probably relatively simple. If it were not, it 

 wouldn't be under the control of a single gene. 

 In the mouse, there are several mutations which 

 are known to produce anemia, i.e., to affect in 

 some way one of the pathways of differentiation 

 in which we are interested. Some of these have 

 already been studied in considerable detail by 

 Dr. Elizabeth Russell and Dr. Seldon Bernstein 

 at The Jackson Laboratory, Bar Harbor, Maine 

 (6). The question is, how do these genes affect 

 the pathways of differentiation of the stem cell? 



Let's imagine what kinds of genetic changes 

 one could see, speaking of gross changes rather 

 than individual steps. Each one of the four sug- 

 gested pathways open to the stem cell, pre- 

 sumably, could be regulated by a separate gene. 

 There might be a gene that regulates self- 

 renewal. For example, there maybe some mole- 

 cule that triggers off the self-renewal division, 

 since there is evidence that it isn't happening 

 all the time (7). Most of the stem cells are not 

 dividing and one may postulate that some of 

 them are triggered off by a gene product. You 

 can imagine four classes of these genes cor- 

 responding to these four possible pathways that 

 this stem cell can go into. You can imagine, 

 also, two general ways that regulation could 

 occur. It could be a property intrinsic in the stem 

 cell itself which regulates it or the regulator 

 could come from outside. Let me just take an 

 example; let's say that regulation might depend 

 on the permeability of the membrane of the cell. 

 That would be something I would call intrinsic 

 to the cell itself. On the other hand, the regulator 

 could be a hormone that comes in from outside 

 and tells the cell to do something. That I would 

 call external because the mutation would then be 

 a mutation that stops the supply of that hormone, 

 whereas the mutation in the first case would be 

 one that alters the membrane. So there are two 

 classes, intrinsic and extrinsic, each of which 

 can be applied to each of the four suggested 

 pathways. 



We have looked at three mutations in some 

 detail so far; and we think that each falls into a 

 different class. One of these is extrinsic in its 

 action and the other two are intrinsic. I'll 

 describe the evidence for this next. 



The mutations we've looked at have been 

 W, SI and/. Let's consider IV and SI first. 

 The animals that we studied were of genotype 

 W/W and Si/Sid, obtained from Drs. Russell 

 and Bernstein. Animals of these two genotypes 

 have very similar phenotypes. Both have a 

 severe macrocytic anemia, the coat color is 

 affected, the animals are sterile and they are 



very radiation-sensitive. Superficially, the 

 phenotypes appear almost identical. However, 

 we know they're different because they map at 

 different genetic sites. If the cells that form 

 colonies in the spleen are a kind of stem cell 

 and the basis for these anemias is a defect in 

 the stem cell, then we should find deficient 

 colony formation when we test the cells from 

 these animals for their ability to form colonies 

 in irradiated normal hosts. 



For W/W^ we found deficient colony for- 

 mation (8), and the subsequent work that's been 

 done indicates that this may be a defect in the 

 ability of the stem cell - the colony-forming 

 cell - to renew itself. Therefore, this is a 

 mutation whose effects are intrinsic to the 

 stem cell. It's apparently a defect at the cel- 

 lular level in the ability of the cell to produce 

 more cells like itself. Because the phenotype 

 was similar, we expected that the stem cells 

 in animals of genotype Sl/Sl d would be similar 

 in their properties to those of animals of geno- 

 type W/W. In fact, they turned out to be very 

 different. When we tested cells from Sl/Sl ^ 

 mice for their ability to form colonies in our 

 standard test system (irradiated normal mice), 

 we found that they formed colonies perfectly 

 well (9). Every attempt that we made to find 

 a defect in the composition of these colonies 

 failed; they are apparently perfectly normal. 

 So then we did some experiments the other 

 way around. We used these animals as recipi- 

 ents for normal cells. We had found previously 

 that W/W animals are good hosts for the 

 growth of normal cells (8). You can put normal 

 stem cells into them and they grow well. In 

 fact, you don't even have to irradiate the hosts 

 first. You can put normal cells into unirradi- 

 ated mice of this genotype and they'll still form 

 spleen colonies. In fact, they'll cure the anemia 

 of the animal permanently (10). Thus, animals 

 of genotype W/W" exhibit a defect at the cellu- 

 lar level, but the host is capable of supporting 

 the growth of normal blood-forming cells. How- 

 ever, if you put normal cells into Sl/Sl ^ mice, 

 they don't form colonies. Even if you irradiate 

 these animals with large doses, they won't 

 support detectable growth of normal cells (9). 

 In this case, the cells seem to be normal, but 

 the mouse does not support their growth nor- 

 mally. Thus, this mutation affects a process 

 which is extrinsic to the stem cells. 



One would expect, if all this is right, that 

 the cells from Sl/Sl '^ would grow in W/W " 

 hosts. In other words, one should be able to 

 take cells from an anemic animal of genotype 



71 



