- Email: [email protected]
A growth-promoting factor for human myeloid leukemia cells from horse serum identified as horse serum transferrin
Biochimica et Biophysica Acta, 1Ol0 (1989) 28-34
28
Elsevier BBA 12387
A growth-promoting factor for human myeloid leukemia cells from horse serum identified as horse serum transferrin K a o r u Y o s h i n a r i , K a t s u m i Y u a s a , F u s a o Iga a n d A k i o M i m u r a Medical Science Laboratory, Asahi Chemical Industry Co., Ltd., Shizuoka-416 (Japan)
(Received 8 August 1988)
Key words: Growth promotiag factor; Transferrin; Myeloid leukemia cell; (Horse serum); (Human cell)
A growth.promoting factor for human myeloid cells was purified to apparent homogeneity from horse serum by a combination of gel filtration, blue Sepharose affinity chromatography, Mono Q anion-exchange chromatography, Mono P chromatofoeusing and sodium dodecyl sulfate polyacrylamide gel electrophoresis. The growth promoter was an iron-bound, single glyeopolypeptide chain with a molecular weight of 84 000, an isoelectric point of 5.4 and an amino terminal sequence of Glu-Gln-Thr-Val-Arg-Trp-Cys-Thr-Val-Ser-Asn-His-Glu-Val-Ser-Lys-. According to the results of the amino acid sequence, iron binding ability and physlcochemical properties, we identified the growth-promoting factor as horse serum transferrin. It was highly active in promoting the proliferation of a human monocytic leukemia cell line, THP-I, as well as of two other human myeloid cell lines, Hi.~60 and K-567. it had the same activity in proliferating THP.I cells as 5% fetal calf serum-supplemented medium. Horse serum transterrin could be substituted for human or bovine serum transferrin. Introduction It has been known for some time that some types of cell grow more efficiently in media supplemented with horse serum than with fetal calf serum or other types of animal serum [1], whereas this is not the case for other types of cell. In addition, horse serum is preferred as opposed to calf serum by some workers as it can be obtained from a closed herd and batch to batch consistency is often greater. Identification of the growthpromoting factors of horse serum itself has been of considerable interest. Over the past few years, the use of serum-free media in cultivating a variety of cells has been reported and such studies have been reviewed by Sato [2]. Recently, proliferation and differentiation of human myeloid leukemia cells in serum-free medium have been studied using HL-60 [3], ML-1 [4] and K-562 [5] cells. In the process of screening for growth-promoting factors in order to construct a serum-free medium for Abbreviations: PBS, phosphate-buffered saline (138 mM sodium chloride/2.7 mM potassium chloride/8 mM dibasic sodium phosphate/1.5 mM monobasic potassium phosphate, pH 7.4); SDS-PAGE, sodium dodecyi sulphate polyacrylamide gel electrophoresis; HS, horse serum; FCS, fetal calf serum; GPF, growth-promoting factor. Correspondence: K. Yoshinari, Medical Science Laboratory, Asahi Chemical Industry Co., Ltd., 2-1, Samejima, Fuji-shi, Shizuoka-416, Japan.
culturing a human monocytic leukemia cell line, THP-1, originally established by Tsuchiya et al. [6], horse serum was found to be more effective in proliferating THP-1 cells than fetal calf serum. Therefore, we have tried to isolate one of the growth factors involved in horse serum and to construct a serum-free medium for culturing THP-1 cells. In this report, we describe the procedures used both to isolate a growth-promoting factor from horse serum and to identify it as horse serum transferrin. Materials and Methods
Materials. Materials were obtained from the following sources: RPMI 1640 medium, fetal calf serum and human serum from Flow Laboratories, VA, U.S.A.; horse serum from the Nippon Biosupplements Center, Tokyo, Japan; bovine insulin, proteinase (from Streptomyces griseus) and Concanavalin-A Sepharose (Con A-Sepharose) 4B from Sigma Chemical Company, MO, U.S.A.; ethanolamine, sodium selenite from Nakarai Chemicals Ltd., Japan; bovine serum albumin (fraction V) from Calbiochem Boehring, CA, U.S.A.; FeC-Test Wako from Wako Pure Chemical Industries, Japan; blue Sepharose CL-6B, Superose 12 HR10/30, Mono Q HR5/5, Mono P HRS/20 and polybuffer 74 for chromatofocusing from Pharmacia Fine Chemicals, Sweden. Cell lines. THP-1 cells (a human monocytic leukemia cell line) were kindly provided by Dr. Tsuchiy.~. THP-1
0167-4889/89/$03.50 © 1989 Elsevier Science Publishers B.V. (Biomedical Division)
29 (AH-01) cells are a subline selected from THP-1, which have been adapted for growth in low serum-containing medium, i.e., RPMI 1640 medium supplemented with a double concentration of amino acids and vitamins, 10 p g / m l bovine insulin, 10 /tg/ml ethanolamine, 0.05 /~g/rnl sodium selenite and 0.1% HS. The HL-60 and K-562 cell lines were maintained in serum-containing medium, i.e., RPMI 1640 medium supplemented with a double concentration of amino acids and vitamins and 5% FCS. All cells were cultured at 37°C in a humidified atmosphere of 5% CO2 in air, and were passaged every 3-4 days at an inoculum size of 2- 105 cells/ml. Cell proliferation assay. For proliferation assays, exponentially growing THP-I-(AH-01) cells were inoculated into 60 mm plastic tissue culture dishes at a density of 1.5.105 cells/ml in serum-free medium (RPMI 1640 medium containing a double concentration of amino acids and vitamins, 5 ~tg/ml bovine insulin, 10 /~g/ml ethanolamine and 0.05/xg/ml sodium selenite) supplemented with any one of the test samples. Cell counts after 3-6 days of cultivation were performed in duplicate with a Coulter Counter (Coulter Electronics Inc., U.S.A.). 1 unt of activity was defined as the reciprocal of the sample dilution causing 50% maximal growth stimulation (net response). Isolation procedure. 0.25 ml of HS heat-inactivated at 56 *C for 30 min was applied to a column of Superose 12 HR10/30, followed by elution of PBS at a constant flow rate of 0.4 ml/min (1 fraction per min). Pooled bioactive fractions were subjected to a blue Sepharose CL-6B (2.6 × 10 cm) affinity column equilibrated with PBS at a flow rate of 30 ml/h. Pooled fractions containing activity which did not bind to the blue Sepharose column were concentrated and buffer-exchanged into 10 mM Hepes-HCl (pH 7.4) using a polysulfone capillary tube. This pool was then applied to a Mono Q HR5/5 anion-exchange column that had been equilibrated with 10 mM Hepes-HCl (pH 7.4), and was eluted at a flow rate of 0.5 ml/min using a linear gradient of sodium chloride the concentration of which was between 0 and 1 M. Fractions of 1 ml were collected. The most active fraction was desalted and applied to a chromatofocusing column of Mono P HR5/20 (0.5 x 20 cm) pro-equilibrated with 25 mM imidazole HCI (pH 7.4) and was eluted with 5-fold diluted polybuffer 74 (pH 4.0, adjusted with HCI) to form a linear pH gradient between pH 7 and 4 at a flow rate of 0.5 ml/min (fractionation was performed on every peak). As a final purification step, polyacrylamide gel electrophoresis in the presence of SDS under non-reducing conditions was performed using 10% polyacrylamide gels according to the method of Laemmli [7]. The total protein content of all samples was determined by the dye-fixation assay (Tonein TP, supplied by Otsuka Assay Laboratories, Japan), using HSA (human serum albumin) as a standard.
Effects of temperature and of chemical and proteinase treatments on growth-promoting activity. The sensitivity of Mono-Q-purified samples to various temperatures and chemical and enzymatic treatments was tested essentially according to the method of Mochizuki [8]. Briefly, aliquo~s of samples were incubated at 56°C, 70°C or 100°C for the indicated length of time. The effects of pH on activity were studied as follows: samples were incubated in different buffers for 6 h at 4 ° C, were dialyzed against PBS and then assayed for cell growth-promoting activity. The pH 2.0, pH 4.0 and pH 6.0 buffers were Na2HPOJcitric acid buffers; the pH 7.4 buffer was PBS, the pH 8.0 buffer was a Tris-HCl buffer and the pH 10.0 buffer was a Na 2CO3/NaHCO3 buffer. The sensitivity to proteinase extracted from Streptomyces griseus was determined by testing samples at a concentration of 0.29 mg/ml in PBS with 400 p g / m l of the proteinase at 37°C for the indicated lengths of time. Growth-promoting activity was directly assayed without removing the proteinase, because it was previously found that at the concentration used in these experiments, it had no effect on the cell proliferation assay. Amino acid sequence analysis. Purified sample was concentrated by ammonium sulfate precipitation (80% saturation), dialyzed against twice-distilled water, lyophylized and finally re-dissolved in water at a concentration of 1 mg/ml. Amino acid sequence analyses were performed using an Applied Bio-systems 470 gasphase protein sequencer (Applied Biosystems, U.S.A.). Firstly, 360 pmol of sample were applied to the sequencer for 25 cycles (Table III, Ept. 1) then, in order to identify the cysteine residues, 130 pmol were carboxymethylated and applied to the sequencer for 21 cycles (Table III, Expt. 2). The phenylthiohydantoins of the amino acids were identified by high-performance liquid chromatography rising a Zorbac ODS C~s column. Iron analyses. Iron concentrations were determined using Wako kits (Wako Pure Chemical Industries, Japan) according to the manufacturer's instructions. Briefly, free ferric irons, to which ferrous ions were reduced by thioglycolic acid, were chelated with a specific reagent, 2-nitro-5-(N-propyl-N-sulfopropylamino)phenol to form a complex with peak absorbance at 750 rim. The iron content of a sample was thus measured spectroscopically by its absorbance at 750 rim.
Results
Growth promoters for human monocytic leukemia cells The growth-promoting abilities of HS, FCS and human serum were compared. As shown in Fig. 1, it was found that with FCS a concentration of at least 1% (i.e., 600 pg protein per ml) was required for full growth of
30 67k 43k 12,4k
15 A
2F
10
e
s U}
e~
i5
1
4
¢,j
vX
e) Z
(1
I
I
1
I
1'0
i
100
I
1()00 3000
Prote~ Conc.( ~glml ) Fig. 1. Proliferative effect of various sera on the growth of THP-1 (AH-01) cells. The proliferation assay was performed as described in Materials and Methods. Samples tested were HS (76 mg/ml, ®), FCS (60 mg,/ml, A), and human serum (84 mg/ml, o) at the indicated concentrations, Proliferation was measured on day 6. Standard de~'iations were less than 10~ of the mean from duplicate determinations.
THP-1 cells, while for HS and for human serum, only 0.02~ (i.e., 15/~g/ml) and 0.005~ (i.e., 4.2/tg/ml) were sufficient, respectively. Although human serum was 3.6-fold more potent than HS in enhancing proliferation of the human cell line, THP-I-(AH-01), HS was chosen as the source of growth promoters to be used in subsequent studies, as with HS, no problems about large-scale supply and pathogenic viruses to humans are encountered. Since HS had a greater THP-1 cell growth-promoting activity than FCS, and since several lots of HS (13 lots tested) were found to have the same level of activity (data not shown), studies were initiated to isolate the growth-promoting substances from HS.
Isolation of growth promotingfactor Elution of heat-inactivated HS (56 o C, 30 min) from a Superose 12 gel-filtration column, showed that the growth-promoting activity had a molecular weight between 30000 and 100000 (Fig. 2). A growth-inhibitory activity was seen in the high-molecular-weight fractions 20 and 21 when the serum had been heat ~l~c,'~vated. These fractions also contained a corresponding protein peak, indicating that the inhibitory effect was caused by aggregation of macromolecules. When pooled, the growth-promoting fractions contained 99~ of the total starting activity and the protein purity was found to be 2.8-fold more than the starting material. Further purificatio~t was achieved by passing the active fractions through the blue Sepharose CL-6B column. Activity eluted with PBS was detected in the unbound portion, 90~ of the material being ~hus reco~ered, with a further purification factor of 2.6. Better resolution was achieved using the Mono Q HRS/5 anion-exchange column. Fig. 3 shows that this step yielded three fractions containing activity (fractions 15-17). The most active fraction, number 16. was
0
'
2~0
40 60 80 Fraction Number Fig. 2. Gel filtration of HS on Superose 12. HS (0.25 ml, 19 mg) was applied to a Superose 12 column under the conditions described in Materials and Methods. Protein elution was monitored continuously tlu'ough its absorbance (Abs.) at 280 nm (solid line). The activity was measured by the THP-I-(AH-01) cell proliferation assay, using a 10 -3 dilution of a test sample. The results obtained on day 3 are ~nown (o o). The broad peak of activity observed (fractions 32-37) was pooled for the next purification step.
further purified by chromatofocusing on the Mono P HR5/20 column. Activity was observed in both peak fractions eluted at pH values of 5.5 + 0.1 and 5.4 + 0.1 (Fig. 4) (no activity in other fractions was observed). Analysis of the most active fraction (pl 5.4) by SDSPAGE under reducing conditions showed that the material consisted of a single silver-stained band with a molecular weight of 84000 (Fig. 5, lane 5). A single band of 84000 was also detected by SDS-PAGE under non-reducing conditions (data not shown), and when excised from the gel and eluted with PBS its activity was found to have an overall yield of 8.4~ (Table I). These data indicated that a growth-promoting factor had thus been purified to homogeneity.
10
!
in
0
0 20 30 Fraction Number Fig. 3. Anion-exchange chromatography on a Mono Q column. The blue Sepharose unbound protein (6.4 mg), dissolved in 10 mM HepesHCi (pH 7.4) was applied to a Mono Q column and was eluted in 1 ml fractions using a linear NaC! gradient, as indicated by the broken line, under the conditions described in Materials arl¢ Methods. Protein elution was monitored through its absorbance (Abs.) at 280 nm (solid line). Activity was measured by the THP-I-(AH-01) cell proliferation assay, at a dilution of 10 -2. The results obtained on day 3 are shown (o o). The most active fraction (number 16) was pooled (six runs) for chromatofocusing.
10
31 TABLE 1
Purificationof growth-promotingfactor Protein was determined by the dye-fixalion assay as described in Materials and Methods. Activity was defined as described in Materials and Methods. Step
Volume (rid)
Portein (mg)
Total activity ( x 104 U)
Spec. at' (U/mg)
Purification (fold)
Overall yield (~)
HS Superose 12 Blue-Sepharose (unbound) Mono Q Mono P SDS-PAGE
4.25 27 71 6.1 5.5 1.0
323
9.4 9.5 8.5 1.9 2.$ 0.8
300 830 2 200 9403 10 500 10400
1 2.8 7.3 31 35 "45
100 99 89 20 21 8.4
!
\
0
=o 0.5
1] 3
38 2.0 1.9 0.77
% v
2: Q.
O3 .0
<
I
~
'
I
I
10
t
20
I
30
Elution Volume ( ml )
Fig. 4. Chromatofocusing on a Mono P column. The desalted activity (0.28 mg protein) obtained by Mono Q chromatography, was applied to a Mono P column and was eluted with a linear pH gradient ( × . . . . . . ×) under the conditions described in Materials and Methods. Protein elution was monitored by its absorbance (Abs.) at 280 nm ( ). Every peak was fractionated. Activity is indicated by the arrows.
Measurement of the growth-promoting activity of samples taken from each purification step showed that at most 0.65 pg of the factor per ml were required to induce maximum growth promotion in THP-I-(AH-01) cells (Fig. 6). As can be seen in Table I and Fig. 6, the overall yield of activity and the final purificatio~l factor were 8.4~ and 35-fold, respectively.
Physicochemical properties of purified growth-promoting factor (GPF) G P F was s t a b l e a t 5 6 - 7 0 ° C for 30 m i n , b u t p r o l o n g e d t r e a t m e n t r e d u c e d a c t i v i t y e v e n at 56 o C. B o i l i n g o f G P F for 5 m i n r e s u l t e d in a c o m p l e t e loss o f a c t i v i t y ( T a b l e II). O n t h e o t h e r h a n d , the results o f the p H s t a b i l i t y a n a l y s i s s h o w e d t h a t G P F was s t a b l e o n l y b e t w e e n p H v a l u e s o f 6 a n d 8. As s h o w n in T a b l e II, t r e a t m e n t of G P F w i t h p r o t e i n a s e for 16 h d e s t r o y e d its b i o l o g i c a l activ~i3, ,=c::~?letely, i n d i c a t i n g t h a t it was a proteinaceous material.
... 1 5 -
"~m
94k.~
SDS-PAGE
~'~"84k 67
"~
& Z
4 3 "~
!
°.Ol
M 1 2
3
4
5
Fig. 5. 10~ SDS-polyacrylamide gel electrophoresis of samples at different stages of purification. Samples at each purification step were analyzed by 10~ SDS PAGE under reducing conditions. (M) molecular weight standards: lane 1, HS; 2, Superose 12 fraction; 3, blue-Sepharose unbound fraction; 4, Mono Q fraction; 5, Mono P fraction. The gel was stained by the silver nitrate method. The molecular weight standards were phosphorylase b (94000), bovine serum albumin (67000) and ovalbumin (43000).
o.1
1
;o
' 100 '
Protein Conc.(/~g/ml)
Fig. 6. Comparison of the ability of GPF preparations at various stages of purification to promote proliferation of the THP-I-(AH-01) cells. Proliferation was measured on day 5 as described in Materials and Methods. Samples tested were HS (e), the Superose 12 fraction (e), the blue-Sepharose unbound fraction (&), the Mono Q fraction (zx), the Mono P fraction (o) and the SDS-PAGE fraction (ra). All assays were performed in duplicate and standard deviations were less than 10¢$ of the mean.
32 TABLE II
TABLE II1
Effects of temperature, pH and proteinase treatment on GPF activity
Amino terminal amino acid sequence
Each treatment was performed as described in Materials and Methods. Cell proliferation was measured on day 5. 100~ and zero residual activities represent cell concentrations reached on day 5 of either 1.2.106 celis/ml or of only the original inoculum size of 1.5.105 ceils/ml, respectively. AH values are means of duplicate determinations and standard deviations were less than 105 of the mean.
In Expt. 1 360 pmol of G P F were applied to an amino acid sequences. In Expt. 2 130 pmol of G P E were carboxylmethylated and applied to a sequencer in order to identify the residue at position 7.
Residual activity
Treatment
(~)
None (pH 7.4)
100
560 C 56 ° C 56°C
60rain
98 100 65
70"C
10rain
10 rain
30 rain
700C 30 rain 70°C 60 rain
98 92 62
Boiling I rain Boiling 2 rain Boiling 5 rain
77 15 0
pH 2.0 pH 4.0 pH 6.0 pH 8.0 pill0.0
4°C 4°C 4°C 4°C 4°C
6h 6h 6h 6h 6h
Proteinase 400 pglml Proteinase 400 p g / m l
37 62 88 88 0
3h 16 h
14 0
Concanavalin A binding ability The possibility that GPF was a glycosylated protein was investigated using Con A-Sepharose 4B beads suspended in PBS. Growth-promoting activity could not be eluted with PBS, but could with PBS containing 0.3 M a-methyl mannoside (data not shown), showing that it had a glycosyl moiety containing mannosyl and/or glucose residues. Amino acid analysis The results are shown in Table III. As position 7 of the amino terminal sequence could not be identified by the first analysis (Table III, Expt. 1), the sample was carboxymethylated in order to determine whether this amino acid was a cysteine residue (Table III, Expt. 2). As a result of the two analyses conducted, a sequence of 16 consecutive amino acids located at the protein's amino terminal end was determined, as shown in Table IV. This sequence was found to share strong sequence homology with the corresponding part of the human transferrin molecule [9] (Table IV); ten out of the 16 were identical (62% homology). In addition, the homology between terminal amino acid residues of GPF and human lactoferrin [10] was 44% (seven residues out of 16).
Position
Amino acid
1 2 3 4 5 6 7 8
Glu Gin Tht Val Arg Trp Cys Thr Vai Ser Asn His Giu Val ~er Lys X Ala X Phe Arg X X Met Lys
9
10 11 12 13 14 15 16 17 18
19 20 21 22 23 24 25
Yield (pmol) Expt. 1
Expt, 2
320 175 48 170 70 _ a _ b 34 140 9.8 78 29 100 77 3.3 46 41 32 24 17 13
70 58 19 70 21 _ 47 12 53 2.7 18 3.9 21.5 27 2.5 10 -
14 1.! .5 4.9
a Identified but not quaatitated. b Not identified.
Iron-binding ability of GPF Due to its structural similarity to human transferrin, the iron-binding ability of G P F was tested using a FeC-Test Wako kit to observe whether those molecules formed a family of iron-binding proteins. It was found to bind to iron with a molar ratio (iron vs. GPF) of 1.73 (-I-0.13 in duplicate determi.--.:::_'_....~. As the same value was also obtained after saturation of this factor TABLE IV
Comparison of the amino terminal sequences of GPF and human transferrin (hTf) Amino acid sequences of G P F and human transferrin are from Table Ill and Ref. 12, respectively. Homologies between G P F and h u m a n transferrin are boxed by a solid line.
GPF
GLu-GLn~Thr'-VaL-Arg-Trp-¢ystThr~VaL-Ser- !
hTf VaL-Pro-Asp-Lys~Thr-VaL-Arg-Trp-Cys~ALa~VaL-Ser-] GLu~H~s-GLu~ALa-Thr~Lys~Cys-GLn-Ser~Phe-Wrg~His~-Lys- I
33 with iron, it was concluded that purified factor was saturated with iron, and contained approximately two iron atoms per molecule. Taking data mentio~ied above into consideration, GPF was identified as horse serum transferrin. In addition, analyses of the minor active fractions in the Mono Q ehromatogram (Fig. 3) revealed that the molecules in fractions 15 and 17 contained iron and protein at molar ratios of 1.68(+0.12) (fraction 15) and 1.81(+0.02)(fraction 17), that their p l values were 5.65 (fraction 15) and 5.35 (~action 17), and that the molecular weight of both proteins was around 84000 (data not shown). Thus, although amino acid sequence analyses of the materials in these two fractions were not performed, it is likely that they are subspecies of horse transferrin. Summation of the datas of the GPFs in all three fractions shows that its concentration in HS amounts to 1.5 mg/rnl; the concentration of human transferrin in plasma has been reported to be 2-4 mg/ml [11].
Biological characterization of GPF Fig. 7 presents the growth curves for THP-I-(AH-01) cells grown in media supplemented with either GPF, HS or FCS. As little as 1 ~g of GPF per ml was able to substitute almost completely for 5% HS, although at day 3 the effects of GPF and HS on the growth of THP-I-(AH-01) cells were slightly different. GPF was also found to support growth of THP-I-(AH-01) cells to
TABLE V
Effects of purified GPF on growth of HL-60 and K-562 cells The proliferation assay was performed as described in Materials and Methods. Inoculation size in this experiment was 2.105 cells/ml. Each cell suspension was supplemented with either 1/xg/ml GPF or 5% FCS. Cell proliferation was measured on day 5. All samples were made in duplicate and were counted with a Coulter counter. Cell line
~1L-60 K-562
Supplement GPF (1/~g/ml)
FCS (5%)
17.0 + 0.6 ( x 10 s cells/ml) 13.1 ±0.2
20.2 ± 0.5 16.1 ±0.3
much the same extent as did 5% FCS. Growth of THP-1 cells continued in this serum-free medium for over 100 days. In order to investigate whether GPF would promote the growth of other myeloid teukemia cell lines, the human promyelocytic cell line, HL-60, and the human erythroleukemia cell line, K-562, were cultivated in serum-free medium supplemented with GPF. As both these cell types had previously been routinely cultured in FCS-supplemented medium, the effects of GPF and 5% FCS on growth were compared. Table V shows that 1 # g / m l of GPF supported the growth of both HL-60 and K-562 cells at rates equal to 85% and 81%, respectively, of those achieved using FCS. Discussion
2°! . - 10
x
5
O~ .O
E z
/
o
10
I
I
I
3
5
7
Incubation Period ( Days )
Fig. 7. Comparative growth curves for THP-I-(AH-01) cells proliferated in medium supplemented with GPF, HS and FCS. THP-I(AH-01) cells were suspended in serum-free medium at a density of 1.5.105 cells/ml and the medium was supplemented with either 1 / t g / m l of GPF ( e e), 5% HS ( o o), 5% FCS (A. . . . . . A) or no additive (zx. . . . . . zx). Cultures were incubated at 3 7 ° C in an atmosphere of 5% CO 2 in air for 7 days. All points are means of duplicates.
This paper presents the results of a study on the purification and characterization of a growth-promoting factor (GPF), identified in horse serum. Comparison of the amino terminal sequences of GPF and human transferrin in this study shows that ten residues out of 16 were identical (62% homology), whereas homology between the portions of human and chicken transferrins sequenced so far have been reported to be only 49.8% [9]. Our study noted further close similarities between GPF and human transferrin in their physicochemical properties, GPF being a single glycoprotein chain iwth a molecular weight of 84000 and a pl of 5.4, and human transferrin having a molecular weight of 80000 and a variable pl of between 5.2 and 5.8. The first purification step, gel filtration, yielded material which eluted at molecular weight positions between 30000 and 100000 (Fig. 2), suggesting that it was smaller than the molecule found electrophoretically to have a molecular weight of 84 000 (Fig. 5). This could also be explained by the report of the tendency of the transferrin molecule to become more compact and spherical after it has bound iron [11,12]. From a biological point of view, Breitman et al. [3] reported that a human promyelocytic leukemia cell line,
34 HL-60, could grow in serum-free medium containing insulin and human transferrin at a rate of 80~;, compared to FCS-containing medium. Our serum-free medium containing bovine insulin and GPF could support the growth of HL-60 cells at a rate of 85% (Table V), compared to 5~ FCS medium. It should be further noted that human transferrin could be replaced by our purified factor, horse serum transferrin, in a serum-free system to support the growth of HL-60 and THP-1 (data not shown). The difference of these growth rates of HL-60 and THP-1 in GPF-supplemented media (85~; vs. 100~) (Table V, Fig. 7) can probably be attributed to adaptation of THP-1 cells to medium supplemented with a low concentration of horse serum. Alternatively, bovine transferrin, at a concentration of 5 /tg/ml, could not support the growth of either THP-1, HL-60 or K562 cells (data not shown). But Tsavaler et al. [13] reported that when growth curves of K562 cells were compared in serum-free medium containing either 300/~g/ml of bovine serum (equivalent to approx. 5~ FCS medium) or 300 ~tg/ml of human transferrin, the growth rates were the same. These facts are consistent with our data shown in Fig. 1, showing that the requirement of FCS was much greater than that of HS or human serum to proliferate THP-1 cells, considering the report [13] that a large amount of bovine sermn transferrin is necessary to effectively defiver iron intracellularly to K562 cells. In addition to human leukemia cells, many types of cell such as a rat pituitary cell fine, GH3 [14], and a human cervical cell line, HeLa [15], were reported to require transferrin as an essential component in serumfree medium. Thus, a growth-promoting factor from horse serum, identified as horse serum transferrin, would be a good tool for comparative study of cell proliferation.
Acknowledgements This work was supported by funds obtained under the Research and Development Project for Basic Technologies for Future Industries from the Ministry of International Trade and Industry of Japan. The authors wish to thank Dr. T. Hitouji for his amino acid sequence analyses. References 1 Endo, H. (1960) Exp. Cell Res. 21, 151-163. 2 Sato, G.H. (1982) in Cold Spring Harbor Conference on Cell Proliferation Vol. 9 (Sato, G., Pardee, A. and Sirbasku, D., eds.), pp. 1-624, Cold Spring Harbor Laboratory, New York. 3 Breitman, T.R., Collins, S.J. and Keene, B.R. (1980) Exp. Cell Res. 126, 494-498. 4 Taketazu, F., Kubota, K., Kajigaua, S., Shionoya, S., Motoyoshi, K., Saito, M., Takaku, F. and Miura, Y. (1984) Cancer Res. 44, 531-535. 50kabe, T., Fujisawa, M. and Takaku, F. (1984) Pro(:. Natl. Acad. Sci. USA 81, 453-455. 6 Tsuchiya, S., Yamabe, M., Yamaguchi, Y., Kobayashi, Y., Konno, T. and Tada, K. (1980) Int. J. Cancer 26, 171-76. 7 Laemmli, U.K. (1970) Nature (Lond.) 227, 680-685. 8 Mochizuki, D., Watson, J. and Gillis, S. (1980) J. Immunol. 125, 2579-2583. 9 MacGillivray, R.T.A., Mendez, E., Shewale, J.G., Sinha, S.K., Lineback-Zinpo, J. and Brew, K. (1983) J. Biol. Chem. 258, 3543-3553. 10 Metz-Boutigue, M.H., Joll~s, J., Mazurier, J., Schoentgen, F., Legrand, D., Spik, G., Montreuil, J. and Joll6s, P. (1984) Cur. J. Biochem. 145, 659-676. 11 Azari, P.R. and Feeney, R.E. (1958) J. Biol. Chem. 232, 293-296. 12 Kornfeld, S. (1969) Biochim. Biophys. Acta 194, 25-33. 13 Tsavaler, L., Stein, B.S. and Sussman, H.H. (1986) J. Cell. Physiol. 128, 1-8. 14 Hayashi, I. and Sato, G.H. (1976) Nature 259, 132-134. 15 Hutchinss, S.E. and Sato, G.H. (1978) Pro¢. Natl. Acad. Sei. USA 75, 901-905.
28
Elsevier BBA 12387
A growth-promoting factor for human myeloid leukemia cells from horse serum identified as horse serum transferrin K a o r u Y o s h i n a r i , K a t s u m i Y u a s a , F u s a o Iga a n d A k i o M i m u r a Medical Science Laboratory, Asahi Chemical Industry Co., Ltd., Shizuoka-416 (Japan)
(Received 8 August 1988)
Key words: Growth promotiag factor; Transferrin; Myeloid leukemia cell; (Horse serum); (Human cell)
A growth.promoting factor for human myeloid cells was purified to apparent homogeneity from horse serum by a combination of gel filtration, blue Sepharose affinity chromatography, Mono Q anion-exchange chromatography, Mono P chromatofoeusing and sodium dodecyl sulfate polyacrylamide gel electrophoresis. The growth promoter was an iron-bound, single glyeopolypeptide chain with a molecular weight of 84 000, an isoelectric point of 5.4 and an amino terminal sequence of Glu-Gln-Thr-Val-Arg-Trp-Cys-Thr-Val-Ser-Asn-His-Glu-Val-Ser-Lys-. According to the results of the amino acid sequence, iron binding ability and physlcochemical properties, we identified the growth-promoting factor as horse serum transferrin. It was highly active in promoting the proliferation of a human monocytic leukemia cell line, THP-I, as well as of two other human myeloid cell lines, Hi.~60 and K-567. it had the same activity in proliferating THP.I cells as 5% fetal calf serum-supplemented medium. Horse serum transterrin could be substituted for human or bovine serum transferrin. Introduction It has been known for some time that some types of cell grow more efficiently in media supplemented with horse serum than with fetal calf serum or other types of animal serum [1], whereas this is not the case for other types of cell. In addition, horse serum is preferred as opposed to calf serum by some workers as it can be obtained from a closed herd and batch to batch consistency is often greater. Identification of the growthpromoting factors of horse serum itself has been of considerable interest. Over the past few years, the use of serum-free media in cultivating a variety of cells has been reported and such studies have been reviewed by Sato [2]. Recently, proliferation and differentiation of human myeloid leukemia cells in serum-free medium have been studied using HL-60 [3], ML-1 [4] and K-562 [5] cells. In the process of screening for growth-promoting factors in order to construct a serum-free medium for Abbreviations: PBS, phosphate-buffered saline (138 mM sodium chloride/2.7 mM potassium chloride/8 mM dibasic sodium phosphate/1.5 mM monobasic potassium phosphate, pH 7.4); SDS-PAGE, sodium dodecyi sulphate polyacrylamide gel electrophoresis; HS, horse serum; FCS, fetal calf serum; GPF, growth-promoting factor. Correspondence: K. Yoshinari, Medical Science Laboratory, Asahi Chemical Industry Co., Ltd., 2-1, Samejima, Fuji-shi, Shizuoka-416, Japan.
culturing a human monocytic leukemia cell line, THP-1, originally established by Tsuchiya et al. [6], horse serum was found to be more effective in proliferating THP-1 cells than fetal calf serum. Therefore, we have tried to isolate one of the growth factors involved in horse serum and to construct a serum-free medium for culturing THP-1 cells. In this report, we describe the procedures used both to isolate a growth-promoting factor from horse serum and to identify it as horse serum transferrin. Materials and Methods
Materials. Materials were obtained from the following sources: RPMI 1640 medium, fetal calf serum and human serum from Flow Laboratories, VA, U.S.A.; horse serum from the Nippon Biosupplements Center, Tokyo, Japan; bovine insulin, proteinase (from Streptomyces griseus) and Concanavalin-A Sepharose (Con A-Sepharose) 4B from Sigma Chemical Company, MO, U.S.A.; ethanolamine, sodium selenite from Nakarai Chemicals Ltd., Japan; bovine serum albumin (fraction V) from Calbiochem Boehring, CA, U.S.A.; FeC-Test Wako from Wako Pure Chemical Industries, Japan; blue Sepharose CL-6B, Superose 12 HR10/30, Mono Q HR5/5, Mono P HRS/20 and polybuffer 74 for chromatofocusing from Pharmacia Fine Chemicals, Sweden. Cell lines. THP-1 cells (a human monocytic leukemia cell line) were kindly provided by Dr. Tsuchiy.~. THP-1
0167-4889/89/$03.50 © 1989 Elsevier Science Publishers B.V. (Biomedical Division)
29 (AH-01) cells are a subline selected from THP-1, which have been adapted for growth in low serum-containing medium, i.e., RPMI 1640 medium supplemented with a double concentration of amino acids and vitamins, 10 p g / m l bovine insulin, 10 /tg/ml ethanolamine, 0.05 /~g/rnl sodium selenite and 0.1% HS. The HL-60 and K-562 cell lines were maintained in serum-containing medium, i.e., RPMI 1640 medium supplemented with a double concentration of amino acids and vitamins and 5% FCS. All cells were cultured at 37°C in a humidified atmosphere of 5% CO2 in air, and were passaged every 3-4 days at an inoculum size of 2- 105 cells/ml. Cell proliferation assay. For proliferation assays, exponentially growing THP-I-(AH-01) cells were inoculated into 60 mm plastic tissue culture dishes at a density of 1.5.105 cells/ml in serum-free medium (RPMI 1640 medium containing a double concentration of amino acids and vitamins, 5 ~tg/ml bovine insulin, 10 /~g/ml ethanolamine and 0.05/xg/ml sodium selenite) supplemented with any one of the test samples. Cell counts after 3-6 days of cultivation were performed in duplicate with a Coulter Counter (Coulter Electronics Inc., U.S.A.). 1 unt of activity was defined as the reciprocal of the sample dilution causing 50% maximal growth stimulation (net response). Isolation procedure. 0.25 ml of HS heat-inactivated at 56 *C for 30 min was applied to a column of Superose 12 HR10/30, followed by elution of PBS at a constant flow rate of 0.4 ml/min (1 fraction per min). Pooled bioactive fractions were subjected to a blue Sepharose CL-6B (2.6 × 10 cm) affinity column equilibrated with PBS at a flow rate of 30 ml/h. Pooled fractions containing activity which did not bind to the blue Sepharose column were concentrated and buffer-exchanged into 10 mM Hepes-HCl (pH 7.4) using a polysulfone capillary tube. This pool was then applied to a Mono Q HR5/5 anion-exchange column that had been equilibrated with 10 mM Hepes-HCl (pH 7.4), and was eluted at a flow rate of 0.5 ml/min using a linear gradient of sodium chloride the concentration of which was between 0 and 1 M. Fractions of 1 ml were collected. The most active fraction was desalted and applied to a chromatofocusing column of Mono P HR5/20 (0.5 x 20 cm) pro-equilibrated with 25 mM imidazole HCI (pH 7.4) and was eluted with 5-fold diluted polybuffer 74 (pH 4.0, adjusted with HCI) to form a linear pH gradient between pH 7 and 4 at a flow rate of 0.5 ml/min (fractionation was performed on every peak). As a final purification step, polyacrylamide gel electrophoresis in the presence of SDS under non-reducing conditions was performed using 10% polyacrylamide gels according to the method of Laemmli [7]. The total protein content of all samples was determined by the dye-fixation assay (Tonein TP, supplied by Otsuka Assay Laboratories, Japan), using HSA (human serum albumin) as a standard.
Effects of temperature and of chemical and proteinase treatments on growth-promoting activity. The sensitivity of Mono-Q-purified samples to various temperatures and chemical and enzymatic treatments was tested essentially according to the method of Mochizuki [8]. Briefly, aliquo~s of samples were incubated at 56°C, 70°C or 100°C for the indicated length of time. The effects of pH on activity were studied as follows: samples were incubated in different buffers for 6 h at 4 ° C, were dialyzed against PBS and then assayed for cell growth-promoting activity. The pH 2.0, pH 4.0 and pH 6.0 buffers were Na2HPOJcitric acid buffers; the pH 7.4 buffer was PBS, the pH 8.0 buffer was a Tris-HCl buffer and the pH 10.0 buffer was a Na 2CO3/NaHCO3 buffer. The sensitivity to proteinase extracted from Streptomyces griseus was determined by testing samples at a concentration of 0.29 mg/ml in PBS with 400 p g / m l of the proteinase at 37°C for the indicated lengths of time. Growth-promoting activity was directly assayed without removing the proteinase, because it was previously found that at the concentration used in these experiments, it had no effect on the cell proliferation assay. Amino acid sequence analysis. Purified sample was concentrated by ammonium sulfate precipitation (80% saturation), dialyzed against twice-distilled water, lyophylized and finally re-dissolved in water at a concentration of 1 mg/ml. Amino acid sequence analyses were performed using an Applied Bio-systems 470 gasphase protein sequencer (Applied Biosystems, U.S.A.). Firstly, 360 pmol of sample were applied to the sequencer for 25 cycles (Table III, Ept. 1) then, in order to identify the cysteine residues, 130 pmol were carboxymethylated and applied to the sequencer for 21 cycles (Table III, Expt. 2). The phenylthiohydantoins of the amino acids were identified by high-performance liquid chromatography rising a Zorbac ODS C~s column. Iron analyses. Iron concentrations were determined using Wako kits (Wako Pure Chemical Industries, Japan) according to the manufacturer's instructions. Briefly, free ferric irons, to which ferrous ions were reduced by thioglycolic acid, were chelated with a specific reagent, 2-nitro-5-(N-propyl-N-sulfopropylamino)phenol to form a complex with peak absorbance at 750 rim. The iron content of a sample was thus measured spectroscopically by its absorbance at 750 rim.
Results
Growth promoters for human monocytic leukemia cells The growth-promoting abilities of HS, FCS and human serum were compared. As shown in Fig. 1, it was found that with FCS a concentration of at least 1% (i.e., 600 pg protein per ml) was required for full growth of
30 67k 43k 12,4k
15 A
2F
10
e
s U}
e~
i5
1
4
¢,j
vX
e) Z
(1
I
I
1
I
1'0
i
100
I
1()00 3000
Prote~ Conc.( ~glml ) Fig. 1. Proliferative effect of various sera on the growth of THP-1 (AH-01) cells. The proliferation assay was performed as described in Materials and Methods. Samples tested were HS (76 mg/ml, ®), FCS (60 mg,/ml, A), and human serum (84 mg/ml, o) at the indicated concentrations, Proliferation was measured on day 6. Standard de~'iations were less than 10~ of the mean from duplicate determinations.
THP-1 cells, while for HS and for human serum, only 0.02~ (i.e., 15/~g/ml) and 0.005~ (i.e., 4.2/tg/ml) were sufficient, respectively. Although human serum was 3.6-fold more potent than HS in enhancing proliferation of the human cell line, THP-I-(AH-01), HS was chosen as the source of growth promoters to be used in subsequent studies, as with HS, no problems about large-scale supply and pathogenic viruses to humans are encountered. Since HS had a greater THP-1 cell growth-promoting activity than FCS, and since several lots of HS (13 lots tested) were found to have the same level of activity (data not shown), studies were initiated to isolate the growth-promoting substances from HS.
Isolation of growth promotingfactor Elution of heat-inactivated HS (56 o C, 30 min) from a Superose 12 gel-filtration column, showed that the growth-promoting activity had a molecular weight between 30000 and 100000 (Fig. 2). A growth-inhibitory activity was seen in the high-molecular-weight fractions 20 and 21 when the serum had been heat ~l~c,'~vated. These fractions also contained a corresponding protein peak, indicating that the inhibitory effect was caused by aggregation of macromolecules. When pooled, the growth-promoting fractions contained 99~ of the total starting activity and the protein purity was found to be 2.8-fold more than the starting material. Further purificatio~t was achieved by passing the active fractions through the blue Sepharose CL-6B column. Activity eluted with PBS was detected in the unbound portion, 90~ of the material being ~hus reco~ered, with a further purification factor of 2.6. Better resolution was achieved using the Mono Q HRS/5 anion-exchange column. Fig. 3 shows that this step yielded three fractions containing activity (fractions 15-17). The most active fraction, number 16. was
0
'
2~0
40 60 80 Fraction Number Fig. 2. Gel filtration of HS on Superose 12. HS (0.25 ml, 19 mg) was applied to a Superose 12 column under the conditions described in Materials and Methods. Protein elution was monitored continuously tlu'ough its absorbance (Abs.) at 280 nm (solid line). The activity was measured by the THP-I-(AH-01) cell proliferation assay, using a 10 -3 dilution of a test sample. The results obtained on day 3 are ~nown (o o). The broad peak of activity observed (fractions 32-37) was pooled for the next purification step.
further purified by chromatofocusing on the Mono P HR5/20 column. Activity was observed in both peak fractions eluted at pH values of 5.5 + 0.1 and 5.4 + 0.1 (Fig. 4) (no activity in other fractions was observed). Analysis of the most active fraction (pl 5.4) by SDSPAGE under reducing conditions showed that the material consisted of a single silver-stained band with a molecular weight of 84000 (Fig. 5, lane 5). A single band of 84000 was also detected by SDS-PAGE under non-reducing conditions (data not shown), and when excised from the gel and eluted with PBS its activity was found to have an overall yield of 8.4~ (Table I). These data indicated that a growth-promoting factor had thus been purified to homogeneity.
10
!
in
0
0 20 30 Fraction Number Fig. 3. Anion-exchange chromatography on a Mono Q column. The blue Sepharose unbound protein (6.4 mg), dissolved in 10 mM HepesHCi (pH 7.4) was applied to a Mono Q column and was eluted in 1 ml fractions using a linear NaC! gradient, as indicated by the broken line, under the conditions described in Materials arl¢ Methods. Protein elution was monitored through its absorbance (Abs.) at 280 nm (solid line). Activity was measured by the THP-I-(AH-01) cell proliferation assay, at a dilution of 10 -2. The results obtained on day 3 are shown (o o). The most active fraction (number 16) was pooled (six runs) for chromatofocusing.
10
31 TABLE 1
Purificationof growth-promotingfactor Protein was determined by the dye-fixalion assay as described in Materials and Methods. Activity was defined as described in Materials and Methods. Step
Volume (rid)
Portein (mg)
Total activity ( x 104 U)
Spec. at' (U/mg)
Purification (fold)
Overall yield (~)
HS Superose 12 Blue-Sepharose (unbound) Mono Q Mono P SDS-PAGE
4.25 27 71 6.1 5.5 1.0
323
9.4 9.5 8.5 1.9 2.$ 0.8
300 830 2 200 9403 10 500 10400
1 2.8 7.3 31 35 "45
100 99 89 20 21 8.4
!
\
0
=o 0.5
1] 3
38 2.0 1.9 0.77
% v
2: Q.
O3 .0
<
I
~
'
I
I
10
t
20
I
30
Elution Volume ( ml )
Fig. 4. Chromatofocusing on a Mono P column. The desalted activity (0.28 mg protein) obtained by Mono Q chromatography, was applied to a Mono P column and was eluted with a linear pH gradient ( × . . . . . . ×) under the conditions described in Materials and Methods. Protein elution was monitored by its absorbance (Abs.) at 280 nm ( ). Every peak was fractionated. Activity is indicated by the arrows.
Measurement of the growth-promoting activity of samples taken from each purification step showed that at most 0.65 pg of the factor per ml were required to induce maximum growth promotion in THP-I-(AH-01) cells (Fig. 6). As can be seen in Table I and Fig. 6, the overall yield of activity and the final purificatio~l factor were 8.4~ and 35-fold, respectively.
Physicochemical properties of purified growth-promoting factor (GPF) G P F was s t a b l e a t 5 6 - 7 0 ° C for 30 m i n , b u t p r o l o n g e d t r e a t m e n t r e d u c e d a c t i v i t y e v e n at 56 o C. B o i l i n g o f G P F for 5 m i n r e s u l t e d in a c o m p l e t e loss o f a c t i v i t y ( T a b l e II). O n t h e o t h e r h a n d , the results o f the p H s t a b i l i t y a n a l y s i s s h o w e d t h a t G P F was s t a b l e o n l y b e t w e e n p H v a l u e s o f 6 a n d 8. As s h o w n in T a b l e II, t r e a t m e n t of G P F w i t h p r o t e i n a s e for 16 h d e s t r o y e d its b i o l o g i c a l activ~i3, ,=c::~?letely, i n d i c a t i n g t h a t it was a proteinaceous material.
... 1 5 -
"~m
94k.~
SDS-PAGE
~'~"84k 67
"~
& Z
4 3 "~
!
°.Ol
M 1 2
3
4
5
Fig. 5. 10~ SDS-polyacrylamide gel electrophoresis of samples at different stages of purification. Samples at each purification step were analyzed by 10~ SDS PAGE under reducing conditions. (M) molecular weight standards: lane 1, HS; 2, Superose 12 fraction; 3, blue-Sepharose unbound fraction; 4, Mono Q fraction; 5, Mono P fraction. The gel was stained by the silver nitrate method. The molecular weight standards were phosphorylase b (94000), bovine serum albumin (67000) and ovalbumin (43000).
o.1
1
;o
' 100 '
Protein Conc.(/~g/ml)
Fig. 6. Comparison of the ability of GPF preparations at various stages of purification to promote proliferation of the THP-I-(AH-01) cells. Proliferation was measured on day 5 as described in Materials and Methods. Samples tested were HS (e), the Superose 12 fraction (e), the blue-Sepharose unbound fraction (&), the Mono Q fraction (zx), the Mono P fraction (o) and the SDS-PAGE fraction (ra). All assays were performed in duplicate and standard deviations were less than 10¢$ of the mean.
32 TABLE II
TABLE II1
Effects of temperature, pH and proteinase treatment on GPF activity
Amino terminal amino acid sequence
Each treatment was performed as described in Materials and Methods. Cell proliferation was measured on day 5. 100~ and zero residual activities represent cell concentrations reached on day 5 of either 1.2.106 celis/ml or of only the original inoculum size of 1.5.105 ceils/ml, respectively. AH values are means of duplicate determinations and standard deviations were less than 105 of the mean.
In Expt. 1 360 pmol of G P F were applied to an amino acid sequences. In Expt. 2 130 pmol of G P E were carboxylmethylated and applied to a sequencer in order to identify the residue at position 7.
Residual activity
Treatment
(~)
None (pH 7.4)
100
560 C 56 ° C 56°C
60rain
98 100 65
70"C
10rain
10 rain
30 rain
700C 30 rain 70°C 60 rain
98 92 62
Boiling I rain Boiling 2 rain Boiling 5 rain
77 15 0
pH 2.0 pH 4.0 pH 6.0 pH 8.0 pill0.0
4°C 4°C 4°C 4°C 4°C
6h 6h 6h 6h 6h
Proteinase 400 pglml Proteinase 400 p g / m l
37 62 88 88 0
3h 16 h
14 0
Concanavalin A binding ability The possibility that GPF was a glycosylated protein was investigated using Con A-Sepharose 4B beads suspended in PBS. Growth-promoting activity could not be eluted with PBS, but could with PBS containing 0.3 M a-methyl mannoside (data not shown), showing that it had a glycosyl moiety containing mannosyl and/or glucose residues. Amino acid analysis The results are shown in Table III. As position 7 of the amino terminal sequence could not be identified by the first analysis (Table III, Expt. 1), the sample was carboxymethylated in order to determine whether this amino acid was a cysteine residue (Table III, Expt. 2). As a result of the two analyses conducted, a sequence of 16 consecutive amino acids located at the protein's amino terminal end was determined, as shown in Table IV. This sequence was found to share strong sequence homology with the corresponding part of the human transferrin molecule [9] (Table IV); ten out of the 16 were identical (62% homology). In addition, the homology between terminal amino acid residues of GPF and human lactoferrin [10] was 44% (seven residues out of 16).
Position
Amino acid
1 2 3 4 5 6 7 8
Glu Gin Tht Val Arg Trp Cys Thr Vai Ser Asn His Giu Val ~er Lys X Ala X Phe Arg X X Met Lys
9
10 11 12 13 14 15 16 17 18
19 20 21 22 23 24 25
Yield (pmol) Expt. 1
Expt, 2
320 175 48 170 70 _ a _ b 34 140 9.8 78 29 100 77 3.3 46 41 32 24 17 13
70 58 19 70 21 _ 47 12 53 2.7 18 3.9 21.5 27 2.5 10 -
14 1.! .5 4.9
a Identified but not quaatitated. b Not identified.
Iron-binding ability of GPF Due to its structural similarity to human transferrin, the iron-binding ability of G P F was tested using a FeC-Test Wako kit to observe whether those molecules formed a family of iron-binding proteins. It was found to bind to iron with a molar ratio (iron vs. GPF) of 1.73 (-I-0.13 in duplicate determi.--.:::_'_....~. As the same value was also obtained after saturation of this factor TABLE IV
Comparison of the amino terminal sequences of GPF and human transferrin (hTf) Amino acid sequences of G P F and human transferrin are from Table Ill and Ref. 12, respectively. Homologies between G P F and h u m a n transferrin are boxed by a solid line.
GPF
GLu-GLn~Thr'-VaL-Arg-Trp-¢ystThr~VaL-Ser- !
hTf VaL-Pro-Asp-Lys~Thr-VaL-Arg-Trp-Cys~ALa~VaL-Ser-] GLu~H~s-GLu~ALa-Thr~Lys~Cys-GLn-Ser~Phe-Wrg~His~-Lys- I
33 with iron, it was concluded that purified factor was saturated with iron, and contained approximately two iron atoms per molecule. Taking data mentio~ied above into consideration, GPF was identified as horse serum transferrin. In addition, analyses of the minor active fractions in the Mono Q ehromatogram (Fig. 3) revealed that the molecules in fractions 15 and 17 contained iron and protein at molar ratios of 1.68(+0.12) (fraction 15) and 1.81(+0.02)(fraction 17), that their p l values were 5.65 (fraction 15) and 5.35 (~action 17), and that the molecular weight of both proteins was around 84000 (data not shown). Thus, although amino acid sequence analyses of the materials in these two fractions were not performed, it is likely that they are subspecies of horse transferrin. Summation of the datas of the GPFs in all three fractions shows that its concentration in HS amounts to 1.5 mg/rnl; the concentration of human transferrin in plasma has been reported to be 2-4 mg/ml [11].
Biological characterization of GPF Fig. 7 presents the growth curves for THP-I-(AH-01) cells grown in media supplemented with either GPF, HS or FCS. As little as 1 ~g of GPF per ml was able to substitute almost completely for 5% HS, although at day 3 the effects of GPF and HS on the growth of THP-I-(AH-01) cells were slightly different. GPF was also found to support growth of THP-I-(AH-01) cells to
TABLE V
Effects of purified GPF on growth of HL-60 and K-562 cells The proliferation assay was performed as described in Materials and Methods. Inoculation size in this experiment was 2.105 cells/ml. Each cell suspension was supplemented with either 1/xg/ml GPF or 5% FCS. Cell proliferation was measured on day 5. All samples were made in duplicate and were counted with a Coulter counter. Cell line
~1L-60 K-562
Supplement GPF (1/~g/ml)
FCS (5%)
17.0 + 0.6 ( x 10 s cells/ml) 13.1 ±0.2
20.2 ± 0.5 16.1 ±0.3
much the same extent as did 5% FCS. Growth of THP-1 cells continued in this serum-free medium for over 100 days. In order to investigate whether GPF would promote the growth of other myeloid teukemia cell lines, the human promyelocytic cell line, HL-60, and the human erythroleukemia cell line, K-562, were cultivated in serum-free medium supplemented with GPF. As both these cell types had previously been routinely cultured in FCS-supplemented medium, the effects of GPF and 5% FCS on growth were compared. Table V shows that 1 # g / m l of GPF supported the growth of both HL-60 and K-562 cells at rates equal to 85% and 81%, respectively, of those achieved using FCS. Discussion
2°! . - 10
x
5
O~ .O
E z
/
o
10
I
I
I
3
5
7
Incubation Period ( Days )
Fig. 7. Comparative growth curves for THP-I-(AH-01) cells proliferated in medium supplemented with GPF, HS and FCS. THP-I(AH-01) cells were suspended in serum-free medium at a density of 1.5.105 cells/ml and the medium was supplemented with either 1 / t g / m l of GPF ( e e), 5% HS ( o o), 5% FCS (A. . . . . . A) or no additive (zx. . . . . . zx). Cultures were incubated at 3 7 ° C in an atmosphere of 5% CO 2 in air for 7 days. All points are means of duplicates.
This paper presents the results of a study on the purification and characterization of a growth-promoting factor (GPF), identified in horse serum. Comparison of the amino terminal sequences of GPF and human transferrin in this study shows that ten residues out of 16 were identical (62% homology), whereas homology between the portions of human and chicken transferrins sequenced so far have been reported to be only 49.8% [9]. Our study noted further close similarities between GPF and human transferrin in their physicochemical properties, GPF being a single glycoprotein chain iwth a molecular weight of 84000 and a pl of 5.4, and human transferrin having a molecular weight of 80000 and a variable pl of between 5.2 and 5.8. The first purification step, gel filtration, yielded material which eluted at molecular weight positions between 30000 and 100000 (Fig. 2), suggesting that it was smaller than the molecule found electrophoretically to have a molecular weight of 84 000 (Fig. 5). This could also be explained by the report of the tendency of the transferrin molecule to become more compact and spherical after it has bound iron [11,12]. From a biological point of view, Breitman et al. [3] reported that a human promyelocytic leukemia cell line,
34 HL-60, could grow in serum-free medium containing insulin and human transferrin at a rate of 80~;, compared to FCS-containing medium. Our serum-free medium containing bovine insulin and GPF could support the growth of HL-60 cells at a rate of 85% (Table V), compared to 5~ FCS medium. It should be further noted that human transferrin could be replaced by our purified factor, horse serum transferrin, in a serum-free system to support the growth of HL-60 and THP-1 (data not shown). The difference of these growth rates of HL-60 and THP-1 in GPF-supplemented media (85~; vs. 100~) (Table V, Fig. 7) can probably be attributed to adaptation of THP-1 cells to medium supplemented with a low concentration of horse serum. Alternatively, bovine transferrin, at a concentration of 5 /tg/ml, could not support the growth of either THP-1, HL-60 or K562 cells (data not shown). But Tsavaler et al. [13] reported that when growth curves of K562 cells were compared in serum-free medium containing either 300/~g/ml of bovine serum (equivalent to approx. 5~ FCS medium) or 300 ~tg/ml of human transferrin, the growth rates were the same. These facts are consistent with our data shown in Fig. 1, showing that the requirement of FCS was much greater than that of HS or human serum to proliferate THP-1 cells, considering the report [13] that a large amount of bovine sermn transferrin is necessary to effectively defiver iron intracellularly to K562 cells. In addition to human leukemia cells, many types of cell such as a rat pituitary cell fine, GH3 [14], and a human cervical cell line, HeLa [15], were reported to require transferrin as an essential component in serumfree medium. Thus, a growth-promoting factor from horse serum, identified as horse serum transferrin, would be a good tool for comparative study of cell proliferation.
Acknowledgements This work was supported by funds obtained under the Research and Development Project for Basic Technologies for Future Industries from the Ministry of International Trade and Industry of Japan. The authors wish to thank Dr. T. Hitouji for his amino acid sequence analyses. References 1 Endo, H. (1960) Exp. Cell Res. 21, 151-163. 2 Sato, G.H. (1982) in Cold Spring Harbor Conference on Cell Proliferation Vol. 9 (Sato, G., Pardee, A. and Sirbasku, D., eds.), pp. 1-624, Cold Spring Harbor Laboratory, New York. 3 Breitman, T.R., Collins, S.J. and Keene, B.R. (1980) Exp. Cell Res. 126, 494-498. 4 Taketazu, F., Kubota, K., Kajigaua, S., Shionoya, S., Motoyoshi, K., Saito, M., Takaku, F. and Miura, Y. (1984) Cancer Res. 44, 531-535. 50kabe, T., Fujisawa, M. and Takaku, F. (1984) Pro(:. Natl. Acad. Sci. USA 81, 453-455. 6 Tsuchiya, S., Yamabe, M., Yamaguchi, Y., Kobayashi, Y., Konno, T. and Tada, K. (1980) Int. J. Cancer 26, 171-76. 7 Laemmli, U.K. (1970) Nature (Lond.) 227, 680-685. 8 Mochizuki, D., Watson, J. and Gillis, S. (1980) J. Immunol. 125, 2579-2583. 9 MacGillivray, R.T.A., Mendez, E., Shewale, J.G., Sinha, S.K., Lineback-Zinpo, J. and Brew, K. (1983) J. Biol. Chem. 258, 3543-3553. 10 Metz-Boutigue, M.H., Joll~s, J., Mazurier, J., Schoentgen, F., Legrand, D., Spik, G., Montreuil, J. and Joll6s, P. (1984) Cur. J. Biochem. 145, 659-676. 11 Azari, P.R. and Feeney, R.E. (1958) J. Biol. Chem. 232, 293-296. 12 Kornfeld, S. (1969) Biochim. Biophys. Acta 194, 25-33. 13 Tsavaler, L., Stein, B.S. and Sussman, H.H. (1986) J. Cell. Physiol. 128, 1-8. 14 Hayashi, I. and Sato, G.H. (1976) Nature 259, 132-134. 15 Hutchinss, S.E. and Sato, G.H. (1978) Pro¢. Natl. Acad. Sei. USA 75, 901-905.