cells which are present at varying levels in human subjects. In fact, CD56 expression alone is not particularly specific to NK cells (6), as it is expressed on activated T cells and NKT cells. Of all the NK cell markers, NKp46 is the most specific to NK cells, but like CD56, it is not expressed on all NK cells (6). Thus, the best approach to identify NK cells is to perform serial negative gating.

In standard fluroresence cytometry, a simple approach is to gate first on lymphocytes by forward and side scatter. Depending on the number of channels available, either a single “dump” channel or sequential negative gating should be performed on CD3, CD19 and/or CD20, and CD14 and/or CD33. Then NK cells can be identified as the remaining cells that express CD56 and/or CD16. As monocytes, in particular, can express CD14 and/or CD33 at low levels, this approach can result in some monocytes within the NK cell gate. This can be avoided by adding a stain for CD7, as NK cells and T cells, but not monocytes, express CD7 (7). Thus, NK cells are CD3-CD19-CD20-CD33-CD14-CD7+ cells that express either CD56 or CD16. In mass cytometry, or CyTOF, more channels are available; however, there is no forward or side scatter to help identify lymphocytes vs. monocytes. Thus, in addition to the scheme recommended above, additional gating to avoid CD56-HLA-DRbright cells and LILRB1bright cells will help to further decrease the number of monocytes inadvertently included in the NK cell gate (6).

Once NK cells have been identified, antibodies to a variety of NK receptors or functions can be used. Beziat and colleagues have recently reported a detailed flow cytometry method to identify KIR gene expression patterns in peripheral blood mononuclear cells (8). CyTOF, with it ability to identify up to 42 parameters simultaneously, is currentlyt the most comprehensive platform to identify expression patterns of multiple classes of NK cell receptors, including KIRs, C-type lectin-like receptors, LILRB1, natural cytotoxicity receptors, and adhesion molecules (3,6). The specific receptors queried should be tailored to the question.

With the new attention to the potential of harnessing NK cells in immunotherapeutic approaches(9,10), it is becoming increasingly apparent that we need to identify which subsets of NK cells should be targeted. For this to be effective, stringent efforts to identify NK cells by flow cytometry, with particular attention to avoiding monocyte inclusion in NK cell gates, will be critical. Given the vast array of surface receptors expressed by NK cells, highly parametric flow cytometry techniques, including new use of highly intense fluorescent dyes (11), or the use of the highly parametric mass cytometry platform (12), will be central to these efforts.


  1. Vivier E, Raulet DH, Moretta A, Caligiuri MA, Zitvogel L, Lanier LL, Yokoyama WM, Ugolini S. Innate or Adaptive Immunity? The Example of Natural Killer Cells. Science. 2011 Jan 6;331(6013):44–9.
  2. Di Santo JP. Functionally distinct NK-cell subsets: Developmental origins and biological implications. Eur. J. Immunol. [Internet]. 2008 Nov;38(11):2948–51. Retrieved from: http://doi.wiley.com/10.1002/eji.200838830
  3. Horowitz A, Strauss-Albee DM, Leipold M, Kubo J, Nemat-Gorgani N, Dogan OC, Dekker CL, Mackey S, Maecker H, Swan GE, M DM, Norman PJ, Guethlein LA, Desai M, Parham P, Blish CA. Genetic and Environmental Determinants of Human NK Cell Diversity Revealed by Mass Cytometry. Sci Transl Med. 2013 Oct 23;5(208):208ra145–5.
  4. Freud AG, Caligiuri MA. Human natural killer cell development. Immunol Rev. 2006 Dec 1;214(1):56–72.
  5. Björkström NK, Ljunggren H-G, Sandberg JK. CD56 negative NK cells: origin, function, and role in chronic viral disease. Trends Immunol. 2010 Nov 1;31(11):401–6.
  6. Strauss-Albee DM, Horowitz A, Parham P, Blish CA. Coordinated regulation of NK receptor expression in the maturing human immune system. J Immunol. 2014 Nov 15;193(10):4871–9. PMCID: PMC4225175
  7. Milush JM, Long BR, Snyder-Cappione JE, Cappione AJ, York VA, Ndhlovu LC, Lanier LL, Michaelsson J, Nixon DF. Functionally distinct subsets of human NK cells and monocyte/DC-like cells identified by coexpression of CD56, CD7, and CD4. Blood. 2009 Nov 26;114(23):4823–31.
  8. Béziat V, Traherne J, Malmberg J-A, Ivarsson MA, Björkström NK, Retière C, Ljunggren H-G, Michaëlsson J, Trowsdale J, Malmberg K-J. Tracing dynamic expansion of human NK-cell subsets by high-resolution analysis of KIR repertoires and cellular differentiation. Eur. J. Immunol. 2014 May 7;44(7):2192–6.
  9. Keener AB. Natural killers:. Nat Med. Nature Publishing Group; 2015 Mar 1;21(3):207–8.
  10. Childs RW, Carlsten M. Therapeutic approaches to enhance natural killer cell cytotoxicity against cancer: the force awakens. Nature Publishing Group. Nature Publishing Group; 2015 May 22;:1–12.
  11. Kling J. Cytometry: Measure for measure. Nature. 2015 Feb 19;518(7539):439–43.
  12. Bendall SC, Nolan GP, Roederer M, Chattopadhyay PK. A deep profiler's guide to cytometry. Trends Immunol. 2012 Jul 1;33(7):323–32. PMCID: PMC3383392

Back to top

Selected Recent Clinical Trial Results

Randomized Trial of Peanut Consumption in Infants at Risk for Peanut Allergy

Clinical Trial: George Du Toit, Graham Roberts, Peter Sayre, et.al. “Randomized Trial of Peanut Consumption in Infants at Risk for Peanut Allergy.” New England Journal of Medicine 2015;372(9):803-813

Disease: Peanut allergy 

Intervention: Consumption of peanuts or avoidance of peanuts until age 5 in infants with eczema and/or egg allergy.

Study design:

  • 640 infants with history of severe eczema and/or egg allergy in UK ages 4-11 months (mean 7.8 months) randomized to consume or avoid peanuts until age 60 months
  • Infants with strongly positive skin prick test to peanuts were excluded.
  • Infants were placed into two cohorts: Those with negative (n=542) or weakly positive (n=98 ) skin prick test.
  • Open-label design
  • All participants had a baseline feeding challenge and were instructed to avoid peanuts (regardless of their randomization assignment) if their feeding challenge was positive, but were included in the ITT analysis. 
  • Consumption group instructed to consume at least 6 g of peanut protein/week (equivalent of about 24 peanuts) until age 60 months
  • Primary outcome was the proportion of participants with peanut allergy at 60 months as measured by an oral food challenge. 


  • In the cohort with a negative baseline skin prick test, 13.7% of avoidance group and 1.9% of consumption group were allergic to peanuts, an 86.1% relative reduction (P<0.001)
  • In the cohort with a positive baseline skin prick test, 35.3% of avoidance group and 10.6% of the consumption group were allergic to peanuts, an 70.0% risk reduction (P=0.004)

Why the Trial is of Interest to the Broader FOCIS community:

The prevalence of peanut allergies in Western countries has doubled in the past 10 years and represents a significant lifestyle and financial burden to individuals and families (1). The mainstay of treatment has been avoidance of peanuts when possible, along with preparation for and treatment of accidental exposures. However, there has been limited evidence to guide prevention of peanut allergy. In 2000, the American Academy of Pediatrics (AAP) published guidelines advising parents to delay the introduction of peanuts to children at high risk of allergies until age 3. In 2008, the AAP revised its guidelines, stating that there was no evidence that delayed introduction of particular foods beyond 4-6 months (the current age at which parents are advised to introduce solid foods) would prevent allergies. The 2008 AAP revision was supported by increasing observational data suggesting that early and sustained oral exposure to peanut protein could reduce the risk of clinical peanut allergy. In particular, a 2008 study conducted by the same investigators as the reviewed study compared Jewish children raised in Israel, where peanut protein is introduced at a median age of 7 months, to Jewish children raised in the UK, where peanut consumption in infancy is discouraged. It was found that school age children in the UK have 10-fold higher rate of peanut allergy than school age children in Israel (2). The reviewed study, the Learning Early About Peanut Allergy (LEAP) Study was launched from the 2008 study and is the first randomized trial to demonstrate that earlier and sustained exposure to peanut protein may prevent peanut allergies in at-risk individuals.
The results of the LEAP study demonstrate that early oral exposure to allergens results in immune tolerance even amongst the cohort of children who were already demonstrating some peanut sensitivity (those with a weakly positive skin prick test). Moreover, the LEAP-On study (Persistence of Oral Tolerance to Peanut) will address the important question whether individuals will remain allergy-free after prolonged cessation of oral peanut exposure. In contrast to this terrific result in allergy, no clear successes have been yet reported using oral antigen therapy in autoimmune disease. An interesting question is whether mimicking this timing, delivery method, and dosing in autoimmunity would result tolerance to autoimmune disease as well.


  1. Prescott SL, Pawankar R, Allen KJ, et al. World Allergy Org J 2013;6:21.
  2. Du Toit G, Katz Y, Sasieni P, et al. J Allergy Clinic Immunol 2008;122:984-91.

Submitted by Sandra Lord, MD, Benaroya Research Institute
Edited by Carla J. Greenbaum, MD, Benaroya Research Institute

Effects Of High-dose Oral Insulin on Immune Responses in Children at High Risk for Type 1 Diabetes: The Pre-point Randomized Clinical Trial

Clinical Trial: Ezio Bonifacio, Anette-G Ziegler, Georgeanna Klingensmith, et. al, Effects of High-Dose Oral Insulin on Immune Responses in Children at High Risk for Type 1 Diabetes: the Pre-POINT Randomized Clinical Trial. JAMA, 2015;313(15):1541-1549

Disease: Islet autoimmunity/Type 1 diabetes (TID)

Drug: Oral insulin

Study design:

  • Double-blind, placebo-controlled, dose-escalation, phase 1/2 clinical pilot study in Germany, Austria, UK, US.
  • 25 autoantibody (insulin, GAD65 and IA-2) negative children ages 2-7 years with a family history of type 1 diabetes and HLA class II susceptibility genotypes.
  • Randomized to receive placebo (n=10) or oral insulin (n=15) once daily for 3-18 months. Oral insulin subjects were randomized to one of 5 different dosing blocks, each block with 3 subjects. Dosing schedules were such that each dosing cohort (2.5 mg, 7.5 mg, 33.5 mg, 67.5 mg) had 6 subjects each. This is depicted in table 1:

Table 1:


  • Outcome measures:
    • Primary: Positive response to any one of the following: serum IgG low-affinity binding to insulin; salivary IgA binding to insulin; CD4+ T cell proliferative responses to insulin
    • Secondary: measurement of pro-regulatory FOXP3 gene signature and pro-inflammatory IFNγ signature in T cells with a response to insulin.


  • Immune responses: Only one subject (at dose 22.5) developed a Salivary IgA response. In the high dose group (67.5 mg), 3/6 developed IgG insulin antibodies and 2 others developed antigen specific CD4+ T cells detected; thus 5/6 subjects in this group had an immune effect of receiving therapy. This was in contrast to 2/10 in placebo group, and 1-2/6 in each of the other dosing groups (p=.02 for trend).
  • Comparing the gene signatures of antigen specific cells (to insulin and proinsulin) pooled from these antibody negative subjects treated with oral insulin to the signatures of antigen specific cells pooled from antibody positive subjects from the BABYDIET study suggested a more pro-regulatory gene signature in those receiving oral antigen therapy.
  • No adverse events.

Why the Trial is of Interest to the Broader FOCIS Community:

The Pre-POINT study was a dose-finding and feasibility study in anticipation of the phase 2 POINT study to determine if oral insulin can delay or prevent TID in those at high risk based on HLA haplotype and family history. The study aimed to correlate specific oral insulin dosing in this islet antibody negative group with desirable and expected immune responses. The results support the plan to use the higher dose in the fully powered POINT trial. Nonetheless, this pilot study highlights the difficulties in making such judgements; few responses were seen overall and whether these are salutatory will depend on the definitive study.

The POINT study and other oral tolerance studies aim to determine whether early, repeated exposure to self-antigens will promote immune tolerance. This concept has gained traction in clinical allergy, where trials have demonstrated that such exposure can prevent allergy (1), but trials have been less convincing in autoimmunity. In TID, previous trials of antigen therapy (including nasal insulin, parenteral insulin, low-dose oral insulin, GAD65, Diapep277) in both the at-risk and new-onset populations have not met their primary endpoints (2-7). The difficulties in conducting such studies and their long duration emphasize the need to select the “right” dose. Two other studies are underway using insulin as antigen therapy in antibody positive relatives; INIT II is testing nasal insulin 440 IU weekly, and Diabetes TrialNet is testing oral insulin at a dose of 7.5 mg daily. The TrialNet study is the follow-on to the Diabetes Prevention Trial oral insulin study in which post-hoc analysis suggested a benefit (3). Selection of dose in these studies anticeeded the technologies used in the Pre-POINT study. While trial results from the TrialNet study will not be available for another year, samples will be assessed for mechanistic studies akin to those used in Pre-POINT to determine whether the dose chosen had an immune effect in this population. In addition to choosing dose, choosing when to administer oral antigen therapy in the course of the disease is another question. The POINT study will enroll antibody negative, genetically at-risk children. In this group, antibodies generally develop very early: 64% of those who develop T1D before puberty will have antibodies by age 2 and 95% by age 5 (8-9). Since the mean age of those in the Pre-Point study was 5 years old, it is possible that these children had an underlying regulatory tendency. The TrialNet and INIT II study enrolls relatives with 2 or more antibodies. These individuals have “islet autoimmunity”. While only ~40% are likely to develop T1D in 5 years, recent data emphasizes that essentially all of these individuals will eventually progress to clinical disease. Much will be learned from comparing trial results and mechanistic assays in using oral antigen therapy to treat before or after antibodies are present.


  1. George Du Toit, et.al. N Engl J Med 2015;372(9):803-8133
  2. Skyler JS et al. Diabetes Care 2005 May: 28(5): 1068-76
  3. Diabetes Prevention Trial-Type 1 Diabetes Study Group: Effects of insulin in relatives of patients with type 1 diabetes mellitus. N Engl J Med 345: 1685–1691, 2002
  4. Wherrett DK et al. Lancet. 2011 Jul 23;378(9788):319-27
  5. Näntö-Salonen K et al. Lancet. 2008 Nov 15;372(9651):1746-55
  6. Schloot NC et al. Clin Immunol. 2013 Dec;149(3):307-16
  7. Ziegler AG et al. JAMA. 2013. Jun 19;309(23):2473-9
  8. Parikka V et al. Diabetologia. 2012 Jul;55(7):1926-36

Submitted by Sandra Lord, MD, Benaroya Research Institute
Edited by Carla J. Greenbaum, MD, Benaroya Research Institute

Mongersen, an Oral Smad7 Antisense Oligonucleotide, and Crohn’s Disease

Clinical Trial: Monteleone G., Neurath M., Ardizzone, et al. Monbersen, an Oral SMAD7 Antisense Oligonucleotide, and Crohn’s Disease. New England Journal of Medicine 2015:372:1104-13

Disease: Active Crohn’s Disease (CD)

Drug: Mongersen, an oral SMAD7 Antisense oligonucleotide. SMAD7 acts as an inhibitor of the immunosuppressive cytokine TGF-β1. Mongerson targets ileal and colonic SMAD7, via an anti-sense mechanism that facilitates SMAD7 messenger RNA degradation.

Study design:

  • 166 randomized, ages 18-75 years, moderate-severe active CD (as defined by Crohn’s Disease Activity Index/CDAI score of 220-400 on a scale 0-600)
  • 2 weeks of treatment, evaluated at days 15, 28, 84
  • 1:1:1:1 randomization: placebo, 10, 40, 160 mg doses
  • Primary outcome: clinical remission at day 15 (CDAI score < 150) and maintenance of remission for at least 2 weeks
  • Secondary outcome: clinical response (reduction of CDAI sore of ≥ 100 points) at 4 weeks.


  • Primary outcome, clinical remission: 65% in 160 mg group; 55% in 40 mg; 12% in 10 mg; 10% in placebo (P<0.001)
  • Secondary outcome, clinical response: 72% in 160 mg (P<0.001); 58% in 40 mg (P<0.001); 37% in 10 mg (P=0.04); 17% in placebo
  • 102/166 (61.4%) with elevated C reactive protein (CRP) at screening. Response rates similar among patients with elevated vs normal baseline CRP.
  • No correlation between normalization of CRP levels and improvement of CDAI score
  • No identified safety issues

Why the Trial is of Interest to the Broader FOCIS Community:

Mongersen belongs to a promising new category of therapy for Crohn’s Disease (CD) and other diseases; namely, antisense oligonucleotide therapy. Antisense oligonucleotides bind to messenger RNA, preventing gene translation and effectively turning the gene off. Antisense therapies have been studied as potential drugs for infections, cancers, inflammatory disorders, and genetic disorders. To date, mongersen is the third antisense drug to be approved by the FDA. Fomivirsen is approved for treatment of cytomegalovirus retinitis, and mipomersen for treatment of homozygous familial hypercholesterolemia.

There is an unmet need for additional therapies for CD, as one-third of patients with CD do not respond to currently available therapies, including anti-TNFα antibodies (infliximab, adalimumab, certolizumab pegol), anti-integrin antibodies (vedolizumab, natalizumab) and immunomodulating drugs (azathioprine, 6-mercaptopurine, methotrexate). In addition, the efficacy of TNFα therapies decreases over time, there is no evidence that the TNFα therapies reduce overall disease progression, and there is a risk of infection and malignancy with long term usage. In CD, TGF-β1 intracellular signaling is blocked by the SMAD7 protein (1;2). TGF-β1 has several anti-inflammatory/pro-regulatory effects, including: inhibition of effector T cell proliferation and differentiation, induction of regulatory T cells, reduced macrophage activation and reduced dendritic-cell maturation (3). Mongersen is a SMAD7 antisense oligonucelotide which facilitates mRNA degradation and decreases SMAD7 to increase TGF-b signaling. It provides specific targeting to the terminal ileum and right colon (the most commonly affected areas in CD) via a pH-dependent coating of the tablet. In this phase 2 trial, participants treated with mongersen experienced 2 week remission rates of 65% and 55% with the 2 highest doses, compared to 10% in the placebo group. By comparison, reported remission rates seen in the large trials of TNFα inhibitors in moderate-severe CD range are 32.5-36% (4;5), and remission rates seen with vedolizumab (an LPAM1 antibody, which selectively blocks gut integrin activity) are 14.5-39% (6). Furthermore, despite only 2 weeks of treatment, the majority of participants (62% and 67% in the 40 mg and 160 mg groups respectively) who were in remission after 2 weeks remained in remission after 12 weeks of follow-up. This contrasts with the rapid relapse typically seen after withdrawal of other anti-inflammatory agents.

There are some important caveats to consider. There was limited evaluation of objective markers of disease activity. Namely, there were no required endoscopic evaluations, either at entry or endpoint; therefore, it’s not known whether improvements in CDAI scores correspond to mucosal healing. The CDAI is somewhat subjective: of the 8 criteria, 3 are patient-reported. Moreover, fecal calprotectin, a stool test for colonic inflammation, was not measured. CRP was measured, but interestingly, there was no correlation between clinical response and normalization of CRP. As the investigators suggest, there may be lag between clinical response and normalization of CRP. In addition, treatment duration and follow up (2 weeks and 12 weeks, respectively) were relatively brief, which should impact any conclusions that are made about safety and efficacy. Although there were no significant safety issues identified, longer-term data is needed. There was one instance of intestinal obstruction, which is a known complication of CD, but given that TGFβ1 can promote fibrosis (7;8), this might be of special interest.

Mongersen’s specific targeting to the terminal ileum and right colon is both a strength and a limitation; although it does not appear to have systemic side effects, it also may not be effective for extra-intestinal manifestations of CD or for perianal disease, the latter of which affects about half of patients with CD.

Despite these caveats, mongersen may represent an entirely new, safe, and effective category of Crohn’s Disease therapy. Antisense therapies have been studied for potential treatment of infections and genetic disorders. To date, mongersen is the third antisense drug to be approved by the FDA. Fomivirsen is approved for treatment of cytomegalovirus retinitis, and mipomersen for treatment of homozygous familial hypercholesterolemia. Follow up studies are needed, to include both longer duration of follow up and additional objective measures of disease activity.


  1. Boirivant M, et al. Gastroenterology 2006;131:1786-1798
  2. Monteleone G, et al. J Clin Invest 2001;108:601-609
  3. Boirivant M et al. Gastroenterology 2006;131:1786-1798
  4. Colombel JF et al. N Engl J Med 2010;362:1383-1395
  5. Hanauer SB, et al. Gastroenterology 2006;130:323-333
  6. Sandborn WJ et al. N Engl J Med 2013;369:711-721
  7. Fichtner-Feigl S et al. Nat Med 2006;12:99-106
  8. Medina C et al. J Pathol 2011;224:461-472

Submitted by Sandra Lord, MD, Benaroya Research Institute
Edited by Carla J. Greenbaum, MD, Benaroya Research Institute

Back to top

Click here to send your comments and suggestions for future Translational Immunology Updates.

Translational Immunology Update is the official publication of the Federation of Clinical Immunology Societies. Scope: Translational Immunology Update aims to provide a knowledge-sharing forum for basic, clinical and translational scientists around all aspects of human immunology including normal and cross-disease physiology, communication between laboratory-based and clinically-based scientists, the lack of negative results that are published and the difficulty for readers to keep up with important literature.

Federation: All members of FOCIS receive Translational Immunology Update. For more information about FOCIS, address correspondence to: Federation of Clinical Immunology Societies, N83 W13410 Leon Road, Menomonee Falls, WI 53051, Phone: (414) 359-1670, Fax: (414) 359-1671, Email: This email address is being protected from spambots. You need JavaScript enabled to view it.
Advertising: Inquiries concerning advertising should be addressed to Sarah Martis, FOCIS Executive Director.
Permissions: Copyright © 2013 Federation of Clinical Immunology Societies
Apart from any fair dealing for the purposes of research or private study, or criticism or review, as permitted under the Copyright, Designs and Patent Act 1988, this publication may be reproduced, stored or transmitted, in any form or by any means, only with the prior permission in writing of the publishers, or in the case of reprographic reproduction, in accordance with the terms and licenses issued by the Copyright Licensing Agency.
Translational Immunology Update is published bimonthly by the Federation of Clinical Immunology Societies. You may opt-out of receiving the publication at any time by clicking the unsubscribe link in the email.