haematopoiesis

Question 1

frequency of HSCs in bone marrow

Answer 1

1:5000

Question 2

markers of HSCs

Answer 2

Defined as c-kit+ Sca-1+ Lin- in mice, CD34+ CD38- in humans

Question 3

path of HSC differentiation

Answer 3

Hemangioblasts  LT-HSCs  ST-HSCs  multipotent progenitor cells  common myeloid and lymphoid progenitors  range of blood cells

Question 4

how many haematopoietic cells generated per day

Answer 4

4 - 5 x 10^11

Question 5

main function of osteoblastic niche

Answer 5

Protects pool of long-term HSCs (LT-HSCs) = possess lifetime repopulating capacity
Helps maintain quiescence by exerting a dominant inhibitory effect on HSC commitment
Found at the endosteal surface (inner surface of trabecular bones)

Why is quiescence important?
Prevents exhaustion of HSC pool so as to permit haematopoiesis throughout life
Genetic integrity of HSC pool is preserved by rigorously controlling entry into the cell cycle and hence the introduction of potentially deleterious mutations
No other tissue must respond so rapidly to environmental challenges and undergo such a dramatic level of expansion involving complex gene rearrangements
Without such a high level of control, the likelihood of transformation is high

Question 6

study showing HSCs home to osteoblastic niche

Answer 6

Xie et al (2008) = real-time imaging showing HSCs home to osteoblastic niche

1. Irradiated mice  transplanted with purified GFP+ HSCs  harvested femurs + tibias after transplantation and used ex vivo real-time imaging to track homing of transplanted HSCs
   - Found that GFP+ HSCs preferentially homed to endosteal region of trabecular bone
2. Confirmed result in live mice
   - Used Scl-TVA transgenic mice = express avian retrovirus receptor in cells with active Scl-promoter-3’-enhancer regulatory elements = predominantly HSCs
   - When mice are injected with avian virus containing luciferase reporter, only Scl-TVA+ cells are susceptible to infection = can be visualised by live-imaging bioluminescence
   - detected strong persistent signals in trabecular bone area of both legs and other regions

Question 7

evidence for osteoblastic niche

Answer 7

Zhang et al (2003) = supports idea of a niche
• Conditional KO of BMP receptor type 1A (BMPr1a) gene using Cre-lox
• Flow cytometry detected 2-fold increase in number of HSCs, BrdU labelling revealed substantial increase in percentage of LT-HSCs (BrdU-negative as not cycling) in KO animals
• BMPR1A not expressed in HSCs  suggests that change in HSC number must result from extrinsic defect
• Then found that number of endosteal osteoblasts increased in KO animals
• Transplantation of WT HSCs into lethally irradiated BMPR1A mutants led to greater expansion of HSCs, when compared to transplanted WT controls
• Evidence that microenvironment regulates stem cells in vivo

Question 8

importance of osteoblastic niche in maintaining HSC integrity

Answer 8

• Also important in maintaining HSC integrity

• Immune privileged to afford protection from inflammatory or autoimmune challenges
• Sequestered nature of niche at endosteal surface offers protection from environmental radiation and mutagens
• Low O2 tension reduces exposure to oxidative stress
• Low blood supply ensures protection from all but the highest concentration of IFNα, buffering HSCs from small fluctuations in plasma cytokine levels  important as IFNα produced on infection initiates increased haematopoiesis  protection ensures quiescence isn’t comprised when increase in cytokine levels is only small

Question 9

evidence of immune privilege in ON

Answer 9

Fujisaki et al (2011) = evidence of immune privilege
• Looked at survival of HSCs in bone marrow transplants to non-irradiated allogeneic (genetically distinct) or syngeneic (genetically identical) recipients
• Persistence of HSCs was similar over first 30 days
• Persistence in allogeneic recipient was due to presence of regulatory T cells in osteoblastic niche  90% of KLS HSCs were found within 20um of Treg cells  protected cells from rejection
• Depletion of Treg cells using monoclonal antibodies specific for CD25 produced selective loss of HSCs in allogeneic but not syngeneic recipients

Question 10

Key cells in ON

Answer 10

SNO cells (spindle-shaped, N-cadherin positive osteoblasts)
• Subset of highly specialised osteoblasts
• Stromal cells that make up osteoblastic niche
• HSCs are anchored and actively tethered to SNO cells, via N-cadherin on SNO cells and α4β1 integrins on HSCs = helps maintain quiescence by preventing entry into cell division

CAR cells (CXCL12 abundant reticulocytes) 
•	Secrete chemokine CXCL12 = important in initiating night-time waves of haematopoiesis

Question 11

key molecules in ON

Answer 11

Secreted factors
1. Angiopoietin 1 (Ang1) released by osteoblasts  interacts with Tie-2 receptor on HSCs  activates expression of B1-integrin and N-cadherin  enhanced adhesion of HSCs to stroma

2. Osteopontin  supports adhesion and negatively HSC proliferation = quiescence
3. RANK and BMP = inhibitory factors

Cell-bound factors
• HSCs are anchored and tethered to SNO cells via interactions between N-cadherin and α4β1 integrins  dissuades HSCs from entering mitosis
• Stem cell factor expressed by SNO cells binds c-KIT on HSCs = important in maintaining viability

Question 12

controversy surrounding N-cadherin

Answer 12

• Wilson et al (2004) = found that c-kit+ Sca-1+ Lin- cells are mixture of N-cadherin positive and N-cadherin negative based on staining with YS antibody, but whether YS-antibody-positive cells are LT-HSCs was unclear

• Created a model in which HSCs lacking c-myc expression  maintain high levels of N-cadherin expression  enhance stem cell niche interaction and promote HSC maintenance
• HSCs with high c-myc expression  repress adhesion molecules  no interaction with SCN  progressive exhaustion of stem cell pool
• Haug et al (2008) = used MNCD2 antibodies to identify N-cadherin + showed that YS polyclonal antibody was unreliable in identifying N-cadherin
• Showed that it was cells with low N-cadherin expression that are able to reconstitute irradiated mice = conflicts Wilson’s results

• Kiel et al (2009)
• Achieved conditional KO of N-cadherin from HSCs and other haematopoietic cells  flanked N-cadherin w LoxP sites, expressed Cre under Mx-1 promoter
- No observable impact on HSC frequency, bone marrow cellularity, lineage composition, ability to sustain haematopoiesis over time
• Then transplanted irradiated WT mice with mixture of WT and N-cadherin KO HSCs  no difference in contribution of 2 populations to chimerism
• Concluded that N-cadherin expression by HSCs is not required for HSC maintenance = may be due to redundancy

Question 13

interplay between two niches

Answer 13

Interplay between osteoblastic and vascular niches
• Dynamic equilibrium exists between two niches
• HSC are released from inhibition of osteoblastic niche in response to demand for differentiated derivates sensed by vascular niche
• When released, HSCs migrate towards sinusoidal blood vessels, along gradient of FGF-4, SDF-1 and increasing oxygen tension  interact directly with endothelial cells, which act as stromal cells of vascular niche
• Stimulus for onset
- Steady state = waves of CXCL12 controlled by circadian rhythms = stimulate release from bone marrow
- Acute infection = IFNa stimulation

Question 14

VN

Answer 14

Vascular niche
• Vascular spaces between the sinusoids = lined with endothelial cells and surrounded by adventitial (CAR) cells
• Supports proliferation, differentiation and mobilisation of ST-HSCs into the bloodstream in response to physiological demand
• Achieved through
- Nutrient-rich environment
- High O2 tension
- Release of factors (e.g. SCF, CXCL12…)

Question 15

endothelial cells in VN

Answer 15

Endothelial cells
• Line medullary vascular sinuses
• Yao et al 2005 - knockout of gp130 cytokine receptor in haematopoietic and endothelial cells using CL
- mice developed bone marrow dysfunction that was accompanied by splenomegaly caused by extramedullary haematopoiesis
- hypocellular marrow contained myeloerythroid progenitors and functional repopulating stem cells. However, long-term bone marrow cultures produced few hematopoietic cells despite continued expression of gp130 in most stromal cells.
- Transplanting gp130-deficient bone marrow into irradiated wild-type mice conferred normal haematopoiesis, whereas transplanting wild-type bone marrow into irradiated gp130-deficient mice did not cure the hematopoietic defects
- Suggest that GP130 expression by endothelial cells important in haematopoiesis

Question 16

CAR cells

Answer 16

CAR cells (CXCL12-abundant reticular cells)
•	Crucial component of niches for HSCs
•	Secrete chemokine CXCL12 = important in initiating night-time waves of haematopoiesis 
•	Omatsu et al (2010) = selectively ablated CAR cells through knockin of DTR-GFP transgene into CXCL12 locus
-	HSCs were reduced in number and size (flow cytometry), were more quiescent than WT HSCs (qRT-PCR revealed reduced expression of cell cycle genes), and had increased expression of genes involved in myeloid fate decision

Question 17

SCF

Answer 17

SCF
• Following migration from osteoblastic niche, HSCs encounter cKIT ligand/stem cell factor  induces proliferation and development to ST-HSCs = repopulating haematopoietic compartment
• SCF can be free or membrane-bound (cleaved by MMP-9)
• Ding et al (2012) = evidence that vascular niche (ECs and MSCs) is principle source of SCF
- Conditional KO of SCF in osteoblasts and stromal cells had no effect on HSC
- Loss of SCF in either endothelial cells or pericytes dramatically reduced HSC numbers
- Double KO lacked any HSC

Question 18

CDKis

Answer 18

CDK inhibitors = differentiation stage-specific regulatory roles = restrain progression through the cell cycle
Scadden (2000)
• produced KO mouse deficient in P21  G1 checkpoint inhibitor
• Quantified populations of cells using colony forming assays (progenitors) and limit-dilution CAFC assays (HSCs)
• found that KO had very little impact at multipotent progenitor stage, bone marrow cellularity or WBCs but that HSCs were vastly overnumbered, massive expansion of this population of cells
• however, they found that these P21 KO HSCs were only able to rescue a lethally irradiated mouse for a certain number of weeks  mice then went in haematological crisis and died
• KO of P21 effectively converted LT-HSCs to ST repopulating cells  inhibited ability to rescue the lethality of mice long-term
• Showed that P21 is very important in controlling entry of LT-HSCs into the cell cycle = exerts dominant inhibitory effect, crucial for maintaining the quiescence of that population and preventing their ultimate exhaustion, dictates size and kinetics

Scadden (2000)
• Produced KO mouse deficient in P27cip1
- Acts as an intrinsic timer for cell proliferation  restricts number of cell divisions
• No change in HSC numbers, but pronounced expansion of MPPs = massively overproduced within the bone marrow and were found in abundance within the peripheral blood = dictates homeostasis at later phases of haematopoiesis than p21

• loss of P27 also conferred significant growth advantage
- shown in Till and McColloch assay
- exposed adult mouse to lethal irradiation, then rescued mouse by IV injection of mixture of multipotent progenitors from WT mouse and P27 KO mouse (40% KO)
- 6 months after rescue, semi-quantitative PCR found >80% of marrow and blood cells in irradiated mice were P27-deficient  KO cells had significant growth advantage, outgrew WT progenitors
• overall, experiments showed that P27 is important in haematopoiesis but more in controlling the proliferation of multipotent progenitors than the HSCs themselves
• important as multipotent progenitors are transit amplifying cells of haematopoietic system, actively proliferating all the time  even their proliferation is constrained, by P27cip1

Question 19

mesenchymal stem cells

Answer 19

Mesenchymal stem cells (MSCs) also contribute to vascular niche = form a reticular network that supports HSCs
• Mendez-Ferrer 2010- identified mesenchymal stem cells as essential to the HSC niche
- MSCs identified as Nestin positive, localise around blood vessels in bone marrow
- Evaluated spatial relationship of MSCs and HSCs in Nes-GFP transgenic mice = immunostained femoral sections with lineage markers, CD48 and CD150  found that HSCs (CD150+, CD48-, Lin-) associated very closely with MSCs (highly significant)
- Express high levels of Scf and CXCL12 as assessed by qPCR

Question 20

cross-antagonism

Answer 20

Cross-antagonism
Co-expression of a pair of a pair of mutually antagonistic TFs
Cells attempt to overcome conflict created by co-expression of TFs with opposite functions  end outcome results in stable phenotype assoictaed with one of potential cell types
relative amount of TFs determines end state

Question 21

GATA1 vs Pu.1

Answer 21

GATA-1 and Pu.1
• Guide cells towards erythrocytic and megakaryocytic lineages, and granulocyte and monocyte lineages respectively
• Physically interact to reciprocally modulate each other’s activity
• High levels of GATA-1 inhibit Pu.1 by actively displacing essential co-factor c-Jun  collapse of monocytic differentiation pathway

• Expression of Pu.1  inhibits GATA-1 from binding to the promoters of target genes, such as α and β globin  lower expression
• GATA1 usually binds to a range of genes to upregulate their expression for differentiation into erythroid cells
- To activate a target gene, GATA-1 recruits a CREB binding protein (CBP), a histone deacetylase  unwinds DNA, making it accessible for transcription
• Pu.1 physically inhibits this interaction by binding GATA-1 and displacing CBP, recruiting Rb and Suv39H instead  drives methylation of Lys9 in H3  resulting recruitment of HP1 strongly represses the target gene = prevents it from ever being expressed
• Conflict resolution therefore drives expression as both cannot exist in the cell.

EXP: RHODES 2005
Aim –KD of Pu.1 facilitates erythroid production
Method – KD Pu.1 in Pu.1-GFP zebrafish using morpholino and looked at changes in myeloid differentiation in real time (morpholino specifically blocked pu.1 not GFP translation = continued visualisation of GFP+ cells = cell-specific lineage marker of myeloid progenitors)
Results – formed Hb producing cells (confocal microscopy) in the location that normally housed myelopoiesis
Limitation – ablation could lead to reactivation of earlier programmes e.g. dedifferentiation followed by redifferentiate rather than just crossing lineages.

EXP: MONTEIRO 2011
Aim – to challenge the paradigm that cross antagonism of gata1 and pu1 determines erythroid vs myeloid fate decision
Method – KO tif1 in zebrafish embryos.
Results – KO resulted in lack of erythroid dvt and myeloid cells - enabled pu.1 to override gata1 expression and result in myeloid fate via loss of GATA1 stimulation by tif1 and loss of pu.1 inhibition by tif1.
Found that tif1 helps to modulate differentiation by differentially controlling levels of gata1 and pu.1 (diagram).
Significance – conclude that differentiation is much more complex than simple mutual antagonism. Tif1 clearly plays important role.

• Pu.1 and GATA-1 form a self-reinforcing positive feedback loop  each actively induce their own expression
• PU.1 is able to interact with promoter of Pu.1 gene to upregulate expression, and likewise for GATA-1!
• Anything that tilts balance just slightly towards either TF will very rapidly polarise cells down either pathway (myeloid for Pu.1, erythroid for GATA-1)

Question 22

CEBPa vs Notch1

Answer 22

C/EBPa (macrophage lineage) and Notch-1 (T cell development)
• Notch 1 = T cell commitment
• C/EBP = myelopoiesis
• Ligation of Notch1  upregulation of Hes1  binds directly to C/EBPa gene  serves as direct repressor of gene expression  prevents cells reverting to monocytic lineage

EXP: DE OBALDIA 2013
Aim – identify role of Hes1 in T cell lineage commitment.
Method – Took multipotent + lymphoid PGs from fetal liver of Hes1 KO mice + looked at development. Then deleted C/EBP with CreLoxp – FACs to look at T cell development
Result – Hes1 deficient multipotent PGs were biased towards myeloid and DCs after notch however Hes1 deficient lymphoid PGs needed further cytokine signalling for diversion into the myeloid lineage. Deletion of C/EBP using creloxp restored development of T cell from Hes1 deficient PGs.
Significance – Hes1 is needed for T cell development.

Question 23

GATA2 vs CEBPa

Answer 23

GATA2 and C/EBPa
• Timing of TF expression influences differentiation in this case
• C/EBPa drives commitment towards granulocyte/macrophage lineage, expression of GATA2 drives mast cell development
• If C/EBPa is expressed first, followed by GATA2, cell becomes specified to eosinophil lineage
• If GATA2 is expressed first, then C/EBPa, cells are specified towards basophils
• Iwasaki et al (2006)
- Enforced expression of GATA-2 into GMPs using GFP-tagged retrovirus vector = exclusively gave risk to pure eosinophil colonies, regardless of cytokines added
- May Giemsa staining revealed all GFP+ cells possessed eosinophilic granules, GMPs failed to give rise to FceRIa+ cells (basophils and mast cells) as detected by flow cytometry, showed upregulation of eosinophil-related genes (e.g. eoPO, IL-5Ra) on western blot
- Then showed that switching order of expression using tagged CEBPa and GATA-2 retroviruses could switch commitment (as shown below) even in common lymphoid progenitors
- Is enforced expression physiological?

Question 24

T cell cross-antagonism

Answer 24

Also seen in T cell lineage
• Naïve CD4+ T cells are not committed to a particular lineage
• Antigenic recognition in chronic manner causes mutual upregulation of antagonistic RORyT and FOXP3 in Th0 cells = co-expressed
• If RORyT wins that battle, cell is committed to Th17 lineage
• If FOXP3 wins, cell is committed to regulatory T cell lineage = profoundly anti-inflammatory, opposite of pro-inflammatory Th17
• Differentiation of 2 cell types driven by TGF-B concentrations in microenvironment = interplay between intrinsic and extrinsic signals!
• Zhou et al (2008) = importance of TGF-B in polarisation towards Treg
- Examined CD4+ T cells from small intestine lamina propria of RORyt-GFP knockin mice  sorted cells using flow cytometry and examined FOXP3 expression in RORyt+ cells using intracellular staining = revealed that 10% of GFP+ cells also expressed FOXP3
- Treated naïve T cells transduced with retroviral vector for RORyt with TGF-B  profound inhibitory effect on IL-17 production, induced sharp increase in FOXP3 expression (measured by intracellular staining)
- RORyt-induced IL-17 expression partially restored on knockdown of FOXP3 (using shRNA vector)
- No in vivo experiments = are these findings physiologically relevant?

• Antigenic recognition in an acute manner leads to mutual upregulation of mutually antagonistic T-bet and GATA-3
• T-bet drives commitment to Th1, GATA3 drives commitment to Th2

Question 25

cytokine mutual antagonism

Answer 25

Cytokines exist as mutually antagonistic pairs, as with TFs
T cell lineage (again)
• Chronic antigen recognition  IL-6 and IL-10 act as cross-antagonistic pair, guiding cells towards Th17 or Treg lineage respectively
• Acute antigen recognition  IL-12 and IL-4 are cross-antagonistic, guide cells towards Th1 and Th2 respectively

In haematopoietic lineage

IL-4 = towards basophils, away from eosinophils 
IL-5 = towards eosinophils, away from basophils 

G-SCF = towards neutrophils, away from macrophages 
M-CSF = the opposite

Question 26

ACh in haematopoiesis

Answer 26

Schloss et al (2022) = novel evidence that B cell-derived ACh influences haematopoiesis
• Generated ChatGFP mice = express GFP under control of choline acetyltransferase promoter
- Detected round GFP+ cells adjacent to marrow vasculature + in blood
• Then produced mice with conditional acetylcholine KO in B cells = Cd19CreChatfl/fl mice
- in blood  more monocytes, neutrophils and B cells no changes in T cell, erythrocyte and platelet numbers
- in bone marrow  more LSK cells and common myeloid progenitors, no changes in lymphoid, erythroid or platelet progenitor populations

Question 27

in vivo evidence of vascular niche

Answer 27

Hooper et al (2009)

• Conditionally deleted VEGFR2 (restricted expression by sinusoidal endothelial cells) in adult mice
• Crossed mice with floxed VEGFR2 allele with mice expressing tamoxifen-inducible Cre gene under ubiquitous Rosa promoter
• No effect on steady state haematopoiesis but impaired reconstitution following sublethal irradiation = significantly slower recovery of WBC and platelet numbers, and SECs