Stem Cells, Vol. 14, No. 4, 452-459,
July 1996
© 1996 AlphaMed Press
Protection of the Small Intestinal Clonogenic Stem Cells from Radiation-Induced Damage by Pretreatment with Interleukin 11 also Increases Murine Survival Time
C.S. Potten
CRC Department of Epithelial Biology, Paterson Institute for Cancer Research, Christie Hospital NHS Trust, Manchester, United Kingdom
Key Words. Intestinal stem cells • Clonogenic cells • Radiation • 5-Fluorouracil • Survival time
Dr. C.S. Potten, CRC Department of Epithelial Biology, Paterson Institute for Cancer Research, Christie Hospital NHS Trust, Wilmslow Road, Manchester, M20 9BX, United Kingdom.
 |
Abstract
|
|---|
The effect of administering recombinant human interleukin 11 in conjunction with cytotoxic insults to the gastrointestinal tract has been studied using the crypt microcolony assay for stem cell function and whole-animal survival time studies. The cytotoxic regimens include single doses of 
rays; single doses of 5-fluorouracil (5-FU) and multiple doses of 5-FU spaced 6 h apart. Interleukin 11 (IL-11) (100 µg/kg) delivered over a period of time prior to cytotoxic exposure afforded protection to the clonogenic cells in the crypts as seen with the microcolony assay and prolonged the animal survival time following radiation exposure. Continuing this dose of IL-11 after cytotoxic exposure afforded little additional protection.
Three doses of 5-FU 6 h apart generated crypt survival curves similar to those obtained after a single dose of rays. IL-11 given prior to two doses of 5-FU effectively abolished the cytotoxic effect of the second dose of 5-FU; i.e., 2.5-3.0 times more crypts survived if IL-11 was administered when the higher 5-FU doses are considered. IL-11 given before a dose of 12 Gy of rays prolonged the survival time of animals by three to four days. This confirms earlier studies demonstrating that protecting clonogenic cells in the crypt survival assay can result in beneficial effects on whole-animal survival times.
|
Introduction
|
|---|
Recombinant human interleukin 11 (rHuIL-11) is a pleiotropic cytokine affecting many cell systems. Originally derived from stromal cells, it shares the GP130 signal transduction pathway with interleukin 6 (IL-6) and other factors. It also shares certain biological properties with IL-6 [1–3]. rHuIL-11 was cloned in 1990 [4] and has been mapped to chromosome 19 [5]. Du et al. [6] described how the propitious administration of interleukin 11 (IL-11) to mice exposed to a lethal regime of 5 fluorouracil (5-FU) and radiation resulted in their predominant survival for up to 30 days, while those not receiving the IL-11 all died within 10 days. The cytotoxic treatment involved a single dose of 150 mg/kg of 5-FU, three days before 6 Gy of 250 kV x-rays. The IL-11 (250 µg/kg) was injected s.c. twice daily, starting on the day of irradiation. The control animals died over the period of 3-10 days, suggesting that damage to the integrity of the gastrointestinal mucosa was responsible. Hendry et al. [7] illustrated that the dependence of animal survival within the LD50/6/7 timeframe was determined by the survival of individual intestinal crypts in the small intestine and that the survival of each individual crypt was determined by the levels of survival of individual clonogenic stem cells in crypts. The latter processes can be studied and assessed by the crypt microcolony assay [8–11]. This is an assay for the regenerative functional competence of intestinal clonogenic stem cells. We have recently demonstrated that administration of IL-11 prior to a range of single doses of radiation can markedly affect the survival curve for these intestinal clonogenic stem cells. At the highest radiation dose, up to four times more clonogens survived in animals pretreated with IL-11 [12].
Here, we demonstrate a similar protective phenomenon when IL-11 is administered prior to treatment with 5-FU. We also demonstrate that the protective action seen with the crypt microcolony assay for radiation is reflected in an extension of the survival time of whole animals.
|
Materials and Methods
|
|---|
Male BDF1 mice between 10 and 12 weeks of age were housed under conventional conditions with standard laboratory chow and water ad libitum. The animal rooms were temperature-, humidity- and light-controlled (12 h light cycle with the lights on at 0700 hours). For the crypt microcolony assay, groups of six mice were used for each dose. In the survival time studies, groups of 20 mice were used. All animal experiments have been performed within the framework of the Animals (Scientific Procedures) Act 1986, UK.
Cytotoxic Exposures
Animals were irradiated, two at a time, in a 137Cs -irradiator at a dose rate of 3.8 Gy/min in the morning (usually between 0900 hours and 1000 hours). For the radiation survival time experiments, a dose of 12 Gy was administered. Various regimes of 5-FU were administered. Doses are expressed on the basis of milligram doses per mouse. The mice weighed an average of 25 g. The highest dose of 15 mg/mouse is therefore equivalent to 600 mg/kg. For the clonogen survival curves, a range of single doses from 1.88 mg/mouse to 15 mg/mouse (75.2-600 mg/kg) were used. Survival curves were also generated for two doses of 5-FU spaced 6 h apart (5-FU x 2), 6 h being the approximate duration of the S phase for the mid-crypt intestinal cells. Three doses of 5-FU spaced 6 h apart have also been studied (5-FU x 3). The 5-FU animal survival time experiment involved a dose of 10 mg/mouse of 5-FU delivered twice with an interval of 6 h between doses. These two treatments (two doses of 10 mg 5-FU or 12 Gy of rays) result in nearly equal levels of approximately 30% crypt survival (Fig. 1
).
View larger version (14K):
[in this window]
[in a new window]
|
Figure 1. The surviving fraction of ileal crypts following single doses of -rays, single doses of increasing size of 5-FU, two doses of increasing size of 5-FU separated by 6 h and three doses of increasing size of 5-FU, each separated by 6 h. Single doses of radiation are shown for comparison. For the 5-FU data, the curves have been fitted using the Puck version of DRFIT [14] Fits are not shown, but the parameters describing these curves are as follows:
5-FU single dose, N = 1 ± 0.54, Do 21.0 ± 15.9 mg/mouse; 5-FU two doses, N = 1.40 ± 0.65, Do 5.4 ± 1.9 mg/mouse; 5-FU three doses, N = 3.68 ± 1.91, Do 2.5 ± 0.6; The single and two-dose curves are significantly different (p = 0.004). Data points are means ± SE of six mice. N is the zero dose extrapolation number, and Do is the reciprocal of the slope of the dose-response curve—a measure of the sensitivity of the target cells.
|
|
IL-11 Administration
2.5 µg/mouse (or approximately 100 µg/kg) of rHuIL-11 (from Genetics Institute, Inc.; Cambridge, MA) were injected s.c. according to one of the following protocols:
Protocol 1: Five injections of IL-11 were administered at 0900 hours and 2100 hours (a.m./p.m.) prior to exposure to the cytotoxic insult (radiation or 5-FU). The last of the five injections was administered immediately after irradiation or 1 h before the first 5-FU injection.
Protocol 2: Five injections of IL-11 were given in the morning and evening following exposure to radiation or 5-FU with the first of these injections being delivered within an hour of cytotoxic exposure.
Protocol 3: Three injections of IL-11 were administered prior to the cytotoxic exposure and continued on an a.m./p.m. delivery basis, for six injections (crypt survival) after exposure (protocol 3A) or for the entire duration of the experiment (protocol 3B).
Protocol 4: Five injections of IL-11 were given prior to the cytotoxic agent and a.m./p.m. injections continued throughout the post-treatment period (survival experiments).
Control animals received an equivalent number and volume of injections of saline containing 0.1% bovine serum albumin (BSA), which served as the vehicle for the IL-11. The dose of IL-11 was selected to be lower than in previous studies [3] and was based on experience in a wide variety of preliminary studies at Genetics Institute, Inc.
Fixation and Histology
For the crypt microcolony assay, the animals were sacrificed four days after cytotoxic exposure, and the entire small intestine was fixed intact in Carnoy's fixative for 30 min prior to storage in 70% ethanol. The ileal region of the small intestine was cut into 10 segments and bundled in millipore tape as described elsewhere [9]. These bundles were embedded and sectioned to provide 10 transverse sections of the ileum stained with hematoxylin and eosin from each mouse (60 transverse sections from each experimental group).
The number of regenerating microcolonies was counted in the circumference of the transverse sections, which represents a convenient unit of length. The size (width) of representative longitudinal sections of microcolonies was assessed using an eyepiece micrometer, and this information was used to correct the raw data for variation in crypt size from experimental group to experimental group [9, 13].
The microcolony technique described originally by Withers and Elkind [8] has been used extensively [10] and the details have been described many times. A variety of computer curve-fitting programs is available. The survival curves generated here have been fitted to the data using the multi-target (Puck) model, DRFIT [14]. The parameters defining these survival curves have been determined with their standard errors, and the saline and IL-11 treated curves have been tested for significant differences. A standard variation ratio F-test was used.
Survival Data
Groups of 20 mice were treated and their well-being assessed four times in each 24-h period for up to 30 days. The numbers of deaths and moribund animals were assessed at each interval. Moribund animals were sacrificed.
The Kruskal-Walis test was used for the survival time data to test for differences between the various groups. The Mann-Whitney U-test was used to test for differences between two groups. Differences were regarded as significant if p < 0.05.
|
Results
|
|---|
5-FU kills primarily S phase cells and since only a portion of the crypt stem cells are in S phase at any given time, a single dose of 5-FU is not very efficient at destroying crypts. As can be seen in Fig. 1
, a single dose of 7.5 mg or higher sterilizes about 30% of the crypts. Increasing the dose in the range 7.5-15 mg has little increased crypt destructive action. The dose response following two doses of 5-FU spaced 6 h apart shows a steadily increasing destruction of crypts at all doses. Three doses of 5-FU delivered 6 h apart are more effective at killing crypts at doses of 5 mg/mouse or above. The response after two doses of 5-FU is somewhat similar to that seen after single doses of -rays. Similar levels of crypt survival (20%-30%) are seen after about 10 mg of 5-FU delivered twice, 7.5 mg of 5-FU delivered three times and a single dose of 12 Gy of -rays.
We have previously shown [12] that IL-11 delivered before irradiation or before and after irradiation shifts the radiation survival curve to the right, indicating a protective effect of IL-11 treatment. There appears to be little difference between the levels of protection afforded by IL-11 administered prior to irradiation and the protection afforded when IL-11 is given both before and after irradiation. In either case, about four times more crypts survive at the highest dose of radiation (18 Gy) in the IL-11 treated groups compared with saline-treated controls [12]. Fig. 2, as well as Figs. 3A and 3B, show that IL-11 given before and after two doses of 5-FU also affords significant protection. The levels of protection increase with increasing 5-FU dose reproducibly to reach a level equivalent to 2.5 to 3 times more crypts surviving in the IL-11 treated groups at the highest 5-FU dose of 15 mg/mouse (Fig. 3A). The IL-11 treated crypts shown in Fig. 2 result in a two-dose response curve very similar to that obtained after a single dose of 5-FU as shown in Fig. 1; i.e., the IL-11 treatment has effectively protected the intestinal clonogens from the effect of the second dose of 5-FU. Fig. 3B shows how the protection factor varies in various replicates of the same experiment. Also shown are cases involving slight variations in IL-11 injection regimes, including one where animals in a reversed light cycle room [15] were used to see whether the circadian rhythms might influence the response. No significant effect of treating at different times of day could be detected in this single experiment.
View larger version (8K):
[in this window]
[in a new window]
|
Figure 2. Surviving fraction of intestinal crypts following two doses of 5-FU, 6 h apart (circles), or two doses of 5-FU 6 h apart with pre- and post-treatment with IL-11 (protocol 3A in Materials and Methods) (squares). Mean ± SE.
|
|
View larger version (13K):
[in this window]
[in a new window]
|
Figure 3A. The protection factor (ratio of the number of surviving crypts in the IL-11-treated groups compared with saline-treated controls) in the 5-FU two-dose experiment. Each point represents the mean of four to six animals in various experiments that have been pooled here. Data from other experiments with radiation as the cytotoxic agent [12] are shown for comparative purposes. Open symbols denote 5-FU experiments, closed symbols denote radiation. The arrow represents the cytotoxic agent in the treatment sequence.
|
|
View larger version (18K):
[in this window]
[in a new window]
|
Figure 3B. Bar diagram showing the protection factor (as defined for Figure 3A) for various experiments involving different IL-11 protocols associated with exposure to two doses of 5-FU 6 h apart. Rev = reverse light cycle; the timing of all injections was performed at the opposite time of the day.
|
|
Twelve Gy of radiation results in the destruction of about 70%-80% of the crypts (i.e., a surviving fraction of about 20%-30%), and this is a dose of whole-body radiation that will kill about 50% of the mice within a period of six or seven days [7]. As can be seen from Fig. 4, all of these animals succumbed to the cytotoxic exposures within a period of 10 or 11 days, probably due to the combined effects of gastrointestinal damage, other mucosal damage (oral mucosae) and hematological cytotoxicity. Fig. 4 shows the results of two independent experiments in which 12 Gy of radiation were given alone, where essentially all the animals die between days 6 and 11. There are no significant differences between either of the 12 Gy alone (crosses and open circles) or the control group with 12 Gy with saline injections (triangles). Also shown are two experiments for the survival time of animals pretreated with IL-11 where all the animals die between days 7 and 14 (closed symbols). There is no significant difference between these two groups. There is an overall shift to the right in the survival time in the IL-11 pretreated animals which represents an approximate three- to four-day prolongation of life. The mean survival times are 9.95 ± 0.40 and 10.85 ± 0.48 days for the two IL-11 treated groups compared with 8.05 ± 0.49 and 7.85 ± 0.39 days for the two 12 Gy alone groups, and 8.20 ± 0.21 for the 12 Gy plus saline group, i.e., a prolongation of two days on the mean survival time for IL-11 treated animals (Table 1). The IL-11 data are significantly different from the radiation alone groups (p < 0.0001). The survival time for animals irradiated with 12 Gy receiving a sequence of saline-BSA injections both before and after irradiation is not different from the 12 Gy alone, indicating that the handling and injection protocols do not account for the extended survival.
These experiments show clearly that pretreatment with IL-11 results in an increased survival of clonogenic cells, and consequently, increased survival of intestinal crypts, which then results in an extension of animal survival time. A dose of 12 Gy reduces the surviving crypts to about one-third, and pretreatment with IL-11 at this dose affords a modest protection, increasing the surviving crypts by a factor of about 1.5, i.e., increasing the surviving fraction from about 0.3 to about 0.45. Thus, this modest change can have a significant effect on the survival time of the animals.
Continued administration of IL-11 following a dose of radiation for the duration of the experiment had no additional benefit on the survival time of the animals (Fig. 5). Somewhat surprisingly, the survival time for animals treated with 10 mg/mouse of 5-FU delivered twice 6 h apart (5-FU x 2) (which results in a similar crypt surviving fraction as seen after 12 Gy of radiation) is extended by a period of three to four days compared with the radiation survival time (Fig. 6). The reasons for this protracted survival time following 5-FU x 2 in the absence of any IL-11 are unclear.
|
Discussion
|
|---|
The results in Fig. 1 demonstrate that for two very different cytotoxic agents with differing-shaped survival curves, dose regimes can be selected that result in comparable levels of crypt killing. The x-ray survival curve which is obtained by delivering single acute doses has a broad shoulder and a crypt survival thereafter which is exponentially related to increasing dose. Such survival curves have been presented on numerous occasions, and many were reviewed by Potten et al. in 1983 [10].
5-FU is a pyrimidine analog which acts as an antimetabolite to uracil. Following intracellular conversion, the products block the conversion of deoxyuridylic acid to thymidylic acid, thus blocking the action of thymidylate synthetase, and hence blocking DNA synthesis. There may also be effects on RNA synthesis. Thus, 5-FU acts initially on a similar population (the S phase cells) to hydroxyurea.
The 5-FU dose-response curve decreases initially following single doses, indicating that some crypts can have all their clonogenic cells destroyed by a single dose of this S phase-acting antimetabolite. However, only about one-third of the crypts appear to be of this category. The majority of crypts requires two doses of 5-FU spaced 6 h apart to destroy all the clonogenic cells in the crypts. Following this treatment, there is a progressive decline in crypt survival to levels of around 5% following the highest doses of 15 mg/mouse. If three doses of 5-FU are given 6 h apart, the extra third injection has little effect at individual doses of or below 3.75 mg. Above this, however, the effect of the third dose becomes apparent. These 5-FU data together with earlier data following multiple injections of hydroxyurea [10] suggest that there may be some heterogeneity among the crypt population in terms of numbers, susceptibility and/or cell cycle status of clonogenic stem cells.
It is clear from Figs. 2 and 3that pretreatment with IL-11 affords considerable protection in terms of the number of crypts surviving the cytotoxic exposure. At comparable levels of crypt damage, comparable levels of protection are observed when -rays or two doses of 5-FU are delivered (Fig. 3A). There are indications from in vitro experiments that IL-11 inhibits cell proliferation [16] by taking cells out of cycle [17].
It is clear that with -rays, pretreatment with IL-11 can alter the time course of survival of the animals by as much as three to four days (or two days for the mean values). This confirms the biological and clinical importance of protection of crypt numbers and the relationship between increased crypt numbers and survival time. The results following radiation also demonstrate that continued application of IL-11 following the cytotoxic exposure seems to afford no additional benefit, which is somewhat surprising in light of the observations of Du et al. [3, 6]. However, this study involved a combination of two cytotoxic agents, radiation and 5-FU both given at lower doses (6 Gy and 150 mg/kg). Furthermore, they used higher doses of IL-11 (250 µg/kg compared with 100 µg/kg used here). Experiments are under way investigating the effects of higher doses of IL-11. Although higher doses of IL-11 given after treatment may have an effect [3, 6], the lower dose used here affords good protection when given before cytotoxic exposure.
It is commonly stated that damage to the gastrointestinal tract, if sufficiently severe, will cause death of the animals within a period of six to seven days. Damage to the hemopoietic system, which is inherently more sensitive, results in death over a period of up to 30 days. These observations form the basis of LD50/6/7 and LD50/30 studies for gut and bone marrow damage respectively. The dose of radiation used here was approximately the LD50/6/7 for these animals, and it was clear that many of the mice that survived beyond this time suffered severe damage to their oral mucosae and their associated glandular and lymphatic elements, and to hematological tissue as well as other tissues and organs. It will be interesting to see whether IL-11 affects the survival time of animals delivered an LD100/6/7 (approximately 1600 rads) where there is an even greater protection ratio for surviving crypts. It will also be informative to see how effective IL-11 is at protecting animals where the oral mucosae and bone marrow are shielded or where a bone marrow transplant is administered. Such experiments are under way.
The shift to longer survival times following two doses of 5-FU that result in the same level of crypt killing (Fig. 6) was surprising. The explanation is unclear, but may be associated with the difference in the mode of administration of the cytotoxic agent, one being a single acute exposure, the other being the delivery of a drug which requires metabolism given as two doses spaced 6 h apart. It may take longer for the cytotoxic effects of such a two-dose drug treatment to take full effect, particularly following high doses which may saturate the drug removal processes. 5-FU would be expected to cause some acute cell death (apoptosis) as a consequence of DNA damage and its S-phase cytotoxicity. However, it may also cause later cell death due to delayed metabolic effects.
It will be interesting to see whether IL-11 given prior to, or prior and after, 5-FU treatment can affect animal survival time. Preliminary experiments generated a variable protective response, but the lack of understanding of the increased survival time after 5-FU alone (Fig. 6, see also above) and the complexity of IL-11 administration in conjunction with multiple cytotoxic exposures make these experiments much more complex. Experiments are under way to investigate these processes further.
The survival of an animal following a cytotoxic insult will depend on the cumulative effects of the damage to various renewing cell populations: oral mucosae, gut, bone marrow, etc. The temporal response and sensitivity of each may vary with different cytotoxic agents. The small intestine with its rapid proliferation will be very sensitive, and with its short transit time (crypt to villus tip) of three to four days, will rapidly express the consequences of cell destruction, the major consequences being loss of mucosal integrity with water and electrolyte imbalance and bacterial sepsis and shock. The more delayed response after 5-FU exposure may be due to such differences in sensitivities.
However, these data clearly show that propitious administration of IL-11 in relation to a cytotoxic insult can afford protection to a significant number of the critical clonogenic stem cells in the crypts of the gut and that the protection of these stem cells results in beneficial changes in the pattern of whole-animal survival following these cytotoxic insults. Further studies will clarify details of the mechanisms by which IL-11 affords such protection.
|
Acknowledgments
|
|---|
This work was supported by the Cancer Research Campaign, UK, and Genetics Institute, Inc., Cambridge, MA, USA. I am extremely grateful to all the scientific officers that assisted with these experiments, in particular to Emma Bowley and Kerry Dye, without whom the studies would not have been possible. I am grateful to Dr. James Keith for helpful discussions and support and to Dr. Steve Roberts for statistical help and advice.
|
Footnotes
|
|---|
Provisionally accepted February 27, 1996.
|
References
|
|---|
-
Turner KJ, Clark SC. Interleukin 11: biological and clinical perspective. In: Mertelsmann R, Hermann F, eds. Hematopoietic Growth Factors in Clinical Applications, 2nd ed. New York: Marcel Dekker, 1994:315-336.
-
Keith JC, Albert L, Sones ST et al. IL-11, a pleiotropic cytokine: exciting new effects of IL-11 on gastrointestinal mucosal biology. In: Murphy MJ, Jr., ed. Polyfunctionality of Hemopoietic Regulators: The Metcalf Forum. Dayton, OH: AlphaMed Press, 1994:79-90.
-
Du XX, Williams DA. Interleukin 11: a multifunctional growth factor derived from the hemopoietic microenvironment. Blood 1994;83:2023–2030.[Abstract/Free Full Text]
-
Paul SR, Bennet F, Calvetti JA et al. Molecular cloning of a cDNA encoding interleukin 11, a stromal cell-derived lymphopoietic and hematopoietic cytokine. Proc Nat Acad Sci USA 1990;87:7512–7516.[Abstract/Free Full Text]
-
McKinley D, Wu Q, Yang-Feng T et al. Genomic sequence and chromosomal location of human interleukin 11 gene (IL-11). Genomics 1992;13:814–819.[Medline]
-
Du XX, Doerschuk CM, Orazi A et al. A bone marrow stromal-derived growth factor, interleukin 11 stimulates recovery of small intestinal mucosal cells after cyto-ablative therapy. Blood 1994;83:33–37.[Abstract/Free Full Text]
-
Hendry JH, Potten CS, Roberts NP. The gastrointestinal syndrome and mucosal clonogenic cells relationship between target cell sensitivities, LD50 and cell survival, and their modification by antibodies. Radiat Res 1983;96:100–112.[Medline]
-
Withers HR, Elkind MM. Micro-colony survival assay for cells of mouse intestinal mucosa exposed to radiation. Int J Radiat Biol 1970;17:261–268.
-
Potten CS, Hendry JH. The microcolony assay in mouse small intestine. In: Potten CS, Hendry JH, eds. Cell Clones: Manual of Mammalian Cell Techniques. Edinburgh: Churchill Livingstone, 1985:50-60.
-
Potten CS, Hendry JH, Moore JV et al. Cytotoxic effects in gastro-intestinal epithelium (as exemplified by small intestine). In: Potten CS, Hendry JH, eds. Cytotoxic Insult to Tissue. Edinburgh: Churchill Livingstone, 1983:105-152.
-
Potten CS, Hendry JH. Clonal regeneration studies. In: Potten CS, Hendry JH, eds. Radiation and Gut. Amsterdam: Elsevier, 1995:45-59.
-
Potten CS. Interleukin 11 protects the clonogenic stem cells in murine small-intestinal crypts from impairment of their reproductive capacity by radiation. Int J Cancer 1995;62:356–361.[Medline]
-
Potten CS, Rezvani M, Hendry JH et al. The correction of intestinal microcolony counts for variation in size. Int J Radiat Biol 1981;40:321–326.
-
Roberts SA. DRFIT: a program for fitting radiation survival models. Int J Radiat Biol 1990;57:1243–1246.[Medline]
-
Potten CS, Al-Barwari SE, Hume WJ et al. Circadian rhythms of presumptive stem cells in three different epithelia of the mouse. Cell Tissue Kinet 1977;10:557–568.[Medline]
-
Booth C, Potten CS. Effects of IL-11 on the growth of intestinal epithelial cells in vitro. Cell Prolif 1995;28:581–594.[Medline]
-
Peterson RI, Wang L, Trepicchio WL et al. Interaction of rhIL-11 with intestinal epithelial cells. Gastroenterology 1995;108 (suppl 1):A-351.
Received December 14, 1995;
accepted for publication March 18, 1996.
Top
Abstract
Introduction
), 12 Gy of 
), and 12 Gy of 
). The number of surviving animals is plotted on each day following irradiation. Each experimental group started with 20 mice.
) both prior to and following the radiation exposure. The 12 Gy curve is the same as shown in Figure 4
) and 12 Gy preceded by five doses of IL-11 (protocol 1) (