Journal of Reproductive Immunology 31 (1946) 51-64 ELSEVIER

oietie steno

e

Received 12 July 1995; revised 26 February 1996; accepted 28 March 1996

Abstract

It has been shown that murine amniotic fluid (MAF) displays immunosuppressive activi- ties. We examined the effect of MAF obtained from normal mice on a murine fetal liver cell (FLC) primary culture in vitro for the detection of the possible existence of cytokines that affect hematopoiesis. MAF promoted proliferation of murine FLC and adult bone marrctv cells. Proliferation-promoting activity of MAF was observed throughout the period between days 12 and 15 of pregnancy, peaking at day 12. Pooled MAF was subjected to purification procedures to isolate the active molecule(s). After ion exchange and size exclusion chro- matography, one of the active molecules was shown to react with anti-mouse stem cell growth factor (SCF) antibody and the molecular weight was determined as 40-45 kDa by sodium dodecylsulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and Western blot. The proliferation-promoting activity of MAF was partially neutrdhsed by the anti-SCF antibody. These results provide evidence that MAF may contain multiple growth factors for FLC. one of which may be a unique molecule related to SCF.

Keyrror&ls: Hematopoiesis; Stem cell growth factor: Amniotic fluid

* Corresponding author, Dept. of Immunology, Nagoya University School of Medicine, 65 Tsurumai-cho, Showa-ku, Nagoya 466, Japan.

0165-0378/96/$15.00 0 1996 Elsevier Science Ireland Ltd. All rights reserved PII SO 165-0378(96)00968-O

52 2. Heidari et al. / Journal of Reproductive Immunology 31 (1996) 51-64

1. Introduction

The formation of amniotic fluid (AF) is a complex process controlled by both mother and fetus. It consists of several different molecules which are secreted by placental and fetal cells (Hunziker and Wegmann, 1986; St&l&n and Wegmann, 1987). Murine amniotic fluid (MAF) has been shown to exhibit transforming growth factor-p (TGF-P)-like activity in stimulating soft agar colony formation by AKR-2B cells (Altman et al., 1990), to inhibit both primary and secondary antibody responses, and to suppress cell-mediated reactions including Con A responses and mixed lymphocyte reactions (Ogra et al., 1974). In a previous study, we examined the effect of MAF on alloreactive cytotoxic T lymphocyte (CTL) responses (Parvin et al,, 1994). We observed that MAF taken from syngeneic C57BL/ 6 pregnant mice or allogeneic C57BL/6 x BALB/c pregnant mice enhanced the anti-H-2d or anti-H-2k CTL responses in a dose-dependent manner. We speculated that MAE may contain not only immunosuppressive elements but also immuno-enhancing factors which upregulate the allogeneic CTL responses.

We have focused our study on the regulation of fetal hematopoietic stem cells by MAF. Earlier investigations have demonstrated that AF affects the development of the gastrointestinal tract (Mulvihill et al., 1986; Morikawa et al., 1988), suggesting the presence of trophic factor(s) in the AF. Morikawa et al. (1994) further reported the presence of a growth factor-like substance in AF in the rat during late gestation, which affects the develop- ment of the colonic goblet cells in the rat fetuses. Significant colony stimulating activity-l (CSA-l)-like activities of normal MAF have also been reported (Jyonouchi et al., 1987). In addition. some reports indicated the presence of epidermal growth factor (EGF) in human AF (Scott et al., 1989; Hoffman and Abramowicz, 1990).

The present study demonstrates that the AF obtained from the second half of the mouse pregnancy induces strong proliferation of fetal hemato- poietic stem cells, and one of the factors displaying this effect in MAF is an apparently unique stem cell growth factor (SCF) molecule.

2. Materials and methods

2. I. Source oj’ MAF

Female and male C57BL/6 (B6) strain of mice were bred in the Labora- tory of Animal Research, Nagoya University School of Medicine (Nagoya,

Japan). AF was obtained from the 12t mice and was centrifuge centrifuged again for 10 min a nents and kept at - 20°C until use.

ecombinant mouse purified by ion e

uced in yeast and

Genzyrne, Cambridge,

tested this antibody on recombinant human hepatocyte growth factor estern blot experiment to detect t y cross reactiv-

ity with other cytokines.

Murine fetal liver at the 12th-15th day of pregnancy was removed aseptica and crushed in Eagle’s minimal essential medium (

harvest ne marrow cells, normal adult murine femurs were with MEM using a syringe with a 21G needle. T fetal liver cells (FLC) or bone marrow cells were suspended in R I- 1640 medium (GIBC0, Grand Island, NY) containing IO’% heat-immobilized (56°C. 30 min) fetal calf serum (FCS), 2000 mM glutamine, 100 U,‘ml of cillin, and 100 /(g/ml streptomycin. They were cultured in 96-welI bottomed plates at a concentration of 1 x 10hsm! (1 x IO”100 pl) in

the medium. Different volumes of MAF were added to the culture and were incubated for 2 days at 37°C in a .5’:/;1 CO-, humidified inc All wells were pulsed with 0.5 /lCi of ~‘~]thymidi~~e for 6 h harvesting onto a glass fiber filter for liquid scintillation counti uptake of [‘Hlthymidine was expressed as the mean countsmin

of triplicate cultures.

54 Z. Heidnri et ul. / Journul of’ Reproductioe In~m~~olo,~~ 31 (1996) 51-64

2.4. Procedures for puriJication

2.4.1. Ion exchange chromatography All subsequent steps were performed at 4°C. MAF was applied to a

DEAE-Sephacel ion exchange column (1.6 x 10 cm, Pharmacia, Upp- sala, Sweden) previously equilibrated with 25 mM Tris-HCl (pH 7.0). The column was washed with starting buffer (25 mM Tris-HCl pH 7.0) and eluted with a gradient from buffer A (25 mM Tris-HCl pH 7.0) to buffer B (25 mM Tris-HCl pH 7.0, containing 1 M NaCl). Fractions were collected and the FLC proliferation assay was carried out to moni- tor all purification steps.

2.4.2. Size exclusion chromatography Fractions showing high bioactivity were concentrated by ultrafiltration

and samples were applied to Superose 6 gel (Pharmacia, Uppsala, Swe- den). The column was eluted with 25 mM Tris-HCl, pH 7.0, containing 0.1 M NaCl. Fractions were collected and assayed on FLC to monitor their bioactivity.

2.5. Western blot analysis

To demonstrate SCF in MAF, MAF obtained from different days of mid-gestation or rSCF as control was mixed with an equal volume of sample buffer (62.5 mM Tris-HCl pH 6.8, 2% SDS, and 10% glyc- erol) as previously described (Nakashima et al., 1993). They were sub- jected to SDS-PAGE and were then transferred to polyvinylidene (Millipore, Bedford, MA) membrane. The protein transferred membrane was incubated with polyclonal rabbit antimouse SCF antibody followed by ‘*‘I-labeled protein A (ICN, Irvine, CA). The autoradiograph was visualised using X-ray film over a l-5 day exposure. The molecular size of the stained protein was estimated by comparing with protein molecu- lar weight standards (Gibco, Gaithersberg, MD).

2.6. Statistical analysis

All numerical data were expressed as mean values + S.D. Statistical analysis of these data was performed using the Wilcoxon test. Values with a confidence level of P < 0.05 were considered significant.

AF on proliferation of murine FLC in primary culture in the presence or absence of Con A was evaluated. significantly promoted the proliferation of FLC as compared wit negative (no MAF) control, whereas Con A did not show any significant proliferation- inducing activity on FLC (Fig. JA). Similar effects were observed when

AF was tested on FLC proliferation in the presence or absence of FCS (Fig. IB). It was noted in these experiments that oliferation of murine FLC in primary culture was definitely promoted by AF, but not by either

mitogen or serum. It was found that MAF obtained from the 12th day of pregnancy had the

highest ability to promote FLC proliferation and this activity decreased as the pregnancy progressed u e end of the second alf of pregnancy ( 2A). The same result was ed when MAF wa tested on adult B proliferation in culture (Fig. 2B).

lmmunoblotting of AF obtained from the second half of pregnancy revealed a definite band with an apparent molecular weight of 40-45 kDa that specifically reacted with anti-SCF antibody which was not developed by control normal mouse serum (Fig. 3A). This antibody was confirmed to react with rSCF of unglycosylated monomer form, which was represented at the 18-20 kDa molecular weight (Fig. 38). No band was revealed when the antibody was tested on rHGF by a Western blot experiment (data not shown).

To determine the possible relation of the active molecule in MAF for hematopoietic growth promotion to SCF, partial purification of the active element was carried out from the pooled MAF. A total of 35 ml AF was collected and centrifuged to remove cellular components. The samples were then applied to the ion exchange chromatography. Fig. 4A shows biological activity of each fraction from the DEAE-Sephacei chromatography. For further purification, the active fractions obtained from a DEAE-Sephacel column (fraction no. 30, 31) were applied to a Superose 6 column and the distribution of the bioactivity in the eluted fractions was determined (Fig. 48). As shown in Fl,. ‘ID. 5, immunoblot analysis against anti-SCF antibody of

56 Z. Heidari et al. / Journal of’ Reproductive Immunology 31 (1996) 51-64

A

Control

Amount of MAF (pl) added to 100 pl culture

No FCS

10% FCS

Amount of MAF (~1) added to 100 pl culture

Fig. 1. Proliferation-promoting effect of MAF on FLC in culture. Murine FLC ceils were cultured with the indicated amounts of MAF in the presence or absence of 5 &ml or 10 {lg/ml ConA (A) or of 10% FCS (B) for [‘Hlthymidine uptake assay. All the values for culture with MAF were significantly (P c 0.01) higher than those for the negative (no MAF) control cultures, irrespective of the presence or absence of Con A or serum. The figure shows a representative of 5 experiments with consistent results.

ay of pregnancy for

Day of pregnancy for MM collection

Fig. 2. MAF obtained from the 12th day of murine pregnancy shows the highest bioactivity. Murine FLC obtained from the 12th day of normal pregnancy (A) and adult BMC (B) were cultured with 5 /‘1 MAF obtained from the 12th-15th day of pregnancy for [%I]thymidine uptake assay. All the values for cultures with MAF are significantly (P < 0.01) higher than those for negative (no MAF) control cultures. The figure shows a representative of 5 experiments.

individual fractions from each step of purification revealed a single band at 40-45 kDa in some (fraction no. 30, 31 from the first step, and fraction no. 10 from the second step) but not all the active fractions. The difference in the molecular weight between the bands developed for MAF and control rSCF might result from post-translational modification of the former.

3.4. Partial neutmlisation of the bioactivity of MAF with anti-SCF antibody

As shown in Fig. 6, a part of the proliferation-promoting activity of MAF on FLC in culture was specifically ncutralised by rabbit anti-mouse SCF antibody compared with normal rabbit serum. The neutralisation occurred in a dose-dependent manner in the range of l-20 pgglml of antibody, and > 50% of the MAF activity was neutralised by 20 /[g/ml of the antibody. Larger amounts of the antibody could not, however, neu- tralise the residual activity of MAF.

4. Discussion

In the present study, it has been shown that MAF obtained from the second half of mouse pregnancy carries the activity to promote proliferation

175-

83 123456

62-

47.5-

325 t,

83- 62-

Fig. 3. Immunoblot analysis of MAF obtained at different days of the :>econd half of pregnancy (A) and control rSCF (B). The MAF proteins were separated by SDS-PAGE and blotted against anti-SCF antibody. (A) Lanes I-6 are MAF obtained from the 12th (lane 2). 13th (lane 3), 14th (lane 4), 15th (lane 5) and 16th (lane 6) day of pregnancy. Note the development of a definite band of 40-45 kDa, which was not developed by control normal mouse serum in place of anti-SCF antibody (lane 1). An additional faint band is seen at the 56-60 kDa position, but this band may not correspond to monomeric SCF because it did not appear in analysis of the purified active component (see Fig. 5).

Fraction number

s d 2, g UO .d -

2 x4 E E 2 e $ 2

0 . . . . . . . . . . . . . . . . . . . Ql 10 18 J

% Fraction number

Fig. 4. Bioassay of fractions of pooled MAF separated by ion exchange chromatography (A) and size exclusion chromatography (B). Samples were eluted as described in Section 2. The proliferation-promoting activity was measured by the FLC proliferation assay for [“HJthymidine uptake. The assay was repeated 3 times with consistent results.

of murine FLC and adult BMC. As fetal liver contains mainly hematopoi- etic stem cells, this activity is likely to be mediated by some growth factor or factors in MAF potentially related to SCF, a ligand for c-kit, whose effect on proliferation of hematopoietic stem cells has been clarified (Zsebo et al,, 1990a,b; McNiece et al., 1991; Vries et al., 1991; Metcalf and Nicola, 1991; Migliaccio et al., 1991; Williams et al., 1992; Bodine et al., 1992). In

60 2. Heidari et al. 1 Journal of Reproducticc bnmunolog~~ 31 (1996) 51-64

our study, the proliferation-promoting activity on FLC and BMC was demonstrated with MAF obtained between days 12- 15 of pregnancy. The activity was highest with MAF obtained from the 12th day of pregnancy (Fig. 2). These results suggest that the MAF during this period promotes the development of the fetal liver by accelerating proliferation of hemato- poietic stem cells. This conclusion is consistent with the previous report that hematopoietic activity occurred during mid-gestation, with erythropoiesis being predominant during day 11 through 16, and by day 18 and thereafter the liver assumes adult function (Yu et al., 1993). Our results also corre- spond to the suggestion of Yu et al. (1993) that MAF contains some hematopoiesis promoting factor(s) through these days.

We have provided evidence that the bioactive molecules in MAF include SCF or SCF-related growth factor, which was previously isolated from another source (Zsebo et al., 1990a). Firstly, our immunoblot assay of MAF revealed a protein band reacting with anti-SCF antibody whose specificity was confirmed on rSCF. Secondly, the bioactive fractions from both DEAE-Sephacel and Superose 6 chromatography separation contain the anti-SCF antibody-reactive protein. Thirdly, the MAF activity was partially neutralised by the anti-SCF antibody. This appears to be the first demon- stration and isolation in MAF of SCF or SCF-related growth factor which shares the common antigenic epitope with the previously reported SCF

1234 1234567 175-

83-

62-

47.5-

32.5-

25-

16.5-

Fig. 5. Immunoblot analysis of bioactive fractions of MAF. Purified fractions showing high bioactivity, which had been obtained from ion exchange chromatography (A) and size exclusion chromatography (B), were subjected to SDS-PAGE followed by immunoblotting against anti-SCF antibody. (A) Lanes 1-4 are for fractions no. 27, 28 (lane l), no. 30 (lane 2), no. 31 (lane 3), no. 33, 34 (lane 4). (B) Lanes l-7 are for fractions no. 7 (lane I), no. 8 (lane 2), no. 9 (lane 3), no. 10 (lane 4), no. 11 (lane 5), no. 12 (lane 6), no. 13 (lane 7). Immunoblotting was repeated 3 times with the same results.

Fig. 6. Partial neutralisation of the proliferation-promoting activity of MAF with anti-SCF antibody. FLC were cultured with 5 /tl MAF in the presence or absence of I-50 ilgjrnl of anti-SCF antibody (for specificity, see Section 2.) or equivalent amount of normal rabbit serum for [3H]thymidine uptake (countsimin) assay. Percent neutralisation of the MAF activity was calculated by dividing countsjmin of the culture with anti-SCF antibody by counts/min of the culture with equivalent amount of normal rabbit serum. The values from the cultures with anti-SCF antibody (P < 0.05 for 1 and 10, and P < 0.01 for 20 and 50 jig/ml of antibody) were significantly lower than those from the cultures with the same concentration of normal rabbit serum. The figure shows a representative of 5 experiments with consistent results.

from rat liver. This conclusion accords with the recent demonstration of the ‘localization of SCF mRNA in human placenta (Saito et al., 1994), human materno-fetal interface (Sharkey et al., 1994) and porcine endometrial uterus (Zhang and Anthony, 1994), all of which could secrete SCF into MAF.

The molecular weight of SCF demonstrated in MAF with the anti-SCF antibody (40-45 kDa) was greater than that of rSCF (18.5 kDa) (Zsebo et al., 1990a), probably because of post-transcriptional modification of the former with glycosylation. Zsebo et al. (1990a) however, reported that SCF secreted from the rat liver represents a molecule of 28-35 kDa by complete purification. Inconsistency of molecular size between the reported SCF secreted by rat liver (28-35 kDa) and SCF in MAF in our study (40-45 kDa) might result from alternate post-translational modificatiora of the molecules. Further studies are in progress to characterize the potentially tissue-specific modification of SCF with N or O-glycosylation.

62 Z. Heidari et al. / Journal of‘ Reproductive I~n~mmoio,gy 31 (1996) 51-64

The failure of complete neutralisation of MAF activity by anti-SCF antibody should be correlated to another observation that some fractions of MAF separated by ion exchange chromatography (fraction no. 28, 34) and by size exclusion chromatography (fraction no. 6-9) of MAF which barely contained an anti-SCF antibody-reactive element but displayed definite or even stronger activity to promote FLC proliferation. This result clearly suggests that MAF contains multiple growth factors effective for promoting proliferation of FLC, in addition to the SCF or SCF-related protein we characterized by use of anti-SCF antibody. Earlier studies reported that MAF contains several growth factors, such as CSA-1 (Jyonouchi et al., 1987) and EGF (Scott et al., 1989; Hoffman and Abramowicz, 1990). Some of these known cjrtokines or as yet uncharacterized new growth factors may be responsible for the remaining activity of MAF. In our experiment the anti-SCF antibody neutralized the MAF activity rather more extensively ( > 50%) than expected from the distribution of SCF in the fractionated MAF (Figs. 5 and 6) as a rather minor component. Earlier papers reported that SCF cooperates with other hematopoietic growth factors greatly to amplify the hematopoiesis i@ernstein et al., 1991; Metcalf and Nicola, 1991). Similar cooperation between the demonstrated SCF or SCF-related protein and other as yet uncharacterised growth factors may be operating in the action of MAF. We are planning to characterize further the molecular nature of the potentially new growth factor(s) other than SCF in MAF.

References

Altman, D.J., Schneider, S.L., Thompson, D.A., Cheng, H.L. and Tomasi, T.B. ( 1990) A transforming growth factor j&(TGF-/&)-like immunosuppressive factor in amniotic fluid and localization of TGF-P, mRNA in the pregnant uterus. J. Exp. Med. 172, 1391- 1401.

Bernstein, I.D., Andrews, R.G. and Zsebo, KM. (1991) Recombinant human stem cell falxor enhances the formation of colonies by CD34+ line-cells, and the generation of colony-forming cell progeny from CD34+ line cells cultured with interleukin-3, granulo- cyte colony-stimulating factor, or granulocyte macrophage colony-stimulating factor. Blood 77, 23 16-2321.

Bodine, D.M., Orlic, D., Birkett, N.C., Siedei, N.E. and Zsebo, K.M. (1992) Stem cell factor increases colony-forming unit-spleen number in vitro in synergy with interleukin-6, and in vivo in Sl/Sl mice as single factor. Blood 79, 913-919.

Hunziker, R.D. and Wegmann, T.G. (1986) Placental immunoregulation. CRC Crit. Rev. 6, 245-285.

Hoffman. G.E. and Abramowicz, J.S. (1990) Epidermal growth factor (EGF) concentration in amniotic fluid and maternal urine during pregnancy. Acta Obstet. Gynecol. Stand. 69, 217-221.

Jyonouchi, H., Voss, R.M and Dood, R.A. (1987) IL l-like activities present in murine amniotic fluid. A significantly larger amount of IL-l,j-like activity is present in the amniotic fluid of autoimmune NZB mice. J. Immunol. 138, 3300-3307.

McNiece, I., Langley, K.E. and Zsebo. K.M. (1991) Recombinant human stem ce]] facto1 synergizes with GM-CSF, G-CSF. IL-3 and El;0 to stimulate human progenitor cells of the myeloid and erythroid lineages. Exp. Hematol. 19, 226-231.

Metcalf. D. and Nicola, N.A. (1991) Direct proliferative actions of stem ce]] factor on murine bone marrow cells in vitro: effects of combination with Colony-stimulating factors_

Proc. Nat]. Acad. Sci. USA 88, 6239-6243. Migliaccio. G., Migliaccio, A.R.. Valincky, J.. Langley, K. and Zsebo, K. ( 1991) Stem cell

factor induces proliferation and differentiation of highly enriched murinc hematOpo&iC

cells. Proc. Nat]. Acad. Sci. LJSA 88. 7420-7424. Morikawa. Y., Shimonaka. H. and Okada. T. ( 1988) Evidence for stimulative effect of

amniotic fluid on the development of colonic goblet cells. Jpn. J. Vet. Sci. 50, ]]09- 1 I I 1.

Morikawa. Y.. Yoshimura, M.. Okada. T. and Ohishi, I. (1994) Growth factor-like sub- stance ICI amniotic fluid in the rat: effect on the development of fetal colonic goblet cells. Biol. Neonate. 66, IOO- 105.

Mulvihill. S.J., Stone, M.M.. Fonkalsrud, E.W. and Debas, H.T. (1986) Trophic effect of amniotic fluid on fetal gastrointestinal development. J. Surg. Res. 40. 291-296.

Nakashima, I., Pu. M.Y.. Hamaguchi. M., Iwamoto. T.. Rahman. S.M.J.. Zhang. Y.H.. Kato, M., Ohkusu. K., Katano, Y., Yosbida. T., Koga. Y.. Isobe, K.I. and Nagase, F. (1993) Pathway of signal delivery to murine thymocytes triggered by co-crosslinking CD3 and Thy-l for cellular DNA fragmentation and growth inhibition. J. Immunol. 151. 351 I-3520.

Ogra, S.S., Murgita, R.A. and Tomasi. T.B. Jr. (1974) Immunosuppressive activity of mouse amniotic fluid. Immunol. Commun. 3, 497.-500.

Parvin. M., Isobe, K.I., Heidari. Z., Goto, S.. Nakashima, I. and Tomoda. Y. (1994) Amniotic fluid enhances allogeneic cytotoxic T cell responses. whereas it suppresses mitogen-stimulated lymphocyte proliferation. Microbial. Irnmunol. 38. 327-330.

Saito. S., Enomoto. M.. Sakakura. S.. Ishii. Y., Sudo, T. and Ichijo. M. (1994) Localization of stem cell factor (SCF) and c-kit mRNA in human placenta] tissue and biological effects of SCF on DNA synthesis in primary cultured cytotrophoblasts. Biochem. Biophys. Res. Commun. 205. 1762- 1769.

Sharkey, A.M., Jokhi, P.P.. King. A.. Loke. Y.W.. Brown. K.D. and Smith. SK. (1994) Expression of c-kit and kit ligand at the human maternofetal interface. Cytokine 6.

195-205. Streilein. J.W. and Wegmann, T.G. (1987) Immunologic privilege in the eye and the fetus.

Immunol. Today 8. 362. Scott, S.M., Buenaflor. G.C and Orth. D.M. (1989) Immunoreactive human epidermal

growth factor concentration in amniotic fluid, umbilical artery and vein serum, and placenta in full-term and preterm infants. Biol. Neonate. 56, 246-251.

Vries, P., Bras& K.A., Eisenman. J.R., Alpert. A.R. and Williams, D.E. (1991) The effect of recombinant mast cc11 growth factor on purified murine hematopoietic stem cells. J. Exp.

Med. 173, i205- 1211. Williams. N., Bertoncello. I., Kavnoudias. H., Zsebo. K.M. and McNiece. 1. (1992) Recom-

binant rat stem cell factor stimulated the amplification and differentiation of fractionated

mouse stem cell populations. Blood 79. 58-64. Yu. H.. Bauer, B., Lipkc, G.K., Philips. R.L. and Zant. G.V. ( 1993) Apoptosis and

hematopoiesis in murine fetal liver. Blood 81. 373.-384. Zhang, Z. and Anthony, R.V. (1994) Porcine stem cell factor/c-kit ligand: its lllO]eCll]ar

Cloning and localization within the uterus. Biol. Reprod. 50. 95-102.

Zsebo. K.M.. Wypq’ch. J.. McNiece. I.K.. Lu. H.S., Smith. K.A.. Karkare. S.B.. Sachdev. R.K.. Yuschenkoff. V.N.. Birkett. NC.. Williams. L.R.. Satyagal, V.N.. Tung. W.. Bosselman. R.,i.. Mendiaz. EA. and Langley. K.E. ( IWOa) Identification. purification and biological characterization of hematopoietie stem cell factor from buffalo rat liver-conditioned medium. Cell 63, 195 201.

Zsebo, K.M.. Williams. D.A., Geissler. E.N.. Broudy, V.C.. Martin. F.H., Atkins, H.L.. Hsu, R.Y . . Brikett. N.C.. Okino. K.H.. Murdock. DC.. Jacobsen. F.W.. Langley, K.E., Smith. K.A.. Takeishi. T.. Cattanach. B.M.. Galli. S.J. and Suggs. S.V. (1990b) Stem cell factor is encoded at the Sl locus of the mouse and is the &and for the C-kit tyrosine kinasr receptor. Cell 63. 213 224.