Lymphohematopoietic Stem Cell Engraftmenta

Lymphohematopoietic Stem Cell Engraftmenta PETER J. QUESENBERRY,b F. MARC STEWART, SUJU ZHONG, HOURI HABIBIAN, CHRISTINA MCAULIFFE, JUDY REILLY, JANE CARLSON, MARK DOONER, SUSIE NILSSON, STEFAN PETERS, GARY STEIN, JANET STEIN, ROB EMMONS, BRIAN BENOIT, IVAN BERTONCELLO, AND PAMELA BECKER University of Massachusetts Medical Center and University of Massachusetts Cancer Center, Worcester, Massachusetts 01605, USA

ABSTRACT: Traditional dogma has stated that space needs to be opened by cytoxic myeloablative therapy in order for marrow stem cells to engraft. Recent work in murine transplant models, however, indicates that engraftment is determined by the ratio of donor to host stem cells, i.e., stem cell competition. One hundred centigray whole body irradiation is stem cell toxic and nonmyelotoxic, thus allowing for higher donor chimerism in a murine syngeneic transplant setting. This nontoxic stem cell transplantation can be applied to allogeneic transplant with the addition of a tolerizing step; in this case presensitization with donor spleen cells and administration of CD40 ligand antibody to block costimulation. The stem cells that engraft in the nonmyeloablated are in G0, but are rapidly induced (by 12 hours) to enter the S phase after in vivo engraftment. Exposure of murine marrow to cytokines (IL-3, IL-6, IL-11 and steel factor) expands progenitor clones, induces stem cells into cell cycle, and causes a fluctuating engraftment phenotype tied to phase of cell cycle. These data indicate that the concepts of stem cell competition and fluctuation of stem cell phenotype with cell cycle transit should underlie any new stem cell engraftment strategy.

Traditionally, it has been considered that space needs to be opened by myeloablative therapy in order for there to be adequate lymphohematopoietic stem cell engraftment. However, previous studies by Micklem, Saxe, Brecher and others have indicated that marrow cells would engraft in nonablated animals.1–5 Stewart et. al.6 extended these observations showing high levels of long-term multilineage engraftment in nonmyeloablated BALB/c mice. A series of studies utilizing a male/female transplantation model in nonmyeloablated BALB/c hosts has shown that engraftment is essentially quantitative and appears to be determined by competition between infused marrow cells and host cells.7–10 These studies show that high levels of engraftment were obtained in marrow, spleen and thymus, that the level of en-

aThis work was supported in part by National Institutes of Health grants No: RO1 DK27424-14, Hematopoietic Cellular Interaction and Lithium, R01 DK49650-04, Repetitive Marrow Transplantation into Normal Mice, P01 DK50222-01A1, Stem Cell Biology and P01 HL56920-02, Hematopoietic Stem Cell Growth and Engraftment. bCorrespondence and requests for materials to: Peter J. Quesenberry, M.D., Director, Cancer Center, University of Massachusetts Medical Center and University of Massachusetts Cancer Center, Two BioTech, Suite 202, 373 Plantation Street, Worcester, MA 01605. Phone, 508/ 856-6956; fax, 508/ 856-1310; e-mail, [email protected]

40

QUESENBERRY et al.: STEM CELL ENGRAFTMENT

41

TABLE 1. Engraftment into nonmyeloablated hosts High levels of chimerism seen in marrow, spleen and thymus. Chimerism persistent out to two years. Chimerism multilineage. Engraftment at stem cell level appears quantitative.

graftment was cell dose related and that high numbers of cells given in one injection gave equal levels of engraftment to the same number of cells divided over multiple injections, although the addition of heparin could enhance engraftment at very high levels of infused cells. Experiments were carried out where 40 million cells were infused on 20 separate occasions into the same mouse over time. This very high level of infusion led to very high levels of marrow engraftment but not to total replacement of marrow. These data are summarized in T ABLE 1. Further studies in this model has shown that highly purified lineage-negative, rhodamine-low, Hoescht-low murine stem cells also engraft well and give high levels of marrow chimerism in nonmyeloablated BALB/c mice.11 Homing studies revealed a probable maximal window of 19 hours for marrow engraftment with fairly rapid clearance of stem cells from blood and lung.12 No early homing was seen in thymus and very low levels in the spleen. Mapping of male cells in female nonmyeloablated hosts utilizing the fluorescence in situ hybridization (FISH) technique on fixed marrow sections revealed that by six weeks postengraftment virtually all of the engrafted male cells were adjacent to the endosteal surface.11 In addition, it was clear that these cells had given rise to bone osteocytes which persisted out to six months post-marrow infusion.12 The above observations suggested that engraftment was dependent upon the ratio of host and donor stem cells, not upon any actions in ‘opening space.’ If this was in fact the case, treatments which could diminish host stem cells without toxicity, should be able to markedly increase the percent of donor chimerism by increasing the competitive advantage of transfused stem cells. This could provide a nontoxic way of augmenting the percentage of donor cells, while minimizing the severe nuetropenia and thrombocytopenia associated with conventional myeloablative conditioning regimines. Accordingly, we assessed the capacity of low doses of irradiation 1) to ablate engrafting stem cells in the host and 2) to enhance donor engraftment.13 Exposure of BALB/c mice to a 100 cGy whole body irradiation markedly decreased the capacity of marrow stem cells to engraft long-term to a level of about 10% of normal. This left enough residual hematopoietic stem/progenitor cells to avoid any significant myeloablation. There were transient and moderate decreases in the white count and platelet count. Mice exposed to 100 cGy and infused with 40 million marrow cells showed high levels of persistent chimerism at 2, 5 and 8 months postinfusion. The chimerism was multilineage and was evidenced in marrow, spleen and thymus. Hosts exposed to 100 cGy whole body irradiation also showed relatively high levels of chimerism when infused with 10 million marrow cells, levels estimated to be those potentially obtainable in a clinical transplant setting. These results indicated that a 100-cGy whole body irradiation was very stem cell toxic, but nonmyelotoxic and allowed for very high levels of long-term donor cell chimerism.

42

ANNALS NEW YORK ACADEMY OF SCIENCES

TABLE 2. Murine allochimerism B6 to BALB/c Low level host treatment High levels of infused stem cells Costimulator blockade

Mismatch at h2 locus 100 cGy 40 × 10 6 B6 marrow cells CD-40 ligand antibody — multiple injections

ALLOCHIMERISM The combination of relatively high levels of infused stem cells and low-level irradiation offers an attractive approach for the nontoxic creation of chimerism. In order for these approaches to be useful for treatment of marrow disorders they must be applicable in an allogenesic stem cell transplantation setting. This introduces the problems of graft-versus-host disease (GVHD) and graft rejection. The keys to establishing long-term stable allochimerism for the treatment of intrinsic marrow diseases, autoimmune disorders or cancer relates to avoiding GVHD and rejection and establishing long-term tolerance. Theoretically low-level host treatment avoids the ‘cytokine storm’ and its purported adverse effects on GVHD. The utilization of high levels of stem cells may also help overcome rejection and GVHD problems. A powerful approach to tolerization has been costimulator blockade of CD40-CD40 ligand or B7-CD28 interactions. Accordingly, we have evaluated the combined use of high levels of stem cells, low-level host irradiation and costimulator blockade as an approach to the nontoxic creation of stable allochimerism (TABLE 2). Thus we evaluated our capacity to establish allochimerism in a completely H2 mismatched murine marrow transplantation model. In one experiment 40 × 106 B6 cells were infused into 100 cGy-treated BALB/c hosts, and engraftment was quantitated utilizing monoclonal antibody marking and

FIGURE 1. Experiment flow chart for allotransplant.

QUESENBERRY et al.: STEM CELL ENGRAFTMENT

43

TABLE 3. B6.SJL: BALB/c allochimerisma Group

Average Marrow Engraftment Determined by FITC-CD45.1d

Treatment

I

40 × 106 BMC + 100 cGyb

II

40 × 106 BMC + 100 cGy + 10 × 106 spleen cellsc

III

+ 100 cGy + 10 × 40 × + CD40 ligand antibody 106 BMC

18 + 7 106 spleen

cellsc

8+6 39 + 4

aB6.SJL as donor is H-2Ks, CD45.1, BALB/c as recipient is H-2Kd, CD45.2 bAll BALB/c host mice were treated with 100 cGy at transplant day. cGroup II & III received 10 × 106 B6.SJL spleen cell i.v. 10 days prior to transplant.

Group III received CD40 ligand as outlined in F IGURE 1. dEngraftment was determined 6 weeks after transplant, using FITC-CD45.1 monoclonal antibody to mark donor cells with analysis by FACS.

fluorescent activated cell sorting at 8 weeks post-marrow infusion (TABLE 3). When 100-cGy exposed BALB/c mice were infused with 40 × 106 marrow cells from B6.SJL mice, engraftment occurred with donor chimerism of 18 ± 7%. If B6.SJL spleen cells (106) were injected intraperitoneally 10 days prior to infusion of 40 × 106 B6.SJL marrow cells to 100-cGy exposed BALB/c hosts, 4 of 5 mice rejected the graft. When CD40-ligand antibody was repetitively administered with and after the spleen cell injection and through the transplant, all mice showed high level chimerism at 8 weeks. The experimental details are outlined in FIGURE 1. Mice in these experiments appeared healthy at the time of sacrifice, with no gross evidence of GVHD. The combination of B6 spleen cell sensitization and CD40-ligand antibody blockade with infusion of B6 marrow into 100-cGy treated BALB/c hosts led to mean donor chimerism levels of 39 ± 4% donor. These data indicate that the nontoxic creation of allochimerism may be a feasible approach to a number of genetic marrow diseases such as thalassemia and sickle cell anemia, autoimmune diseases and cancer.

PHENOTYPE OF ENGRAFTING STEM CELL The stem cell which engrafts in vivo in either irradiated or normal hosts is quiescent as determined by either in vitro tritiated thymidine or in vivo hydroxyurea suicide experiments. However, when these quiescent stem cells are infused into normal nonirradiated hosts, the majority rapidly enter active cell cycle, with approximately 50% being in S phase by 12 hours postinfusion as determined by in vivo hydroxyurea suicide.14 This is consistent with previous results, utilizing different approaches, by Hendrikx et al.15 In vitro cytokine stimulation of marrow cells results in maintenance or expansion of progenitors with stimulation of cell cycle transit. When BALB/c male marrow cells were exposed to interleukin-3 (IL-3), interleukin-6 (IL-6), interleukin-11 (IL-11) and steel factor in liquid culture for 48 hours progenitors (colony-forming cells (CFC) and high proliferative potential colony-forming cells (HPP-CFC)) were maintained or expanded, and cell cycle progression was stimulated.16,17 However, at 48 hours of cytokine culture, engraftability into either normal or irradiated hosts was markedly impaired.16,17 Lineage-negative,

44

ANNALS NEW YORK ACADEMY OF SCIENCES

TABLE 4. Critical concepts in stem cell engraftment strategies Engraftment is determined by the ratio of host to donor stem cells. Therapies which nontoxically reduce host stem cells will increase donor chimerism. Stem cells giving long-term chimerism show marked fluctuation with cell cycle transit. Donor cell priming and costimulator blockade facilitate tolerance and stable allochimerism.

rhodamine-low, Hoechst-low stem cells were purified from whole BALB/c marrow and incubated in liquid culture with IL-3, IL-6, IL-11 and steel factor, and their cell cycle transit was determined using tritium labeling and cell doublings.18 The first cell cycle showed a 16–20-hour time interval from dormancy to S phase, and the first population doubling occurred at 36–40 hours; subsequent population doublings occurred every 12 hours indicating a markedly shortened G1 phase. When engraftability at 2 or 6 months post-marrow infusion was determined for BALB/c male cells at varying times out to 80 hours in liquid culture, it was found that the capacity to engraft showed a fluctuating phenotype, with marked changes in engraftment being observed over 2–4-hour intervals.19 There was an initial loss of long-term engraftment at an average of 33 hours in cytokine culture corresponding with late S, and early G2 (as previously determined); a recovery of engraftment levels to that seen with input marrow was then seen at 40 hours of culture. These data suggest that bone marrow stem cells show a plastic reversible phenotype apparently linked to cell cycle transit. In toto, these data indicate that certain critical concepts (T ABLE 4) should form the intellectual base for stem cell transplant approaches in either the autologous or allogeneic setting. REFERENCES 1. M ICKLEM , H.S., C.M. CLARKE , E.P. E VANS & C.E. F ORD . 1968. Fate of chromosome-marked mouse bone marrow cells transfused into normal syngeneic recipients. Transplantation 6: 299. 2. T AKADA , A., Y. T AKADA & J.L. AMBRUS . 1970. Proliferation of donor spleen and marrow cells in the spleens and bone marrows of unirradiated and irradiated adult mice. Proc. Soc. Exp. Biol. Med. 136: 222. 3. T AKADA , Y. & A. T AKADA . 1971. Proliferation of donor hematopietic cells in irradiated and unirradiated host mice. Transplantation 12: 334. 4. B RECHER , G., J.D. A NSELL , H.S. M ICKLEM , J.H. T JIO & E.P. CRONKITE . 1982. Special proliferative sites are not needed for seeding and proliferation of transfused bone marrow cells in normal syngeneic mice. Proc. Natl.Acad. Sci. USA 79: 5085. 5. SAXE , D.F., S.S. BOGGS & D.R. B OGGS . 1984. Transplantation of chromosomally marked syngeneic marrow cells into mice not subjected to hematopoietic stem cell depletion. Exp. Hematol. 12: 277. 6. STEWART , F.M., R. C RITTENDEN , P.A. L OWRY , S. P EARSON-W HITE & P.J. QUESEN BERRY. 1993. Long-term engraftment of normal and post-5-fluorouracil murine marrow into normal nonmyeloablated mice. Blood 81: 2566-2571. 7. R AMSHAW , H.S., S.S. R AO , R.B. C RITTENDEN , S.O. P ETERS , H.U. W EIER & P.J. QUE SENBERRY . 1995. Engraftment of bone marrow cells into normal unprepared hosts: effects of 5-fluorouracil and cell cycle status. Blood 86(3): 924–929. 8. R AMSHAW , H., R.B. C RITTENDEN , M. D OONER , S.O. P ETERS , S.S. RAO & P.J. Q UESEN BERRY . 1995. High levels of engraftment with a single infusion of bone marrow cells into normal unprepared mice. Biol. Blood Marrow Trans. 1: 74-80.

QUESENBERRY et al.: STEM CELL ENGRAFTMENT

45

9. R AO , S.S., S.O. P ETERS , R.B. C RITTENDEN , F.M. STEWART , H.S. RAMSHAW & P.J. Q UESENBERRY . 1997. Stem cell transplantation in the normal nonmyeloablated host: relationship between cell dose, schedule and engraftment. Exp. Hematol. 25: 114– 121. 10. B LOMBERG , M.E., S.S. R AO , J.L. R EILLY , C.Y. T IARKS , S.O. P ETERS , E.L.W. KITTLER & P.J. Q UESENBERRY . 1998. Repetitive bone marrow transplantation in nonmyeloablated recipients. Exp. Hematol. 26: 320–324. 11. N ILSSON , S., M. D OONER , C. TIARKS , W. H EINZ -U LRICH & P.J. QUESENBERRY . 1997. Potential and distribution of transplanted hematopoietic stem cells in a nonablated mouse model. Blood 89: 4013–4020. 12. N ILSSON , S., M.S. D OONER & P.J. Q UESENBERRY . Unpublished observations. 13. STEWART , F.M., S. Z HONG , J. W UU , C.C. H SIEH, S.K. N ILSSON & P.J. Q UESENBERRY . 1998. Lymphohematopoietic engraftment in minimally myeloablated hosts. Blood 91: 3681–3687. 14. N ILSSON , S.K., M.S. D OONER & P.J. Q UESENBERRY . 1997. Synchronized cell-cycle induction of engrafting long-term repopulating stem cells. Blood 90: 4646–4650. 15. H ENDRIKX , P.J., A.C.M. M ARTENS, A. H AGENBEEK , J.F. K EIJ & J.W.M. V ISSER . 1996. Homing of fluorescently labeled hemopoietic stem cells. Exp. Hemato.l 24: 129. 16. PETERS , S.O., E.L. K ITTLER , H.S. R AMSHAW & P.J. Q UESENBERRY . 1995. Murine marrow cells expanded in culture with IL-3, IL-6, IL-11, and SCF acquire an engraftment defect in normal hosts. Exp. Hematol. 23: 461–469. 17. PETERS , S.O., E.L.W. K ITTLER , H.S. RAMSHAW & P.J. Q UESENBERRY . 1996. Ex vivo expansion of murine marrow cells with interleukin-3, interleukin-6, interleukin-11, and stem cell factor leads to impaired engraftment in irradiated hosts. Blood 87: 30–37. 18. R EDDY , G.P.V., C.Y. T IARKS , L. P ANG & P.J. QUESENBERRY . 1997. Synchronization and cell cycle analysis of pluripotent hematopoietic progenitor stem cells. Blood 90: 2293–2299. 19. H ABIBIAN , H.K., S.O. P ETERS , C.C. H SIEH, J. W UU , K. V ERGILIS , C.I. G RIMALDI , J. R EILLY , J.E. C ARLSON , A.E. F RIMBERGER , F.M. S TEWART & P.J. Q UESENBERRY . 1998. The fluctuating phenotype of the lymphohematopoietic stem cell with cell cycle transit. J. Exp. Med. 188: 393–398.

DISCUSSION H.E. B ROXMEYER (Indiana University): Could tell us a little more about your anti-CD40 ligand studies? Why did you chose that ligand? Did you look at antiCD40? Did you look at anti-41BB or antibodies to any other members of the tumor necrosis factor (TNF) receptor family? P.J. Q UESENBERRY (University of Massachusetts Medical Center): We are doing B7 experiments with CD40 ligand antibody. We picked it based on work in our diabetes group at the University of Massachusetts. They have been doing extensive studies on skin grafting. They found that they can go to a mismatched combination with spleen priming and CD40 ligand antibody and obtain prolonged takes of skin grafts. We are very excited about this data. However, I found out in San Diego that we have been scooped, because Megan Sikes has very similar data. This area is moving very fast. We shall hear more later in this meeting about potential clinical application. R. H OFFMAN (University of Illinois College of Medicine): We have been trying to do similar type experiments in larger animals, basically in baboons. The degree of

46

ANNALS NEW YORK ACADEMY OF SCIENCES

allochimerism has been disturbingly low with what one would call minimal radiation doses or no radiation doses at all. I am perplexed by the discrepancies between the data in the mouse and those in larger animals. Is there some intrinsic difference between the baboon system and the murine system that would explain this? The experience in clinical human allogeneic transplantation would suggest, however, that some sort of preparative regimen is required for engraftment. So how would you explain this? Q UESENBERRY : As we look more and more it appears that the preparative regimen alters host-donor stem cell ratios rather than creating space. You can obtain high rates of donor chimerism with minimal to no treatment if you infuse enough stem cells. So I do not agree. I do not think you need much in the way of preparative regimens. It depends on the number of stem cells, and again I think we shall hear about that later. If you use very low numbers of stem cells (most previous experiments used low levels of stem cells), then you need aggressive preparative regimens. If you have high levels of stem cells, you need less aggressive to no preparative treatment. I shall give a disturbing trivial explanation why primate or any other animal system might be very different, and that is colony infections. We just had experience with a parvovirus infection effecting engraftment. One would have to look carefully at other factors within any animal system for engraftment. I have been very impressed with how different infections can totally wipe you out, so that may be a potential factor in the primate model. H OFFMAN : No, that is not a factor, because with myeloablative doses of radiation we get engraftment. Q UESENBERRY : That is not convincing, because with myeloablative doses you may get engraftment with parvovirus, too. So, depending on your system, a current infection may or may not have a major effect on the result. Y. R EISNER (Weizmann Institute of Science): A comment on the stem cell dose effect. As you know we have been using it for many years in allogeneic transplants. One of the confusing issues when you look into allogeneic as opposed to the autologous setting, is that these CD34 cells can also tolerize and interfere with the rejection mediated by host T cells. Later, at the end of this meeting I shall show data about this. In early studies we found indications, based on the superiority of myeloablative agents vs immunosuppressive drugs, in enhancing T cell-depleted bone marrow allografts, that there is a competition between stem cells that play a role in engraftment. But also there is a major factor here of tolerizing the immune system. Altogether, in our primate experiments we find it more difficult to achieve engraftment with megadoses of stem cells. For example, in mice exposed to 650 rads we were able to achieve very nice donor type chimerism with 40 million T cell-depleted allogeneic bone marrow cells, whereas in monkeys we are currently using 700 rads and are still failing to achieve donor chimerism. We believe this difference is quantitative and not qualitative. We may have to find new approaches to deal with the more vigorous immune systems that are likely to be found in the outbred primate and in man. Perhaps your suggestion to use anti-CD40 ligand may be useful in combination with radiation, but you might find that 100 rads in the monkey is still not enough. You might have to use 700 rads plus anti-CD40 ligand to get where you want to be.

QUESENBERRY et al.: STEM CELL ENGRAFTMENT

47

Q UESENBERRY : I agree with that. And that may be a double blockade, the addition of a B7 block to CD40 blockade. Your other comment is very appropriate. I think the data in kidney transplant in animal models are impressive. Marrow chimerism may be able to get around a lot of the rejection problems with the kidney and also suggests very interesting strategies for bone marrow itself. C. J. E AVES (Terry Fox Laboratory): Have you actually tried the bromodeoxyuridine (BrdU) experiment posttransplant to ask whether the rate of turnover is any different in a posttransplant scenario from what it is normally. We know the change in accumulation of stem cells is different, but is this explained either partly or wholly by a change in turnover? Q UESENBERRY : Actually we are doing those studies, but I do not have data. I would say from the hydroxyurea experiments we have done, and also from Visser’s experiments, that when you infuse dormant marrow cells most of them go through cycle within a couple of days. Their kinetics would be very different from other people’s studies with chronic BrdU. But the experiments to directly test that are appropriate and are going to be very interesting. B. T OROK -S TORB (Fred Hutchinson Cancer Research Center): I do not know how the monkey studies are being done, what the preparative regimen is, or how the cells are prepared. However, in our transplant program in Seattle we have cloned 6 human leukocyte antigen (HLA)-identical transplants with peripheral blood stem cells with only 200 cGy of irradiation and 35 days total of immunosuppression and have established stable chimerism in all these patients. We believe as you do that we are not making marrow space with the total body irradiation (TBI), but that we are establishing tolerance. Q UESENBERRY : We have done one patient at 100 cGy, and we are just measuring chimerism; but we have one tumor regression in that setting. It is a very exciting area. T OROK -STORB : These 6 transplants have been done on an outpatient basis. The people were never hospitalized, and the one who is out 7 months is all donor now.