SEARCH

SEARCH BY CITATION

Keywords:

  • pH;
  • ex vivo expansion ;
  • erythroid differentiation

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. References

Physiological parameters such as pH and oxygen tension probably play significant roles in the regulation of haemopoiesis in the bone marrow microenvironment, but these roles have yet to be characterized in detail. We have found that changes in culture pH (0.2 units) can cause significant changes in the culture composition of mature cells and colony-forming cells (CFCs), especially in the presence of erythropoietin (Epo). Peripheral blood (PB) CD34+ cells cultured at different pH values (7.15–7.6) were characterized using total cell counts, colony assays, morphological analysis, haemoglobin staining, flow cytometry, immunocytochemical staining, and Western blots. Cultures performed at high (7.6) pH contained greater numbers of haemoglobin-positive and band-3-positive cells, and acquired these erythroid differentiation markers sooner than standard (7.35) and low (7.1) pH cultures. Flow cytometry using CD71 and CD45RA antigens also indicated that erythroid differentiation proceeds faster at high pH and is blocked at an intermediate stage by low pH. Morphological data confirmed that high pH cultures had been shifted towards late-stage erythroid compartments as compared to low and standard pH cultures. These findings have important implications both in elucidating the regulatory role of pH in the bone marrow microenvironment and for the design of in vitro systems to study the development of erythroid cells.

It has been recognized since early this century that pH buffering is an important and complex control system in mammals. Cells subjected to different pH environments exhibit distinct responses in terms of differentiation and protein secretion patterns. In haemopoietic cells, changes in intracellular pH have been associated with activation of lymphocytes, neutrophils and platelets ( Ozaki, 1992). Also, the proliferation of, and erythropoietin (Epo) secretion by, murine macrophages have been shown to be pH-dependent, with optimal values at higher than physiological pH (7.6–8.0) ( Rich, 1988). Although the connection between pH and haemopoietic cells is more difficult to make in vivo because of interactions with CO2, O2 and bicarbonate, there are suggestions of a link between blood pH and the generation and function of erythroid cells. One such condition is end-stage renal failure, in which both acidosis (low blood pH) and anaemia (shortage of red blood cells) are primary symptoms ( Segal et al, 1988 ). In addition, Yang (1992) and Yang et al (1995 ) have demonstrated that some respiratory failure patients who exhibit acidosis also have impaired function of the primary ion channel found in erythroid cells.

In order to provide insight into these pathophysiological conditions and to further understand the effects of culture pH on the erythroid lineage, we have conducted experiments to examine in detail the effects of culture pH on the ex vivo differentiation of erythroid lineage cells. Since bone marrow pH in vivo is likely to be as low as 7.1, and dramatic effects on haemopoietic cell lines were noted at pH 7.6 ( Endo et al, 1994 ), we chose to examine three pH values spanning this range (7.1, 7.4, 7.6). In previous work we have demonstrated that culture pH has a significant effect on haemopoietic cells ( McAdams et al, 1997 ). These studies demonstrated important changes in cloning efficiency with pH and provided preliminary indications of changes in the differentiation programme of haemopoietic (particularly erythroid) progenitors cultured ex vivo. In the current studies we used the previously developed culture system to perform a detailed analysis of erythroid differentiation markers and draw new conclusions about the effects of culture pH on the kinetics of erythroid differentiation. These studies were performed with CD34+ progenitors, thus extending our investigation of pH effects to this clinically important cell population. Based both on our in vitro results and an examination of the bone marrow microvasculature, we propose that pH may be an important regulator of erythroid proliferation within the bone marrow microenvironment.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. References

Cells and cell separation procedures

Apheresis products collected from patients with nonhaematological cancers following stem-cell-mobilization regimens consisting of treatment with granulocyte CSF (G-CSF) with or without chemotherapy (hereafter referred to as peripheral blood mononuclear cells (PB MNC)) were obtained from Response Oncology (Memphis, Ten.). All cells were obtained after informed consent under protocols approved by the respective Institutional Review Boards. After approximately 48 h in autologous serum at ambient temperature, PB MNC were selected for CD34+ cells using the MiniMACS system (Miltenyi Biotec, Sunnyvale, Calif.) according to the manufacturer's instructions.

Growth factors

Interleukin-3 (IL-3) and IL-6 were donated by Novartis (East Hanover, N.J.). Stem cell factor (SCF) was donated by Amgen (Thousand Oaks, Calif.). Granulocyte-macrophage colony stimulating factor (GM-CSF) was purchased from Immunex (Seattle, Wash.), G-CSF from Amgen, and Epo from Ortho Biotech (Raritan, N.J.).

Methylcellulose colony assays

Colony assays were conducted by plating cells at 1000 cells/ml. The medium consisted of Iscove's modified Dulbecco's medium (IMDM) containing 1.1% methylcellulose (Dow, Midland, Mich.), 30% fetal bovine serum (FBS; Hyclone, Logan, Utah), 2% bovine serum albumin (BSA; Intergen, Purchase, N.Y.), 100 μM 2-mercaptoethanol, 50 μg/ml gentamycin sulphate (GIBCO, Gaithersburg, Md.), 5.0 ng/ml IL-3, 10 ng/ml IL-6, 4 ng/ml GM-CSF, 1.5 ng/ml G-CSF, 50 ng/ml SCF, and 3 U/ml Epo. Aliquots of 1.0 ml were plated in duplicate in 35 mm suspension culture dishes (Nunc, Naperville, Ill.) and incubated for 13–15 d at 37°C under fully humidified conditions in an atmosphere of 5% CO2 and 5% O2. Colonies containing >50 cells were scored as CFU-GM, BFU-E or CFU-Mix according to standard criteria.

pH-adjusted liquid cultures

Serum-containing (12.5% FBS and 12.5% horse serum) human long-term medium (HLTM) was prepared as previously described ( Sandstrom et al, 1997 ). All cultures were performed with 1 ml HLTM supplemented with 50 ng/ml SCF, 5 ng/ml IL-3, 10 ng/ml IL-6 and 3 U/ml Epo. Medium was adjusted to the proper pH by addition of 1.0 N HCl or 1.0 N NaOH. PB CD34+ cells were plated at 5000/ml in 1 ml in the centre eight wells of 24-well plates containing sterile water in the outside 16 wells. pH measurements were taken on days 0, 6 and 10. Cells were cultured at 37°C under fully humidified conditions in an atmosphere of 5% CO2 and 5% O2. Cultures were fed by the addition of 1 ml of medium on day 6, and by one half medium exchange on days 12 and 18 for extended cultures.

Benzidine staining assay

Staining was performed according to the method of Gopalakrishnan & Anderson (1979).

pH measurements

pH was measured by sacrificing the entire contents of a replicate well into a 17 × 100 mm polypropylene tube and measuring immediately on a Corning 340 pH meter (Corning, N.Y.).

Flow cytometry analysis

Cells were evaluated for the expression of CD45RA-PE and CD71-FITC using the protocol by Bender et al (1994 ). Briefly, cells were labelled with monoclonal antibodies and fixed in 1% paraformaldehyde, and 10 000 events were collected using a FACScan flow cytometer (Becton Dickinson, San Jose, Calif.).

Western blots

Reduced cells lysates (15 μg total protein per lane for haemoglobin blots, 25 μg total protein per lane for band three blots) were run on SDS-Page gels, transferred to nitrocellulose and incubated with primary (mouse anti-human band 3 and rabbit anti-human haemoglobin (polyclonal), Sigma) and secondary antibodies (peroxidase-conjugated goat anti-mouse IgG and peroxidase-conjugated mouse anti-rabbit IgG, Sigma) according to the protocol described by Sambrook et al (1997 ). Bound antibody was visualized using an ECL Western detection kit (Amersham Life Sciences, Arlington Heights, Ill.). Blots were scanned and analysed using NIH Image software (Bethesda, Md.). Protein concentrations were determined using a Bio-Rad Protein Assay Kit (Bio-Rad Laboratories, Hercules, Calif.).

Morphological analysis

Cytospin slides were prepared by centrifuging 15 000 cells in cytospin funnels at 1000 rpm for 5 min using a Shandon Cytospin 3 (Pittsburgh, Pa.). Cells were then stained with Wright-Giemsa stain (Harleco, Gibbstown, N.J.) for 30 s, followed by a phosphate buffer rinse for 1 min. The slides were then evaluated for the presence of granulocyte lineage, monocyte lineage, pronormoblast, basophilic normoblast, polychromatic normoblast, and orthochromatic normoblast cells.

Immunocytochemical staining

Cytospins of cells were stained for band 3 using a Histomark Streptavidin–horseradish peroxidase (HRP) kit (Kirkegard & Perry, Gaithersburg, Md.). The staining procedure was performed at 25°C. Slides were soaked in a 15 : 1 mixture of phosphate-buffered saline (PBS) and 30% H2O2 for 10 min to block endogenous peroxidase and were then washed with PBS. Slides were then incubated with normal goat serum for 15 min in a humidified environment to block non-specific staining. Excess serum was subsequently removed, and slides were incubated with anti-band 3 antibody (1 : 500 dilution in PBS) for 1 h. Slides were then incubated with biotinylated goat anti-mouse secondary antibody for 30 min, streptavidin–peroxidase for 30 min, and a chromogen solution (peroxidase chromogen kit, Biomeda, Foster City, Calif.) for 2–10 min for colour development, with a PBS wash performed prior to each step. Slides were rinsed with distilled water and then stained with haematoxylin (Biomeda) for 1 min. Slides were rinsed for a final time with tap water and then air-dried, coated with Crystal Mount (Biomeda), and dried overnight before applying a coverslip using Accumount (Scientific Products). Approximately 200 cells per slide were counted to determine the percentage of band-3-positive cells in the cultures.

Cell count and size analysis

Nucleated cell counts were performed by diluting cell samples in cetrimide and counting the released nuclei on a Coulter Multisizer (Coulter, Hialeah, Fla.) Cell size distributions were determined using the Coulter Multisizer by diluting cell samples in Isoton II (Coulter).

Statistical analysis

Table 1. Table I. PB CD34+ cultures in HLTM with IL-3/IL-6/SCF/Epo (n = 12). Abbreviations: SEM standard error of the mean; ND, no data were obtained. Absolute value multipliers: Cells (×10−5); CFU-GM (×10−3); BFU-E (×10−3). Mean relative values are obtained by normalizing each value to the value obtained at intermediate pH in each individual experiment and averaging these normalized values.* Significant (P < 0.007; Student's t-test) difference between the indicated data point and the corresponding data point at intermediate pH. Thumbnail image of

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. References

pH-adjusted liquid culture

The results for 12 replicate cultures of PB CD34+ cells harvested on day 10 are presented in 11 Table I. Table I indicates that lower pH significantly reduces (by about 60%) total cell production, whereas higher pH has no significant effect on total cells in cultures supplemented with SCF/IL-3/IL-6/Epo. The number of erythroid-lineage cells in each culture was assessed by benzidine staining ( 1 Table I), which was used to detect the presence of haemoglobin. A summary of the stage at which various markers of erythroid differentiation (including haemoglobin) are expressed is presented in 2 Table II. As expected for the growth factor combination of SCF/IL-3/IL-6/Epo, the culture was predominantly (68% at pH 7.15) erythroid in nature. Low pH caused a statistically significant reduction in erythroid cells to 52%, while high pH increased erythroid cells to 80% of the culture. The decrease in the number of BFU-E detected at low pH conditions was not statistically significant, but the number of BFU-E ( 1 Table I) was reduced >3-fold by high pH. Similar results were seen for CFU-GM colonies ( 1 Table I); no change at low pH, and a significant drop at high pH. Taken together, the results for total cells and progenitor cells yield highly reproducible and statistically significant differences in cloning efficiency ( 3 Table III). At low pH, the drop in total cells combined with maintenance of colony forming cells results in an approximately doubled cloning efficiency (5%) over that in standard pH culture conditions (2.7%). At high pH, the cloning efficiency (1%) is less than half of that of standard pH conditions.

Table 2. Table II. Summary of erythroid differentiation parameters. Telen, 1990; Jandl, 1987b; Loken et al, 1987 ; Okumura et al, 1992 −, Not expressed; ±, either not expressed or expressed at very low levels; +, expressed at moderate levels; ++, expressed at high levels. Data summarized from multiple sources ( Telen, 1990; Jandl, 1987b; Loken et al, 1987 ; Okumura et al, 1992 ). Thumbnail image of
Table 3. Table III. Cloning efficiencies of PB CD34+ cultures in HLTM supplemented with IL-3/IL-6/SCF/Epo (n = 12). * Statistically significant (P < 0.001) compared to the intermediate pH value. Thumbnail image of

Morphological examination of cells from liquid cultures

Cells from three liquid cultures were Wright-Giemsa stained for morphological examination. Cells were classified into six categories: monocyte lineage (M), granulocyte lineage (G), pronormoblast (E1), basophilic normoblast (E2), polychromatic normoblast (E3) and orthochromatic normoblast (E4). The results (Fig 1) indicated that in low pH cultures, cells were approximately 43% erythroid, with pronormoblasts (12%) and basophilic normoblasts (21%) as the most predominant erythroid compartments. In intermediate pH cultures, cells were 76% erythroid with basophilic normoblasts (29%) and polychromatic normoblasts (33%) the most abundant subtypes. In high pH cultures, cells were 84% erythroid and orthochromatic normoblasts (38%) were the most common erythroid subtype. The morphological data presented here matched well with the benzidine staining data described in the previous section. Furthermore, the results were extended to individual subtypes of erythroid progenitors, showing a predominance of immature subtypes at low pH, and a shift towards more mature subtypes as pH was increased to intermediate and high pH conditions.

image

Figure 1. , pronormoblast (least mature erythroid); E2, basophilic normoblast; E3, polychromatic normoblast; E4, orthochromatic normoblast (most mature erythroid)

Download figure to PowerPoint

Flow cytometry analysis

The antigens CD45RA and CD71 have proved useful for classifying erythroid cells by flow cytometry. Among CD34+ cells, CD45RACD71+ cells have been identified as erythroid progenitor cells ( Mayani et al, 1993 ), and can be clearly identified in Fig 2 as the subpopulation in the lower right quadrant of the day 0 cell analysis. Cells in the erythroid lineage are CD45RA negative, and do not up-regulate expression of this antigen during differentiation. The CD71 antigen is up-regulated during the intermediate stages of erythroid differentiation, then down-regulated in the final stages ( Okumura et al, 1992 ). Thus, as the cultures became progressively more erythroid in nature, proportionally fewer cells were found in the upper two quadrants (non-erythroid cells), and more in the lower right quadrant (intermediate stage erythroid cells). Then, as erythroid differentiation proceeded, proportionally greater numbers of cells were found in the lower left quadrant (mature erythroid cells). The data in Fig 2 are from one representative experiment (of two examined) from the 12 erythroid-specific PB CD34+ cell cultures (data presented in Table I). Fig 2 indicates that by day 7, cultures performed at all three pH values were predominantly erythroid in nature. One feature of interest at day 7 was that the high pH culture already contained a significant fraction of mature CD71 erythroid cells (48%), whereas low and intermediate pH cultures contained very few mature erythroid cells (2% and 8%, respectively). The most interesting feature of the day 14 data was the fact that while the intermediate pH culture had progressed to the mature erythroid stage, the low pH culture had not yet advanced to the CD71 stage of differentiation. Thus, the flow cytometry data suggest that pH altered the rate of erythroid differentiation, accelerating it at high pH, and slowing it down at low pH.

image

Figure 2. Fig 2. Flow cytometry analysis for CD45RA and CD71 antigen content at days 0, 7 and 14 for one representative erythroid culture inoculated with PB CD34+ cells 5000 cells/ml in HLTM + SCF/IL-3/IL-6/Epo at low (7.15), intermediate (7.4) or high (7.6) pH.

Download figure to PowerPoint

Kinetic evaluation of erythroid differentiation markers

The benzidine staining results in Fig 3 provide a clear indication of the altered kinetics of erythroid differentiation that was suggested by the flow cytometry data in Fig 2. On day 6 in the high pH cultures, the proportion of haemoglobin-containing cells was already significantly elevated above those in the cultures at lower pH. On day 8, the intermediate pH culture began to express haemoglobin and at day 14 caught up with the high pH culture. The low pH culture exhibited a very slow rise in haemoglobin-containing cells that did not reach 50% even by day 18, again indicating some type of block in erythroid differentiation as suggested by the flow cytometry analysis.

image

Figure 3. /IL-6/Epo at low, intermediate or high inoculation pH, as indicated.

Download figure to PowerPoint

The presence of band 3, an antigen up-regulated later in erythroid differentiation ( Sieff et al, 1982 ) was tracked for 14 d by immunostaining of cytospin slides (Fig 4). Consistent with the above results, the high pH culture was predominantly (80%) band 3 positive by day 14, the intermediate pH culture expressed an intermediate amount, and expression of band 3 was very small in the low pH cultures.

image

Figure 4. + cells at 5000 cells/ml in HLTM + SCF/IL-3/IL-6/Epo at low, intermediate or high inoculation pH, as indicated.

Download figure to PowerPoint

Finally, it is known from previous work that as cells differentiate along the erythroid lineage, their diameter steadily decreases ( Loken et al, 1987 ). Since visual examination suggested that the average cell diameter decreased in high pH cultures, the Coulter Multisizer was used to track the size distributions of cells cultured at the three pH levels (data not shown). The high pH culture had the smallest average cell diameter, which is consistent with the above evidence that this culture contained more differentiated erythroid cells than did cultures at lower pH values. Cultures at intermediate and low pH had larger average cell diameters, indirectly indicating that these cultures contained fewer and/or less differentiated erythroid cells. From the cell size distributions it was apparent that several distinct subpopulations developed in the intermediate and high pH cultures, whereas there was a single symmetric peak for the low pH culture. This observation would fit the hypothesis that erythroid differentiation is blocked at an intermediate stage in the low pH culture.

Western blots

Total cells from day 10 cultures at each pH were lysed to obtain protein for immunoblotting with band 3 and haemoglobin antibodies (Fig 5). The band 3 blot confirmed the results obtained with the cytospin procedure: that band 3 expression was maximal at high pH, lower at intermediate pH, and essentially absent at low pH. The haemoglobin blot confirmed that significantly more haemoglobin-expressing cells were produced in intermediate and high pH cultures than at low pH. Densitometry results (legend, Fig 5) indicated that there was a large difference (6.2-fold) between the amount of haemoglobin contained in cells in the low and intermediate pH cultures, even though the difference in the percentage of haemoglobin-containing cells was relatively small ( Table I; 1.3-fold). These data suggest that in addition to increasing the percentage of haemoglobin-expressing cells, increased pH many also increase the quantity of haemoglobin protein per cell. This was consistent with our earlier work in methylcellulose culture, in which we observed that as pH was increased, erythroid colonies took on a progressively deeper shade of red ( McAdams et al, 1997 ). This was also consistent with the greater haemoglobin content expected as erythroid cells proceeded through differentiation ( 2Table II).

image

Figure 5.  301 (high pH).

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. References

Changes in culture pH have been shown to affect haemopoietic differentiation in several systems. For the myeloid lineage, Fischkoff et al (1984 ) and Fischkoff & Rossi (1990) demonstrated that HL-60 (a human myeloid leukaemia cell line) cells, which normally differentiate into monocytes when cultured at pH 7.2, will instead differentiate into eosinophils when cultured at pH 7.6. For the erythroid lineage, Endo et al (1994 ) investigated the effects of pH on the human erythroid cell lines KU-812 and K562. Their experiments indicated that a significantly greater proportion of cells acquire a mature erythroid phenotype (haemoglobin-positive) when cultured at pH 7.6 than when cultured at lower pH values. A progressive increase in differentiation was noted as pH increased, indicating gradual changes rather than a step-change phenomenon.

Our initial studies investigating the effects of culture pH on ex vivo haemopoietic cultures ( McAdams et al, 1997 ) led us to hypothesize that the rate and/or extent of differentiation of cells of the erythroid lineage accelerated with increasing extracellular pH, over the range 7.1–7.6. In this report we have provided data supporting this hypothesis by examining multiple markers of erythroid differentiation in the culture of CD34+ peripheral blood progenitor cells. The culture data in Table I show that culture at low pH resulted in fewer total cells without significantly affecting the number of BFU-E. Increasing the pH, however, resulted in a significant drop in BFU-E content, and an increase in the percentage of haemoglobin-containing cells. These data suggest that the generation of more mature erythroid cells from BFU-E progenitors is inhibited by low pH, whereas at high pH it is accelerated, resulting in a loss of BFU-E. An alternative explanation is that the low pH conditions promote greater destruction of mature erythroid cells rather than inhibiting their production. However, our previous study ( McAdams et al, 1997 ) utilizing both pH-adjusted methylcellulose cultures and liquid cultures (same culture conditions as in these studies) failed to document the presence of any (even non-viable) mature erythroid cells, or increased cellular debris that one would expect to accompany the destruction hypothesis. A further examination of the cellular phenotype in these cultures was performed using morphological (Fig 1) and flow cytometry (Fig 2) analyses. Both analyses indicated that the distribution of erythroid cell types was more immature at low pH, and more mature at high pH. The flow cytometry analysis, in particular, demonstrated a kinetic effect in that similar patterns of change occurred, but at different rates in different pH environments. The time course profile for benzidine-positive cells in Fig 3 provided further confirmation of this change in the rate of differentiation. The use of band 3 as an additional erythroid marker (Fig 4) supports the hypothesis, showing increased expression at high pH and essentially no expression at low pH. Together, these results indicate that a low pH-induces block or bottleneck in erythroid differentiation occurs at a post-BFU-E stage after the cell has up-regulated CD71 and begun to accumulate haemoglobin, but before it acquires late erythroid markers such as band 3 ( Table II). The morphological analysis suggests that this is most probably somewhere within the basophilic normoblast stage. The Western blots (Fig 5) were consistent with our assessment of the location of the low pH-induced block, showing that the low pH cultures contained a small amount of haemoglobin, but no band 3 protein.

Both the structure of the bone marrow ( Adler, 1984) and measurements of pH in various tissues strongly suggest that substantial pH gradients exist in the bone marrow microenvironment. In normal brain tissue, pH values range from 6.7 to 7.5 ( Gerweck & Seetharaman, 1996). Measurements of normal subcutaneous tissue using a sensitive fluorescence ratio imaging technique have found an average pH value of 7.23 ( Martin & Jain, 1994). In addition, the same study documented steep pH gradients as the distance from blood vessels (pH 7.4) was increased to 10 μm (pH 7.25) or 30 μm (pH 7.1). The latter two values correspond to about one or three cell diameters in distance from the blood vessel. Structural studies of bone marrow show that a significant amount of the bone marrow extravascular space is >50 μm from the nearest sinus ( Adler, 1984; Weiss, 1976).

We propose that pH plays a role in regulating the development of erythroid cells within the bone marrow. The later stages of erythroid differentiation are often accomplished within erythroblastic islands, in which a ‘nurse’ macrophage cell assists in the maturation and enucleation of the erythroid cells with which it maintains intimate contact in the bone marrow microenvironment ( Tavassoli & Yoffey, 1983). Since mature erythroid cells lack motility, these erythroblastic islands are located immediately adjacent to the sinus wall ( Jandl, 1987a), thus facilitating the egress of mature erythroid cells from the bone marrow. We suggest two possible physiological roles in vivo of our observations that reduced pH slows erythroid development. One possible role may be to regulate the rate at which erythroid development occurs within the erythroblastic island. Due to the large size (>50 μm) of the erythroblastic island ( Wickramasinghe, 1975), erythroid cells at the interior are likely to be exposed to a lower pH than are the cells at the periphery. This would lead to an increasing rate of differentiation of the erythroid cells radially outward from the macrophage, which may facilitate the release of the most mature erythroid cells at the periphery into the circulation.

Another role may be to allow the primitive erythroid progenitor, which may be generated away from the sinus wall, time to reach an appropriate macrophage near the sinus before beginning its differentiation programme. Maturation of erythroid progenitors far into the bone marrow interior may be disadvantageous because of the lack of motility of mature erythroid cells ( Jandl, 1987a). The lower pH further away from the sinuses would tend to slow down or inhibit the erythroid maturation process. The primitive erythroid cells would gradually migrate or be pushed towards the sinuses by the general outward flow ( Lord, 1990) of haemopoietic cells within the bone marrow environment. As the erythroid progenitors encounter increasing pH values near the sinus, they would accelerate their programme of differentiation at a more appropriate location in the bone marrow with the assistance of macrophages. This is in contrast to granulocytic development — in which granulocytic post-progenitors are formed deeper in the bone marrow and are actively motile, propelling themselves towards the sinuses as they differentiate ( Jandl, 1987a). This may account for the less profound effects of culture pH on the granulocytic lineage in comparison to the erythroid lineage ( McAdams et al, 1997 ).

The observation that the rate and/or extent of erythroid differentiation is sensitive to changes in pH has implications in several areas. First, in studies characterizing the effects of other agents and conditions on the erythroid lineage, it is very important to control and monitor pH throughout the course of an experiment. This may be particularly important in cultures utilizing stromal cells or high-density cultures in monolayers, since the metabolic output of such cultures may produce significant changes in local pH values even if the bulk medium pH remains unaltered ( Akatov et al, 1985 ). Second, this report demonstrates that pH gradients within bone marrow tissue may be an important mechanism in regulating erythroid development. Further studies on the effects of pH on haemopoietic cells may provide additional insight into these regulatory mechanisms.

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. References

We thank Chet Cudak, Kevin Longin and Dr Bonnie Hazelton at Response Oncology (Memphis, Ten.) for providing peripheral blood apheresis samples. We thank Dr Xiaoing Qiao for the morphological analysis of cytospin slides, and Wen Lu for the Western blots. We acknowledge the donation of SCF by Amgen and of IL-3 and IL-6 by Novartis. Supported in part by National Institutes of Health Grant HL48276. T.A.M. was partially supported by a Department of Defense Breast Cancer Training Grant USAMRMC DAMD17-94-J-4466.

References

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. References
  • 1
    Adler, S.S. (1984) Blood circulation of bone marrow. Blood Vessels and Lymphatics in Organ Systems (ed. by D. E. Abramson and P. B. Dobrin), p. 705. Academic Press, San Deigo.
  • 2
    Akatov, V.S., Lezhnev, E.I., Vexler, A.M., Kublik, L.N. (1985) Low pH value of pericellular medium as a factor limiting cell proliferation in dense cultures. Experimental Cell Research, 160, 412 418.
  • 3
    Bender, J.G., Unverzagt, K., Walker, D. (1994) Guidelines for determination of CD34+ cells by flow cytometry: application to the harvesting and transplantation of peripheral blood stem cells. Hematopoietic Stem Cells (ed. by E. Wunder, H. Sovalat, P. R. Hénon and S. Serke), p. 31. AlphaMed Press, Dayton.
  • 4
    Endo, T., Ishibashi, Y., Okana, H., Fukumaki, Y. (1994) Significance of pH on differentiation of human erythroid cell lines. Leukemia Research, 18, 49 54.
  • 5
    Fischkoff, S.A., Pollak, A., Gleich, G.J., Testa, J.R., Misawa, S., Reber, T.J. (1984) Eosinophilic differentiation of the human promyelocytic leukemia cell line, HL-60. Journal of Experimental Medicine, 160, 179 196.
  • 6
    Fischkoff, S.A. & Rossi, R.M. (1990) Lineage directed HL-60 cell sublines as a model system for the study of early events in lineage determination of myeloid cells. Leukemia Research, 14, 979 988.
  • 7
    Gerweck, L.E. & Seetharaman, K. (1996) Cellular pH gradient in tumor versus normal tissue: potential exploitation for the treatment of cancer. Cancer Research, 56, 1194 1198.
  • 8
    Gopalakrishnan, T.V. & Anderson, W.F. (1979) Mouse erythroleukemia cells. Methods in Enzymology, Vol. 58. Cell Culture (ed. by W. B. Jakoby and I. H. Pastan), p. 506. Academic Press, New York.
  • 9
    Jandl, J.H. (1987a) Blood: Textbook of Hematology, pp. 1–21. Little, Brown and Company, Boston.
  • 10
    Jandl, J.H. (1987b) Blood: Textbook of Hematology, pp. 49–107. Little, Brown and Company, Boston.
  • 11
    Loken, M.R., Shah, V.O., Dattilio, K.L., Civin, C.I. (1987) Flow cytometric analysis of human bone marrow. I. Normal erythroid development. Blood, 69, 255 263.
  • 12
    Lord, B.I. (1990) The architecture of bone marrow cell populations. International Journal of Cell Cloning, 8, 317 331.
  • 13
    Martin, G.R. & Jain, R.K. (1994) Noninvasive measurement of interstitial pH profiles in normal and neoplastic tissue using fluorescence ratio imaging microscopy. Cancer Research, 54, 5670 5674.
  • 14
    Mayani, H., Dragowska, W., Lansdorp, P.M. (1993) Characterization of functionally distinct subpopulations of CD34+ cord blood cells in serum-free long-term cultures supplemented with hematopoietic cytokines. Blood, 82, 2664 2672.
  • 15
    McAdams, T.A., Miller, W.M., Papoutsakis, E.T. (1997) Variations in culture pH affect the cloning efficiency and differentiation of progenitor cells in ex vivo haemopoiesis. British Journal of Haematology, 97, 889 895.
  • 16
    Okumura, N., Tsuji, K., Nakahata, T. (1992) Changes in cell surface antigen expressions during proliferation and differentiation of human erythroid progenitors. Blood, 80, 642 650.
  • 17
    Ozaki, Y. (1992) Effect on pH on blood cells. Japanese Journal of Clinical Medicine, 50, 2112 2118.
  • 18
    Rich, I.N. (1988) The macrophage as a production site for hemato-poietic regulator molecules: sensing and responding to normal and pathophysiological signals. Anticancer Research, 8, 1015 1040.
  • 19
    Sambrook, J., Fritsch, E.F., Maniatis, T. (1997) Molecular Cloning: a Laboratory Manual. Cold Spring Harbor Laboratory Press, New York.
  • 20
    Sandstrom, C.E., Bender, J.G., Papoutsakis, E.T., Miller, W.M. (1995) Effects of CD34+ cell selection and perfusion on ex vivo expansion of peripheral blood mononuclear cells. Blood, 86, 958 970.
  • 21
    Segal, G.M., Eschbach, J.W., Egrie, J.C., Stueve, T., Adamson, J.W. (1988) The anemia of end-stage renal disease: hematopoietic progenitor cell response. Kidney International, 33, 983 988.
  • 22
    Sieff, C., Bicknell, D., Caine, G., Robinson, J., Lam, G., Greaves, M.F. (1982) Changes in cell surface antigen expression during hemopoietic differentiation. Blood, 60, 703 713.
  • 23
    Tavassoli, M. & Yoffey, J.M. (1983) Bone Marrow Structure and Function, pp. 47–64. Alan R. Liss, New York.
  • 24
    Telen, M.J. (1990) Expression of human erythrocyte membrane protein antigens during erythroid development. Erythroid Cells (ed. by J. R. Harris), p. 27. Plenum Press, New York.
  • 25
    Weiss, L. (1976) The haematopoietic microenvironment of the bone marrow: an ultrastructural study of the stroma in rats. Anatomical Record, 186, 161 184.
  • 26
    Wickramasinghe, S.N. (1975) Erythropoiesis. Human Bone Marrow, p. 162. Blackwell Scientific Publications, Oxford.
  • 27
    Yang, X. (1992) Erythrocyte membrane band 3 protein and HCO3/Cl exchange function in cor pulmonale patients. Chung Hua I Hsueh Tsa Chih (Taipei), 72, 277 279.
  • 28
    Yang, X., Mao, B., Qian, G. (1995) A study on erythrocyte membrane band 3 protein anion transport function in chronic respiratory failure patients. Chung Hua Nei Ko Tsa Chih, 34, 531 535.