Bone Marrow Transplantation, (1998) 21, 7984 1998 Stockton Press All rights reserved 02683369/98 $12.

00

Cadaveric bone marrow and spleen cells for transplantation

G Soderdahl1, C Tammik2, M Remberger2 and O Ringden1,2
Departments of 1Transplantation Surgery and 2Clinical Immunology, Huddinge Hospital, Huddinge, Sweden

Summary: Cadaveric bone marrow was harvested from 20 braindead donors to determine optimal conditions for procurement for transplantation. The number of nucleated cells obtained from 1 ml of bone marrow was signicantly higher in vertebrae (87 20 106, mean s.e.m.) than in the sternum (10.2 3.8 106) or ribs (4.9 2.0 106). Viability of cells was not signicantly affected by storage temperature (4C or 20C) or duration of storage (672 h). In addition to bone marrow, spleen cells were harvested from three cadaveric donors. The mean yield from 1 g of spleen tissue was 4.4 106 nucleated cells. Using magnetic beads, we removed 96% of T lymphocytes without affecting the total yield of stem cells from cadaveric bone marrow. Using CD34-positive cell selection, we obtained a 99.6% T cell depletion efciency, but with a loss of 60% of CD34-positive cells. Using optimized techniques, we obtained an estimated mean yield of 5.5 1010 mononuclear cells from the whole thoracic and lumbar vertebral column. With a mean fraction of CD34-positive cells of 2.1 0.3%, recovery and purity were not affected by site of sample, temperature or donor age. In contrast, the CD34-positive fraction in spleen preparations was 0.41 0.06%. When analyzing the number of colony-forming units (CFU-GM, BFU-E and CFUGEMM), we found no signicant differences between cadaveric bone marrow and bone marrow aspirates from living donors. However, cells harvested from the spleen gave signicantly fewer CFUs than did bone marrow from living donors. We conclude that bone marrow from cadaveric donors can be harvested and procured with a high degree of viability and good function. With an appropriate technique of harvesting and procurement, it seems feasible to recover enough stem cells for transplantation. Keywords: bone marrow; spleen cells; transplantation; cadaveric; hematopoietic stem cells

alternative to volunteer donors. In addition, bone marrow, harvested together with solid organs, can even be used to modulate the immune response of the organ recipient. The discovery of microchimerism after transplantation of solid organs has been considered to explain mechanisms of longterm allograft acceptance without immunosuppression.57 Thus, donor-specic bone marrow augmentation after solid organ transplantation has been thought to promote extended chimerism and possibly to produce tolerance.8,9 Tolerance induction has been described with living donor leukocyte infusions in mice,10 dogs11 and rhesus monkeys.12 Hitherto, extension of these results to human transplantations has failed to induce tolerance although improved renal graft survival was seen.13 Several methods for harvesting cadaveric bone marrow have been described.14,15 However, little is known about optimization of the number of stem cells required for successful engraftment in the recipient. This paper describes a rapid method for the harvesting and procurement of cadaveric bone marrow. Isolation conditions have also been studied to optimize the yield of progenitor cells. Finally, the quality of the processed cadaveric bone marrow has been analyzed and compared with bone marrow aspirated from living donors. Materials and methods Bone marrow was taken from 20 cadaveric organ donors. In ve, kidneys only were taken, in eight cases, liver and/or pancreas were also taken and in the remaining seven, thoracic organs were removed as well. Informed consent was given by relatives. Around 70% of those approached gave consent. University of Wisconsin (UW) solution16 was used for organ preservation in all cases. The median age of the donors was 47 years with a range between 467 years (nine males and 11 females). The cause of death in seven cases was trauma and the other 13 deaths were due to intracranial hemorrhage. For comparison, bone marrow aspirated from the iliac crest of 10 living donors, median age 28 years (7 61 years), eight males and two females was analyzed. Bone marrow harvest Bone marrow was harvested from vertebral bodies, sternum body and ribs. In three cases, sternal samples were taken before perfusion of organs with UW solution. To obtain marrow from vertebral bodies, the outer cortex was removed with an electrical bone saw and the trabecular bone containing red marrow was extracted, using a Gigli saw or a chisel. Trabecular samples were cut into small

In the eld of bone marrow transplantation, only one third of candidates for allogeneic bone marrow transplantation have an HLA-identical sibling donor.1 Therefore, marrow from unrelated donors is increasingly employed.24 Bone marrow from cadaveric donors used as a source may be an
Correspondence: Dr O Ringden, Dept of Clinical Immunology, Huddinge Hospital, F79, S-141 86 Huddinge, Sweden Received 27 June 1997; accepted 31 August 1997

Cadaveric bone marrow G Soderdahl et al

80

pieces and transported to the laboratory in heparinized RPMI 1640 medium either on ice (+4C) or at room temperature (+20C). For sternal and rib samples, whole sections of bone marrow were removed with an electric saw and thereafter cut into smaller pieces. Bone marrow procurement Samples were analyzed immediately or stored for a maximum of 3 days at room temperature (+20C) or in a refrigerator (+4 +8C) before separation. Cell suspensions were passed through a sterile 70-m cell strainer (Falcon; Becton Dickinson, San Jose, CA, USA) to remove bone fragments before layering on a FicollHypaque gradient and then centrifuged at 500 g for 20 min. The interface containing mononuclear cells was collected and the cells were washed and resuspended in RPMI 1640, supplemented with 10% human AB serum. Viability of bone marrow cells was tested with trypan blue exclusion. Spleen cell procurement Spleens were taken from cadaveric organ donors, then cut, emulsied through a cell strainer and washed.17 The cells were separated on a FicollHypaque gradient and were kept frozen in liquid nitrogen before use. FACS analysis of cell surface markers The method for FACS analysis on Ficoll-separated cells has been described previously.18 Briey, after cells were counted, they were diluted in PBS to 106 cells/ml. To 0.1 ml cell suspension, 10 l of conjugated monoclonal antibodies (mabs) were added; CD45 FITC/CD14 PE (Leucogate; Becton Dickinson), r1/r2a Simultest Control (Becton Dickinson), CD34 FITC (HPCA-2; Becton Dickinson) and CD19 PE (HD-37; Dako, Copenhagen, Denmark), CD3 FITC (UCHT-1, Dako A/S) and CD 56 PE (Leu-19, Becton Dickinson). Cell suspensions were incubated for 10 min, at room temperature in the dark. After washing once and resuspension in PBS, the cells were analyzed in a FACSort (Becton Dickinson), evaluating 10 20 000 cells per sample. After running the negative control, 100 l propidium iodide (1:1000) were added to the cells and after incubation for a further 5 min, the cells were analyzed to evaluate viability. The percentages of the various cell populations were calculated in an ungated cell population. Assay for hematopoietic precursors A commercial kit Stem cell CFU kit (GIBCO BRL, Life Technology, NY, USA) was employed to study colony formation and was used according to the manufacturers instructions. This semi-solid system mimics the extracellular matrix produced by stromal cells. Triplicate cultures with 12.5 103 cells/well, 0.5 ml in each well, were incubated for 10 days. Colonies were identied as BFU-E (burst-forming unitserythroid cells) containing densely packed hemoglobinized cells, CFU-GM (colony-forming unitsgranulocytes, macrophages) and CFU-GEMM

(colony-forming unitsgranulocytes, erythroid cells, macrophages, megakaryocytes). T cell depletion Prepared bone marrow cells were resuspended in RPMI + 2% FCS to a nal concentration of 14 107/ml. Magnetic bead (Dynal, Oslo, Norway)-coupled mabs directed against CD2, CD3 and CD8 were added to a mean of four beads/T cell and incubated for 30 min at +4C on a rocking platform. Cells with surface-bound beads were removed with a magnetic device and the supernatant containing uncoupled cells was collected. This procedure was performed twice to achieve maximal clearance of magnetic beads. CD34-positive cell selection For CD34 cell selection a commercially available system was used according to the manufacturers instruction (Ceprate, LC-34; Cellpro, Bothell, WA, USA). In this system, cells are incubated with a biotin-labelled anti-CD34 antibody and labelled target cells are captured with avidincoated beads, using a continuous ow column. Bound target cells are released by gentle agitation.

Results Immune markers and hematopoiesis Table 1 shows the fractions of various populations of hematopoietic cells, assayed by FACS analysis. The number of colony-forming units in the analyzed samples is also shown. Three of the sternal samples were collected in an early phase of the donor operation before procurement of organs and perfusion with UW solution. However, there were no differences in viability, yield of nucleated cells or cell populations in these samples as compared to those taken from the sternal body at the end of the donor operation. The fraction of CD34-positive cells was not affected by temperature or storage for less than 3 days (Figure 1). Patients above 47 years of age had 2.0 0.4 ( s.e.)% CD34-positive cells in the marrow, as compared to 2.2 0.4 in donors below this age (NS). Viability and yield The viability of mononucleated cells immediately after gradient centrifugation ranged between 8499%, with a mean value of 93.5 1.0% (mean s.e.m.). No signicant difference was observed between samples from the vertebrae, rib or sternum (data not shown). Viability was not signicantly affected by storage temperature or duration of storage (maximum 3 days), although we found a tendency to decreased viability when samples were stored for a prolonged period at room temperature (Figure 2). The yield of nucleated cells per ml of bone marrow was found to be higher from vertebrae than from sternal body or rib (P 0.001) (Figure 3). The mean yield from one vertebral body was calculated as 2.6 0.7 109 mononuclear cells. Since samples from vertebral bodies proved

Cadaveric bone marrow G Soderdahl et al

Table 1

Fraction of immune markers and number of progenitors of bone marrow from various sources Fraction of mononucleated cells % n CD3 7.0 0.8 7.1 1.5 5.0 0.6 8.3 3.1 21.1 1.7 11.2 0.5 CD34 2.1 0.2 2.1 0.4 1.9 0.7 3.0 0.3 0.4 0.1 1.3 0.3 CD19 5.4 0.5 3.9 1.0 ND 4.8 0.9 ND 4.2 0.5 CD56 0.6 0.1 1.3 0.6 2.2 1.9 0.9 0.4 ND 1.4 0.1 n 11 3 1 3 10 No. of colonies/105 cells CFU-GM 373 91 642 274 ND 200 99 49 638 102 BFU-E 560 114 394 216 ND 136 192 91 708 260 CFU-GEMM 67 36 144 8 ND 96 3.0 2.9 128 32

81

Vertebrae Sternuma Sternumb Rib Spleen LD

17 7 3 3 3 10

ND = not done; LD = living donors. The results are expressed as mean s.e.m. a Samples taken after perfusion of organs with UW solution. b Samples taken before perfusion of organs with UW solution.

CD34-positive cell kinetics

2.8 2.6 2.4
Refr, n = 3 RT, n = 3

100 80 60

% CD34

2.2 2.0 1.8 1.6 1.4 1.2 Day 1 Day 2 Day 3

106

40

P < 0.001
20 0 Vertebrae Sternum Rib Spleen

Figure 1 CD34-positive cells (mean s.e.) in cadaveric bone marrow stored for 1 to 3 days in a refrigerator (+4 to +8C) , and at room temperature (+20C) .

Figure 3 Yield of mononuclear cells/g from various sources. Results are expressed as mean 106. Vertebrae, n = 17; sternum, n = 7; rib, n = 3; spleen, n = 3.

100

Refr RT

Colony-forming units During the analysis for colony-forming units (CFU-GM, BFU-E and CFU-GEMM), no signicant differences were seen when samples were stored for 3 days on ice (+4C) or at room temperature (+20C) (Figure 4). When the number of colony-forming units from cadaveric bone marrow was compared with preparations of bone marrow from living donors, there were no signicant differences between either of the assays (Table 1).

95

90

85

80 Day 1 Day 2 Day 3

Figure 2 Effect of temperature and duration of storage on viability of nucleated cells. The results are expressed as mean s.e.m. RT = room temperature (+20C), n = 5; refrigerator (REFR) = +4 to +8C, n = 4.

T cell depletion Depletion with anti CD3, CD4 and CD8-coupled magnetic beads, to remove T lymphocytes, reduced the total number of mononucleated cells by 14 3.5%, with no detectable loss of CD34-positive cells. The fraction of CD3-positive cells decreased from 8.6 4.1% to 0.36 0.23%, which indicates a 96 2.0% depletion efciency. When T cells were depleted with the Ceprate method of selection for CD34-positive cells, 99.6% of the T cells were depleted, but with a loss of 60% of the CD34-positive cells as well (Table 2).

to be superior as regards yield, a calculation of the maximal yield from one donor was performed. This was based on the harvest of 10 vertebral bodies, each containing 30 g of bone marrow with a mean yield of 8.7 107 mononucleated cells/ml. This resulted in a total number of 2.6 1010 mononuclear cells and 5.5 108 CD34-positive cells.

Cadaveric bone marrow G Soderdahl et al

82
1200 1100 1000 900 BFU-E CFU-GEMM CFU-GM

Colonies/ 105 cells

800 700 600 500 400 300 200 100 0 Day 1 RT Day 1 +4C Day 2 RT Day 2 +4C Day 3 RT Day 3 +4C

Figure 4 Effect of temperature and duration of storage on the number of colony-forming units. The results are expressed as mean of three tests. RT = room temperature; or refrigerator = +4 to +8C.

Table 2

Comparison of T cell depletion using magnetic beads (Dynabeads) with the use of CD34-positive cell selection (Cellpro CC-34) Dynabeads Before After Yield % Before Cellpro CC-34 After Yield %

No. of nuclear cells No. of CD34+ cells CD3+ cells

156 3.3 2.5 13.0

137 7.0 2.5 0.4

84 100 3

250 3.6 13.0

2.8 1.3 0.1

1.3 40 0.9

The number of cells are expressed as mean 106 s.e.m.

Spleen cell preparations The fractions of the various hematopoietic cells analyzed by ow cytometry are shown in Table 1. The fraction of CD34-positive cells was signicantly lower in spleen cells, as compared to bone marrow harvested from vertebral bodies (P 0.003). There were also fewer CFUs in spleen cell preparations than in bone marrow from living donors (CFU-GM; P = 0.02). The mean yield from 1 g of spleen tissue was 44 106 mononuclear cells. Assuming a mean weight of 100 g, the total obtainable number of CD34positive cells from a single spleen would be estimated at 19 106. Discussion It was earlier shown that cadaveric vertebral bodies are a suitable source of bone marrow for allogeneic transplantation.14 The method described includes resection of the thoracolumbar column en bloc for further processing in the laboratory. In this paper, we describe a modied method which leaves intervertebral discs and adjacent cortical bone in place to give additional strength for the column, which

is advantageous for ethical reasons. This method is relatively fast and easily performed after removal of the abdominal organs allowing excellent exposure of the thoracovertebral column. The mean number of nucleated cells obtained from a vertebral body is 8.7 107/ml bone marrow, which is somewhat lower than has earlier been reported. However, some of the samples taken at the beginning of the series showed extremely low values, probably because of technical difculties with procurement and inexperience in the harvesting procedure. Using data from the 10 most recent donors, a total yield per donor of 6.2 108 CD34-positive cells was obtained which seems sufcient for transplantation. A point of practical value is that the loss of cells during the rst 36 h is not a crucial problem and the start of procurement can be postponed until the day after the donor operation. Between 1 and 4 106 CD34 cells are sufcient for engraftment using unrelated peripheral blood progenitor cells.19 Only a few CD34 cells are lost during cryopreservation.20 Thus, 6.2 108 CD34 cells would be enough for two adult recipients with 70 kg body weight, giving 4 106 CD34-positive cells/kg. An interesting question arises when comparing cadaveric

Cadaveric bone marrow G Soderdahl et al

donor (CD) bone marrow with bone marrow aspirated from living donors (LD). We saw no differences in viability of the cells after the procurement. The fact that the fraction of CD34-positive cells was smaller in the LD marrow than in the CD marrow probably indicates that aspirated marrow is diluted with peripheral blood. When quality, as measured by colony-forming units, of CD marrow is compared with bone marrow aspirated from living donors, we see a tendency towards lower amounts in the cadaveric marrow, but the difference is not signicant. For conventional allogeneic bone marrow transplantation, the establishment of registers of voluntary living donors has become an important alternative to HLA identical siblings in the search for a well-matched graft.24 However, another potential source is cadaveric bone marrow, procured from organ donors, then frozen and stored for future transplantations. In Sweden, the annual number of organ donors is about 15 per 106 population. If bone marrow were to be routinely harvested and stored, a bank acquiring 120 cases a year could be established. As compared to the four million donors available in various registers, it can be argued that this number is negligible, but due to local variations in the most common HLA patterns, it could provide an additional source when searching for a well-matched graft, for instance in ethnic minorities. As regards costs, the method for harvest and procurement of cadaveric bone marrow is relatively simple and fast, when used together with organ retrieval. There is a not inconsiderable cost for freezing and storage. However, it should be noted that no additional costs would be incurred by HLA-typing and viral screening, since these are done in any case for organ donors. We therefore think that the procurement of cadaveric bone marrow from brain-dead donors is justied. It does not seem worth storing the spleen to be used for stem cell transplantation because an average spleen contains only 1.9 107 CD34+ cells. This would give a dose of only 0.3 106 CD34+ cells/kg to an average adult. In organ transplantation, despite recent advances in immunosuppression, rejection remains the most common cause of allograft failure, and immunosuppressive agents are a signicant cause of morbidity and mortality in clinical transplantation. The possibility of tolerance induction and avoidance of long-term immunosuppression would represent a major step in the development of organ transplantation. Donor-specic bone marrow infusion, as an adjunct to solid organ transplantation, has been observed to promote microchimerism and induction of tolerance in animal models. In man, such infusions have also been associated with establishment of chimerism and in vitro evidence of decreased antidonor reactivity in mixed lymphocyte cultures, and so far only one study showed an improvement in graft survival.5,6,13 However, in none of the trials has preconditioning of the patient prior to bone marrow infusion been performed. The reason for this is fear of toxicity, neutropenia, infections and severe GVHD due to HLA-incompatibility. T cell depletion of transplanted bone marrow has proved to be effective in preventing acute GVHD in HLA-identical siblings.21,22 If T cell depletion is combined with conventional immunosuppressive agents, the GVHD reaction, even

in non-HLA identical matches, may be controlled when a combination of allogeneic and T cell-depleted autologous marrow is given after ablative therapy.7,23,24 This would then seem to be a possible way to achieve macrochimerism and tolerance towards the HLA antigens of the infused bone marrow. Time constraints do not permit giving ablative and immunosuppressive conditioning prior to organ transplantation. Furthermore, there may be very high risks of infections if organ and bone marrow transplantation are performed simultaneously. We would therefore suggest that conditioning and bone marrow transplantation is delayed for some weeks until the surgical wounds have healed and the patient is in a stable clinical condition. We conclude that bone marrow from cadaveric donors for transplantation can be procured with a high degree of engraftment potential. Our data indicate that the yield of stem cells is better when marrow is taken from vertebrae instead of from the ribs or sternal body. With this technique, it seems possible to isolate a sufcient number of the cells required for bone marrow transplantation.

83

Acknowledgements
This study was supported by grants from the Childrens Cancer Foundation (1994-060), the Swedish Cancer Foundation (0070 B95-09XCC), the Belven Foundation, the Swedish Medical Research Council (B96-16X-05971-16C), the FRF Foundation and the Tobias Foundation.

References
1 Thomas ED, Storb R, Clift RA et al. Bone-marrow transplantation. Parts I and II. New Engl J Med 1975; 292: 832843 and 895902. 2 Marks DI, Cullis JO, Ward KN et al. Allogenic bone marrow transplantation for chronic myeloid leukemia using sibling and volunteer unrelated donors. A comparison of complications in the rst 2 years. Ann Intern Med 1993; 119: 207214. 3 Ringden O, Remberger M, Persson U et al. Similar incidence of graft-versus-host disease using HLA-A, -B and -DR identical unrelated bone marrow donors as with HLA-identical siblings. Bone Marrow Transplant 1995; 15: 619625. 4 Downie TR, Hows JM, Gore SM et al. A survey of use of unrelated donor bone marrow transplantation at 46 centers worldwide, 198393. International Unrelated Search and Transplant (IMUST) Study. Bone Marrow Transplant 1995; 15: 499503. 5 Rao AS, Thomson AW, Shapiro R, Starzl TE. Chimerism after whole organ transplantation: its relationship to graft rejection and tolerance induction. Curr Opin Nephrol Hypertens 1994; 3: 589595. 6 Wren SM, Nalesnik M, Hronakes ML et al. Mixed chimerism to induce tolerance for solid organ transplantation. J Pediatr Surg 1991; 26: 439443. 7 Sykes M, Sachs DH. Mixed allogenic chimerism as an approach to transplantation tolerance. Immunol Today 1988; 9: 2327. 8 Fontes P, Rao AS, Demetris AJ et al. Bone marrow augmentation of donor-cell chimerism in kidney, liver, heart, and pancreas islet cell transplantation. Lancet 1994; 344: 151155. 9 Shapiro R, Rao AS, Fontes P et al. Combined kidney/bone

Cadaveric bone marrow G Soderdahl et al

84

10

11 12 13

14

15 16 17

marrow transplantation evidence of augmentation of chimerism. Transplantation 1995; 59: 306309. Monaco AP, Wood ML. Studies on heterologous antilymphocyte serum in mice. VII. Optimal cellular antigen for induction of immunologic tolerance with antilymphocyte serum. Transplant Proc 1970; 2: 489496. Caridis DT, Liegeois A, Barrett I, Monaco AP. Enhanced survival of canine renal allografts of ALS-treated dogs given bone marrow. Transplant Proc 1973; 5: 671674. Thomas FT, Carver FM, Foil MB et al. Long-term incompatible kidney survival in outbred higher primates without chronic immunosuppression. Ann Surg 1983; 198: 370378. Barber WH, Mankin JA, Laskow DA et al. Long-term results of a controlled prospective study with transfusion of donorspecic bone marrow in 57 cadaveric renal allograft recipients. Transplantation 1991; 51: 7075. Rybka WB, Fontes PA, Rao AS et al. Hematopoietic progenitor cell content of vertebral body marrow used for combined solid organ and bone marrow transplantation. Transplantation 1995; 59: 871874. Mugishima H, Terasaki P, Sueyoshi A. Bone marrow from cadaver donors for transplantation. Blood 1985; 65: 392396. Southard JH, Belzer FO. Organ preservation. Annu Rev Med 1995; 46: 235247. Ringden O. Activation of human lymphocyte subpopulations by rabbit anti-human 2 microglobulin and by lipopolysaccharide. Scand J Immunol 1976; 5: 891900.

18 Loken M. Immunouorescence techniques. In: Molamed MR, Lindmo T, Mendelsohn ML (eds). Flow Cytometry and Sorting, 2nd edn. New York: Wiley-Liss, 1990, p 341. 19 Ringden O, Potter MN, Oakhill A et al. Transplantation of peripheral blood progenitor cells from unrelated donors. Bone Marrow Transplant 1996; 17 (Suppl. 2): S62S64. 20 Ek S, Ringden O, Markling L et al. Effects of cryopreservation on subsets of fetal liver cells. Bone Marrow Transplant 1993; 11: 395398. 21 Prentice HG, Blacklock HAA, Janossy G et al. Depletion of T lymphocytes in donor marrow prevents signicant graftversus-host disease in matched allogeneic leukaemic marrow transplant recipients. Lancet 1984; i: 472475. 22 Ringden O, Pihlstedt P, Markling L et al. Prevention of graftversus-host disease with T cell depletion or cyclosporin and methotrexate. A randomized trial in adult leukemic marrow recipients. Bone Marrow Transplant 1991; 7: 221226. 23 Sykes M, Chester CH, Sachs DH. Protection from graftversus-host disease in fully allogenic chimeras by prior administration of T cell-depleted syngenic bone marrow. Transplantation 1988; 46: 327330. 24 Moses RD, Orr KS, Bacher JD et al. Cardiac allograft survival across major histocompatibility complex barriers in the rhesus monkey following T-lymphocyte-depleted autologous marrow transplantation. II Prolonged allograft survival with extensive marrow T cell depletion. Transplantation 1989; 47: 435444.