Osteoclasts (OCLs) are multinucleated phagocytes derived from monocytic progenitors and specialized in bone resorption (Madel et al., 2019). Similar to other cells of monocytic origin, they are also innate immunocompetent cells and heterogeneous in their phenotype, function, and origin (Ibáñez et al., 2016; Jacome-Galarza et al., 2019; Madel et al., 2020; Madel et al., 2019; Yahara et al., 2020). OCLs derived from steady-state bone marrow (BM) cells or from BM CD11b⁺ monocytic cells (MN-OCLs) promote tolerance by inducing CD4⁺ and CD8⁺ regulatory T cells (tolerogenic OCLs [t-OCLs]) (Ibáñez et al., 2016; Kiesel et al., 2009). In contrast, in the context of bone destruction linked to inflammatory bowel disease (IBD) or when derived from dendritic cells (DC-OCLs), OCLs induce Tnf-α-producing CD4⁺ T cells (inflammatory OCLs [i-OCLs]) (Ibáñez et al., 2016; Madel et al., 2020). OCLs associated with inflammation can be identified by expression of Cx3cr1 (fractalkine receptor) and the proportion of Cx3cr1⁺ OCLs increases in osteoporosis, IBD, and after Rank-L treatment (Ibáñez et al., 2016; Madel et al., 2020). However, Cx3cr1 is only expressed in approximately 20% of i-OCLs (Ibáñez et al., 2016; Madel et al., 2020), which highlights their heterogeneity while limiting the possibility to analyze them in the context of pathological bone loss and urges the identification of novel markers.

Current anti-resorptive therapies aim to globally inhibit OCLs, without considering their recently established diversity. In the long term, they result in poor bone remodeling which may increase the risk of atypical fractures (Reyes et al., 2016). Therefore, an in-depth characterization of OCLs associated with healthy versus inflammatory bone resorption would allow the identification of distinct characteristics that could help to specifically target i-OCL.

i-OCLs arise under the control of persistent high levels of Rank-L, IL-17, and TNFα mainly produced by CD4⁺ T cells that play a major role in pathological osteoclastogenesis observed in osteoporosis and IBD (Cenci et al., 2000; Ciucci et al., 2015; Ibáñez et al., 2016; Li et al., 2011). Interestingly, the emergence of such osteoclastogenic CD4⁺ T cells is associated with gut dysbiosis and increased intestinal permeability (Jones et al., 2018; Li et al., 2016). In line with this, bacterial probiotics such as Lactobacillus and Bifidobacteria have been shown to effectively reduce osteoporotic bone loss (Li et al., 2016; Britton et al., 2014; Ohlsson et al., 2014; Ibáñez et al., 2019b), but their specific effect on i-OCLs remains unknown.

Here, using a comparative RNAseq approach performed on sorted pure mature MN-OCLs and DC-OCLs as models of t-OCLs and i-OCLs, respectively, as already established (Ibáñez et al., 2016; Madel et al., 2020), we showed that the two OCL populations are distinctly equipped to respond to different signals that can modulate their differentiation. In particular, the pattern recognition receptors (PRRs) Dectin-1, Tlr2, and Mincle involved in the response to fungi (Li et al., 2019; Sancho and Reis e Sousa, 2012) are overexpressed in i-OCLs. Administration of the probiotic yeast Saccharomyces boulardii CNCM I-745 (Sb), which is used in the treatment of gastrointestinal disorders for its anti-inflammatory properties and its capacity to restore the gut microbiota (Czerucka and Rampal, 2019; Terciolo et al., 2019), significantly reduces bone loss and inflammatory parameters in vivo in ovariectomized (OVX) mice. In vitro, Sb derivates as well as low doses of agonists of the PRRs overexpressed in i-OCLs specifically inhibit the differentiation of these cells without affecting t-OCLs. These data open perspectives on targeting specific OCL populations and provide evidence for the protective effect of a probiotic yeast on inflammatory bone resorption. Our study unveils very new insights into the regulation and modulation of i-OCLs and enables a better understanding of the molecular mechanisms involved in inflammation-induced bone erosions.

The present study demonstrates that i-OCLs and t-OCLs are two distinct OCL populations with specific molecular signatures that differ in their preferential use of differentiation pathways and their capacity to sense the environment through the PRRs Dectin-1, Tlr2, and Mincle. Based on these specificities, we could show that targeting these receptors reduces bone loss in OVX mice and inhibits specifically the differentiation of i-OCLs while sparing t-OCLs.

OCL heterogeneity has been neglected for a long time, and the mechanisms regulating the emergence and function of i-OCLs are still largely unknown. Our transcriptomic profiling shows that, while being comparable in their expression of bone resorption-associated genes, t-OCLs and i-OCLs differ in their expression of genes associated with immune responses, which provides additional evidence for their previously reported divergent immune properties (Ibáñez et al., 2016). It also highlights differences in genes related to OCL differentiation. Osteoclastogenesis is regulated by two main pathways, the classical Rank-associated and the co-stimulatory Ig-like receptor-associated pathways, both of which are required to promote OCL differentiation (Humphrey and Nakamura, 2016; Koga et al., 2004; Seeling et al., 2013). Increased Rank-L levels as well as stimulation of FcγR and ITAM signaling are involved in pathological bone resorption in rheumatic diseases such as rheumatoid arthritis (Herman et al., 2008; Ochi et al., 2007). However, to date these pathways have not been described to be preferentially used by specific populations of OCLs. Here, we show that the implication of the co-stimulatory pathway in diseases associated with high bone resorption could be linked to its upregulation during i-OCL differentiation, which suggests that specific targeting of this pathway and its related molecules could limit inflammatory bone loss with a minimal impact on OCLs involved in physiological bone remodeling.

Remarkably, Tlr2, Dectin-1, and Mincle are increased in i-OCLs and OCLs from OVX mice compared to t-OCLs and to those from SHAM mice. These receptors are expressed in myeloid cells and sense microbial structures from bacteria, fungi, or parasites, including lipoproteins, glycoproteins, peptidoglycan for Tlr2, β-glucans for Dectin-1 and Tlr2, and α-mannose for Mincle (Sancho and Reis e Sousa, 2012; Savva and Roger, 2013). Moreover, they interact with each other and with Fcγ receptors for the induction of inflammatory responses (Gantner et al., 2003; Sato et al., 2003). Very few studies have investigated the effect of microbial β-glucans and α-mannose on osteoclastogenesis. Tlr2 stimulation was shown to reduce osteoclastogenesis from non-committed OCL progenitors but to increase OCL survival and differentiation of Rank-L-primed pre-OCLs (Souza and Lerner, 2019; Takami et al., 2002). High concentrations of curdlan (in the range of µg/mL) have been previously reported to decrease the overall differentiation of OCLs from Dectin-1-transfected RAW cells or from BM cells by upregulating Mafb and reducing Nfatc1 expression (Zhu et al., 2017). Our results extend these data by revealing that with much lower doses of curdlan (range of ng/mL), only i-OCL formation is inhibited whereas other OCLs are not affected, in line with the huge difference in Dectin-1 and Tlr2 expression by their respective progenitors. Interestingly, here we show that the yeast probiotic Sb significantly reduces bone loss in OVX mice. Sb has proven its beneficial probiotic effects in several gastrointestinal disorders. It favors regeneration of the integrity of the epithelial barrier in colitis and restores the gut microbiota after antibiotic treatment or diarrhea by increasing SCFA-producing bacteria such as Lachnospiraceae, Bacteroides, Ruminococcus, and Prevotellaceae (Moré and Swidsinski, 2015; Terciolo et al., 2019; Yu et al., 2017). In line with this, our results demonstrate that Sb improves the gut barrier function in the context of OVX-induced osteoporosis. It also normalizes the concentration of SCFAs and lactate in OVX mice, which strongly suggests that Sb favors bacteria species producing these metabolites. Indeed previous studies reported that the diversity of intestinal microbiota is lower in OVX mice than in SHAM mice, and its composition is altered (Li et al., 2021; Zaiss et al., 2019). LAB as well as SCFA-producing bacteria have a beneficial effect on bone (Zaiss et al., 2019) which probably participates in the reduced bone loss observed in Sb-treated OVX mice. Moreover, gut dysbiosis and increased intestinal permeability reported in osteoporosis (He et al., 2020; Xu et al., 2020) induce activation of osteoclastogenic CD4⁺ T cells over-producing Rank-L and Tnf-α (D’Amelio et al., 2008; Li et al., 2016). Thus, the reduction in BM Tnf-α-producing CD4⁺ T cells observed after Sb treatment in OVX mice is likely due to its beneficial effect on the gut, as described for bacterial probiotics (Li et al., 2016). As Tnf-α⁺ CD4⁺ T cells are major inducers of i-OCLs (Ciucci et al., 2015; Ibáñez et al., 2016; Madel et al., 2019), this effect probably also participates in the decreased proportion of i-OCLs observed in Sb-treated OVX mice. In addition, in various models of gut disorders, Sb has been described to increase the production of anti-inflammatory cytokines such as IL-10 and IL-4 (Czerucka and Rampal, 2019; Pais et al., 2020) that are potent inhibitors of OCL formation (Fujii et al., 2012; Park-Min et al., 2009). Despite we did not detect significant difference in serum IL-10 concentration after Sb treatment in OVX mice (data not shown), we could not rule out local cytokine modification in the BM.

In addition to its systemic effects on inflammation and microbiota, Sb is also likely to directly affect i-OCL differentiation. The yeast cell wall components, ß-glucans, are well known to cross the gut barrier and translocate to the blood from where they can disseminate to organs (Isnard et al., 2021; Rice et al., 2005) and can directly modulate cells from the monocytic family (Ibáñez et al., 2019a), which includes circulating OCL progenitors, through stimulation of CLRs. Consequently, in vivo they can possibly exert the same direct inhibitory effect on the differentiation of i-OCLs as observed in vitro. This is confirmed by our in vitro analysis showing a specific inhibition of i-OCL differentiation compared to t-OCLs by Sb, as well as low doses of agonists of Tlr2, Dectin-1, and Mincle, which is in line with the much higher expression of these receptors in BM-derived DCs than in BM MNs.

The inhibitory effect of Mincle stimulation observed here is contrasting with a recent publication reporting that sensing of necrotic cells through Mincle by OCL progenitors stimulates OCL formation (Andreev et al., 2020). Clec4e KO mice have increased bone mass (Andreev et al., 2020), a result confirmed by our study. However, we also observed a similar or even higher OCL formation from progenitors from Clec4e KO compared to control mice. Mincle is not expressed by osteoblasts (Andreev et al., 2020), but the increased bone mass observed in Clec4e KO mice could be related to indirect effects of Mincle on osteoblasts and OCLs through myeloid cells that are well-known regulators of bone cell differentiation and activity (Yahara et al., 2021). Moreover, Mincle KO mice display reduced CD4⁺IL-17⁺ T cells (Martínez-López et al., 2019), which may also contribute to a reduced OCL differentiation and increased bone mass. It is therefore likely that depending on whether Mincle is stimulated by exogenous microbial signals (as in our study) or endogenous necrotic signals (as in Andreev’s study), this receptor has divergent effects on osteoclastogenesis. These observations revealed the complex effects of Mincle on osteoclastogenesis, as already reported for other monocytic cells (Patin et al., 2017).

The kinase Syk plays a major role in Dectin-1 and Mincle signaling pathways, and indeed, Syk is rapidly phosphorylated upon curdlan stimulation in BM-DCs. On the other hand, Syk is required for efficient OCL differentiation (Zou et al., 2007), including from DCs as shown here. However, in DCs stimulated with curdlan in osteoclastogenic medium, the expression of Syk decreases with time together with Nfatc1, a master gene of OCL differentiation, and Ctsk, a main marker of OCLs required for their activity, which participates in reducing i-OCL formation. These results are in agreement with the literature showing in the same conditions a degradation of Syk after treatment with high doses of curdlan (Yamasaki et al., 2014). In DCs, the interaction between Tlr2 and Dectin-1 is involved in the stimulation of NF-κB and in the production of inflammatory cytokines such as Tnf-α and IL-12 (Gantner et al., 2003; Sancho and Reis e Sousa, 2012). This strongly induces DC activation and therefore potent phagocytic and anti-microbial responses (Gantner et al., 2003; Sancho and Reis e Sousa, 2012). Accordingly, we show that treatment with curdlan or with Sb-CM induces DC maturation while it decreases the proportion of Rank⁺ cells expressing Csf1r and FcgrII/III, all of which are required for OCL formation. Thus, stimulation of DCs with the PRR agonists has a dual effect, i.e., increasing their maturation and simultaneously reducing their ability to give rise to i-OCLs.

In conclusion, we here demonstrated that i-OCLs differ from t-OCLs in the control of their differentiation and in their capacity to sense their environment and in particular to respond to stimuli through CLR and TLR activation. Based on these properties, we demonstrated that the probiotic yeast Sb has a beneficial effect on bone loss in osteoporotic mice by restoring gut barrier integrity and reducing osteoclastogenic CD4⁺ T cells thereby reducing indirectly i-OCLs. Moreover, it also directly targets i-OCL progenitors to interfere with their differentiation.

These insights open new interesting perspectives for the treatment of pathological bone resorption by demonstrating that the Ig-like receptor costimulatory pathway and the associated PRR pathway, rather than the Rank pathway, could represent efficient therapeutic targets to specifically impact on inflammatory osteoclastogenesis while maintaining physiological OCL resorption. The dose of Sb used in mice is much higher than that recommended for humans. However, this is also true in studies evaluating probiotic lactobacillus strains in osteoporosis that have shown protective effects at much lower doses in humans than in mice (Britton et al., 2014; Nilsson et al., 2018; Jansson et al., 2019). Therefore, Sb administration in osteoporotic patients would represent a feasible possibility. These novel insights and regulatory mechanisms mediated by yeast probiotics could provide new therapeutic options to overcome the global inhibition of OCLs and the resulting impaired bone quality associated with current anti-resorptive therapies.

RNAseq data have been deposited in ENA (European Nucleotide Archive) accession number PRJEB42043.

1. 1. Madel MB
   2. Garchon HJ
   3. Wakkach A
   4. Blin-Wakkach C

   (2022) ENA

   ID PRJEB42043. Targeting inflammatory osteoclasts with probiotic yeast S. boulardii reduces bone loss in osteoporosis.

   https://www.ebi.ac.uk/ena/browser/view/PRJEB42043

1. 1. Anders S
   2. Pyl PT
   3. Huber W

   (2015) HTSeq--a python framework to work with high-throughput sequencing data

   Bioinformatics 31:166–169.

   https://doi.org/10.1093/bioinformatics/btu638

   • PubMed
   • Google Scholar
2. 1. Andreev D
   2. Liu M
   3. Weidner D
   4. Kachler K
   5. Faas M
   6. Grüneboom A
   7. Schlötzer-Schrehardt U
   8. Muñoz LE
   9. Steffen U
   10. Grötsch B
   11. Killy B
   12. Krönke G
   13. Luebke AM
   14. Niemeier A
   15. Wehrhan F
   16. Lang R
   17. Schett G
   18. Bozec A

   (2020) Osteocyte necrosis triggers osteoclast-mediated bone loss through macrophage-inducible C-type lectin

   The Journal of Clinical Investigation 130:4811–4830.

   https://doi.org/10.1172/JCI134214

   • PubMed
   • Google Scholar
3. 1. Britton RA
   2. Irwin R
   3. Quach D
   4. Schaefer L
   5. Zhang J
   6. Lee T
   7. Parameswaran N
   8. McCabe LR

   (2014) Probiotic L. reuteri treatment prevents BONE LOSS in a menopausal ovariectomized mouse model: probiotics suppress estrogen deficiency-induced bone loss

   Journal of Cellular Physiology 229:1822–1830.

   https://doi.org/10.1002/jcp.24636

   • Google Scholar
4. 1. Cenci S
   2. Weitzmann MN
   3. Roggia C
   4. Namba N
   5. Novack D
   6. Woodring J
   7. Pacifici R

   (2000) Estrogen deficiency induces bone loss by enhancing T-cell production of TNF-alpha

   The Journal of Clinical Investigation 106:1229–1237.

   https://doi.org/10.1172/JCI11066

   • PubMed
   • Google Scholar
5. 1. Ciucci T
   2. Ibáñez L
   3. Boucoiran A
   4. Birgy-Barelli E
   5. Pène J
   6. Abou-Ezzi G
   7. Arab N
   8. Rouleau M
   9. Hébuterne X
   10. Yssel H
   11. Blin-Wakkach C
   12. Wakkach A

   (2015) Bone marrow th17 TNFα cells induce osteoclast differentiation, and link bone destruction to IBD

   Gut 64:1072–1081.

   https://doi.org/10.1136/gutjnl-2014-306947

   • PubMed
   • Google Scholar
6. 1. Czerucka D
   2. Rampal P

   (2019) Diversity of saccharomyces boulardii CNCM I-745 mechanisms of action against intestinal infections

   World Journal of Gastroenterology 25:2188–2203.

   https://doi.org/10.3748/wjg.v25.i18.2188

   • PubMed
   • Google Scholar
7. 1. D’Amelio P
   2. Grimaldi A
   3. Di Bella S
   4. Brianza SZM
   5. Cristofaro MA
   6. Tamone C
   7. Giribaldi G
   8. Ulliers D
   9. Pescarmona GP
   10. Isaia G

   (2008) Estrogen deficiency increases osteoclastogenesis up-regulating T cells activity: a key mechanism in osteoporosis

   Bone 43:92–100.

   https://doi.org/10.1016/j.bone.2008.02.017

   • PubMed
   • Google Scholar
8. 1. Daria Zenkova VK

   (2018) Phantasus

   Bioconductor 3.16:B9.

   https://doi.org/10.18129/B9.BIOC.PHANTASUS

   • Google Scholar
9. 1. Dobin A
   2. Davis CA
   3. Schlesinger F
   4. Drenkow J
   5. Zaleski C
   6. Jha S
   7. Batut P
   8. Chaisson M
   9. Gingeras TR

   (2013) STAR: ultrafast universal RNA-seq aligner

   Bioinformatics 29:15–21.

   https://doi.org/10.1093/bioinformatics/bts635

   • PubMed
   • Google Scholar
10. 1. Dweep H
    2. Gretz N

    (2015) MiRWalk2.0: a comprehensive atlas of microrna-target interactions

    Nature Methods 12:697.

    https://doi.org/10.1038/nmeth.3485

    • PubMed
    • Google Scholar
11. 1. Fujii T
    2. Kitaura H
    3. Kimura K
    4. Hakami ZW
    5. Takano-Yamamoto T

    (2012) IL-4 inhibits TNF-α-mediated osteoclast formation by inhibition of RANKL expression in TNF-α-activated stromal cells and direct inhibition of TNF-α-activated osteoclast precursors via a T-cell-independent mechanism in vivo

    Bone 51:771–780.

    https://doi.org/10.1016/j.bone.2012.06.024

    • PubMed
    • Google Scholar
12. 1. Gantner BN
    2. Simmons RM
    3. Canavera SJ
    4. Akira S
    5. Underhill DM

    (2003) Collaborative induction of inflammatory responses by dectin-1 and toll-like receptor 2

    The Journal of Experimental Medicine 197:1107–1117.

    https://doi.org/10.1084/jem.20021787

    • PubMed
    • Google Scholar
13. 1. Halper J
    2. Madel MB
    3. Blin-Wakkach C

    (2021) Differentiation and phenotyping of murine osteoclasts from bone marrow progenitors, monocytes, and dendritic cells

    Methods in Molecular Biology 2308:21–34.

    https://doi.org/10.1007/978-1-0716-1425-9_2

    • PubMed
    • Google Scholar
14. 1. He J
    2. Xu S
    3. Zhang B
    4. Xiao C
    5. Chen Z
    6. Si F
    7. Fu J
    8. Lin X
    9. Zheng G
    10. Yu G
    11. Chen J

    (2020) Gut microbiota and metabolite alterations associated with reduced bone mineral density or bone metabolic indexes in postmenopausal osteoporosis

    Aging 12:8583–8604.

    https://doi.org/10.18632/aging.103168

    • PubMed
    • Google Scholar
15. 1. Herman S
    2. Müller RB
    3. Krönke G
    4. Zwerina J
    5. Redlich K
    6. Hueber AJ
    7. Gelse H
    8. Neumann E
    9. Müller-Ladner U
    10. Schett G

    (2008) Induction of osteoclast-associated receptor, a key osteoclast costimulation molecule, in rheumatoid arthritis

    Arthritis and Rheumatism 58:3041–3050.

    https://doi.org/10.1002/art.23943

    • PubMed
    • Google Scholar
16. 1. Humphrey MB
    2. Nakamura MC

    (2016) A comprehensive review of immunoreceptor regulation of osteoclasts

    Clinical Reviews in Allergy & Immunology 51:48–58.

    https://doi.org/10.1007/s12016-015-8521-8

    • PubMed
    • Google Scholar
17. 1. Ibáñez L
    2. Abou-Ezzi G
    3. Ciucci T
    4. Amiot V
    5. Belaïd N
    6. Obino D
    7. Mansour A
    8. Rouleau M
    9. Wakkach A
    10. Blin-Wakkach C

    (2016) Inflammatory osteoclasts prime TNFα-producing CD4(+) T cells and express CX3 CR1

    Journal of Bone and Mineral Research: The Official Journal of the American Society for Bone and Mineral Research 31:1899–1908.

    https://doi.org/10.1002/jbmr.2868

    • PubMed
    • Google Scholar
18. 1. Ibáñez L
    2. Pontier-Bres R
    3. Larbret F
    4. Rekima A
    5. Verhasselt V
    6. Blin-Wakkach C
    7. Czerucka D

    (2019a) Saccharomyces boulardii strain CNCM I-745 modifies the mononuclear phagocytes response in the small intestine of mice following salmonella typhimurium infection

    Frontiers in Immunology 10:643.

    https://doi.org/10.3389/fimmu.2019.00643

    • PubMed
    • Google Scholar
19. 1. Ibáñez L
    2. Rouleau M
    3. Wakkach A
    4. Blin-Wakkach C

    (2019b) Gut microbiome and bone

    Joint Bone Spine 86:43–47.

    https://doi.org/10.1016/j.jbspin.2018.02.008

    • PubMed
    • Google Scholar
20. 1. Iborra S
    2. Izquierdo HM
    3. Martínez-López M
    4. Blanco-Menéndez N
    5. Reis e Sousa C
    6. Sancho D

    (2012) The DC receptor DNGR-1 mediates cross-priming of CTLs during vaccinia virus infection in mice

    The Journal of Clinical Investigation 122:1628–1643.

    https://doi.org/10.1172/JCI60660

    • PubMed
    • Google Scholar
21. 1. Isnard S
    2. Lin J
    3. Bu S
    4. Fombuena B
    5. Royston L
    6. Routy JP

    (2021) Gut leakage of fungal-related products: turning up the heat for HIV infection

    Frontiers in Immunology 12:656414.

    https://doi.org/10.3389/fimmu.2021.656414

    • PubMed
    • Google Scholar
22. 1. Jacome-Galarza CE
    2. Percin GI
    3. Muller JT
    4. Mass E
    5. Lazarov T
    6. Eitler J
    7. Rauner M
    8. Yadav VK
    9. Crozet L
    10. Bohm M
    11. Loyher PL
    12. Karsenty G
    13. Waskow C
    14. Geissmann F

    (2019) Developmental origin, functional maintenance and genetic rescue of osteoclasts

    Nature 568:541–545.

    https://doi.org/10.1038/s41586-019-1105-7

    • PubMed
    • Google Scholar
23. 1. Jansson PA
    2. Curiac D
    3. Lazou Ahrén I
    4. Hansson F
    5. Martinsson Niskanen T
    6. Sjögren K
    7. Ohlsson C

    (2019) Probiotic treatment using a mix of three lactobacillus strains for lumbar spine bone loss in postmenopausal women: a randomised, double-blind, placebo-controlled, multicentre trial

    The Lancet Rheumatology 1:e154–e162.

    https://doi.org/10.1016/S2665-9913(19)30068-2

    • Google Scholar
24. 1. Jones RM
    2. Mulle JG
    3. Pacifici R

    (2018) Osteomicrobiology: the influence of gut microbiota on bone in health and disease

    Bone 115:59–67.

    https://doi.org/10.1016/j.bone.2017.04.009

    • PubMed
    • Google Scholar
25. 1. Kiesel JR
    2. Buchwald ZS
    3. Aurora R

    (2009) Cross-presentation by osteoclasts induces FOXP3 in CD8+ T cells

    Journal of Immunology 182:5477–5487.

    https://doi.org/10.4049/jimmunol.0803897

    • PubMed
    • Google Scholar
26. 1. Koga T
    2. Inui M
    3. Inoue K
    4. Kim S
    5. Suematsu A
    6. Kobayashi E
    7. Iwata T
    8. Ohnishi H
    9. Matozaki T
    10. Kodama T
    11. Taniguchi T
    12. Takayanagi H
    13. Takai T

    (2004) Costimulatory signals mediated by the ITAM motif cooperate with RANKL for bone homeostasis

    Nature 428:758–763.

    https://doi.org/10.1038/nature02444

    • PubMed
    • Google Scholar
27. 1. Lawenius L
    2. Colldén H
    3. Horkeby K
    4. Wu J
    5. Grahnemo L
    6. Vandenput L
    7. Ohlsson C
    8. Sjögren K

    (2022) A probiotic mix partially protects against castration-induced bone loss in male mice

    The Journal of Endocrinology 254:91–101.

    https://doi.org/10.1530/JOE-21-0408

    • PubMed
    • Google Scholar
28. 1. Lézot F
    2. Chesneau J
    3. Navet B
    4. Gobin B
    5. Amiaud J
    6. Choi Y
    7. Yagita H
    8. Castaneda B
    9. Berdal A
    10. Mueller CG
    11. Rédini F
    12. Heymann D

    (2015) Skeletal consequences of RANKL-blocking antibody (IK22-5) injections during growth: mouse strain disparities and synergic effect with zoledronic acid

    Bone 73:51–59.

    https://doi.org/10.1016/j.bone.2014.12.011

    • PubMed
    • Google Scholar
29. 1. Li JY
    2. Tawfeek H
    3. Bedi B
    4. Yang X
    5. Adams J
    6. Gao KY
    7. Zayzafoon M
    8. Weitzmann MN
    9. Pacifici R

    (2011) Ovariectomy disregulates osteoblast and osteoclast formation through the T-cell receptor CD40 ligand

    PNAS 108:768–773.

    https://doi.org/10.1073/pnas.1013492108

    • PubMed
    • Google Scholar
30. 1. Li JY
    2. Chassaing B
    3. Tyagi AM
    4. Vaccaro C
    5. Luo T
    6. Adams J
    7. Darby TM
    8. Weitzmann MN
    9. Mulle JG
    10. Gewirtz AT
    11. Jones RM
    12. Pacifici R

    (2016) Sex steroid deficiency-associated bone loss is microbiota dependent and prevented by probiotics

    The Journal of Clinical Investigation 126:2049–2063.

    https://doi.org/10.1172/JCI86062

    • PubMed
    • Google Scholar
31. 1. Li TH
    2. Liu L
    3. Hou YY
    4. Shen SN
    5. Wang TT

    (2019) C-Type lectin receptor-mediated immune recognition and response of the microbiota in the gut

    Gastroenterology Report 7:312–321.

    https://doi.org/10.1093/gastro/goz028

    • PubMed
    • Google Scholar
32. 1. Li R
    2. Boer CG
    3. Oei L
    4. Medina-Gomez C

    (2021) The gut microbiome: a new frontier in musculoskeletal research

    Current Osteoporosis Reports 19:347–357.

    https://doi.org/10.1007/s11914-021-00675-x

    • PubMed
    • Google Scholar
33. 1. Love MI
    2. Huber W
    3. Anders S

    (2014) Moderated estimation of fold change and dispersion for RNA-Seq data with deseq2

    Genome Biology 15:550.

    https://doi.org/10.1186/s13059-014-0550-8

    • PubMed
    • Google Scholar
34. 1. Madel MB
    2. Ibáñez L
    3. Rouleau M
    4. Wakkach A
    5. Blin-Wakkach C

    (2018) A novel reliable and efficient procedure for purification of mature osteoclasts allowing functional assays in mouse cells

    Frontiers in Immunology 9:2567.

    https://doi.org/10.3389/fimmu.2018.02567

    • PubMed
    • Google Scholar
35. 1. Madel M-B
    2. Ibáñez L
    3. Wakkach A
    4. de Vries TJ
    5. Teti A
    6. Apparailly F
    7. Blin-Wakkach C

    (2019) Immune function and diversity of osteoclasts in normal and pathological conditions

    Frontiers in Immunology 10:1408.

    https://doi.org/10.3389/fimmu.2019.01408

    • PubMed
    • Google Scholar
36. 1. Madel MB
    2. Ibáñez L
    3. Ciucci T
    4. Halper J
    5. Rouleau M
    6. Boutin A
    7. Hue C
    8. Duroux-Richard I
    9. Apparailly F
    10. Garchon HJ
    11. Wakkach A
    12. Blin-Wakkach C

    (2020) Dissecting the phenotypic and functional heterogeneity of mouse inflammatory osteoclasts by the expression of cx3cr1

    eLife 9:e54493.

    https://doi.org/10.7554/eLife.54493

    • PubMed
    • Google Scholar
37. 1. Martínez-López M
    2. Iborra S
    3. Conde-Garrosa R
    4. Mastrangelo A
    5. Danne C
    6. Mann ER
    7. Reid DM
    8. Gaboriau-Routhiau V
    9. Chaparro M
    10. Lorenzo MP
    11. Minnerup L
    12. Saz-Leal P
    13. Slack E
    14. Kemp B
    15. Gisbert JP
    16. Dzionek A
    17. Robinson MJ
    18. Rupérez FJ
    19. Cerf-Bensussan N
    20. Brown GD
    21. Bernardo D
    22. LeibundGut-Landmann S
    23. Sancho D

    (2019) Microbiota sensing by mincle-syk axis in dendritic cells regulates interleukin-17 and -22 production and promotes intestinal barrier integrity

    Immunity 50:446–461.

    https://doi.org/10.1016/j.immuni.2018.12.020

    • PubMed
    • Google Scholar
38. 1. McCabe LR
    2. Irwin R
    3. Schaefer L
    4. Britton RA

    (2013) Probiotic use decreases intestinal inflammation AND increases BONE density in healthy male but not female mice

    Journal of Cellular Physiology 228:1793–1798.

    https://doi.org/10.1002/jcp.24340

    • PubMed
    • Google Scholar
39. 1. Merck E
    2. Gaillard C
    3. Gorman DM
    4. Montero-Julian F
    5. Durand I
    6. Zurawski SM
    7. Menetrier-Caux C
    8. Carra G
    9. Lebecque S
    10. Trinchieri G
    11. Bates EEM

    (2004) OSCAR is an fcrgamma-associated receptor that is expressed by myeloid cells and is involved in antigen presentation and activation of human dendritic cells

    Blood 104:1386–1395.

    https://doi.org/10.1182/blood-2004-03-0850

    • PubMed
    • Google Scholar
40. 1. Mócsai A
    2. Humphrey MB
    3. Van Ziffle JAG
    4. Hu Y
    5. Burghardt A
    6. Spusta SC
    7. Majumdar S
    8. Lanier LL
    9. Lowell CA
    10. Nakamura MC

    (2004) The immunomodulatory adapter proteins DAP12 and fc receptor gamma-chain (fcrgamma) regulate development of functional osteoclasts through the syk tyrosine kinase

    PNAS 101:6158–6163.

    https://doi.org/10.1073/pnas.0401602101

    • PubMed
    • Google Scholar
41. 1. Moré MI
    2. Swidsinski A

    (2015) Saccharomyces boulardii CNCM I-745 supports regeneration of the intestinal microbiota after diarrheic dysbiosis - a review

    Clinical and Experimental Gastroenterology 8:237–255.

    https://doi.org/10.2147/CEG.S85574

    • PubMed
    • Google Scholar
42. 1. Nilsson AG
    2. Sundh D
    3. Bäckhed F
    4. Lorentzon M

    (2018) Lactobacillus reuteri reduces bone loss in older women with low bone mineral density: a randomized, placebo-controlled, double-blind, clinical trial

    Journal of Internal Medicine 284:307–317.

    https://doi.org/10.1111/joim.12805

    • PubMed
    • Google Scholar
43. 1. Ochi S
    2. Shinohara M
    3. Sato K
    4. Gober HJ
    5. Koga T
    6. Kodama T
    7. Takai T
    8. Miyasaka N
    9. Takayanagi H

    (2007) Pathological role of osteoclast costimulation in arthritis-induced bone loss

    PNAS 104:11394–11399.

    https://doi.org/10.1073/pnas.0701971104

    • PubMed
    • Google Scholar
44. 1. Ohlsson C
    2. Engdahl C
    3. Fåk F
    4. Andersson A
    5. Windahl SH
    6. Farman HH
    7. Movérare-Skrtic S
    8. Islander U
    9. Sjögren K

    (2014) Probiotics protect mice from ovariectomy-induced cortical bone loss

    PLOS ONE 9:e92368.

    https://doi.org/10.1371/journal.pone.0092368

    • PubMed
    • Google Scholar
45. 1. Pais P
    2. Almeida V
    3. Yılmaz M
    4. Teixeira MC

    (2020) Saccharomyces boulardii: what makes it tick as successful probiotic?

    Journal of Fungi 6:78.

    https://doi.org/10.3390/jof6020078

    • PubMed
    • Google Scholar
46. 1. Park-Min KH
    2. Ji JD
    3. Antoniv T
    4. Reid AC
    5. Silver RB
    6. Humphrey MB
    7. Nakamura M
    8. Ivashkiv LB

    (2009) Il-10 suppresses calcium-mediated costimulation of receptor activator NF-kappa B signaling during human osteoclast differentiation by inhibiting TREM-2 expression

    Journal of Immunology 183:2444–2455.

    https://doi.org/10.4049/jimmunol.0804165

    • PubMed
    • Google Scholar
47. 1. Patin EC
    2. Orr SJ
    3. Schaible UE

    (2017) Macrophage inducible C-type lectin as a multifunctional player in immunity

    Frontiers in Immunology 8:861.

    https://doi.org/10.3389/fimmu.2017.00861

    • PubMed
    • Google Scholar
48. 1. Patro R
    2. Duggal G
    3. Love MI
    4. Irizarry RA
    5. Kingsford C

    (2017) Salmon provides fast and bias-aware quantification of transcript expression

    Nature Methods 14:417–419.

    https://doi.org/10.1038/nmeth.4197

    • PubMed
    • Google Scholar
49. 1. Reyes C
    2. Hitz M
    3. Prieto-Alhambra D
    4. Abrahamsen B

    (2016) Risks and benefits of bisphosphonate therapies

    Journal of Cellular Biochemistry 117:20–28.

    https://doi.org/10.1002/jcb.25266

    • PubMed
    • Google Scholar
50. 1. Rice PJ
    2. Adams EL
    3. Ozment-Skelton T
    4. Gonzalez AJ
    5. Goldman MP
    6. Lockhart BE
    7. Barker LA
    8. Breuel KF
    9. Deponti WK
    10. Kalbfleisch JH
    11. Ensley HE
    12. Brown GD
    13. Gordon S
    14. Williams DL

    (2005) Oral delivery and gastrointestinal absorption of soluble glucans stimulate increased resistance to infectious challenge

    The Journal of Pharmacology and Experimental Therapeutics 314:1079–1086.

    https://doi.org/10.1124/jpet.105.085415

    • PubMed
    • Google Scholar
51. 1. Sancho D
    2. Reis e Sousa C

    (2012) Signaling by myeloid C-type lectin receptors in immunity and homeostasis

    Annual Review of Immunology 30:491–529.

    https://doi.org/10.1146/annurev-immunol-031210-101352

    • PubMed
    • Google Scholar
52. 1. Sato M
    2. Sano H
    3. Iwaki D
    4. Kudo K
    5. Konishi M
    6. Takahashi H
    7. Takahashi T
    8. Imaizumi H
    9. Asai Y
    10. Kuroki Y

    (2003) Direct binding of toll-like receptor 2 to zymosan, and zymosan-induced NF-kappa B activation and TNF-alpha secretion are down-regulated by lung collectin surfactant protein A

    Journal of Immunology 171:417–425.

    https://doi.org/10.4049/jimmunol.171.1.417

    • PubMed
    • Google Scholar
53. 1. Savva A
    2. Roger T

    (2013) Targeting Toll-like receptors: promising therapeutic strategies for the management of sepsis-associated pathology and infectious diseases

    Frontiers in Immunology 4:387.

    https://doi.org/10.3389/fimmu.2013.00387

    • PubMed
    • Google Scholar
54. 1. Seeling M
    2. Hillenhoff U
    3. David JP
    4. Schett G
    5. Tuckermann J
    6. Lux A
    7. Nimmerjahn F

    (2013) Inflammatory monocytes and fcγ receptor IV on osteoclasts are critical for bone destruction during inflammatory arthritis in mice

    PNAS 110:10729–10734.

    https://doi.org/10.1073/pnas.1301001110

    • PubMed
    • Google Scholar
55. 1. Souza PPC
    2. Lerner UH

    (2019) Finding a Toll on the route: the fate of osteoclast progenitors after Toll-like receptor activation

    Frontiers in Immunology 10:1663.

    https://doi.org/10.3389/fimmu.2019.01663

    • PubMed
    • Google Scholar
56. 1. Takami M
    2. Kim N
    3. Rho J
    4. Choi Y

    (2002) Stimulation by toll-like receptors inhibits osteoclast differentiation

    Journal of Immunology 169:1516–1523.

    https://doi.org/10.4049/jimmunol.169.3.1516

    • PubMed
    • Google Scholar
57. 1. Terciolo C
    2. Dapoigny M
    3. Andre F

    (2019) Beneficial effects of Saccharomyces boulardii CNCM I-745 on clinical disorders associated with intestinal barrier disruption

    Clinical and Experimental Gastroenterology 12:67–82.

    https://doi.org/10.2147/CEG.S181590

    • PubMed
    • Google Scholar
58. 1. Wells CA
    2. Salvage-Jones JA
    3. Li X
    4. Hitchens K
    5. Butcher S
    6. Murray RZ
    7. Beckhouse AG
    8. Lo YLS
    9. Manzanero S
    10. Cobbold C
    11. Schroder K
    12. Ma B
    13. Orr S
    14. Stewart L
    15. Lebus D
    16. Sobieszczuk P
    17. Hume DA
    18. Stow J
    19. Blanchard H
    20. Ashman RB

    (2008) The macrophage-inducible C-type lectin, mincle, is an essential component of the innate immune response to candida albicans

    Journal of Immunology 180:7404–7413.

    https://doi.org/10.4049/jimmunol.180.11.7404

    • PubMed
    • Google Scholar
59. 1. Xu Z
    2. Xie Z
    3. Sun J
    4. Huang S
    5. Chen Y
    6. Li C
    7. Sun X
    8. Xia B
    9. Tian L
    10. Guo C
    11. Li F
    12. Pi G

    (2020) Gut microbiome reveals specific dysbiosis in primary osteoporosis

    Frontiers in Cellular and Infection Microbiology 10:160.

    https://doi.org/10.3389/fcimb.2020.00160

    • PubMed
    • Google Scholar
60. 1. Yahara Y
    2. Barrientos T
    3. Tang YJ
    4. Puviindran V
    5. Nadesan P
    6. Zhang H
    7. Gibson JR
    8. Gregory SG
    9. Diao Y
    10. Xiang Y
    11. Qadri YJ
    12. Souma T
    13. Shinohara ML
    14. Alman BA

    (2020) Erythromyeloid progenitors give rise to a population of osteoclasts that contribute to bone homeostasis and repair

    Nature Cell Biology 22:49–59.

    https://doi.org/10.1038/s41556-019-0437-8

    • PubMed
    • Google Scholar
61. 1. Yahara Y
    2. Ma X
    3. Gracia L
    4. Alman BA

    (2021) Monocyte/macrophage lineage cells from fetal erythromyeloid progenitors orchestrate bone remodeling and repair

    Frontiers in Cell and Developmental Biology 9:622035.

    https://doi.org/10.3389/fcell.2021.622035

    • PubMed
    • Google Scholar
62. 1. Yamasaki T
    2. Ariyoshi W
    3. Okinaga T
    4. Adachi Y
    5. Hosokawa R
    6. Mochizuki S
    7. Sakurai K
    8. Nishihara T

    (2014) The dectin 1 agonist curdlan regulates osteoclastogenesis by inhibiting nuclear factor of activated T cells cytoplasmic 1 (nfatc1) through syk kinase

    The Journal of Biological Chemistry 289:19191–19203.

    https://doi.org/10.1074/jbc.M114.551416

    • PubMed
    • Google Scholar
63. 1. Yu L
    2. Zhao XK
    3. Cheng ML
    4. Yang GZ
    5. Wang B
    6. Liu HJ
    7. Hu YX
    8. Zhu LL
    9. Zhang S
    10. Xiao ZW
    11. Liu YM
    12. Zhang BF
    13. Mu M

    (2017) Saccharomyces boulardii administration changes gut microbiota and attenuates D-galactosamine-induced liver injury

    Scientific Reports 7:1359.

    https://doi.org/10.1038/s41598-017-01271-9

    • PubMed
    • Google Scholar
64. 1. Zaiss MM
    2. Jones RM
    3. Schett G
    4. Pacifici R

    (2019) The gut-bone axis: how bacterial metabolites bridge the distance

    The Journal of Clinical Investigation 129:3018–3028.

    https://doi.org/10.1172/JCI128521

    • PubMed
    • Google Scholar
65. 1. Zhu X
    2. Zhao Y
    3. Jiang Y
    4. Qin T
    5. Chen J
    6. Chu X
    7. Yi Q
    8. Gao S
    9. Wang S

    (2017) Dectin-1 signaling inhibits osteoclastogenesis via IL-33-induced inhibition of nfatc1

    Oncotarget 8:53366–53374.

    https://doi.org/10.18632/oncotarget.18411

    • PubMed
    • Google Scholar
66. 1. Zou W
    2. Kitaura H
    3. Reeve J
    4. Long F
    5. Tybulewicz VLJ
    6. Shattil SJ
    7. Ginsberg MH
    8. Ross FP
    9. Teitelbaum SL

    (2007) Syk, c-Src, the alphavbeta3 integrin, and ITAM immunoreceptors, in concert, regulate osteoclastic bone resorption

    The Journal of Cell Biology 176:877–888.

    https://doi.org/10.1083/jcb.200611083

    • PubMed
    • Google Scholar

Decision letter after peer review:

Thank you for submitting your article "Specific targeting of inflammatory osteoclastogenesis by the probiotic yeast S. boulardii CNCM I-745 reduces bone loss in osteoporosis" for consideration by eLife. Your article has been reviewed by 3 peer reviewers, and the evaluation has been overseen by a Reviewing Editor and Mone Zaidi as the Senior Editor. The following individual involved in the review of your submission has agreed to reveal their identity: Marco Ponzetti (Reviewer #1).

The reviewers have discussed their reviews with one another, and the Reviewing Editor has drafted this to help you prepare a revised submission.

The general objective of this work is the dissection of osteoclast diversity; in particular, the authors intend to identify the specific features and properties that distinguish inflammatory and steady-state (tolerogenic) osteoclasts. To this end, the authors perform a transcriptional analysis of inflammatory and tolerogenic osteoclasts and identify the pattern recognition receptors TLR2, Dectin-1, and Mincle as differentially expressed genes. Agonists of these receptors or yeast probiotics regulating the elicited mechanisms in vitro and in vivo caused a specific inhibition of the differentiation of inflammatory rather than tolerogenic osteoclasts, thus highlighting the preferential use of different differentiation pathways by the two distinct osteoclast populations.

The project is based on the previous knowledge and know-how of the authors on this peculiar skeletal cell population. The work is well conceived; the experiments are clearly designed and exploit state-of-the-art technologies. The results confirm the heterogeneity of osteoclasts and provide new insights in this respect. The in vitro and in vivo studies suggest that osteoclast heterogeneity can be purposedly modulated; which might be useful and advisable for therapeutic purposes. Overall, the work provides hints for further implementation and future broad applications to diseases featuring pathological bone loss. The osteoclasts are generated in vitro in the presence of M-CSF and RANKL to induce tolerogenic osteoclasts or GM-CSF / Il-4 to generate inflammatory osteoclasts. The demonstration of these cell populations in the S.b. treated mice in vivo is not present. The author tried to tackle this, by analyzing the differentiation potential of bone marrow progenitor cells of S.b. treated animals, which provides some information. The effect on tolerogenic osteoclasts could have been further evaluated, whether they are not affected at all, or whether there are also effects.

Essential revisions:

1. Please briefly describe biocodex in the disclosure section and provide details of the connection, including the fact that biocodex is one of the funding bodies of the work. 3g/kg thrice a week seems like quite a high dosage, do the authors expect Sb to be effective also at lower dosages in humans? The authors could probably provide references where high-dose probiotics in mice then proved effective at lower doses in humans in the Discussion section. Do the authors have any data on the effectiveness of Sb in IBD-dependent bone loss? Could any of the identified markers be used to sort mature inflammatory osteoclasts from tolerogenic osteoclasts? (see reviewer 1's comments)

2. It would be helpful to understand the localization of the different cell types in vivo. Could the authors use a co-staining approach for osteoclast markers and the PPRs to identify this subpopulation in vivo? That would certainly strengthen the message of the manuscript. (see reviewer 2's comments)

3. By evaluating the expression profiling it was not clear, whether an unfiltered GO analysis of DEGs would reveal C-type lectin receptor signaling as well, or whether other discriminating factors would be superior. As I understood in Figure 1e there was already a pre-selection of osteoclast/resorptive associated genes. (see reviewer 2's comments)

4. The functional studies of the inhibition of PPRs seemed to decrease also strongly in cell numbers in general (Figure 4 a-d), suggesting that also progenitor cells are affected. The authors should address this through the analysis of earlier time points. (see reviewer 2's comments)

5. There are already publications on PPRs and their function on osteoclasts, e.g. for the Mincle and Dectin-1 and in these publications, the effects are claimed to be more general on osteoclasts. In part, even opposite effects are reported. The effects of silencing of these receptors on the "tolerogenic cells" and the consequence on resorption should be therefore more clear. (see reviewer 2's comments)

6. The authors talk about decreased osteoclast numbers in the histomorphometry, but the osteoclast area is shown in Figure 2f. The authors should show the numbers of multinucleated cells as well. (see reviewer 2's comments)

7. The failure to rescue cortical parameters by S.b. is intriguing, the authors should discuss it. (see reviewer 2's comments)

8. There is an apparent strong trend that osteocytes are increased in supplementary figure 2c. The authors should discuss it since they also mention trends (p=0.09) elsewhere. (see reviewer 2's comments)

9. Page 9, lines 313-314: in line with the references here provided, the authors may quantify these cytokines in their experimental conditions. (see reviewer 3's comments)

10. Please show the level of gene silencing obtained with siRNA targeting TLR2 (and control siRNA as well). (see reviewer 3's comments)

11. In the supplementary figure 2, please provide also cortical porosity data (panel a) and representative images of Osterix and Sost IHC staining for the control groups SHAM and SHAM+Sb. Please detail how osteoblast and osteocyte count was performed (how many sections per mouse? Quantization on the entire tissue section or a number of fields?) (see reviewer 3's comments)

12. Pages 12-13, Primary cell culture and osteoclast differentiation: please specify the duration of the osteoclast differentiation assay and the frequency of PRR agonist supplementation into the culture. Also, please indicate at which time point OCL activity was measured. (see reviewer 3's comments)

13. In general, we suggest the authors present all the data as scatter plots with bars (where applicable), as already done in a number of panels. (see reviewer 3's comments)

14. A point-by-point response to all three reviewers' comments.

Reviewer #1 (Recommendations for the authors):

The report presented by Madel/Halper and colleagues is very well-written, logical, easy to read, and with a very good flow and pace. All the experiments presented are scientifically sound and the conclusions are not overstated. As a person working in the field, I agree with the authors that osteoclast heterogeneity is an under-investigated and underestimated issue and that more research is needed to understand how the different sub-populations of osteoclasts participate in bone loss in pathological conditions. In this report, the authors first explore osteoclast heterogeneity using RNAseq and miRNA analyses, and then, find that specific pattern-recognition receptors are overexpressed in inflammatory osteoclasts compared to tolerogenic osteoclasts, and that this difference can also be observed in vivo in OVX mice, they proceed to administer a commercially-available strain of the probiotic yeast S. boulardii to OVX mice, observing prevention of the ovariectomy-induced bone loss, with an effect that is not oestrogen-mediated since uterus weight is comparable between treated and untreated mice. The report deserves publication in eLife, I only have a few comments that I hope will improve the quality of the manuscript even more.

Figure 2d type size is inconsistent and in general in the same figure sometimes the font size varies.

Figure 4e 10μg goes inside the space of the figure.

Supp figure 2b missing measuring units for obs and ocs.

Line 400 should read {greater than or equal to}.

Line 468 please specify how the region of interest was selected and the software and algorithms used for analysis (e.g. bruker CTan, thresholding method global).

Line 479 for how long?

A table collecting all antibodies used (also with RRID) and relative concentrations would be helpful as a supplementary. The same could be said for primer sequences/Taqman assay ID.s.

Line 533 what is the concentration of saponin 1X?

Please briefly describe biocodex in the disclosure section and provide details of the connection, including the fact that biocodex is one of the funding bodies of the work.

3g/kg thrice a week seems like quite a high dosage, do the authors expect Sb to be effective also at lower dosages in humans? The authors could probably provide references where high-dose probiotics in mice then proved effective at lower doses in humans in the Discussion section.

Do the authors have any data on the effectiveness of Sb in IBD-dependent bone loss?

Could any of the identified markers be used to sort mature inflammatory osteoclasts from tolerogenic osteoclasts?

Line 241 maybe the authors could show this data in a supplementary figure.

Did the authors evaluate the serum levels of IL4 and IL10 in OVX mice treated with Sb? Up to the authors' discretion but it would be interesting to know this, since if they are affected, they may play a part in the phenotype rescue observed, and if they are not, it would be even more interesting because everything would be caused by β-glucans or other compounds contained in the probiotics.

Were there any side effects to the probiotic treatment?

Figure 2D and all others: authors should also show the statistical comparisons between Sham-Sb and OVX-Sb.

Usually, osteoclasts are evaluated histomorphometrically using Oc.S/BS and Oc.N/BS – is there a specific reason why the authors used Tracp+ Area/Bone surface for these analyses?

It would be interesting to perform GSEA on these datasets as well, which might lead to further findings and useful data for the authors.

Again, congratulations to the authors for an interesting and well-done report.

Reviewer #2 (Recommendations for the authors):

1. It would be helpful to understand the localization of the different cell types in vivo. Could the authors use a co-staining approach for osteoclast markers and the PPRs to identify this subpopulation in vivo? That would certainly strengthen the message of the manuscript.

2. By evaluating the expression profiling it was not clear, whether an unfiltered GO analysis of DEGs would reveal C-type lectin receptor signaling as well, or whether other discriminating factors would be superior. As I understood in Figure 1e there was already a pre-selection of osteoclast/resorptive associated genes.

3. The functional studies of the inhibition of PPRs seemed to decrease also strongly in cell numbers in general (Figure 4 a-d), suggesting that also progenitor cells are affected. The authors should address this through the analysis of earlier time points.

4. There are already publications on PPRs and their function on osteoclasts, e.g. for the Mincle and Dectin-1 and in these publications, the effects are claimed to be more general on osteoclasts. In part, even opposite effects are reported. The effects of silencing these receptors on the "tolerogenic cells" and the consequence on resorption should be therefore more clear.

5. The authors talk about decreased osteoclast numbers in the histomorphometry, but the osteoclast area is shown in Figure 2f. The authors should show the numbers of multinucleated cells as well.

6. The failure to rescue cortical parameters by S.b. is intriguing, the authors should discuss it.

7. There is an apparent strong trend that osteocytes are increased in supplementary figure 2c. The authors should discuss it since they also mention trends (p=0.09) elsewhere.

8. In Figure 3a-d single data points are missing and should be added.

Reviewer #3 (Recommendations for the authors):

Suggestions to the authors:

– Additional data:

Page 9, lines 313-314: in line with the references here provided, the authors may quantify these cytokines in their experimental conditions.

Please show the level of gene silencing obtained with siRNA targeting TLR2 (and control siRNA as well).

In supplementary figure 2, please provide also cortical porosity data (panel a) and representative images of Osterix and Sost IHC staining for the control groups SHAM and SHAM+Sb. Please detail how osteoblast and osteocyte count was performed (how many sections per mouse? Quantization on the entire tissue section or a number of fields?).

– Method description:

Pages 12-13, Primary cell culture and osteoclast differentiation: please specify the duration of the osteoclast differentiation assay and the frequency of PRR agonist supplementation into the culture. Also, please indicate at which time point OCL activity was measured.

– Figure legends: Actually, we think referring to the supplementary figures in the legend to the main figures is confounding. We think the legend should describe only what is displayed in the corresponding figure.

– Text edits:

Page 3, line 60: HETEROGENEOUS instead of ETEROGENEOUS; line 78: i-OCLs (plural) instead of i-OCL; line 87: you can use the acronym i-OCLs here since it has been already introduced earlier.

Page 4, line 108: maybe use ANALYSIS instead of APPROACH; line 110: "and representing": we think AND can be deleted; line 112: "between BOTH OCL subsets": we think "between THE TWO OCL subsets" is more appropriate.

Page 5, line 150: IN HERE should be HEREIN; line 155: SHAM OCLs: we would prefer "OCLs derived from SHAM mice".

Page 7, line 219: "to a lesser extent" this can be deleted from our point of view; in fact, in both treatment groups (SHAM and OVX) the OCL % seems to be reduced by 50%; line 221: the term KINASE can be deleted here.

Page 8, line 247: the proportion OF; line 249: "Lastly,…(Figure 4g)" we are not convinced this statement is correctly placed here, since it refers to the result of an experiment described earlier, while the present paragraph deals with the effect of Sb-CM.

Page 9, line 283: "recognized by" can be replaced by FOR, in line with the construction used in the previous line; line 293: "in A several OF gastrointestinal disorders" should be "in several gastrointestinal disorders", or "in a SERIES (VARIETY or similar term) of"; line 312: "Sb IT has been described" delete IT.

Page 10, line 344: use DEMONSTRATED instead of IDENTIFIED; line 349: please check this line.

Page 12, line 365: "assigned into two groups" we suggest "DIVIDED INTO two groups"; lines 371-372: "french ministry of health. Higher education and research" all initials in capital letters.

Page 13, line 404 ALIZARINE should be ALIZARIN.

Page 15, line 499: SERIC should be SERUM.

Page 24, line 834: "FACS plots" should be "histograms".

Page 25, line 855: if we get it correct from the figure, it seems that the number of replicates is n{greater than or equal to}5 (not n=5); lines 860-861: n=?. Figure 4, panels a-c: please indicate in the legend what is represented in the graphs: mean {plus minus} s.d.? In the text and in the figures please confirm to the unit mL (not ml; at present, both are used here).

Figure 4, panel b: ZYMOZAN should be ZYMOSAN; panel h: it is not necessary to indicate the name of the specific PRR agonist used on top of each graph because it is indicated under each graph.

Figure 5, panel c and Figure 6, panel d: MHC2 should be MHC-II.

Legend to supplementary figure 2: invert cortical bone surface/bone volume and cortical thickness, in accordance with the order of the graphs in the figure; the last line: conditionned should be conditioned.

– Figures:

In general, we suggest the authors present all the data as scatter plots with bars (where applicable), as already done in a number of panels.

Figure 1, panel j, first graph: normalized to…?

Figure 2: we think it could be interesting to see a representative plot, to visualize double positive cells.

Figure 3, panels e and f: use "SHAM+Sb" and "OVX+Sb", in line with the other panels and figures.

Reviewer #1 (Recommendations for the authors):

The report presented by Madel/Halper and colleagues is very well-written, logical, easy to read, and with a very good flow and pace. All the experiments presented are scientifically sound and the conclusions are not overstated. As a person working in the field, I agree with the authors that osteoclast heterogeneity is an under-investigated and underestimated issue and that more research is needed to understand how the different sub-populations of osteoclasts participate in bone loss in pathological conditions. In this report, the authors first explore osteoclast heterogeneity using RNAseq and miRNA analyses, and then, find that specific pattern-recognition receptors are overexpressed in inflammatory osteoclasts compared to tolerogenic osteoclasts, and that this difference can also be observed in vivo in OVX mice, they proceed to administer a commercially-available strain of the probiotic yeast S. boulardii to OVX mice, observing prevention of the ovariectomy-induced bone loss, with an effect that is not oestrogen-mediated since uterus weight is comparable between treated and untreated mice. The report deserves publication in eLife, I only have a few comments that I hope will improve the quality of the manuscript even more.

We thank the reviewer for it's positive comments.

Figure 2d type size is inconsistent and in general in the same figure sometimes the font size varies.

Figure 4e 10μg goes inside the space of the figure.

Supp figure 2b missing measuring units for obs and ocs.

All these points have been corrected in the corresponding figures.

Line 400 should read {greater than or equal to}.

This has been corrected.

Line 468 please specify how the region of interest was selected and the software and algorithms used for analysis (e.g. bruker CTan, thresholding method global).

The requested information has been added in the material and methods section.

Line 479 for how long?

The information has been added in the material and methods section.

A table collecting all antibodies used (also with RRID) and relative concentrations would be helpful as a supplementary. The same could be said for primer sequences/Taqman assay ID.s.

The key resources table has been provided with the revised manuscript.

Line 533 what is the concentration of saponin 1X?

The concentration is 1 g/100 mL. The information has been added in the material and method.

Please briefly describe biocodex in the disclosure section and provide details of the connection, including the fact that biocodex is one of the funding bodies of the work.

As requested, in the competing interest section we have indicated that Biocodex is a pharmaceutical laboratory and we have provided the requested information.

3g/kg thrice a week seems like quite a high dosage, do the authors expect Sb to be effective also at lower dosages in humans? The authors could probably provide references where high-dose probiotics in mice then proved effective at lower doses in humans in the Discussion section.

We agree with the reviewer that the dosage of Sb we used in vivo (~65 mg/mouse corresponding to ~3.10⁹ CFU/mouse/day, 3 times a week for 4 weeks) is much higher than the one used in human (200 mg/person/day, corresponding to ~10¹⁰ CFU). However, this range of dose is frequently used in studies evaluating the effects of Sb on chronic diseases in mouse (as for instance in 2 diabetic models in which mice received 0.5×10⁸ CFU of Sb/mouse/day for 8 weeks) (https://doi.org/10.1016/j.lfs.2022.120616) or 120mg/mouse/day for 4 weeks (DOI: 10.1128/mBio.01011-14).

Interestingly, in the context of osteoporosis, equivalent differences in the dosage of bacterial probiotics used in human and mouse have been reported. For instance, Lactobacillus strains were shown to reduce bone loss in osteoporotic at 1x10⁹ CFU/ml of drinking water/day (corresponding to about 4,5x10⁹ CFU/mouse/day) for 6 weeks, and this protective effect has been confirmed in human with a dose of 10¹⁰ CFU/ patient /day for 1 year (https://doi.org/10.1016/S2665-9913(19)30068-2). Same result was reported for Lactobacillus reuteri which is protective in mouse at a dose of 3x10⁸ CFU/3 times a week for 4 weeks (https://doi.org/10.1002/jcp.24636) and which reduces bone loss in osteoporotic patients at dose of 1x10¹⁰ CFU/patient/day for 1 year (DOI: 10.1111/joim.12805 ).

Therefore, based on such studies showing protective effect of probiotics with much lower doses in human compared to mouse, we can assume that lower doses of Sb would also be effective in human, even though the clear answer to this question would require a clinical trial in osteoporotic patients. This has been discussed in the Discussion section of our revised manuscript.

Do the authors have any data on the effectiveness of Sb in IBD-dependent bone loss?

Unfortunately, we don't have any data on the effect of Sb on IBD-dependent bone loss.

Could any of the identified markers be used to sort mature inflammatory osteoclasts from tolerogenic osteoclasts?

From our transcriptomic data we indeed identified markers, including those described here, that we are using to sort and analyse inflammatory and non-inflammatory osteoclasts for new specific properties. However, these new data will be the subject of another manuscript that is in preparation. Therefore we prefer not to include them in our manuscript as we believe it's not essential for this study. We hope the reviewer will understand.

Line 241 maybe the authors could show this data in a supplementary figure.

The requested data on cell apoptosis have been included in Figure 4—figure supplement 1a-c.

Did the authors evaluate the serum levels of IL4 and IL10 in OVX mice treated with Sb? Up to the authors' discretion but it would be interesting to know this, since if they are affected, they may play a part in the phenotype rescue observed, and if they are not, it would be even more interesting because everything would be caused by β-glucans or other compounds contained in the probiotics.

We performed the requested experiment by dosing serum concentration of IL-10 and IL-4 by ELISA. The results showed no significant difference in IL-10 concentration between OVX and OVX+Sb mice (see Author response image 1). This result has been added as data not shown. IL-4 was not detected in any of the conditions.

Author response image 1

Download asset Open asset

Were there any side effects to the probiotic treatment?

We did not observed any side effects of the probiotic in the treated mice.

Figure 2D and all others: authors should also show the statistical comparisons between Sham-Sb and OVX-Sb.

Figures 2 and 3 have been modified as requested to show the statistical comparisons between Sham-Sb and OVX-Sb.

Usually, osteoclasts are evaluated histomorphometrically using Oc.S/BS and Oc.N/BS – is there a specific reason why the authors used Tracp+ Area/Bone surface for these analyses?

In our hands, TRAcP staining in vivo is usually very intense, which makes it impossible to always see the limit between 2 OCLs and all nuclei, and thus to properly count individually these cells. That's why in all our previous studies, we have always evaluated OCL area and not their number. This approach has already been validated in a number of publications (for instance = Mansour et al., Cell research 2011. DOI:10.1038/cr.2011.21; Mansour et al., J Exp Med 2012. DOI:10.1084/jem.20110994; Moukengue B et al. EBioMedicine. 2020. doi: 10.1016/j.ebiom.2020.102704; Guihard P et al. Am J Pathol. 2015. doi: 10.1016/j.ajpath.2014.11.008; Lamoureux F, et al. Nat Commun. 2014. doi: 10.1038/ncomms4511; Gobin B, et al. Int J Cancer. 2015. doi: 10.1002/ijc.29040; Gobin B et al. Cancer Lett. 2014. doi: 10.1016/j.canlet.2013.11.017; Velasco CR, et al. Glycobiology. 2011. doi: 10.1093/glycob/cwr002).

However, to comply with the reviewer's comment, we modified the text not to talk about OCL number.

It would be interesting to perform GSEA on these datasets as well, which might lead to further findings and useful data for the authors.

This analysis has been performed (see Author response table 1). As it gave results equivalent to those presented in figure 1c, we did not include it in the revised manuscript.

Again, congratulations to the authors for an interesting and well-done report.

We thank the reviewer for it's positive comments.

Reviewer #2 (Recommendations for the authors):

1. It would be helpful to understand the localization of the different cell types in vivo. Could the authors use a co-staining approach for osteoclast markers and the PPRs to identify this subpopulation in vivo? That would certainly strengthen the message of the manuscript.

We agree with the reviewer that this is an interesting question. However, we have a new article in preparation in which the specific localization and new properties of inflammatory and tolerogenic osteoclasts will be analyzed in detail. Thus, we prefer not to include part of these new data in the present manuscript so as not to limit the interest of this new paper. We hope the reviewer will understand our position.

2. By evaluating the expression profiling it was not clear, whether an unfiltered GO analysis of DEGs would reveal C-type lectin receptor signaling as well, or whether other discriminating factors would be superior. As I understood in Figure 1e there was already a pre-selection of osteoclast/resorptive associated genes.

The reviewer is right, the analysis in former Figure 1e is done on a list of genes preselected from the Kegg pathway mmu04380 (osteoclast differentiation). In line with the reviewer's suggestion, we performed similar analysis on the totality of the genes differentially express between MN-OCLs and DC-OCLs, which also highlighted the c-lectin like receptor pathway. This new analysis has been added in Figure 1d.

3. The functional studies of the inhibition of PPRs seemed to decrease also strongly in cell numbers in general (Figure 4 a-d), suggesting that also progenitor cells are affected. The authors should address this through the analysis of earlier time points.

As requested by the reviewer, we performed a viability assay by measuring the apoptotic cells (Annexin –V+ PI+) by FACS 24 and 48 hours after the addition of agonists. Our results revealed that neither the agonists, nor the Sb-conditioned medium, induced cell apoptosis. These results have been added in Figure 4—figure supplement 1a-d.

4. There are already publications on PPRs and their function on osteoclasts, e.g. for the Mincle and Dectin-1 and in these publications, the effects are claimed to be more general on osteoclasts. In part, even opposite effects are reported. The effects of silencing these receptors on the "tolerogenic cells" and the consequence on resorption should be therefore more clear.

We fully agree with the comment of reviewer. It's true that in the literature, it has been reported that curdlan reduces global osteoclast differentiation through Dectin-1. However, this effect is observed only with high concentrations (range of 10 µg/mL and more) of curdlan (https://doi.org/10.18632/oncotarget.18411) whereas the effect we observed specifically on i-OCLs is already achieved with 10 ng/mL. It has also been reported that the Dectin-1 protein is expressed in BM OCL progenitors but at low levels (https://doi.org/10.1074/jbc.M114.551416) and that Dectin-1 activation by curdlan efficiently reduces OCL formation from RAW cells only when cells are transfected with Dectin-1 (https://doi.org/10.1074/jbc.M114.551416). Although these studies have already been discussed in our manuscript, we have expanded this point in the revised version (see "discussion").

Regarding Mincle, we also agree that opposite effects have been reported in the literature and indeed stimulation of Mincle by necrotic osteocytes stimulates OCL differentiation (doi: 10.1172/JCI134214). We therefore performed in vitro experiments on BM cells from Mincle KO mice and microscanner analysis of these mice. For the microscanner, we obtained results similar to those of Andreev et al. 2020, with an increased bone mass in Mincle (Clec4e) KO mice. However, for the in vitro experiments, our results are divergent with this study. We performed osteoclastogenic differentiation in vitro from total BM cells, BM-derived DCs and BM monocytes, and we observed a tendency to an increased osteoclastogenesis, whatever the OCL progenitor we used. This discrepancy may be explain by the use of a different protocol for osteoclastogenesis as we start the differentiation adding M-CSF and RANKL together, while Andreev et al. 2020 add M-CSF first and 2 days later they add RANKL. Furthermore, we tested the effect of the Mincle agonist GlcC14C18 on osteoclastogenesis from Mincle KO DCs and our results show that the inhibition of osteoclastogenesis is reduced compared to WT confirming the implication of Mincle in this inhibition.

Therefore, these observations revealed the complex effects of Mincle on osteoclastogenesis, as already shown on other monocytic cells. These new results have been added in in Figure 4-Suppl Figure 2 and commented in the discussion.

Lastly, as requested, we performed additional experiment to assess the effect of agonist on the resorption capacity of tolerogenic OCLs, which did not reveal any significant effect. The data were added in Figure 4—figure supplement 1.

5. The authors talk about decreased osteoclast numbers in the histomorphometry, but the osteoclast area is shown in Figure 2f. The authors should show the numbers of multinucleated cells as well.

As explained for reviewer 1, in our hands, TRAcP staining in vivo is usually very intense, which makes it impossible to always see the limit between 2 OCLs and all nuclei, and thus to properly count individually these cells. That's why in all our previous studies, we have always evaluated OCL area and not their number. This approach has already been validated in a number of publications (for instance = Mansour et al., Cell research 2011. DOI:10.1038/cr.2011.21; Mansour et al., J Exp Med 2012. DOI:10.1084/jem.20110994; Moukengue B et al. EBioMedicine. 2020. doi: 10.1016/j.ebiom.2020.102704; Guihard P et al. Am J Pathol. 2015. doi: 10.1016/j.ajpath.2014.11.008; Lamoureux F, et al. Nat Commun. 2014. doi: 10.1038/ncomms4511; Gobin B, et al. Int J Cancer. 2015. doi: 10.1002/ijc.29040; Gobin B et al. Cancer Lett. 2014. doi: 10.1016/j.canlet.2013.11.017; Velasco CR, et al. Glycobiology. 2011. doi: 10.1093/glycob/cwr002).

However, to comply with the reviewer's comment, we modified the text not to talk about OCL number.

6. The failure to rescue cortical parameters by S.b. is intriguing, the authors should discuss it.

Such a lack of effect on cortical bone has also been reported in OVX mice treated with bacterial probiotics (McCabe et al., 2013, DOI 10.1002/jcp.24340; Britton et al., 2014, DOI 10.1002/jcp.24636 ; Lawenius et al., 2022, DOI 10.1530/JOE-21-0408 …). The reasons are not clear but it could be due to a too short period of treatment, a starting point too late after surgery, or a preferential location of certain OCL subtypes in the trabecular versus cortical bone.

This point has been commented in the "results" section.

7. There is an apparent strong trend that osteocytes are increased in supplementary figure 2c. The authors should discuss it since they also mention trends (p=0.09) elsewhere.

The comparison of osteocyte numbers between the OVX and OVX+Sb groups yielded a p=0.22, far from significance. In contrast, the pVal mentioned by the reviewer (Figure 3c-d) are lower, despite not reaching significance (p=0.09 and 0.08 for butyrate and lactate, respectively). Moreover, the different metabolites we measured all showed a similar pattern of variation, in particular an increase after Sb treatment in OVX mice. It's because they all varied the same way that we discussed this tendency in this figure.

8. In Figure 3a-d single data points are missing and should be added.

Single data points have been added in all the figures.

Reviewer #3 (Recommendations for the authors):

Suggestions to the authors:

– Additional data:

Page 9, lines 313-314: in line with the references here provided, the authors may quantify these cytokines in their experimental conditions.

As explained for reviewer 1, We performed the requested experiment by dosing by ELISA serum concentration of IL-10 and IL-4. The results showed no significant difference in IL-10 concentration between OVX and OVX+Sb mice (see Author response image 1). This result has been added as data not shown.

IL-4 was not detected in any of the conditions.

Please show the level of gene silencing obtained with siRNA targeting TLR2 (and control siRNA as well).

As requested, these data have been added in Figure 4—figure supplement 1e.

In supplementary figure 2, please provide also cortical porosity data (panel a) and representative images of Osterix and Sost IHC staining for the control groups SHAM and SHAM+Sb. Please detail how osteoblast and osteocyte count was performed (how many sections per mouse? Quantization on the entire tissue section or a number of fields?).

Unfortunately, it is not possible to provide data on cortical porosity with the parameters that have been recorded in the µCT facility. And we don't have anymore bone sample left to repeat the anaysis for cortical porosity. We hope the reviewer will understand.

The images of Osterix and Sost staining has been added for SHAM and SHAM+Sb groups in Figure 2-supplement figure 2. The requested information on cell count and quantization has been added in the material and method section.

– Method description:

Pages 12-13, Primary cell culture and osteoclast differentiation: please specify the duration of the osteoclast differentiation assay and the frequency of PRR agonist supplementation into the culture. Also, please indicate at which time point OCL activity was measured.

The requested information has been added in the material and method section.

– Figure legends: Actually, we think referring to the supplementary figures in the legend to the main figures is confounding. We think the legend should describe only what is displayed in the corresponding figure.

This is a request of the journal. It is stated in the author guide that "Figure Supplements should be referred to in the legend of the associated primary figure".

Page 7, line 219: "to a lesser extent" this can be deleted from our point of view; in fact, in both treatment groups (SHAM and OVX) the OCL % seems to be reduced by 50%; line 221: the term KINASE can be deleted here.

We removed "to a lesser extend". The other points have been modified in the text.

Page 8, line 247: the proportion OF; line 249: "Lastly,…(Figure 4g)" we are not convinced this statement is correctly placed here, since it refers to the result of an experiment described earlier, while the present paragraph deals with the effect of Sb-CM.

This sentence also refers to Sb-CM, which, we agree, was not clearly indicated. This has been corrected. Morevoer, we deleted the SB-CM data from figure 4g to put it as a new figure (6b) to have all the data on Sb-CM in the same figure.

Page 25, line 855: if we get it correct from the figure, it seems that the number of replicates is n{greater than or equal to}5 (not n=5); lines 860-861: n=?. Figure 4, panels a-c: please indicate in the legend what is represented in the graphs: mean {plus minus} s.d.? In the text and in the figures please confirm to the unit mL (not ml; at present, both are used here).

Figure 4, panel b: ZYMOZAN should be ZYMOSAN; panel h: it is not necessary to indicate the name of the specific PRR agonist used on top of each graph because it is indicated under each graph.

Figure 5, panel c and Figure 6, panel d: MHC2 should be MHC-II.

Legend to supplementary figure 2: invert cortical bone surface/bone volume and cortical thickness, in accordance with the order of the graphs in the figure; the last line: CONDITIONNED should be CONDITIONED.

All these points have been modified in the text and figures.

– Figures:

In general, we suggest the authors present all the data as scatter plots with bars (where applicable), as already done in a number of panels.

We followed the reviewer's recommendation and presented the data as requested.

Figure 1, panel j, first graph: normalized to…?

"Normalized to mode" is a classical representation of FACS data on histograms. It's a normalization made by the sofware (FlowJo) to harmonize the height of the histograms that could change among the graphs depending on the number of cells that have been acquired in each sample.

Figure 2: we think it could be interesting to see a representative plot, to visualize double positive cells.

Unfortunately, in these experiments the markers were analyzed independently so it's not possible to show double positive cells.

Author details

1. Maria-Bernadette Madel

   1. Université Côte d’Azur, CNRS, LP2M, Nice, France
   2. LIA ROPSE, Laboratoire International Associé Université Côte d’Azur - Centre Scientifique de Monaco, Nice and Monaco, France

   Contribution

   Formal analysis, Investigation, Methodology, Validation, Writing – original draft, Writing – review and editing

   Contributed equally with

   Julia Halper

   Competing interests

   No competing interests declared

2. Julia Halper

   1. Université Côte d’Azur, CNRS, LP2M, Nice, France
   2. LIA ROPSE, Laboratoire International Associé Université Côte d’Azur - Centre Scientifique de Monaco, Nice and Monaco, France

   Contribution

   Formal analysis, Investigation, Methodology, Validation, Writing – review and editing

   Contributed equally with

   Maria-Bernadette Madel

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0003-3967-6633

3. Lidia Ibáñez

   Department of Pharmacy, Cardenal Herrera-CEU University, Valencia, Spain

   Contribution

   Formal analysis, Investigation, Methodology, Writing – review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-2217-8817

4. Lozano Claire

   IRMB, Université Montpellier, Montpellier, France

   Contribution

   Formal analysis, Investigation, Supervision

   Competing interests

   No competing interests declared

5. Matthieu Rouleau

   1. Université Côte d’Azur, CNRS, LP2M, Nice, France
   2. LIA ROPSE, Laboratoire International Associé Université Côte d’Azur - Centre Scientifique de Monaco, Nice and Monaco, France

   Contribution

   Formal analysis, Investigation, Methodology, Writing – review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-0075-9880

6. Antoine Boutin

   1. Université Côte d’Azur, CNRS, LP2M, Nice, France
   2. LIA ROPSE, Laboratoire International Associé Université Côte d’Azur - Centre Scientifique de Monaco, Nice and Monaco, France

   Contribution

   Formal analysis, Investigation, Methodology

   Competing interests

   No competing interests declared

7. Adrien Mahler

   1. Université Côte d’Azur, CNRS, LP2M, Nice, France
   2. LIA ROPSE, Laboratoire International Associé Université Côte d’Azur - Centre Scientifique de Monaco, Nice and Monaco, France

   Contribution

   Formal analysis, Funding acquisition

   Competing interests

   No competing interests declared

8. Rodolphe Pontier-Bres

   1. LIA ROPSE, Laboratoire International Associé Université Côte d’Azur - Centre Scientifique de Monaco, Nice and Monaco, France
   2. Centre Scientifique, de Monaco, Monaco

   Contribution

   Formal analysis

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-4139-5913

9. Thomas Ciucci

   Laboratory of Immune Cell Biology, Center for Cancer Research, National Cancer Institute, National Institutes of Health, Bethesda, United States

   Present address

   David H. Smith Center for Vaccine Biology and Immunology, Department of Microbiology and Immunology, University of Rochester Medical Center, Rochester, United States

   Contribution

   Conceptualization, Writing – review and editing, Writing – original draft

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-5828-0207

10. Majlinda Topi

    1. Université Côte d’Azur, CNRS, LP2M, Nice, France
    2. LIA ROPSE, Laboratoire International Associé Université Côte d’Azur - Centre Scientifique de Monaco, Nice and Monaco, France

    Contribution

    Formal analysis

    Competing interests

    No competing interests declared

11. Christophe Hue

    Université Paris-Saclay, UVSQ, INSERM, Infection et inflammation, Montigny-Le-Bretonneux, France

    Contribution

    Writing – review and editing

    Competing interests

    No competing interests declared

12. Jerome Amiaud

    Inserm, Universite de Nantes, Nantes, France

    Contribution

    Formal analysis, Investigation

    Competing interests

    No competing interests declared

13. Salvador Iborra

    Department of Immunology, Ophthalmology and ENT. School of Medicine, Universidad Complutense de Madrid, Madrid, Spain

    Contribution

    Writing – review and editing, Validation

    Competing interests

    No competing interests declared

14. David Sancho

    Immunobiology Laboratory, Centro Nacional de Investigaciones Cardiovasculares (CNIC), Madrid, Spain

    Contribution

    Writing – review and editing, Investigation

    Competing interests

    No competing interests declared

15. Dominique Heymann

    Université de Nantes, Institut de Cancérologie de l’Ouest, Saint Herblain, France

    Contribution

    Conceptualization, Supervision, Writing – review and editing

    Competing interests

    No competing interests declared

16. Henri-Jean Garchon

    1. Université Paris-Saclay, UVSQ, INSERM, Infection et inflammation, Montigny-Le-Bretonneux, France
    2. Genetics Division, Ambroise Paré Hospital, AP-HP, Boulogne-Billancourt, France

    Contribution

    Formal analysis, Funding acquisition, Supervision, Writing – review and editing, Provided and participated in experiments on CD11c-Syk mutant mice

    Competing interests

    No competing interests declared

17. Dorota Czerucka

    1. LIA ROPSE, Laboratoire International Associé Université Côte d’Azur - Centre Scientifique de Monaco, Nice and Monaco, France
    2. Centre Scientifique, de Monaco, Monaco

    Contribution

    Funding acquisition, Writing – review and editing, Provided and participated in experiments on CD11c-Syk mutant mice

    Competing interests

    received a research grant from Biocodex. Biocodex had no role in the design of the study, in the analysis and interpretation of the data and in the preparation of the manuscript

18. Florence Apparailly

    IRMB, Université Montpellier, Montpellier, France

    Contribution

    Formal analysis, Funding acquisition, Supervision, Writing – review and editing

    Competing interests

    No competing interests declared

19. Isabelle Duroux-Richard

    IRMB, Université Montpellier, Montpellier, France

    Contribution

    Formal analysis, Investigation, Writing – review and editing

    Competing interests

    No competing interests declared

20. Abdelilah Wakkach

    1. Université Côte d’Azur, CNRS, LP2M, Nice, France
    2. LIA ROPSE, Laboratoire International Associé Université Côte d’Azur - Centre Scientifique de Monaco, Nice and Monaco, France

    Contribution

    Conceptualization, Supervision, Methodology, Writing – review and editing

    Contributed equally with

    Claudine Blin-Wakkach

    Competing interests

    No competing interests declared

21. Claudine Blin-Wakkach

    1. Université Côte d’Azur, CNRS, LP2M, Nice, France
    2. LIA ROPSE, Laboratoire International Associé Université Côte d’Azur - Centre Scientifique de Monaco, Nice and Monaco, France

    Contribution

    Conceptualization, Formal analysis, Funding acquisition, Methodology, Project administration, Supervision, Writing – original draft, Writing – review and editing

    Contributed equally with

    Abdelilah Wakkach

    For correspondence

    claudine.blin@univ-cotedazur.fr

    Competing interests

    received a research grant from Biocodex. Biocodex had no role in the design of the study, in the analysis and interpretation of the data and in the preparation of the manuscript

    "This ORCID iD identifies the author of this article:" 0000-0002-2621-3907

• 575

  Page views

• 88

  Downloads

• 0

  Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.