Shiga Toxin 2 Induces
Macrophage-Granulocyte Colonies from
Human Bone Marrow and Cord Blood Stem
Cells
Shin Chiyoda, Tae Takeda and Yosuke Aoki
Infect. Immun. 2002, 70(9):5316. DOI:
10.1128/IAI.70.9.5316-5318.2002.
These include:
REFERENCES
CONTENT ALERTS
This article cites 21 articles, 1 of which can be accessed free at:
http://iai.asm.org/content/70/9/5316#ref-list-1
Receive: RSS Feeds, eTOCs, free email alerts (when new
articles cite this article), more»
Information about commercial reprint orders: http://journals.asm.org/site/misc/reprints.xhtml
To subscribe to to another ASM Journal go to: http://journals.asm.org/site/subscriptions/
Downloaded from http://iai.asm.org/ on June 11, 2014 by guest
Updated information and services can be found at:
http://iai.asm.org/content/70/9/5316
INFECTION AND IMMUNITY, Sept. 2002, p. 5316–5318
0019-9567/02/$04.00⫹0 DOI: 10.1128/IAI.70.9.5316–5318.2002
Copyright © 2002, American Society for Microbiology. All Rights Reserved.
Vol. 70, No. 9
Shiga Toxin 2 Induces Macrophage-Granulocyte
Colonies from Human Bone Marrow and
Cord Blood Stem Cells
Shin Chiyoda,1 Tae Takeda,2 and Yosuke Aoki3*
Nagasaki Red Cross Blood Center, Nagasaki,1 and Department of Infectious Diseases Research, National Children’s
Medical Research Center,2 and Department of Food and Health Sciences, Faculty of
Human Life Sciences, Jissen Women’s University,3 Tokyo, Japan
Received 23 April 2002/Returned for modification 17 May 2002/Accepted 27 May 2002
into ␣-medium (Flow Lab) with a Pasteur pipette and made
into single-cell suspensions by repeated pipetting. Umbilical
cord blood was obtained at the time of delivery after uncomplicated full-term pregnancy. Buffy coat cells obtained from
umbilical cord blood were suspended in ␣-medium. Informed
consent was obtained from all subjects.
Methylcellulose culture was carried out according to the
method of Iscove et al. (13). One milliliter of culture mixture
containing 2 ⫻ 105 nucleated bone marrow cells or 2 ⫻ 105 to
4.5 ⫻ 105 cord blood mononuclear cells, ␣-medium, 1.35%
methylcellulose, 30% fetal bovine serum, 1% deionized bovine
serum albumin, 10⫺4 M 2-mercaptoethanol, growth factors,
and various amounts of Stx was plated on each 35-mm culture
dish. Cultures were incubated at 37°C in a humidified 4.6%
CO2 air incubator for 16 days. All cultures were conducted in
triplicate.
Distinct groups of cells containing 40 cells or more were
counted as colonies. Individual colonies were stained using the
May-Giemsa method to identify cell types within each colony.
Granulocyte-macrophage (GM) colonies were defined as those
containing mainly granulocytes, and macrophage (M⌽) colonies contained M⌽ almost exclusively.
HL-60 cells (established from human acute myelogenous
leukemia cells) and Jurkat cells (established from human acute
lymphocytic leukemia cells) were cultured in RPMI 1640 medium (104 cells/ml) in the presence or absence of Stx1 or Stx2
for 72 h, and the proliferation status of each cell line was
determined by the Celltiter 96 aqueous nonradioactive cell
proliferation assay method (Promega Co. Ltd.). Experiments
were each conducted at least three times, with similar results.
Statistical analysis was performed with Student’s t test.
As shown in Table 1, adding Stx2 in culture resulted in the
appearance of both GM and M⌽ colonies, with predominantly
M⌽ colonies. In contrast, adding granulocyte colony-stimulating factor (G-CSF), interleukin-1␤ (IL-1␤), or IL-3 with stem
cell factor (SCF) induced GM colonies more predominantly
than M⌽ colonies. Adding G-CSF to the bone marrow cell
culture containing Stx2 markedly increased the number of GM
colonies above that in the culture containing only Stx2 or
G-CSF alone (P ⬍ 0.001). Almost the same synergistic effect
In 10 to 14% of patients infected with Escherichia coli O157:
H7, hemolytic uremic syndrome (HUS) develops 1 week after
the onset of diarrhea (8, 9, 14, 17). HUS is an acute, mainly
pediatric illness characterized by acute hemolytic anemia with
typical fragmented red blood cells, thrombocytopenia, and
neuropathy. Highly elevated granulocytosis has been documented in many cases (15), including four cases seen by Gasser
et al. (10), who first reported HUS in 1955. The occurrence of
granulocytosis is not surprising, because many patients with
bacterial infections such as those from E. coli, Shigella, Salmonella, or pneumococci usually show a rising granulocyte count
in the peripheral blood. The level of granulocytosis in patients
with E. coli O157:H7 infection, however, is extraordinarily high
and correlates well with the severity of the illness. Among E.
coli O157:H7 virulence factors, Shiga toxin (Stx) is a major
pathogenic factor in HUS development. Using a mouse model,
we demonstrated that injecting Stx2 into the peritoneal cavity
of mice caused marked (sevenfold) granulocytosis in the peripheral blood. Elevated granulopoiesis and suppressed erythropoiesis have been observed in the bone marrow of mice
injected with Stx2 (5). In order to clarify the mechanism behind
granulocytosis in HUS, we studied the direct effect of Stx’s on
bone marrow stem cell differentiation by using human bone
marrow cell culture. The effect on stem cells in the cord blood
was also examined.
Stx2 was purified from a recombinant strain carrying the
Stx2 gene by a method described elsewhere (21). Stx1 was
purified from E. coli O157:H7 by the method of Noda et al.
(16). The amount of lipopolysaccharide in the Stx preparations, which was determined by using a Limulus amebocyte
lysate (Pregel-M; Teikokuzoki Co. Ltd., Tokyo, Japan), was
less than 2.5 pg in 1 ng of purified Stx.
Bone marrow cells were obtained by sternal puncture with a
heparinized plastic syringe from a healthy volunteer and centrifuged at 170 ⫻ g for 10 min. Buffy coat cells were aspirated
* Corresponding author. Mailing address: Faculty of Human Life
Sciences, Jissen Women’s University, Osakaue 4-1-1, Hino-city, Tokyo,
Japan 191-8510. Phone: 81-42-585-8892. Fax: 81-42-585-8892. E-mail:
[email protected].
5316
Downloaded from http://iai.asm.org/ on June 11, 2014 by guest
Addition of Shiga toxin 2 to human bone marrow or cord blood cell culture induced macrophage-granulocyte
colonies. Although Shiga toxin 2 alone induced colonies mainly composed of macrophages, it induced colonies
mainly consisting of granulocytes when combined with physiological doses of interleukin-1␤, granulocyte
colony-stimulating factor, or stem cell factor with interleukin-3.
VOL. 70, 2002
NOTES
TABLE 1. Effect of Stx’s on colony formation from human
bone marrow stem cells
M␾
No addition
Stx1 (10)
Stx1 (50)
Stx1 (100)
Stx2 (10)
Stx2 (50)
Stx2 (100)
G-CSF (100)
G-CSF (100) ⫹ Stx1(50)
Stx2 (50) ⫹ G-CSF (100)
Stx2 (50) ⫹ G-CSF (100) ⫹ Stx1 (50)
SCF (5,000) ⫹ IL-3 (500)
Stx2 (50) ⫹ SCF (5,000) ⫹ IL-3 (500)
IL-1␤ (100)
Stx2 (50) ⫹ IL-1␤ (100)
IL-1␣ (50)
Stx2 (50) ⫹ IL-1␣ (50)
0
0
0
0
0
2.3 ⫾ 2.0
3.7 ⫾ 0.6
15 ⫾ 3.0
17 ⫾ 2.0
26 ⫾ 3.5
25 ⫾ 2.5
16 ⫾ 2.5
25 ⫾ 4.2
2.3 ⫾ 0.7
15 ⫾ 1.7
1.3 ⫾ 1.0
1.3 ⫾ 1.0
0
0
0
0
7.7 ⫾ 1.5
12 ⫾ 1.5
12 ⫾ 2.5
5.0 ⫾ 2.0
5.7 ⫾ 0.6
2.0 ⫾ 1.0
1.3 ⫾ 1.5
1.7 ⫾ 0.6
3.3 ⫾ 2.3
1.3 ⫾ 0.6
3.3 ⫾ 0.6
1.3 ⫾ 1.0
2.0 ⫾ 1.0
Cytokine or Stx added
(pg/dish)
No addition
Stx1 (10)
Stx1 (50)
Stx1 (100)
Stx2 (10)
Stx2 (50)
Stx2 (100)
G-CSF (100)
Stx2 (10) ⫹ G-CSF (100)
SCF (5,000) ⫹ IL-3 (500)
Stx2 (50) ⫹ SCF (5,000) ⫹ IL-3 (500)
IL-1␤ (100)
Stx2 (50) ⫹ IL-1␤ (100)
IL-1␣ (50)
Stx2 (50) ⫹ IL-1␣ (50)
a
No. of colonies/disha
GM
0
0
0
0
4.0 ⫾ 0
6.0 ⫾ 1.0
9.3 ⫾ 0.6
14 ⫾ 1.0
23.7 ⫾ 2.1
12 ⫾ 1.0
28 ⫾ 2.0
4.3 ⫾ 0.8
21 ⫾ 1.2
0.7 ⫾ 0.6
1.3 ⫾ 1.0
M␾
0
0
0
0
10 ⫾ 2.0
20 ⫾ 0.6
17 ⫾ 1.2
4.7 ⫾ 1.2
3.0 ⫾ 1.0
11 ⫾ 2.3
8.0 ⫾ 2.6
1.3 ⫾ 0.6
2.7 ⫾ 1.2
0.7 ⫾ 0.6
1.0 ⫾ 0
Results are the mean ⫾ standard deviation of triplicate experiments.
Results are the mean ⫾ standard deviation of triplicate experiments.
was obtained when IL-1␤ was added to culture containing Stx2
(P ⬍ 0.001) or IL-3 with SCF instead of IL-1␤ (P ⬍ 0.05), as
shown in Table 1. Addition of cytokines and Stx2 had almost
the same effect on colony formation of human cord blood stem
cells as for human bone marrow stem cells (Table 2). Addition
of IL-1␣ alone to bone marrow or cord blood stem cell culture
caused negligible colony formation, and the colony-stimulating
effect of Stx2 was rather inhibited by IL-1␣. No colonies developed when various amounts of Stx1 were added to the
culture (Tables 1 and 2). Addition of Stx1 to the culture containing G-CSF failed to change the number of colonies induced by G-CSF, and the presence of Stx1 in the culture
containing both Stx2 and G-CSF did not alter the number of
colonies induced by both Stx2 and G-CSF (Table 1). The
culture containing Stx1(100 pg/dish) failed to increase the
number of dead cells compared with controls. The proliferation status of both HL-60 and Jurkat cells did not change with
the addition of Stx1 or Stx2 (Fig. 1). These results indicate that
Stx1 has no toxic effect on bone marrow stem cells. Heat-
denatured Stx2 (boiled for 10 min) had no colony-stimulating
activity in either human bone marrow or cord blood cell cultures.
Recently, it was shown showed that intraperitoneal injection
of sublethal doses of Stx2 into mice caused marked granulocytosis in peripheral blood (5). Accordingly, we assayed in vitro
colony formation to analyze the direct effect of Stx2 on human
bone marrow and cord blood stem cells. We found that adding
Stx2 to human bone marrow cell culture induced mainly M⌽
colonies. We further found that it induced a large number of
GM colonies when cytokines such as G-CSF, IL-1␤, or SCF
plus IL-3 were added. These results strongly suggest that Stx2
induces granulocytosis in the peripheral blood by directly or
indirectly stimulating bone marrow stem cells to develop into
GM colonies in vivo. Elevated granulocytosis in patients with
bacterial infections is believed due to endotoxin (lipopolysaccharide), which is a cell wall pleiotropic component. However,
no report has, to our knowledge, postulated the possibility that
bacterial exotoxin (protein) is a factor in granulocytosis. This is
the first report that Stx2 stimulates granulopoiesis directly or in
FIG. 1. Effects of Stx1 and -2 on the proliferation status of bone marrow-derived cells. Jurkat cells (white) and HL-60 cells (gray) were cultured
in the presence of Stx1 (A) or Stx2 (B). Bars indicate the standard deviation (n ⫽ 3).
Downloaded from http://iai.asm.org/ on June 11, 2014 by guest
GM
a
TABLE 2. Effect of Stx’s on colony formation from
cord blood stem cells
No. of colonies/disha
Cytokine and/or Stx added
(pg/dish)
5317
5318
NOTES
INFECT. IMMUN.
REFERENCES
1. Aoki, Y. 1986. The mechanism of activation of human natural killer cells by
medullasin. Jpn. J. Cancer Res. 77:1018–1026.
Editor: J. T. Barbieri
2. Aoki, Y., and R. Machinami. 1982. Medullasin and inflammation. Asian
Med. J. 26:15–37.
3. Aoki, Y., and R. Machinami. 1983. Role of medullasin in granulocytes in the
development of inflammation. I. Phlogistic activity and the effect on functions of macrophages and granulocytes. Arthritis Rheum. 26:1002–1010.
4. Aoki, Y., M. Sumiya, and Y. Oshimi. 1982. Medullasin enhances human
natural killer cell activity. J. Clin. Investig. 69:1223–1230.
5. Aoki, Y., Y. Yoshida, and T. Takeda. 1999. Shiga toxin 2 promotes the stem
cell differentiation into granulocytes in the bone marrow causing granulocytosis in the peripheral blood. J. Infect. 39:97–98.
6. Boerlin, P., S. A. Mcween, F. Boerlin-Petzold, J. B. Wilson, R. P. Johnson,
and C. L. Glyles. 1999. Associations between virulence factors of Shiga
toxin-producing Escherichia coli and disease in humans. J. Clin. Microbiol.
37:497–503.
7. Bolande, R. P., and B. S. Kaplan. 1985. Experimental studies on the hemolytic uremic syndrome. Nephron 39:228–236.
8. Carter, A. O., A. A. Borczyk, A. A. K. Carlson, B. Harvey, J. A. Hockin, M. A.
Karmali, C. Krishman, D. A. Korn, and H. L. Lior. 1987. A severe outbreak
of Escherichia coli O157:H7-associated hemorrhagic colitis in a nursing
home. N. Engl. J. Med. 317:1496–1501.
9. Fong, J. C. S., J.-P. de Chadere’vian, and B. S. Kaplan. 1982. The hemolytic
uremic syndrome: current concept and management. Pediatr. Clin. North
Am. 29:835–856.
10. Gasser, C., E. Gauthier, A. Steck, R. E. Sieberman, and R. Oechslir. 1955.
Hemolytic uremic syndrome: bilateral renal cortex necrosis by acute acquired hemolytic anemia. Schweiz. Med. Wochenenschr. 38:905–909.
11. Gianantonio, C. A., M. Vitacco, F. Mendilahazu, A. Ruffy, and J. Mendilahazu. 1964. The hemolytic uremic syndrome. J. Pediatr. 64:478–491.
12. Hase-Yamazaki, T., and Y. Aoki. 1995. Stimulation of human lymphocytes by
cathepsin G. Cell. Immunol. 160:24–32.
13. Iscove, N. N., F. Sieber, and K. H. Winterhalter. 1974. Erythroid colony
formation in cultures of mouse and human bone marrow: analysis of the
requirement for erythropoietin by gel filtration and affinity chromatography
on agarose-concanavalin A. J. Cell. Physiol. 83::309–320.
14. Kaplan, B. S., P. D. Thompson, and J.-P. de Chadare’vian. 1976. The hemolytic uremic syndrome. Pediatr. Clin. North Am. 23:761–777.
15. Martin, D. S., I. Walters, U. Matthei, J. Mitchel, T. Dillon, and M. Barrat.
1989. Polymorphonuclear leukocyte count in childhood hemolytic uremic
syndrome. Pediatr. Nephrol. 3:130–134.
16. Noda, M., T. Yotsudo, N. Nakabayashi, T. Hirayama, and Y. Takeda. 1987.
Purification and some properties of Shiga toxin from Escherichia coli
O157:H7 that is immunologically identical to Shiga toxin. Microb. Pathol.
2:339–349.
17. Ostroff, S. M., J. M. Kobayashi, and J. A. Lewis. 1989. Infections with
Escherichia coli O157:H7 in Washington state. JAMA 262:355–359.
18. Shumwy, C. N., K. L. Telplan, C. F. Piel, and R. H. Philibbs. 1966. The
hemolytic uremic syndrome. Pediatr. Clin. North Am. 13:295–314.
19. Vedanarayanan, V. V., B. S. Kaplan, and J. C. S. Fong. 1987. Neutrophil
function in an experimental model of hemolytic uremic syndrome. Pediatr.
Res. 21:252–256.
20. Yamazaki, T., and Y. Aoki. 1998. Cathepsin G enhances human natural killer
cell activity. Immunology 93:115–121.
21. Yotsudo, T., N. Nakabayashi, T. Hirayama, and Y. Takeda. 1987. Purification and some properties of a vero toxin from Escherichia coli O157:H7 that
is immunologically unrelated to Shiga toxin. Microb. Pathol. 3:21–30.
Downloaded from http://iai.asm.org/ on June 11, 2014 by guest
combination with cytokines such as G-CSF, IL-1␤, or SCF plus
IL-3, causing marked granulocytosis in the peripheral blood
Leukocytes contain several agents that damage tissue, especially vascular endothelial cells. Infiltration of neutrophils in
the kidney was mentioned in early reports on HUS (11, 18).
Bolande and Kaplan (7) studied leukocytes in the buffy coat of
patients with HUS soon after onset and noted that, in some
cases, a tactile relationship existed between leukocyte cell processes and altered red blood cell walls and that leukocytes had
abnormal morphologies with cytoplasmic projections. Vedanarayanan et al. (19) reported neutrophil function in an experimental rabbit model using an endotoxin-derived modified
generalized Schwartzman reaction to understand the role of
neutrophils in HUS. They speculated that neutrophil activation may be a mechanism of renal injury in this model. Characteristic functions of granulocytes are derived from proteases
in lysozomes. In addition to bactericidal activity and tissue
destruction in inflammation, granulocyte proteases, especially
neutral proteases, have a broad spectrum of functions. Both
medullasin and cathepsin G enhance DNA synthesis of human
lymphocytes (2, 12). They also enhance human NK cell activity
(1, 4, 20). Medullasin induces inflammation by injuring endothelial cells in vessels and accumulating both M⌽ and granulocytes when injected into rabbit or guinea pig skin (3). Intraperitoneal injection of Stx2 into mice enhanced medullasin
levels in granulocytes (unpublished data), and so elevated medullasin in granulocytes would appear to be an important risk
factor for HUS.
Surprisingly enough, Stx1 did not show the same functions as
Stx2. We previously believed that Stx1 and Stx2 have similar
functions, although one paper reported that Stx2 had more
potent toxicity than Stx1 (6). Our results suggest differences in
the pathogenic mechanism between these two toxins. The primary structures of both have 55% homology in the A and B
subunits. The responsive structure of Stx2 in granulocytosis is
to be analyzed in a future study.