Acta haemat. 56: 107-115 (1976)
Colony Formation by Canine Hemopoietic Cells in vitro Inhibition by Polymorphonuclear Leukocytes1
P. K ovacs, C. B ruch and T. M. F liedner Department of Clinical Physiology University of Ulm, Ulm
Key Words. Bone marrow culture • Canine leukocytes • Cell culture • Colony formation • Polymorphonuclear leukocytes Abstract. In soft agar cultures of canine blood leukocytes, an inhibition of col ony formation was observed relative to the size of the inoculum. Analysis of the cellular composition of the inoculum suggested that this inhibition was associated with the number of polymorphonuclear leukocytes present. Removal of phagocytic cells by the iron ingestion method or selective destruction of granulocytes by freezing in the presence of dimethyl sulfoxide eliminated the inhibitory action on colony formation. In mixed cultures of canine bone marrow and autologous blood leukocytes, a similar inhibition of colony formation was observed. The results pre sented indicate that polymorphonuclear leukocytes, if present in a concentration exceeding 2.5X10*/ml of inoculum, inhibit in vitro granulocytic/monocytic colony formation.
In vitro growth of granulocytic and/or macrophage colonies in soft agar cultures [1, 10] is influenced, in part, by the interaction of cells being plated. Such interactions may enhance or inhibit colony formation in cultures of human bone marrow cells [7]. It has also been reported that macrophages and granulocytes inhibit colony formation by mouse and rat hemopoietic stem cells [8, 9, 11]. Our present experiments prov ide evidence that colony formation by canine peripheral blood leuko cytes and bone marrow cells is inhibited when excessive number of poly morphonuclear leukocytes (PMN) are present in the culture.
Downloaded by: East Carolina University - Laupus Library 150.216.68.200 - 1/16/2019 7:28:17 AM
1 Research work supported by the European Atomic Energy Community (Euratom), the Deutsche Forschungsgemeinschaft (Sonderforschungsbereich 112) and executed while Dr. K ovacs was a Visiting Fellow of the International Atomic Energy Agency.
108
K ovacs/B ruch /F liedner
Materials and Methods Beagles of both sexes were used as bone marrow cell and blood leukocyte do nors. Bone marrow was aspirated from the iliac crest and anticoagulated with EDTA. After sedimentation with dextran (molec. wt. approximately 250,000) the nucleated-cell-rich plasma was removed and centrifuged. The cells were washed and the mononuclear and polymorphonuclear cells counted differentially in a Neubauer chamber at 500 X magnification. Venous blood was anticoagulated with heparin and the leukocytes were isolated by dextran sedimentation, then washed and counted differentially as for the bone marrow cells. In some of the experi ments, leukocytes were collected from peripheral blood by leukapheresis using an NCI-IBM experimental blood cell separator, as described elsewhere [10]. Cultures were set up in MEM for Spinner cultures (Gibco), supplemented with vitamins, sodium pyruvate, amino acids and 20%> horse serum (Flow Labs); then agar was added to a final concentration of 0.3°/o, as described in detail elsewhere [10]. Various numbers of cells were plated in triplicate or quadruplicate in 35-mm plastic Petri dishes (1 ml medium in each). As PMN are generally considered un able to proliferate, the size of the inoculum was given in terms of mononuclear cells, though the number of accompanying PMN was also determined and record ed. Serum from dogs killed 10 days after 1,200 rad whole body X-irradiation served as the source of colony stimulating activity and was added to each dish be fore plating in amounts of 0.3 and 0.2 ml for bone marrow and blood leukocyte cultures, respectively. These quantities of serum supported maximal growth of col onies, i.e., increased amounts failed to increase the number of colonies. Cultures were incubated in desiccators containing 3% COa in humidified air, according to F irket [5], for 7 days at 37 °C. Colonies were defined as aggregates of at least 15 cells and were counted under a dissecting microscope at 30 X magnification. For the removal of phagocytes, cells were suspended in MEM and colloidal iron carbonyl was added. After incubation for 30 min at 37 °C, cells that had in gested iron were removed by means of a magnet outside the tube. For the selective destruction of granulocytes, cells were suspended in MEM containing 10°/o dime thyl sulfoxide, then frozen and thawed, as described elsewhere [2],
The number of colonies formed in cultures of bone marrow cells was found to be a linear function of the number of cells inoculated (mono nuclear) within broad limits (fig. la). In bone marrow cultures, no sign of inhibition of colony growth could be observed, even in plates containing as many as 1,000 colonies. On the other hand, in cultures of blood leu kocytes, an increase in inoculum size above certain limits failed to in crease the number of colonies. In fact, if the size of the inoculum was too large, it resulted in a complete lack of colony formation. Such an
Downloaded by: East Carolina University - Laupus Library 150.216.68.200 - 1/16/2019 7:28:17 AM
Result
Inhibition of Colony Formation by Leukocytes
109
Fig. 1. Number of colonies per dish plotted against the number of mononu clear cells (MNC) inoculated. Vertical bars indicate;}; 1 SE. a In cultures of bone marrow cells, b In cultures of blood leukocytes.
Downloaded by: East Carolina University - Laupus Library 150.216.68.200 - 1/16/2019 7:28:17 AM
experiment is presented in figure lb, where the inoculum response curve seemed to remain linear up to 1X106 mononuclear cells per dish (form ing about 60 colonies), but no colony formation was observed in plates containing 2X10® or 4X106 mononuclear cells. The ratio of PMN to mononuclear cells was 2.4 in this experiment. In terms of the number of PMN accompanying the mononuclear cells, the inoculum response curve was linear until 2.4X106 PMN per inoculum were present; no colony formation occurred when 4.8 X106 and 9.6 X 10* PMN were present. As the above phenomenon might be explicable in terms of an inhibi tion by PMN present in high numbers, attempts were made to correlate colony formation with varying numbers of PMN in cultures containing a constant number (0.5 X106) of mononuclear blood leukocytes. The re sults from 85 cultures are presented in figure 2. In the 75 cultures show ing colony formation, the number of accompanying PMN per dish never exceeded 3.5X10° and only in five cases was higher than 2.5X10®. In all of the remaining ten cultures showing no colony formation, the ino culum contained more than 2.5 X 10® and in seven of these cultures even more than 3.5 X 10® PMN. Figure 3 shows the results of 18 inoculum response experiments with blood leukocytes. As the plating efficiency was not uniform for the va rious experiments, the number of colonies was expressed relative to those formed with 0.5X10® mononuclear leukocytes in each experiment. The
110
K ovacs/B ruch /F liedner 50
40
30
20
••
• •
f
0.5
1
•
2
3 45
10
x106 PMN/dish
Fig. 2. Number of colonies per dish plotted against the number of PMN per dish in cultures set up from various blood leukocyte samples. In each culture 0.5X10* MNC were plated accompanied by varying numbers of PMN as indicated on the abscissa.
xIO6 MNC/dish
Downloaded by: East Carolina University - Laupus Library 150.216.68.200 - 1/16/2019 7:28:17 AM
Fig. 3. Relative colony number in cultures containing less ( • = mean+SE; number in parenthesis indicates the number of separate experiments) and more (x = results of separate experiments) than 3XIO6 PMN plotted against the number of MNC plated. Colony count in cultures inoculated with 0.5XIO6 MNC was taken as the unit in each experiment.
Inhibition of Colony Formation by Leukocytes
x106 MNC/dish 0 0.1 0.2
0.4
0.6
0.3
1.0
111
1.2
colony counts of cultures which contained more than 3X10« PMN are indicated by symbols different than those showing the number of colo nies in cultures with less than 3X106 PMN. In most cultures with more than 3X10« PMN, no colonies grew; in the few cultures that showed colony formation, the number of colonies was always below the linear regression derived from the colony counts of cultures with less than 3X10« PMN. Four million mononuclear blood leukocytes in a culture dish usually meant high numbers of PMN and colonies failed to form. Growth of nu merous colonies was observed, however, when the same number of mononuclear leukocytes from the same cell suspension was plated after removal of the phagocytic cells by iron carbonyl or after selective inacti vation of granulocytes by freezing and thawing (table I). To see whether colony formation by bone marrow cells was similarly inhibited by the presence of high numbers of PMN, a constant number of dog bone marrow mononuclear cells (105) was plated together with increasing numbers of autologous blood leukocytes. The small number of colonies formed in cultures inoculated with an identical number of blood leukocytes without any bone marrow cells, was subtracted from the
Downloaded by: East Carolina University - Laupus Library 150.216.68.200 - 1/16/2019 7:28:17 AM
Fig. 4. Number of colonies of bone marrow origin in mixed cultures of a con stant inoculum of bone marrow containing 10s MNC and of varying numbers of blood leukocytes as indicated on the abscissa. Mean+SE.
112
K ovàcs/B ruch /F liedner
Table I. Effect of the removal of phagocytes or of freezing and thawing on the number of colonies in cultures of blood leukocytes Treatment of leukocytes before plating
Million cells plated per dish mono polymorpho nuclear nuclear
Number of dishes
Colonies per dish (mean ± SE)
No treatment Removal of phagocytes Freezing and thawing
4.0 4.0 4.0
3 3 2
0 126± 10 362 ± 12
7.1 0.3 destructed
colony count of the corresponding mixed culture in order to determine the net number of colonies of bone marrow origin. The number of bone marrow colonies was not reduced by the presence of less than 3X10° PMN in a culture dish. However, colony formation was substantially diminished in cultures containing 4X10° PMN; colonies failed to grow with more than 4X108 PMN per dish (fig. 4).
As peripheral blood leukocytes of normal dogs contain only about 15 in vitro colony-forming units (CFU) per 106 cells [10], culture inocula with one to three million leukocytes produced, in some cases, too few colonies to allow reliable quantification of CFU in peripheral blood. In order to improve the accuracy of the estimation, it would have been necessary to increase the number of colonies counted. A simple solution to this problem could have been to increase the inoculum size; in such a way, the number of colonies per plate in cultures of canine bone marrow could have been increased far above those easy to count. An increase in the inoculum size above certain limits in cultures of canine peripheral blood leukocytes, however, resulted in a complete inhibition of colony formation. This phenomenon might be explained by hypothesizing the presence of an inhibitory cell among blood leukocytes, the effect of which is manifest when the number of such cells in the culture exceeds a critical value. By increasing the inoculum size, this critical value may be reached, thereby resulting in inhibition of colony formation. As for the identity of this postulated inhibitory cell, PMN have been reported to inhibit colony formation in other species [3, 7-9, 11].
Downloaded by: East Carolina University - Laupus Library 150.216.68.200 - 1/16/2019 7:28:17 AM
Discussion
113
Therefore, the question arises, whether the lack of colony formation in some canine leukocyte cultures might be related to the number of PMN present in the inoculum. In cultures inoculated with a constant number (0.5X10°) of mononuclear leukocytes, some colonies always grew when the number of accompanying PMN did not exceed 2.5X10°, while no colony formation occurred in cultures containing more than 3.5X10° PMN. This suggests that the absence of colony formation may indeed be associated with the presence of big numbers of PMN in a culture and that the critical number for inhibition lies between 2.5 and 3.5X10° PMN per dish (for a 1.2-ml culture). Data presented in figure 3 show that, within each experiment, the plating efficiency (number of colonies per mononuclear cell plated) of leukocyte cultures was independent of the inoculum size in plates containing less than 3X10° PMN; it was greatly diminished in cultures with more than 3X10“ PMN. This sup ports the view that the lack of colony formation observed with higher inoculum size in cultures of peripheral blood leukocytes may be related to the high number of PMN present. In plates containing 4X10“ mononuclear cells with 7X10° accompa nying PMN, no colonies grew at all. However, numerous colonies were formed from the same number of mononuclear leukocytes from the same leukocyte suspension when the number of accompanying PMN was substantially reduced, either by removal of phagocytes or by selective de struction of PMN by freezing before plating. This observation provides direct evidence for the assumption that the presence of high numbers of PMN in a culture inhibits colony formation. The lack of such a phenomenon in cultures of canine bone marrow cells may be explained as follows. The frequency of CFU in canine bone marrow cells is about 100 times higher than in canine peripheral blood leukocytes. For this reason, increasing the inoculum size results in too high a number of colonies in a bone marrow culture to be counted; this occurs with an inoculum cell count far below that containing a critical number of PMN. Data presented in figure 4, however, show that colony formation by canine bone marrow cells is definitely susceptible to the in hibition by the presence of high numbers of added blood PMN. The above inhibitory effect of PMN may be explained (a) by assum ing that PMN consume a component in the medium necessary for colo ny formation, or (b) by assuming that canine PMN contain and/or pro duce a factor that inhibits colony formation. Other experiments by our group have shown that a medium conditioned by high numbers of can
Downloaded by: East Carolina University - Laupus Library 150.216.68.200 - 1/16/2019 7:28:17 AM
Inhibition of Colony Formation by Leukocytes
114
K ovacs/ Bruch /F liedner
ine leukocytes inhibits colony formation [3], and C hervenick and LoB uglio [4] have observed a similar effect with a medium conditioned by human granulocytes; these results support the second assumption. Our observations reveal a possible pitfall in the estimation of CFU in canine peripheral blood. On the basis of our results, cultures showing no colonies cannot be used as evidence for the lack of CFU in the leuko cytes plated, without stating that the concentration of PMN in the cul ture did not exceed the threshold value for inhibition. They also raise questions about the nature of the factor produced by polymorphonuclear cells that is capable of inhibiting proliferation of granulocytic precursor cells and about its physiological role, if any, in regulating granulocytic production by inhibitory mechanisms, as has been recently postulated [ 6 ].
Acknowledgements. The skillful technical assistance of Miss E. R über is grate fully acknowledged. Dr. W. Ross was kind enough to help in the final stages of preparing the manuscript.
References 1 Bradley, T. R. and M etcalf, D.: The growth of mouse bone marrow cells in vitro. Aust. J. exp. Biol. med. Sci. 44: 287-300 (1966). 2 Bruch , C.; H erbst, E. W.; C alvo, W., et a!.: Freezing of blood leukocytes foi transplantation into lethally irradiated dogs; in W einer , O ldham et Schwarzenberg La cryoconservation des cellules normales et neoplastiques, pp. 51-62 (Inserm librairie Ophrys, Paris 1973). 3 Bruch , C.: Studies on the inhibitory effect of granulocytes on human granulocytopoiesis in agar cultures. Exp. Hemat. (submitted). 4 C hervenick , P. A. and L o -B uglio , A. F.: Human blood monocytes. Stimula tors of granulocyte and mononuclear colony formation in vitro. Science 178:
Downloaded by: East Carolina University - Laupus Library 150.216.68.200 - 1/16/2019 7:28:17 AM
164-166 (1972).
5 F irket , H.: A very simple trick to produce controlled C 0 2 concentrations in the gas phase overlying cell cultures. Experientia 25: 671-672 (1969). 6 F uedner T. M.: Kinetik und Regulationsmechanismen des Granulozytenumsatzes. Schweiz, med. Wschr. 104: 98-107 (1974). 7 H askill, I. S.; M c K night , R. D., and G albraith, P. R.: Cell-cell interaction in vitro. Studied by density separation of colony-forming, stimulating, and in hibiting cells from human bone marrow. Blood 40: 394-399 (1972). 8 H eit , W.; K ern , P.; K ubanek, B., and H eimpei., H.: Some factors influencing granulocytic colony formation in vitro by human white blood cells. Blood 44: 511-515 (1974).
Inhibition of Colony Formation by Leukocytes
115
9 Ichikawa, Y.; P luznik , D. H., and Sachs, L.: Feedback inhibition of the de velopment of macrophage and granulocyte colonies. I. Inhibition by macro phages. Proc. natn. Acad. Sci. USA 58: 1480-1486 (1967). 10 KovÂcs, P.; Bruch , C.; H erbst , E., and F liedner , T. M.: In vitro colony forming units (CFUc) collected from dog blood. Their concentration before, during, and after single and repeated leukaphereses. Blood (submitted). 11 P aran, M.; I chikawa, Y., and Sachs, L.: Feedback inhibition of the develop ment of macrophage and granulocyte colonies. II. Inhibition by granulocytes. Proc. natn. Acad. Sci. USA 62: 81-87 (1968). 12 P luznik , D. H. and Sachs, L.: The cloning of normal ‘mast’ cells in tissue cul ture. J. cell. comp. Physiol. 66: 319-324 (1965).
Downloaded by: East Carolina University - Laupus Library 150.216.68.200 - 1/16/2019 7:28:17 AM
Prof. Dr. T. M. F liedner , Abteilung für Klinische Physiologie der Universität Ulm, Oberer Eselsberg, M 24, 309, D-7900 Ulm!Donau (FRG)
Colony Formation by Canine Hemopoietic Cells in vitro Inhibition by Polymorphonuclear Leukocytes1
P. K ovacs, C. B ruch and T. M. F liedner Department of Clinical Physiology University of Ulm, Ulm
Key Words. Bone marrow culture • Canine leukocytes • Cell culture • Colony formation • Polymorphonuclear leukocytes Abstract. In soft agar cultures of canine blood leukocytes, an inhibition of col ony formation was observed relative to the size of the inoculum. Analysis of the cellular composition of the inoculum suggested that this inhibition was associated with the number of polymorphonuclear leukocytes present. Removal of phagocytic cells by the iron ingestion method or selective destruction of granulocytes by freezing in the presence of dimethyl sulfoxide eliminated the inhibitory action on colony formation. In mixed cultures of canine bone marrow and autologous blood leukocytes, a similar inhibition of colony formation was observed. The results pre sented indicate that polymorphonuclear leukocytes, if present in a concentration exceeding 2.5X10*/ml of inoculum, inhibit in vitro granulocytic/monocytic colony formation.
In vitro growth of granulocytic and/or macrophage colonies in soft agar cultures [1, 10] is influenced, in part, by the interaction of cells being plated. Such interactions may enhance or inhibit colony formation in cultures of human bone marrow cells [7]. It has also been reported that macrophages and granulocytes inhibit colony formation by mouse and rat hemopoietic stem cells [8, 9, 11]. Our present experiments prov ide evidence that colony formation by canine peripheral blood leuko cytes and bone marrow cells is inhibited when excessive number of poly morphonuclear leukocytes (PMN) are present in the culture.
Downloaded by: East Carolina University - Laupus Library 150.216.68.200 - 1/16/2019 7:28:17 AM
1 Research work supported by the European Atomic Energy Community (Euratom), the Deutsche Forschungsgemeinschaft (Sonderforschungsbereich 112) and executed while Dr. K ovacs was a Visiting Fellow of the International Atomic Energy Agency.
108
K ovacs/B ruch /F liedner
Materials and Methods Beagles of both sexes were used as bone marrow cell and blood leukocyte do nors. Bone marrow was aspirated from the iliac crest and anticoagulated with EDTA. After sedimentation with dextran (molec. wt. approximately 250,000) the nucleated-cell-rich plasma was removed and centrifuged. The cells were washed and the mononuclear and polymorphonuclear cells counted differentially in a Neubauer chamber at 500 X magnification. Venous blood was anticoagulated with heparin and the leukocytes were isolated by dextran sedimentation, then washed and counted differentially as for the bone marrow cells. In some of the experi ments, leukocytes were collected from peripheral blood by leukapheresis using an NCI-IBM experimental blood cell separator, as described elsewhere [10]. Cultures were set up in MEM for Spinner cultures (Gibco), supplemented with vitamins, sodium pyruvate, amino acids and 20%> horse serum (Flow Labs); then agar was added to a final concentration of 0.3°/o, as described in detail elsewhere [10]. Various numbers of cells were plated in triplicate or quadruplicate in 35-mm plastic Petri dishes (1 ml medium in each). As PMN are generally considered un able to proliferate, the size of the inoculum was given in terms of mononuclear cells, though the number of accompanying PMN was also determined and record ed. Serum from dogs killed 10 days after 1,200 rad whole body X-irradiation served as the source of colony stimulating activity and was added to each dish be fore plating in amounts of 0.3 and 0.2 ml for bone marrow and blood leukocyte cultures, respectively. These quantities of serum supported maximal growth of col onies, i.e., increased amounts failed to increase the number of colonies. Cultures were incubated in desiccators containing 3% COa in humidified air, according to F irket [5], for 7 days at 37 °C. Colonies were defined as aggregates of at least 15 cells and were counted under a dissecting microscope at 30 X magnification. For the removal of phagocytes, cells were suspended in MEM and colloidal iron carbonyl was added. After incubation for 30 min at 37 °C, cells that had in gested iron were removed by means of a magnet outside the tube. For the selective destruction of granulocytes, cells were suspended in MEM containing 10°/o dime thyl sulfoxide, then frozen and thawed, as described elsewhere [2],
The number of colonies formed in cultures of bone marrow cells was found to be a linear function of the number of cells inoculated (mono nuclear) within broad limits (fig. la). In bone marrow cultures, no sign of inhibition of colony growth could be observed, even in plates containing as many as 1,000 colonies. On the other hand, in cultures of blood leu kocytes, an increase in inoculum size above certain limits failed to in crease the number of colonies. In fact, if the size of the inoculum was too large, it resulted in a complete lack of colony formation. Such an
Downloaded by: East Carolina University - Laupus Library 150.216.68.200 - 1/16/2019 7:28:17 AM
Result
Inhibition of Colony Formation by Leukocytes
109
Fig. 1. Number of colonies per dish plotted against the number of mononu clear cells (MNC) inoculated. Vertical bars indicate;}; 1 SE. a In cultures of bone marrow cells, b In cultures of blood leukocytes.
Downloaded by: East Carolina University - Laupus Library 150.216.68.200 - 1/16/2019 7:28:17 AM
experiment is presented in figure lb, where the inoculum response curve seemed to remain linear up to 1X106 mononuclear cells per dish (form ing about 60 colonies), but no colony formation was observed in plates containing 2X10® or 4X106 mononuclear cells. The ratio of PMN to mononuclear cells was 2.4 in this experiment. In terms of the number of PMN accompanying the mononuclear cells, the inoculum response curve was linear until 2.4X106 PMN per inoculum were present; no colony formation occurred when 4.8 X106 and 9.6 X 10* PMN were present. As the above phenomenon might be explicable in terms of an inhibi tion by PMN present in high numbers, attempts were made to correlate colony formation with varying numbers of PMN in cultures containing a constant number (0.5 X106) of mononuclear blood leukocytes. The re sults from 85 cultures are presented in figure 2. In the 75 cultures show ing colony formation, the number of accompanying PMN per dish never exceeded 3.5X10° and only in five cases was higher than 2.5X10®. In all of the remaining ten cultures showing no colony formation, the ino culum contained more than 2.5 X 10® and in seven of these cultures even more than 3.5 X 10® PMN. Figure 3 shows the results of 18 inoculum response experiments with blood leukocytes. As the plating efficiency was not uniform for the va rious experiments, the number of colonies was expressed relative to those formed with 0.5X10® mononuclear leukocytes in each experiment. The
110
K ovacs/B ruch /F liedner 50
40
30
20
••
• •
f
0.5
1
•
2
3 45
10
x106 PMN/dish
Fig. 2. Number of colonies per dish plotted against the number of PMN per dish in cultures set up from various blood leukocyte samples. In each culture 0.5X10* MNC were plated accompanied by varying numbers of PMN as indicated on the abscissa.
xIO6 MNC/dish
Downloaded by: East Carolina University - Laupus Library 150.216.68.200 - 1/16/2019 7:28:17 AM
Fig. 3. Relative colony number in cultures containing less ( • = mean+SE; number in parenthesis indicates the number of separate experiments) and more (x = results of separate experiments) than 3XIO6 PMN plotted against the number of MNC plated. Colony count in cultures inoculated with 0.5XIO6 MNC was taken as the unit in each experiment.
Inhibition of Colony Formation by Leukocytes
x106 MNC/dish 0 0.1 0.2
0.4
0.6
0.3
1.0
111
1.2
colony counts of cultures which contained more than 3X10« PMN are indicated by symbols different than those showing the number of colo nies in cultures with less than 3X106 PMN. In most cultures with more than 3X10« PMN, no colonies grew; in the few cultures that showed colony formation, the number of colonies was always below the linear regression derived from the colony counts of cultures with less than 3X10« PMN. Four million mononuclear blood leukocytes in a culture dish usually meant high numbers of PMN and colonies failed to form. Growth of nu merous colonies was observed, however, when the same number of mononuclear leukocytes from the same cell suspension was plated after removal of the phagocytic cells by iron carbonyl or after selective inacti vation of granulocytes by freezing and thawing (table I). To see whether colony formation by bone marrow cells was similarly inhibited by the presence of high numbers of PMN, a constant number of dog bone marrow mononuclear cells (105) was plated together with increasing numbers of autologous blood leukocytes. The small number of colonies formed in cultures inoculated with an identical number of blood leukocytes without any bone marrow cells, was subtracted from the
Downloaded by: East Carolina University - Laupus Library 150.216.68.200 - 1/16/2019 7:28:17 AM
Fig. 4. Number of colonies of bone marrow origin in mixed cultures of a con stant inoculum of bone marrow containing 10s MNC and of varying numbers of blood leukocytes as indicated on the abscissa. Mean+SE.
112
K ovàcs/B ruch /F liedner
Table I. Effect of the removal of phagocytes or of freezing and thawing on the number of colonies in cultures of blood leukocytes Treatment of leukocytes before plating
Million cells plated per dish mono polymorpho nuclear nuclear
Number of dishes
Colonies per dish (mean ± SE)
No treatment Removal of phagocytes Freezing and thawing
4.0 4.0 4.0
3 3 2
0 126± 10 362 ± 12
7.1 0.3 destructed
colony count of the corresponding mixed culture in order to determine the net number of colonies of bone marrow origin. The number of bone marrow colonies was not reduced by the presence of less than 3X10° PMN in a culture dish. However, colony formation was substantially diminished in cultures containing 4X10° PMN; colonies failed to grow with more than 4X108 PMN per dish (fig. 4).
As peripheral blood leukocytes of normal dogs contain only about 15 in vitro colony-forming units (CFU) per 106 cells [10], culture inocula with one to three million leukocytes produced, in some cases, too few colonies to allow reliable quantification of CFU in peripheral blood. In order to improve the accuracy of the estimation, it would have been necessary to increase the number of colonies counted. A simple solution to this problem could have been to increase the inoculum size; in such a way, the number of colonies per plate in cultures of canine bone marrow could have been increased far above those easy to count. An increase in the inoculum size above certain limits in cultures of canine peripheral blood leukocytes, however, resulted in a complete inhibition of colony formation. This phenomenon might be explained by hypothesizing the presence of an inhibitory cell among blood leukocytes, the effect of which is manifest when the number of such cells in the culture exceeds a critical value. By increasing the inoculum size, this critical value may be reached, thereby resulting in inhibition of colony formation. As for the identity of this postulated inhibitory cell, PMN have been reported to inhibit colony formation in other species [3, 7-9, 11].
Downloaded by: East Carolina University - Laupus Library 150.216.68.200 - 1/16/2019 7:28:17 AM
Discussion
113
Therefore, the question arises, whether the lack of colony formation in some canine leukocyte cultures might be related to the number of PMN present in the inoculum. In cultures inoculated with a constant number (0.5X10°) of mononuclear leukocytes, some colonies always grew when the number of accompanying PMN did not exceed 2.5X10°, while no colony formation occurred in cultures containing more than 3.5X10° PMN. This suggests that the absence of colony formation may indeed be associated with the presence of big numbers of PMN in a culture and that the critical number for inhibition lies between 2.5 and 3.5X10° PMN per dish (for a 1.2-ml culture). Data presented in figure 3 show that, within each experiment, the plating efficiency (number of colonies per mononuclear cell plated) of leukocyte cultures was independent of the inoculum size in plates containing less than 3X10° PMN; it was greatly diminished in cultures with more than 3X10“ PMN. This sup ports the view that the lack of colony formation observed with higher inoculum size in cultures of peripheral blood leukocytes may be related to the high number of PMN present. In plates containing 4X10“ mononuclear cells with 7X10° accompa nying PMN, no colonies grew at all. However, numerous colonies were formed from the same number of mononuclear leukocytes from the same leukocyte suspension when the number of accompanying PMN was substantially reduced, either by removal of phagocytes or by selective de struction of PMN by freezing before plating. This observation provides direct evidence for the assumption that the presence of high numbers of PMN in a culture inhibits colony formation. The lack of such a phenomenon in cultures of canine bone marrow cells may be explained as follows. The frequency of CFU in canine bone marrow cells is about 100 times higher than in canine peripheral blood leukocytes. For this reason, increasing the inoculum size results in too high a number of colonies in a bone marrow culture to be counted; this occurs with an inoculum cell count far below that containing a critical number of PMN. Data presented in figure 4, however, show that colony formation by canine bone marrow cells is definitely susceptible to the in hibition by the presence of high numbers of added blood PMN. The above inhibitory effect of PMN may be explained (a) by assum ing that PMN consume a component in the medium necessary for colo ny formation, or (b) by assuming that canine PMN contain and/or pro duce a factor that inhibits colony formation. Other experiments by our group have shown that a medium conditioned by high numbers of can
Downloaded by: East Carolina University - Laupus Library 150.216.68.200 - 1/16/2019 7:28:17 AM
Inhibition of Colony Formation by Leukocytes
114
K ovacs/ Bruch /F liedner
ine leukocytes inhibits colony formation [3], and C hervenick and LoB uglio [4] have observed a similar effect with a medium conditioned by human granulocytes; these results support the second assumption. Our observations reveal a possible pitfall in the estimation of CFU in canine peripheral blood. On the basis of our results, cultures showing no colonies cannot be used as evidence for the lack of CFU in the leuko cytes plated, without stating that the concentration of PMN in the cul ture did not exceed the threshold value for inhibition. They also raise questions about the nature of the factor produced by polymorphonuclear cells that is capable of inhibiting proliferation of granulocytic precursor cells and about its physiological role, if any, in regulating granulocytic production by inhibitory mechanisms, as has been recently postulated [ 6 ].
Acknowledgements. The skillful technical assistance of Miss E. R über is grate fully acknowledged. Dr. W. Ross was kind enough to help in the final stages of preparing the manuscript.
References 1 Bradley, T. R. and M etcalf, D.: The growth of mouse bone marrow cells in vitro. Aust. J. exp. Biol. med. Sci. 44: 287-300 (1966). 2 Bruch , C.; H erbst, E. W.; C alvo, W., et a!.: Freezing of blood leukocytes foi transplantation into lethally irradiated dogs; in W einer , O ldham et Schwarzenberg La cryoconservation des cellules normales et neoplastiques, pp. 51-62 (Inserm librairie Ophrys, Paris 1973). 3 Bruch , C.: Studies on the inhibitory effect of granulocytes on human granulocytopoiesis in agar cultures. Exp. Hemat. (submitted). 4 C hervenick , P. A. and L o -B uglio , A. F.: Human blood monocytes. Stimula tors of granulocyte and mononuclear colony formation in vitro. Science 178:
Downloaded by: East Carolina University - Laupus Library 150.216.68.200 - 1/16/2019 7:28:17 AM
164-166 (1972).
5 F irket , H.: A very simple trick to produce controlled C 0 2 concentrations in the gas phase overlying cell cultures. Experientia 25: 671-672 (1969). 6 F uedner T. M.: Kinetik und Regulationsmechanismen des Granulozytenumsatzes. Schweiz, med. Wschr. 104: 98-107 (1974). 7 H askill, I. S.; M c K night , R. D., and G albraith, P. R.: Cell-cell interaction in vitro. Studied by density separation of colony-forming, stimulating, and in hibiting cells from human bone marrow. Blood 40: 394-399 (1972). 8 H eit , W.; K ern , P.; K ubanek, B., and H eimpei., H.: Some factors influencing granulocytic colony formation in vitro by human white blood cells. Blood 44: 511-515 (1974).
Inhibition of Colony Formation by Leukocytes
115
9 Ichikawa, Y.; P luznik , D. H., and Sachs, L.: Feedback inhibition of the de velopment of macrophage and granulocyte colonies. I. Inhibition by macro phages. Proc. natn. Acad. Sci. USA 58: 1480-1486 (1967). 10 KovÂcs, P.; Bruch , C.; H erbst , E., and F liedner , T. M.: In vitro colony forming units (CFUc) collected from dog blood. Their concentration before, during, and after single and repeated leukaphereses. Blood (submitted). 11 P aran, M.; I chikawa, Y., and Sachs, L.: Feedback inhibition of the develop ment of macrophage and granulocyte colonies. II. Inhibition by granulocytes. Proc. natn. Acad. Sci. USA 62: 81-87 (1968). 12 P luznik , D. H. and Sachs, L.: The cloning of normal ‘mast’ cells in tissue cul ture. J. cell. comp. Physiol. 66: 319-324 (1965).
Downloaded by: East Carolina University - Laupus Library 150.216.68.200 - 1/16/2019 7:28:17 AM
Prof. Dr. T. M. F liedner , Abteilung für Klinische Physiologie der Universität Ulm, Oberer Eselsberg, M 24, 309, D-7900 Ulm!Donau (FRG)
