Near-optimal Conditions for the In Vitro Culture of Hemopoietic Progenitor Cells in Bone Marrow from the Rat

Abstract

In vitro techniques for the culture of hemopoietic stem cells and committed hemopoietic progenitor cells in rat bone marrow have not been adequately described in the literature. In the present investigations, and using commercially available hemopoietic cytokines and growth factors, the conditions required to perform long-term bone marrow culture (LTBMC) using rat femoral bone marrow were studied, in conjunction with the colony-forming unit cell assay (CFU-C), to quantify the number of progenitor cells. CFU-C production by LTBMCs, set up using Iscove’s modified Dulbecco’s medium supplemented with fetal calf serum and horse serum, ceased after week 2 of culture. However, the duration of CFU-C production was significantly increased by supplementing LTBMCs with the cytokine recombinant mouse stem cell factor or recombinant rat stem cell factor.

Introduction

The hemopoietic system of laboratory animals and man is particularly sensitive to the toxic effects of many drugs (e.g., chloramphenicol, chloroquine, penicillamine, busulphan, gold salts) and chemicals (e.g., benzene, lindane, arsenic, alcohol). Exposure to such agents may have severe hematological consequences (Catovsky and Hoffbrand 1999; Heimpel 2000; Jandl 1996; Young and Alter 1994; Young and Vincent 1980). In vitro bone marrow culture assays are useful in assessing toxicity and injury to hemopoietic stem cells and progenitor cell populations in both experimental animals and man (Gasper 2000; Schofield 1986; Williams 2000). Culture techniques for the quantification of stem and progenitor cells using human bone marrow, as well as mouse, cat, dog, horse, and sheep bone marrow, are well defined and recorded (Bagby and Heinrich 2000; Gengozian 2000; Kohn et al. 1995; Swardson and Kociba 1994; Swardson and Kociba 1996). However, in vitro culture techniques have not been adequately described for bone marrow samples from the rat, and indeed, a recent literature search has produced no relevant publications in this field. Considering the importance of the rat in preclinical drug safety evaluation studies in the pharmaceutical industry, and the widespread use of this species in experimental toxicology investigations and in other areas of biomedical research (Mutai 2000), it would be extremely useful if in vitro culture techniques for bone marrow from the rat were routinely available.

Nevertheless, some short-term culture assays have been described using rat bone marrow to identify stem cells and committed progenitor cells (Khaldoyanidi et al. 1997; Lucas et al. 1999). However, these assays have tended to be performed using feeder cells under agar as a source of hemopoietic cytokines and growth factors (Khaldoyanidi et al. 1997); alternatively, spleen-or lung-conditioned media in semisolid methylcellulose have been used (Kim et al. 2001; Kimura et al. 1986).

Using the long-term bone marrow culture (LTBMC) system described by Dexter and others (Brühl et al. 1988; Dexter and Lajtha 1974; Dexter et al. 1977; Gartner and Kaplan 1980) as a starting point, in recent studies with myelotoxic drugs (Gibson et al. 2003; Molyneux et al. 2005; Turton, Fagg et al. 2006), near-optimal conditions for the long-term culture of rat femoral bone marrow have now been identified. In addition, a short-term culture method has also been developed to quantify the number of committed hemopoietic progenitor cells in rat bone marrow using defined, commercially available, hemopoietic cytokines and growth factors. This last (short-term) culture technique was recently used to assess changes in the numbers of progenitor cells, from both myeloid and erythroid lineages, in the rat following the administration of the myelotoxic agent chlorambucil (Molyneux et al. 2004).

It is considered that the identification of optimal culture conditions for the LTBMC of rat bone marrow cells, using defined media, would allow further investigations to be carried out on the myelotoxic effects of anticancer agents and other drugs and chemicals on early rat hemopoietic progenitor cells and on rat bone marrow stromal cells. Therefore, using these methods, it would be possible to quantify the number of hemopoietic stem and progenitor cells in animals treated with experimental agents. In addition, it would provide a technique to assess bone marrow stroma formation and allow the evaluation of the potential toxic effects of drugs and chemicals on rat bone marrow cells, and on the bone marrow microenvironment. Furthermore, the development of in vitro clonogenic assays for rat bone marrow cells would allow investigations to be conducted on the hemotoxicity of agents in vitro.

Femoral bone marrow samples from the outbred Hanover Wistar rat and the inbred F344 rat were investigated in the present studies. Both of these stocks/strains are widely used in toxicologic investigations (Festing 1979, 1990).

Materials and Methods

Bone Marrow Collection

Twelve adult (eight- to ten-week-old) female outbred Hanover Wistar rats and twelve inbred F344 rats (B & K Universal Ltd., Hull, UK) were sacrificed by the intraperitoneal injection of pentobarbitone sodium (Sagatal, Rhône Mérieux Ltd., Harlow, UK). The bone marrow (BM) contents of the left and right femora were each aspirated into 5 mL Iscove’s modified Dulbecco’s medium (IMDM; Life Technologies, Paisley, UK) supplemented with 10% fetal calf serum (FCS; PAA Laboratories GmbH, Linz, Austria).

Long-term Bone Marrow Culture

Bone marrow cell suspensions in IMDM supplemented with 10% FCS were counted by hemocytometer using trypan blue exclusion of dead cells (Turton, Sones et al. 2006). Long-term bone marrow cultures (LTBMCs) were set up using 12 x 10⁶ bone marrow cells in IMDM with the osmolality adjusted to 340 mOsm/kg (IMDM₃₄₀; Invitrogen Ltd., Paisley, UK), supplemented with 10⁶M hydrocortisone succinate (Sigma Chemical Co., Poole, UK), 1% glutamine (Sigma), 100 IU of penicillin and streptomycin (Gibco, Paisley, UK), and 20% FCS. Cultures were set up in six-well plates, in duplicate, with or without the addition of 100 ng/mL recombinant mouse (rm) stem cell factor (rmSCF; R & D Systems Europe Ltd., Abingdon, UK) or 100 ng/mL recombinant rat (rr) SCF (rrSCF, Amgen UK Ltd., Cambridge, UK). Cultures were maintained at 33°C in 5% CO₂ in air.

Colony-forming Unit Cell Assay

At weekly intervals, 50% of the nonadherent cells were removed from the LTBMCs and the volume replaced with fresh medium, which included rmSCF or rrSCF, or which did not include either SCF. Nonadherent cells were counted using trypan blue exclusion and cultured at 10⁵ cells in 1 mL IMDM containing 30% FCS, 1% bovine serum albumin (Sigma), 10⁻⁴M β-mercaptoethanol (Sigma), and 0.9% methylcellulose (Stem Cell Technologies Inc., London, UK). Cultures were set up in duplicate in 35 mm dishes (Nunclon, Loughborough, UK), with the following growth factors added to each dish: 4 IU recombinant human erythropoietin (rhEpo; Janssen-Cilag Ltd., High Wycombe, UK), 50 ng recombinant mouse interleukin 3 (rmIL-3; R & D Systems Europe Ltd.), 50 ng rmSCF or 50 ng rrSCF, 50 ng recombinant human interleukin 6 (rhIL-6; Novartis Pharmaceuticals Ltd., Langley, UK), and 50 ng recombinant human granulocyte colony stimulating factor (rhG-CSF; Amgen UK Ltd.). The cultures were incubated at 37°C in 5% CO₂ in air for fourteen days. On day 14, granulocyte–macrophage colony-forming units (CFU-GM), erythroid burst-forming units (BFU-E), and colonies containing both granulocyte–macrophage and erythroid elements (CFU-GEM) were counted. The numbers of CFU-GM, BFU-E, and CFU-GEM colonies were added together, and the total number of colony-forming unit cells (CFU-C) per well was calculated.

Statistical Analysis

A one-way analysis of variance (ANOVA) followed by Tukey’s highest significance test for post hoc pairwise multiple comparison was used to compare the number of CFU-GM and erythroid colonies produced by fresh rat bone marrow cells cultured with rhEpo, rmIL-3,rhIL-6,rhG-CSF, and either rmSCF or rrSCF.

Nonadherent cells harvested from LTBMC grown without SCF, or with the addition of either rmSCF or rrSCF, were compared at each time point using a one-way ANOVA followed by Tukey’s highest significance test for post hoc pairwise multiple comparison.

All statistical analysis was carried out using GraphPad Prism Version 4.00 for Windows (GraphPad Software, San Diego, CA, USA).

Results

The number of committed progenitor cells produced by 10⁵ fresh rat bone marrow cells was assessed using the colony-forming unit cell (CFU-C) assay. Cultures were supplemented with rhEpo, rmIL-3,rhIL-6,rhG-CSF, and either rmSCF or rrSCF (Table 1). The production of CFU-GM colonies in cultures from Hanover Wistar rat bone marrow was comparable when cultures were supplemented with either rmSCF (40.8 colonies) or rrSCF (38.7 colonies). The number of erythroid colonies produced was, however, significantly higher in cultures supplemented with rrSCF (10.8 colonies, rrSCF; 4.3 colonies, rmSCF; p < .001). The number of CFU-GM colonies produced by F344 bone marrow was higher in cultures supplemented with rmSCF (38.0 colonies, rmSCF; 24.0 colonies, rrSCF; p < .001); however, the number of erythroid colonies was larger in cultures supplemented with rrSCF (48.4 colonies, rrSCF; 19.3 colonies, rmSCF; p < .001).

The addition of rmSCF profoundly improved Hanover Wistar rat LTBMCs, compared with cultures without SCF (Table 2); the number of CFU-GM colonies was significantly increased at all time points from week 1 to week 5 of culture (p < .01 or p < .001). The production of erythroid colonies was also sustained for a longer period of time (until week 5) in cultures supplemented with rmSCF in comparison with cultures without SCF (Table 2). However, this increase in the number of erythroid colonies in cultures with rmSCF was statistically significant only at weeks 1 and 2 of culture (p < .001 at both time points).

The addition of rrSCF had a similar overall effect on Hanover Wistar LTBMCs, as did supplementation with rmSCF (Table 2). It is seen that production of CFU-GM colonies was sustained until week 5 of culture, and the number of colonies generated was significantly increased at weeks 1, 2, and 3 in comparison with cultures without SCF (p < .01 or p < .001). Erythroid colony production was also improved by the addition of rrSCF compared with cultures lacking SCF (Table 2).

Cultures set up with rmSCF and with rrSCF were generally comparable in the number of colonies produced (Table 2). However, LTBMCs supplemented with rmSCF gave rise to significantly more CFU-GM colonies at week 4 of culture, and to significantly more erythroid colonies at week 2 of culture (p < .05 for both CFU-GM and erythroid colonies).

Using F344 rat bone marrow, LTBMCs also performed significantly better with the addition of either rmSCF or rrSCF compared to cultures without SCF (Table 3). Compared with LTBMCs without SCF, cultures supplemented with rmSCF produced significantly more CFU-GM colonies at week 1, 2, 3, and 4 and significantly more erythroid colonies at week 1 and 2. The addition of rrSCF did not improve colony output in F344 LTBMCs to the same extent as that seen with rmSCF (Table 3). However, both CFU-GM and erythroid colonies were significantly increased in cultures supplemented with rrSCF at weeks 1 and 2 of culture in comparison with cultures without SCF. In terms of the numbers of CFU-GM colonies and erythroid colonies produced in F344 rat LTBMCs supplemented with rmSCF or rrSCF, statistical analysis showed no significant differences in the number of colonies generated by either of the two SCFs.

Discussion

The present studies, designed to identify the optimal conditions for LTBMC of rat femoral bone marrow cells, demonstrate that the number of committed progenitor cells can be quantified when the marrow is cultured in semisolid methylcellulose supplemented with defined, commercially available cytokines and growth factors.

In general, the long-term culture of bone marrow from the outbred Hanover Wistar rat and the inbred F344 rat gave similar results (Tables 2 and 3 ). However, the addition of rmSCF or rrSCF appeared to be essential; rmSCF and rrSCF both improved the duration of the LTBMCs and increased the number of colonies produced weekly in the CFU-C assay. These results find a parallel in the study of Dunlop et al. (1993), who reported that rrSCF enhanced the proliferation of lineage-negative mouse hemopoietic progenitor cells. Similar findings have been reported in studies with SCF in canine bone marrow cultures (Sandmeier et al. 1996; Shull et al. 1992).

The culture of rat bone marrow using long-term and short-term culture assays would allow the measurement of changes in both committed progenitor cells and in more primitive bone marrow progenitor cells. This would therefore provide techniques to measure the toxic effects of experimental agents on the bone marrow compartment of rats, which has not been possible previously. It is considered that the present studies on rat bone marrow culture complement the well-defined and widely used techniques already available for the in vitro culture of marrow from other species, such as man, mouse, cat, dog, and horse (Gengozian 1998; Gengozian and Legendre 1995; Kohn et al. 1995; Swardson and Kociba 1994; Swardson and Kociba 1996; Whitwam et al. 1998). Furthermore, the present experiments would allow a second rodent species to be considered in comparative studies with bone marrow cultures, whereas in the past only mouse bone marrow could be used. The present techniques will also be useful in a number of areas of hematological research, and in particular in investigations on the assessment of drug- and chemical-induced bone marrow toxicity, and in the wider fields of investigative toxicology and experimental pathology.

In conclusion, although it is appreciated that the present investigations have resulted in the development of a technique that allows the setting up of rat LTBMCs, with the production of CFU-Cs, the time period involved is only up to a maximum of four to five weeks (Tables 2 and 3 ). Therefore, it is reasonable to point out that the technique requires further development to extend this present four- to five-week period of culture. Also, although the longevity of the present rat LTBMCs compares with the period of culture for mouse LTBMCs (Dunlop et al. 1993), the current technique in the rat does leave room for further improvement. Nevertheless, the present approach permits worthwhile hematological investigations with rat bone marrow to be carried out, and it is hoped that other research workers in the field will consider the present methods as a starting point and that they will carry out experiments to further improve the technique.