Let me start with a short introduction by summarizing what we know about dendritic cells.
Dendritic cells (DCs) function in the immune system. Their job is to ‘sample’ the environment and to present these ‘samplings’ on their surface to other cells of the immune system (they are called antigen presenting cells – APCs). DCs are found in small quantities scattered throughout tissues that come in contact with the outside world. Two types of very well known DCs today are plasmacytoid DCs (pDCs – type I interferon producing) and conventional DCs (cDCs), which could divide into 2 or 3 more subsets based on different expression of surface markers. Another type is migratory DCs, for example Langerhans cells. The DC lineage is very heterogeneous and could originate from common lymphoid or common myeloid progenitors. Until recently, it was unclear whether common a DC progenitor exist or not. Recent studies published in Nature Immunology described, for the first time, a common progenitor for cDCs and pDCs in mouse bone marrow.
Today I’m interviewing the senior first author one of the paper: Identification of clonogenic common Flt3+M-CSFR+ plasmacytoid and conventional dendritic cell progenitors in mouse bone marrow – Dr. Nobuyuki Onai.
Dr. Onai is the assistant professor of Immunology at University of Akita (Japan).
Dear Dr. Onai I’d like to thank you for this interview. Firstly, I’d like to ask you some general questions about DCs.
1. Can you explain in simple terms what we know today about the DC lineage? What we still don’t know? And what did you find out in your study?
NO: DCs are heterogeneous cells of antigen presenting cell. In steady state spleen and lymph node, DCs are divided into two main population, plasmacytoid DC (pDC) and conventional DCs (cDC) subsets.
In our study, we identified common dendritic progenitors (CDP)
Many studies have been done DC subset ontogeny in the second lymphoid organs; however, we don’t know the DC ontogeny in the other organ such as lung or mucosa.
And also, we don’t know whether there are similar CDP counterparts in men.
2. Because pDCs and cDCs could arise from common lymphoid (CLP) as well as myeloid progenitors (CMP), it seems like only the DC lineage breaks all canonical schemes of hematopoiesis, in which multipotent progenitors divide into 2 main branches – myeloid and lymphoid. Is that what a lot of studies showed before?
ON: Yes, many studies showed in mice and two studies have been in men. In the mice, pDC and cDC subsets are generated from CMP, CLP, GMP, and pro-T1 cells (DN1 cells) (Traver et all., 2000; Manz et al., 2001 ; Wu et al., 2001 ; Shigematus et al., 2004 ; Karsunky et al., 2005). In men, Galy et al (1995) originally reported that human CLP gave rise to DCs, and later, Chicha et al., (2004) showed that both human CMP and CLP give rise to both pDC and cDC.
3. You isolate common DC progenitor (CDP) based on high expression of Flt3 and M-CSF-Receptor within lineage negative bone marrow cell population. How did you get the idea for your study? What were the clues that led to your discovery?
ON: We were trying to identify CDP based on the relevant cytokine receptor expression which is important for DC development. Flt3 is one candidate receptor because Flt3-ligand deficient mice severely reduced no. of pDC and cDC. And Flt3L is only cytokine to induce both pDC and cDC development in vitro. Another cytokine receptor such as GM-CSF and M-CSF might be candidate.
GM-CSF frequently used in both laboratory research and clinical applications for generating cDC, but role of GM-CSF in steady-state DC development might be dispensable since both GM-CSF null mice and GM-CSF receptor null mice have all DC subsets in lymphoid organs.
M-CSF and M-CSFR deficient mice have relatively normal amounts of DC, but do not develop monocyte and Langerhans cells. And, only antibodies against M-CSFR were available. Thus, we use Flt3 and M-CSFR.
To identify the DC restricted progenitors, we need to exclude defined HSC and progenitors such as LT-HSC, ST-HSC, MPP, CMP, and CLP, which are within lineage negative c-kit high population (including HSC compartments and myeloid progenitors) or c-kit intermediate IL-7R positive fraction (including CLP), and progenitors might be just down stream of CMP or CLP, but not lineage positive.
Thus, we focused on the lineage negative c-kit intermediate/low IL-7R negative fraction.
4. What is the place of the common DC progenitor that you described in the hematopoietic hierarchy? Is it in the same place with CMP and CLP (new separate branch arising from the multipotent progenitor) or is it downstream of them?
ON: We showed that lineage negative c-kit posi Sca-1 posi fraction and Flt3 positive CMP fraction clearly give rise to CDP in vivo, but Flr3 positive CLP give rise to few CDP.
Thus, CDP is it downstream then, not same place.
5. I was wondering how the cDC progenitor that you described is highly restricted to DC lineage. Did you check for NK cell markers as well? I know that early NK progenitors are highly dependent on Flt3-ligand – which was the cytokine that you used to generate all DC from cDC. Maybe NK cell and DC- lineage shared a common progenitor? It can be checked by adding IL-2 or IL-15 in your cell culture of CDP.
We checked NK1.1 and DX5 expression of offspring from CDP in vitro and in vivo, we could not detect these NK cell markers expression in them. Thus, we believe CDP do not give rise to NK cells in vitro and in vivo. But we did not try to add IL-2 or IL-15 in our culture.
Yes, for example pro-T1 cells (DN1 cells) in the thymus, gave rise to T cells, DC, and NK cells, but we don’t know NK and DC share common progenitor.
6. Another research group (Naik, et al) published a description of quite a similar CDC progenitor in the same issue of Nature Immunology. Can you comment about their study? What did they do differently from you?
ON: Naik identified our CDP like cells (they called pro-DC cells) using in vitro Flt3L high dose culture system derived from lineage negative BM cells.
Pro-DC gave rise to pDC, CD8+ DC, and CD8- DC on a single cell level in vitro and on a population level, pDC, CD8+ DC, and CD8- DC in the spleen and BM of the recipient mice.Their work is nicely complementary to our finding.
7. Some studies suggest that monocytes and their progenitors can also contribute to the DC lineage and in some conditions (inflammation) DCs originates from monocytes. Miriam Merad in her commentary pointed out that your study as well as Naik’s study “do not support the tentative theory that the macrophage DC precursor (MDP) (described by Fogg et al. in 2006) is a major pathway for DC development in lymphoid organs in vivo”. What do you think about it? What is the role of MDPs in DC lineage?
ON: Yes, this is important question. We would like to do direct comparison of CDP and MDP in DC lineage, however, to isolate MDP, we need CX3CR1-GFP knock-in mice, which are not commercially available.
8. Is it really important to know how many clearly defined progenitors there are in one lineage? How far should we go in discovering the specifics of each lineage?
ON: This is also really important, but I do not know.
9. How could our basic knowledges about different DC progenitors be translated to clinic? Can we manipulate by this lineage for patient benefits in case of immunological disorders, infectious diseases and cancer?
ON: First of all, our finding CDP might be useful for understanding critical molecular or instructive events on DC commitment from progenitors or pDC vs cDC commitments.
If we find human counterparts of CDP, it may help to guide modulation of the DC compartment in clinical setting. But I am not sure how feasible to manipulate CDP for patient benefit. Of course, I would like to try it, if possible.
Interviewers: Alexey Bersenev & Joanna Balcerek