Expression of lectin-like transcript-1 in human tissues

Background: Receptor-ligand pairs of C-type lectin-like proteins have been shown to play an important role in cross talk between lymphocytes, as well as in immune responses within concrete tissues and structures, such as the skin or the germinal centres. The CD161-Lectin-like Transcript 1 (LLT1) pair has gained particular attention in recent years, yet a detailed analysis of LLT1 distribution in human tissue is lacking. One reason for this is the limited availability and poor characterisation of anti-LLT1 antibodies. Methods: We assessed the staining capabilities of a novel anti-LLT1 antibody clone (2H7), both by immunohistochemistry and flow cytometry, showing its efficiency at LLT1 recognition in both settings. We then analysed LLT1 expression in a wide variety of human tissues. Results: We found LLT1 expression in circulating B cells and monocytes, but not in lung and liver-resident macrophages. We found strikingly high LLT1 expression in immune-privileged sites, such as the brain, placenta and testes, and confirmed the ability of LLT1 to inhibit NK cell function. Conclusions: Overall, this study contributes to the development of efficient tools for the study of LLT1. Moreover, its expression in different healthy human tissues and, particularly, in immune-privileged sites, establishes LLT1 as a good candidate as a regulator of immune responses.

Receptor-ligand pairs of C-type lectin-like proteins have been shown to play an important role in cross-talk between lymphocytes and in immune responses within tissues. Three examples have been well characterised in humans. These are the NKp65-Keratinocyte associated C type lectin (KACL), the NKp80-Activated induced C-type lectin (AICL) and the CD161-Lectin-Like Transcript 1 (LLT1), which are involved in skin immunobiology 1 , cross-talk between Natural Killer (NK) cells and monocytes 2 and modulation of T, NK and B cell immune responses 3-6 , respectively. Amongst these, the CD161-LLT1 pair has been the focus of attention of several recent studies 6-10 . LLT1 has been described as a multi-functional protein 11 , and to fully elucidate the functional consequences of its interactions with its receptor, CD161, a comprehensive characterisation of LLT1 distribution is needed. The current published literature presents inconsistencies, which may partially be due to the activation state of the cells tested and the different anti-LLT1 antibodies used. Indeed, LLT1 has been shown to be upregulated upon different forms of activation 6,12-15 .
Tissues within the body display varying antigenic profiles, and the expression of specific molecules is involved in the maintenance of tissue function. Tissue grafts placed in particular anatomical structures can avoid rejection for long periods of time 16 . This observation led to the notion of immune-privilege, believed to be an evolutionary adaptation to protect essential organs from harmful inflammatory responses. At first, it was thought that antigens did not have access to immune-privileged sites, thus avoiding a response. However, more recent evidence suggested that the maintenance of immuneprivilege relies on active rather than passive mechanisms 17,18 . Some examples include: a lack of lymphatic drainage, low expression of MHC class I molecules, local production of immunosuppressive cytokines, as well as enhanced expression of inhibitory surface molecules 19 . Immunologically privileged sites include the brain, the eyes, the placenta, the fetus and the testes. Although there has been abundant research regarding the mechanisms behind effective suppression of inflammatory responses in immune-privileged structures, further studies are required to fully elucidate and understand them 20 .
The main aim of this study was to broadly characterise the expression of LLT1 within the human body. We screened a wide variety of human cell types and tissues using our novel monoclonal antibody, clone 2H7, and described LLT1 expression in circulating B cells and monocytes. The presence of LLT1 could also be observed in B cells in tonsils, as previously described 6,9,21 , but not in Kupffer cells in the liver or alveolar macrophages in the lung. Furthermore, LLT1 could be detected in several healthy human tissues, but it was remarkably prevalent in immune-privileged sites, such as brain, placenta and testes. We also confirmed the previously described phenomenon that LLT1 inhibits NK cell function 4,5,14,22 .
Overall, the current study contributes to the development of effective tools for the study of LLT1. We characterised the strong expression of this C-type lectin in B cells, monocytes and immuneprivileged tissues; thus, postulating a role for LLT1 in cross talk between lymphocytes and immune tolerance.

Tissues
A series of normal paraffin-embedded human tissues comprising samples of tonsil, liver and lung were obtained from Proteogenix.
Tonsils were also obtained following routine tonsillectomy from the ENT Department at the John Radcliffe Hospital, Oxford. Ethical approval was obtained from the John Radcliffe Hospital, and written informed consent was obtained from all subjects.
Formalin-fixed, paraffin-embedded healthy and tumour tissue arrays were obtained from AMS Biotechnology. FACS analysis was performed on Miltenyi Biotec MACSQuant cytometer and analyzed with FlowJo Version 9.6.2 software (TreeStar).

Immunohistochemistry
Tissue deparaffinisation was performed using Histo-Clear (National Diagnostics) and ethanol (Sigma Aldrich; 100%, 90% and 70%). Heat mediated antigen retrieval was achieved using Dako target retrieval solution (Dako). Endogenous peroxidase activity was blocked using 3% H 2 O 2 (5 min × 2; Alfa Aesar) and 0.1% sodium azide (15 min; Sigma Aldrich) in water. Non-specific binding was blocked by incubating the sample for 30 min at RT with 0.5% blocking reagent (PerkinElmer) in PBS. The 2H7 mAb or IgG2A isotype control (R&D Systems) (3 μg/ml) were added and incubated overnight at 4ºC. The sample was then incubated with horse anti-mouse polymer horseradish peroxidase (HRP)-conjugated (Vector Laboratories, Catalog No. MP-7402) for 30 min at RT. ImmPACT DAB peroxidase substrate (Vector Laboratories) was added and incubated for 2-10 min. The reaction was stopped with running deionised water. The section was covered with hematoxylin (Vector Laboratories) for 45 seconds and rinsed with deionised water. Samples were then dehydrated by serial passage through 70%, 90% and 100% ethanol followed by Histo-Clear. Samples were allowed to dry and mounted with VectaMount mounting media (Vector Laboratories). For analysis of immunohistochemical staining, images were acquired on a DSS1 Coolscope Slide Scanner (Nikon).
For immunofluorescent staining, the following primary antibodies were used: anti-LLT1 (R&D Systems, Catalog No. AF3480, goat polyclonal) and anti-CD68 (DAKO, Catalog No. M0876, clone PG-M1, mouse IgG3, k). They were diluted in blocking buffer and incubated for 30 min at RT. Anti-goat HRP-conjugated polymer was added followed by a 30 min incubation at RT. Anti-mouse HRP-conjugated antibody was added followed by a 30 min incubation at RT. Fluorescein Amplification Reagent (PerkinElmer) was diluted 1/300 in new Tyramide amplification buffer, added and incubated in the dark for 15 min. Slides were mounted with ProLong® Gold Antifade Reagent with DAPI (Invitrogen). For immunofluorescent microscopy, images were acquired on an Olympus Fluoview FV1000 microscope (Olympus) and analyzed using Fiji (ImageJ v1.47h); National Institute of Health, USA).

Data analysis
Graphs and statistical analysis were performed using GraphPad Prism Version 6.0a (GraphPad Software) and Adobe Illustrator CS4 14.0.0.

LLT1 is expressed on tonsillar B cells and circulating B cells and monocytes, but not on lung and liver-resident macrophages
There are still many inconsistences in the published data regarding the distribution of LLT1 in human tissues and cell types. Some past studies reported LLT1 expression in resting PBMCs, whereas others could only detect it after activation 13-15 .
We assessed the presence of this C-type lectin in resting and activated PBMCs. We analysed different cell subsets ( Figure 1A) and detected abundant expression of LLT1 in resting monocytes (60-80%) and B cells (15-30%) ( Figure 1B). These results fit with the previously characterised expression of LLT1 in B-cell derived Raji cells 5,13,23 and monocyte-derived THP-1s 23 . Interestingly, the receptor ligand pair LLT1-CD161 was expressed on PBMCs in an exclusive manner ( Figure 1B). While monocytes and B cells expressed LLT1, their levels of CD161 expression were null. On the contrary, all the other subsets tested expressed CD161 to a certain extent, although they did not express its ligand, LLT1.
We next studied LLT1 presence in tissue-resident B cells and macrophages. We and others have shown the expression of LLT1 in tissue resident germinal centre B cells 6,9 . We demonstrated that the 2H7 mAb recognises LLT1 on germinal centre B cells, both immunohistologically and by flow cytometry (Figure 1C) 6 . Thus, the 2H7 mAb is a good tool for studying the distribution of LLT1 in tissue through immunohistochemical staining. Expression of this C-type lectin on tissue-resident macrophages had previously only been addressed in the joints of rheumatoid arthritis (RA) patients, which were positive for LLT1 10 . We wanted to assess the expression of LLT1 in macrophages resident in other tissues. In order to do so, we performed immunofluorescent staining of lung and liver sections, using LLT1 and the macrophage marker CD68. CD68 expressing alveolar macrophages could be detected in the lung, as well as CD68 expressing Kupffer cells in the liver. However, both cell types were negative for LLT1 ( Figure 1D; Dataset 1 34 ), suggesting that terminally differentiated macrophages do not express this C-type lectin. Nonetheless, LLT1+ cells could be detected in both tissues, suggesting that these LLT1+ cells may, for example, be epithelial cells, but further work is needed for their full characterisation.

LLT1 on activated PBMCs
The expression of LLT1 on activated PBMCs was assessed. Stimulation with PMA/ionomycin for 24 and 48h had no significant effect on CD4+ T cells (Figure 2A). Minimal levels of LLT1 were observed on CD8+ T cells after stimulation, although this did not reach significance ( Figure 2C). However, the expression levels were very low and of questionable biological relevance. Similar results were seen using PHA ( Figure 2B and D).
Interestingly, the percentage and levels of LLT1 on B cells initially dropped after 24h, but increased after 48h ( Figure 2E) upon stimulation with PMA/ionomocyin; a similar trend was observed with PHA ( Figure 2F).
Monocytes also lowered LLT1 expression upon activation after both 24 and 48h stimulation with PMA/ionomycin ( Figure 2G), but not after PHA stimulation ( Figure 2H).

LLT1 is expressed in healthy human tissue and some tumours
There have been limited attempts to characterise the distribution of LLT1 within human tissues. In this study, we screened a wide variety of human healthy tissues using the 2H7 antibody clone. A representative stain of each tissue tested is shown in Figure 3. LLT1 could be detected in a wide variety of tissues, such as the gallbladder and the digestive tract (glandular cells), as well as in the kidneys (cells in tubules) or the lung (pneumocytes). We also compared the expression pattern of LLT1 in healthy and tumour human tissues (Supplementary Figure 1). Although LLT1 upregulation has been shown in glioblastoma and prostate cancer 22,24 , our results did not support this being a common trend in all cancerous tissues. Most likely, changes in LLT1 expression upon malignant transformation are tissue-dependent.

LLT1 is expressed in immune-privileged sites
Although LLT1 could be detected in different human tissues (Figure 3), its expression was strikingly high in immune-privileged sites (Figure 4). Cells in the seminiferous ducts within the testes, trophoblastic cells in the placenta and neurons strongly expressed LLT1. Purkinje cells, a large type of neuron that resides in the cerebellum and release the neurotransmitter gamma-aminobutyric (GABA) was also found to be positive for LLT1. A key feature of immune privilege is low expression of MHC class I molecules, which protects certain tissues from excessive and damaging inflammatory T cell responses 19 . However, downregulation of MHC class I molecules results in increased susceptibility to NK cell killing. Therefore, we next tested the effect of LLT1 on NK cell cytotoxic effector functions.

LLT1 inhibits NK cell degranulation
A role for LLT1 in suppression of NK cell function has been described 4,5,14,22 . Here, we confirmed that the presence of LLT1 reduces NK cell degranulation. NK cell surface expression of CD107a was reduced when NK cells were cultured with target cells, the 300.19 cell line transfected with LLT1, as compared to controls ( Figure 5A and B). Figure 5C shows expression levels of LLT1 on target cells, confirming very high levels of this C-type lectin in the transfected 300.19-LLT1 cells, as expected. In summary, our data suggests a plausible role for LLT1 in immune-regulation and, particularly, in negative modulation of NK cell responses in immune-privileged sites.

Discussion
In humans, there are three well-characterised NKC-encoded receptor-ligand pairs: these are the CD161-LLT1, NKp65-KACL and NKp80-AICL. The expression of CD161 has been widely studied. It has been described on the vast majority of NK cells 25 , on different innate-like T cell subset,s such as NKT cells 3 , mucosalassociated invariant T (MAIT) cells 26 , γδ T cells 27,28 and in other T cell subgroups, both in the CD4+ and CD8+ compartments. CD161 defines cell populations with shared transcriptional and functional features across different human T cell lineages 8 . In contrast, the expression and localisation of LLT1 has been much less studied.
LLT1 was first described on NK, T and B cells 29 , although some subsequent studies showed different results 12,13 . We showed LLT1 expression on circulating B cells and monocytes, confirming the results obtained in previous research 12 . It is important to note that the current literature still presents inconsistencies regarding LLT1 distribution in PBMCs, which could be due to the use of different antibodies as well as the diverse activation state of the cells tested 5,6,12,13 .
LLT1 has been shown in joint-resident macrophages of RA patients 10 ; however, we could not detect LLT1 on macrophages from the liver or the lung. These results could be explained by the state of activation of macrophages, suggesting an increase of LLT1 expression under inflammatory conditions. PMA/ionomycin and mitogen (PHA) stimulation of PBMCs demonstrated that a broad range of cell types could, to a limited degree, express some LLT1, which was dependent on the duration of the stimuli. In particular, B cells showed a bi-phasic expression pattern.
We showed expression of LLT1 in different healthy human tissues ( Figure 3) and, particularly, in immune-privileged sites (Figure 4). It is believed that immune-privilege is the result of an evolutionary process that confers special immune tolerance to certain structures 19 . Organs, such as the eye, the brain or the placenta, present the exceptional capacity of preventing classical inflammatory responses that could be highly detrimental or even fatal. This  singular immune status is linked to low expression levels of MHC class I molecules, which subsequently lead to increased susceptibility to killing by NK cells. We and others have shown that the presence of LLT1 results in decreased NK cell function ( Figure 5) 4,5,14 . Therefore, LLT1 could play a prominent role in keeping NK cells under control in immune-privilege sites, thus preventing damage of low-expressing MHC class I tissues. This hypothesis matches with the role described for the murine version of LLT1, mOCIL. The distribution of mOCIL differs substantially from its human homolog, as it is believed to be expressed almost ubiquitously, similarly to MHC class I molecules 30,31 .
A high degree of homology has been described for the mouse and human forms of the CLEC2D protein 32 , suggesting that these antibodies could be reacting with both mouse (Clr-b) and human (LLT1) forms. The distribution of Clr-b has been widely studied, in contrast to the human one. However, so far, there is no particular mention of the presence of Clr-b in mouse B cells, although it has been described in nearly all haematopoietic cells and abundant mouse tissues, with some exceptions (i.e. the brain) 30,31 .
We found LLT1 and CD161 to be expressed in different subgroups of lymphocytes as well as monocytes (Figure 1). We also described Click here to access the data. expression of LLT1 in various human healthy tissues and particularly in immune-privileged sites (Figure 3 and Figure 4). It is tempting to speculate that this pair of C-type lectins is involved in the crosstalk between distinct LLT1 and CD161 expressing immune cell types, such as B cells and T cells or monocytes and NK cells. The closely related pair of C-type lectins NKp80-AICL follows this same pattern: they are expressed on NK cells and monocytes, respectively, playing a role in reciprocal cell activation 2 . The other well-described human C-type lectin pair is NKp65-KACL, which is expressed mainly on NK cells and keratinocytes, respectively 33 .
Thus, this particular pair illustrates a very different case, as it is involved in the immune surveillance of a specific tissue (i.e. the skin). This framework could also apply to LLT1 and CD161, both in terms of lymphocyte/monocyte interaction and interaction between NK cells and immune-privileged sites, although these hypotheses require further investigation.
Overall, we have contributed to the development and optimisation of tools necessary for the study of LLT1. Its striking expression in immune-privileged sites, as well as its presence in different immune cell types establishes LLT1 as an excellent candidate for immune-regulation. A detailed understanding of LLT1 distribution, regulation and function will give great insights into our knowledge on how immune-privilege works, as well as helping us to comprehend tissue-specific immune responses during inflammation. Author contributions AL, CBW, and PK designed, performed, and analyzed experiments and wrote the manuscript; GJF provided the anti-LLT1 Abs and the 300 cell lines; LG and AP contributed to specific experiments; and CBW and PK designed experiments and provided overall guidance.

Competing interests
No competing interests were disclosed.

Grant information
This work was supported by grants from the NIHR Biomedical Research Centre, Oxford, and NIHNIAID U19 Bio-defense Programme (NIH NIAID 1U19AI082630-01) (CBW), Wellcome Trust Senior Fellowship WT091663MA (PK) and Obra Social La Caixa (AL).
The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. 1.

Ondřej Vaněk
Jan Bláha Department of Biochemistry, Faculty of Science, Charles University, Prague, Czech Republic The present study by A. Llibre is describing novel monoclonal antibody, clone 2H7, that is able to et al. detect protein LLT1 in various healthy as well as tumour human tissues, as shown by immunohistochemistry, fluorescence microscopy and flow cytometry. LLT1 was shown to have broad expression pattern with the highest expression levels detected in circulating B cells and monocytes and, surprisingly, also in immune-privileged sites -brain, placenta, testes. This observation supports its role in inhibition of NK cell cytotoxicity mediated by its interaction with inhibitory NK cell receptor NKR-P1. LLT1 inhibitory properties were confirmed in the present study using NK cell degranulation assay.
Overall, this study is of general interest, mainly to the scientists studying directly this receptor:ligand interaction pair or other closely related pairs. Given the fact that LLT1 and especially NKR-P1 are linked to multiple human immune pathologies or other diseases, detailed characterization of LLT1 expression is adding valuable piece of information to this field.
However, as already pointed out by Prof. Lanier in his peer review, I would also argue that for maximising impact of this work and its possible benefit to the scientific community, the availability of this novel antibody should be addressed, as well as the information whether it is blocking the LLT1:NKR-P1 interaction or not. Also, as already mentioned in preceding review, the use of blocking antibody (possibly 2H7) should have been included in degranulation assay, as it is a standard control setup in such experiments. This would also directly show blocking capabilities of 2H7 clone.
We have read this submission. We believe that we have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.
No competing interests were disclosed.