Modulation of Hematopoietic Stem Cell Homing and Engraftment by CD26

See allHide authors and affiliations

Science  13 Aug 2004:
Vol. 305, Issue 5686, pp. 1000-1003
DOI: 10.1126/science.1097071


Hematopoietic stem cell homing and engraftment are crucial to transplantation efficiency, and clinical engraftment is severely compromised when donor-cell numbers are limiting. The peptidase CD26 (DPPIV/dipeptidylpeptidase IV) removes dipeptides from the amino terminus of proteins. We present evidence that endogenous CD26 expression on donor cells negatively regulates homing and engraftment. By inhibition or deletion of CD26, it was possible to increase greatly the efficiency of transplantation. These results suggest that hematopoietic stem cell engraftment is not absolute, as previously suggested, and indicate that improvement of bone marrow transplant efficiency may be possible in the clinic.

The efficiency of hematopoietic stem cell (HSC) transplantation is important when donor-cell numbers are limiting. For example, since the first cord blood transplants (13), the use of cord blood has been mainly restricted to children, not adults, as a result of apprehension about limited cell numbers. Attempts at ex vivo expansion of stem cells for clinical transplantation have not been encouraging (4, 5). An alternative means to enhance engraftment is to increase HSC homing efficiency to bone marrow (BM) niches. Recently it was suggested that HSCs engrafted mice with absolute efficiency (68). However, if all HSCs homed with absolute efficiency and engraftment, problems of limiting donor cells would not be a concern for clinical transplantation (3). Thus, enhancement of homing and engraftment of HSC is needed if advances in transplantation with limiting numbers of HSC are to be realized. On the basis of our work implicating CD26 in granulocyte colony-stimulating factor (G-CSF)–induced mobilization of HSCs and hematopoietic progenitor cells (HPCs) (911), we investigated the involvement of CD26 in homing and engraftment. Inhibition or deletion of CD26 on donor cells enhanced short-term homing, long-term engraftment, competitive repopulation, secondary transplantation, and mouse survival, which suggests that CD26 is a novel target for increasing transplantation efficiency.

Mouse bone marrow HSCs were defined as cells within the Sca-1+lin population (12). Using chemotaxis assays, we previously established that Diprotin A (Ile-Pro-Ile) is a specific inhibitor of CD26 (10). We show here that Diprotin A–treated C57BL/6 Sca-++lin BM cells exhibited twofold increases in CXCL12-induced migration (Fig. 1A). CD26-deficient (CD26–/–) Sca-1+lin BM cells had up to threefold greater migratory response, compared with control Sca-1+lin BM cells (Fig. 1A). Diprotin A treatment of CD26–/– cells did not further enhance chemotaxis (Fig. 1A). Thus, in vitro migration of Sca-1+lin HSC cells to CXCL12 was enhanced by specific inhibition and even more by the absence of CD26 peptidase activity.

Fig. 1.

Inhibition/loss of CD26 increases CXCL12 chemotaxis and CXCR4-dependent short-term homing of Sca-1+lin mouse BM cells. (A) Diprotin A–treated C57BL/6 Sca-1+lin cells (squares) and CD26–/– BM cells (triangles) had enhanced CXCL12-induced migratory response compared with untreated C57BL/6 cells (circles) (P < 0.01). Diprotin A treatment of CD26–/– cells (diamonds) had no further effect. n = 8 samples. (B) Sorted Sca-1+lin C57Bl/6 cells pretreated with 5mM Diprotin A for 15 min and CD26–/– cells have increased short-term homing into BoyJ recipient mice (P < 0.05; n = 10 mice; total from two experiments). (C) Sca-1+lin cells within donor LDBM pretreated with CD26 inhibitors (Diprotin A or Val-Pyr) or transplantation of CD26–/– cells significantly increases short-term homing of donor cells (P < 0.01; n = 6 mice; total from two experiments). (D) Increased homing efficiency of Sca-1+lin C57BL/6 HSC within LDBM cells noted with Diprotin A treatment or with CD26–/– cells is reversible by their treatment with CXCR4 antagonist AMD3100 for 15 min before transplant. AMD3100 also reduces homing efficiency of C57BL/6 donor cells in the absence of CD26 inhibition (P < 0.05; n = 5 mice). (E) CD26 peptidase activity (U/1000 cells; 1U = 1 pmol p-nitroanilide per min) of C57BL/6 BM cells is rapidly lost with 15 min inhibitor treatment (P < 0.01); recovery begins within 4 hours.

Short-term homing experiments used congenic C57Bl/6 (CD45.2+) and BoyJ (CD45.1+) cells to assess recruitment of transplanted HSCs to BM (13). Treatment of 1 × 104 to 2 × 104 sorted Sca-1+lin BM C57Bl/6 donor cells with CD26 inhibitor (Diprotin A) for 15 min before transplant resulted in ninefold increases in homing efficiency in BoyJ recipients compared with untreated cells (Fig. 1B). Transplantation of sorted CD26–/– Sca-1+lin BM cells (14) resulted in 11-fold increases in homing efficiency (Fig. 1B). This suggests that inhibition, or loss of CD26 activity, significantly increases homing of sorted Sca-1+lin HSCs in vivo. Pretreatment of 20 × 106 low-density (LD) BM donor cells with CD26 inhibitors resulted in 1.5-fold increases in homing efficiency of C57BL/6 Sca-1+lin cells (within the LDBM donor population) into BoyJ recipient BM 24 hours after transplant (Fig. 1C). Transplantation of CD26–/– cells provided a 2.6-fold increase in homing efficiency (Fig. 1C). Thus, inhibition or loss of CD26 activity in the total LDBM donor unit (containing differentiated cells and progenitors) increases in vivo homing of Sca-1+lin HSCs within the LDBM fraction. The use of LDBM cells more accurately represents clinical protocols than does the use of sorted Sca-1+lin HSCs. Differences in homing efficiency between sorted Sca-1+lin cells and LDBM cells may be partially explained by larger numbers of Sca-1+lin donor cells (3 × 104) contained within the 20 × 106 cell LDBM donor unit or by accessory cells contained within the LDBM, but not sorted, cell population.

Treatment of 10 × 106 LDBM donor cells with CXCR4 antagonist AMD3100 (15) for 15 min before transplantation reversed increases in homing efficiency of CD26-inhibited or deleted Sca-1+lin cells (Fig. 1D). AMD3100 treatment itself reduced homing efficiency compared with control cells (Fig. 1D). Migration data from treatment with AMD3100 and in vitro CD26–/– HSC/HPC, combined with our previous studies (10), suggest that CXCL12 is a logical downstream target of enhanced transplant efficiency. This is consistent with an important role for CXCL12 in migration (16), mobilization (1719), homing, and engraftment of HSCs (3, 2022), holding HSCs and HPCs in the bone marrow (23), and enhancing cell survival, an additional component of HSC engraftment (24, 25).

Although CD26 peptidase activity is rapidly lost after treatment with CD26 inhibitors (Diprotin A or Val-Pyr), recovery begins within 4 hours after treatment (Fig. 1E), which might explain homing and enhancement differences of inhibitor-treated, compared with CD26–/–, donor cells (Fig. 1, B and C). As reported for CD34+ cells, cytokine treatment (with interleukin-6 and stem cell factor) of Sca-1+lin cells resulted in increased CXCR4 expression (fig. S1) (26). Cytokine treatment did not affect CD26 expression or activity (fig. S1), which suggests that these cytokines do not regulate CD26.

Transplant efficiency requires consideration of long-term donor HSC engraftment. Transplants were performed with 5 × 105 LDBM C57Bl/6 (CD45.2+) or CD26–/– (CD45.2+) cells into lethally irradiated congenic BoyJ (CD45.1+) recipients. CD26–/– donor cells made a significantly greater contribution to peripheral blood (PB) leukocytes 6 months after transplant (Fig. 2A and fig. S2). This was especially apparent at limiting cell dilutions (Fig. 2A) and correlated with changes in mouse survival (Fig. 2, B and C). At day 60, 0% survival was observed with 2.5 × 104 control cells (Fig. 2B); 80% survival was observed with an equivalent number of CD26–/– cells (Fig. 2C). Transplantation of 2.5 × 104 normal cells was below that required for mouse survival. Recipient survival is dependent on short- and long-term reconstitution of marrow. The absence of surviving mice in this group by day 21 suggests that loss of short-term reconstitution may be responsible for lethality at this donor-cell dose. At limiting transplanted donor cells, long-term engraftment and mouse survival increased with CD26–/– donor cells. At nonlimiting donor-cell numbers (2 × 105), improvement is also observed in engraftment and survival with CD26–/– cells at day 60, which suggests that long-term reconstitution is also targeted.

Fig. 2.

CD26–/– donor cells have enhanced long-term engraftment at limiting cell dilutions and increase recipient survival. (A) CD26–/– donor cells into congenic BoyJ recipient mice significantly increases noncompetitive long-term engraftment 6 months after transplant (P < 0.01; n = 3 to 5 mice). (B and C) Mouse survival following limiting cell dilutions. Increased survival is observed 60 days after transplant in recipients receiving low numbers of CD26–/– cells (C) versus control C57BL/6 cells (B) (P < 0.01; n = 3 to 5 mice). (D) Increased donor-cell contribution to formation of PB leukocytes during noncompetitive long-term engraftment assays was also observed with CD26 inhibitor treatment (Diprotin A or Val-Pyr) 6 months after transplant (P < 0.05; n = 5 mice). (E) Even greater donor contribution to chimerism was observed in 6-month secondary transplants receiving cells from primary mice engrafted with CD26-inhibited cells (P < 0.01; n = 5 mice).

At nonlimiting cell doses and in a non-competitive assay, treatment of C57BL/6 donor cells with either CD26 inhibitor resulted in a one-third increase in donor-cell contribution to leukocyte formation in lethally irradiated BoyJ recipients relative to untreated cells (Fig. 2D). In secondary transplanted recipient mice, a threefold increase in donor-cell contribution to PB leukocytes was seen with CD26 inhibition (Fig. 2E and fig. S3). Increases in secondary repopulating HSCs compared with repopulating HSCs in primary recipients indicates an increased homing/engraftment of self-renewing stem cells with CD26 inhibition.

Competitive repopulating HSC assays provide the most functional assessment of HSC by direct comparison of engraftment from experimental donor cells (CD45.2+) relative to constant numbers of competitor cells (CD45.1+) (13, 27). Six months after transplant, increased donor contribution to chimerism was observed with Diprotin A or Val-Pyr treatment relative to cotransplanted cells (Fig. 3A and fig. S4A). At limiting donor-cell numbers (1.25 × 105 and 0.625 × 105), no significant increases in donor contribution were observed with CD26 inhibitor treatment (Fig. 3A). However, CD26–/– donor cells significantly enhanced chimerism at all donor-cell numbers measured (Fig. 3A). Even greater increases in donor-cell contribution were observed with CD26 inhibitor–treated and CD26–/– donor cells in secondary transplanted BoyJ recipients 4 months after transplant (Fig. 3B); this result was more striking when CD26–/– donor cells were used (Fig. 3B and fig. S4B). It is unlikely that some increases in CD26–/– donor-cell engraftment are the result of increased HSC cell numbers in this population. Numbers of Sca-1+lin cells (2.69 ± 0.49 × 104 per femur pair in C57BL/6 and 2.75 ± 0.15 × 104 per femur pair in CD26–/–) and CFU-GM, BFU-E, and CFU-GEMM in PB, BM, and spleen (11) are comparable in CD26–/– and control C57BL/6 mice. Cycling status of CD26–/– cells was examined because HSCs/HPCs not in G0/G1 are reported to manifest decreased homing and engraftment (28). No significant differences were seen in cycling status between CD26–/– and control C57BL/6 Sca-1+lin BM cells, either when freshly isolated or after 24 hours of preincubation with growth factors (fig. S5).

Fig. 3.

CD26-inhibited and CD26–/– cells have increased competitive repopulation and engraftment of secondary repopulating HSCs compared with control cells. (A) Increased donor-cell chimerism in competitive repopulation assays was observed with Diprotin A or Val-Pyr treatment of 5 × 105 or 2.5 × 105 donor cells in direct competition with 5 × 105 competitor recipient BoyJ cells (P < 0.01). CD26–/– cells had a greatly enhanced contribution to chimerism at all donor-cell numbers (P < 0.01; n = 5 mice). (B) An even greater increase in donor-cell contribution was observed with CD26 inhibitor–treated donors and CD26–/– donors in secondary transplanted recipient BoyJ mice (P < 0.01; n = 5).

Enhancing transplant efficiency has clinical implication but is being debated. Recent reports suggest that HSCs engraft mice with absolute efficiency (7, 8). One report (7) was heavily influenced by mathematical correction factors, and the other (8) addressed single-cell transplants by a subset of HSCs among competitor cells that themselves could save the lethally irradiated recipient. Contrary to this is the reality of the clinical situation (3) and studies in which injection of HSCs directly into BM showed enhanced engraftment compared with intravenous administering of cells (2931). Removal of endogenous CD26 activity on donor HSCs increased homing and engraftment. Thus, improvement in transplant efficiency is possible. Further advancement may require more effective use of CD26 inhibitors, which may translate into the use of HSCs for clinical transplantation from sources containing limiting cell numbers, such as cord blood.

Supporting Online Material

Materials and Methods

Figs. S1 to S5


References and Notes

View Abstract

Navigate This Article