Physiological Migration of Hematopoietic Stem and Progenitor Cells Douglas E. Wright, Amy J. Wagers, Anjali Pathak Gulati, Frances L. Johnson, and Irving L. Weissman

Supplementary Material

Supplemental Figure 1. Effect of parabiosis surgery on the number of circulating LT-HSC. Lethally irradiated CD45.1 mice were co-injected with a radioprotective dose of CD45.1 BM (2 x 10⁵ cells) and with 250l of whole blood obtained from CD45.2 untreated control animals or from CD45.2 parabiotic pairs at 24 hours, 72 hours, or 10 days following surgical joining. Both members of the donor parabiotic pairs were CD45.2, and blood from both parabionts was pooled prior to injection. Peripheral blood of recipient mice was analyzed 33 weeks following cell transfer for the presence of CD45.2 cells. Recipient mice were considered to be engrafted by CD45.2 donor HSC if they exhibited frequencies of donor-derived granulocytes at least 2 SD from the mean background signal of unmanipulated CD45.1 animals. Each cross represents a single engrafted recipient mouse. The number of mice with above-background percentages of donor-derived granulocytes and the total number of recipients per group are indicated for each timepoint. Heavy horizontal bars represent means for each group. Kruskal-Wallis one-way nonparametric analysis of variance and Mann-Whitney U tests yielded P-values >0.05 for all groups. Based on the 100% engraftment of mice transplanted with 250 l control CD45.2 blood, and assuming a seeding efficiency of blood-borne HSC equivalent to BM HSC (10%), one may estimate the steady-state frequency of HSC in normal blood to be >10 cells per 250 l, or >100 blood-borne HSC per mouse (based on an average blood volume of 2.5 ml).

Medium version | Full size version

---

Supplemental Figure 2. Enrichment of CFU-S₁₂ activity in the Sca-1^- fraction of PB. Total blood leukocytes from normal donors were fractionated into Sca-1 and Sca-1^- populations by magnetic cell separation. Cells were stained as in (26) with PE-conjugated anti-Sca-1 mAb, followed by anti-PE microbeads (Miltenyi) and separated by using the AutoMACS cell separation system (Miltenyi). Sca-1⁺ and Sca-1^- cells were injected intravenously into irradiated recipients, and spleen colonies were counted 12 days later. (A) FACS profile of Sca-1⁺ (red histogram) vs. Sca-1^- (blue histogram) populations. (B) Frequency of day 12 spleen colonies in mice receiving Sca-1⁺or Sca-1^-PB cells. Data are presented as the mean (± SD) number of spleen colonies per million injected leukocytes (P<0.05).

Medium version | Full size version

---

Supplemental Figure 3. Stability of HSC frequency in PB. CD45.2 donor animals were bled via the tail vein once every 7 days for 5 weeks, and blood was injected directly into irradiated recipients, together with 2 x 10⁵ radioprotective CD45.1 BM cells, to determine HSC activity. Reconstitution of recipients with donor blood cells was monitored by assessing the frequency of CD45.2⁺ granulocytes in the PB following transplant. Data plotted as column graphs on the left axis represent the mean percentage (± SD) of CD45.2⁺ Mac-1⁺Gr-1⁺ leukocytes in the PB of recipients >12 weeks following transplant, for each group of recipients (i.e. those transplanted with blood obtained on week 1 of serial bleeding, week 2, week 3, etc.). The overlayed line graph, plotted on the right axis, shows the mean count of leukocytes/l blood of the CD45.2 donor on each of the days of bleeding for transplantation. Donor cell engraftment was not significantly different among any of the 5 timepoints (P>0.05).

Medium version | Full size version

---

Supplemental Figure 4. In vivo trafficking of intravenously injected blood-borne HSC and progenitors. Unconditioned non-transgenic animals were injected intravenously with 1.16 x 10⁶ Lin-GFP⁺ blood cells, and sacrificed 5 minutes after injection. The recipients were extensively perfused with 10 mM EDTA/1xPBS to collect blood, and BM, spleen, kidney, lung, and liver were harvested. Single cell suspensions were prepared from these tissues (for kidney, lung and liver, following collagenase treatment) and analyzed for the presence of GFP⁺ cells by FACS. Approximately 96% of the input cells were cleared from the blood and GFP⁺ cells were found in the BM, liver, spleen and lung. In these assays, 1-2% of injected HSC/progenitor cells cleared from circulation trafficked to BM. Extrapolating from these observations, we estimate the rate of blood to BM homing of circulating HSC to be 250-500 cells per day.

Medium version | Full size version