CELL DIFFERENTIATION AND DENSITY SENSING 
Richard H. Gomer, Ph.D., Assistant Investigator 
Little is known about the molecular mechanisms 
that regulate the differentiation of cells into various 
cell types or the mechanisms that allow cells to 
sense how many of a given cell type are in an or- 
ganism. Dr. Gomer' s laboratory is studying 
cell-type differentiation and density-sensing mech- 
anisms used by the simple eukaryote Dictyostelium 
discoideum. 
Dictyostelium normally grows as isolated amoe- 
bae that multiply by fission. In the wild, the amoe- 
bae feed on bacteria on decaying leaves and the soil 
surface. When the cells starve, they stop dividing 
and aggregate using relayed pulses of cAMP as the 
chemoattractant. The cells form aggregation streams 
that flow and coalesce into an aggregation center. 
The aggregate then elongates into a slug that uses 
phototaxis and thermotaxis to crawl to an open area, 
whereupon it differentiates into a fruiting body con- 
sisting of a mass of spore cells held up by an ~ 2-mm 
column of stalk cells. Spores can then be dispersed 
by the wind or insects to start new colonies of Dic- 
tyostelium. 
Development of Techniques 
for Gene Isolation 
The initial decision for a Dictyostelium cell to 
differentiate into a precursor of either a spore or 
stalk cell is based on the cell-cycle phase of each 
cell at the time of starvation. For cells that are 
starved in S or early G2 phase {Dictyostelium does 
not have a Gj phase), one sister differentiates into a 
prestalk cell, while the other sister becomes a null 
cell (a cell that does not express either a character- 
istic prestalk or a prespore antigen) . Cells starved in 
late G2 or M phase similarly differentiate into either 
a prespore or a null cell. 
Until recently there have been no general meth- 
ods available to identify the gene causing a given 
mutation in Dictyostelium. Using available tech- 
niques for transformation, homologous recombina- 
tion, gene disruption, and antisense repression, Dr. 
Gomer's laboratory has begun to develop general 
methods to identify genes in Dictyostelium, and to 
use the methods specifically to isolate genes in- 
volved in the cell-cycle-dependent cell-type differ- 
entiation mechanism. Using aggregate and fruiting 
body morphology for the initial screen, three differ- 
ent methods were tried. First, overexpression trans- 
formants were generated by using extrachromo- 
somal vectors containing ~ 1 0-kb regions of random 
fragments of genomic DNA; this method has proved 
to be unsatisfactory. 
As a second method, random homologous integra- 
tion was tried in a manner similar to P-element mu- 
tagenesis in Drosophila to generate transformants 
with altered cell-type ratios. Several transformants 
with abnormal morphologies were identified. These 
include no aggregation, aggregation but no fruiting 
body formation, and aggregates with multiple long, 
string-like projections. Some of these appear to have 
abnormal prestalk:prespore cell-type ratios, as as- 
sayed by immunofluorescence. Dr. Gomer's labora- 
tory has isolated the DNA regions flanking the inser- 
tion site from one of these transformants and is 
testing whether the insertion caused the phenotype. 
As a third approach, libraries of cDNA fragments 
were made for shotgun antisense transformation. 
Unlike in higher eukaryotes, antisense works well in 
Dictyostelium. From a morphological screen of 
2,800 antisense transformants, 25 having unusual 
developmental morphologies were identified. 
These include failure to aggregate, failure to pro- 
ceed beyond aggregation, very small aggregates that 
do not form fruiting bodies, and tiny fruiting bodies. 
The antisense DNA was isolated from the 25 trans- 
formants, cloned, and used to retransform Dictyo- 
stelium. Currently, 4 of the 25 specific cDNAs that 
cause an abnormal developmental morphology have 
been identified. The success of this technique indi- 
cates that it will be a powerful general method to 
isolate genes rapidly in Dictyostelium. For one of 
the four cloned cDNA antisense transformants, cells 
aggregate but then do not proceed to form a fruiting 
body. This transformant has an abnormal cell-type 
ratio. This cDNA fragment is currently being se- 
quenced, and a larger fragment of the cDNA has 
been isolated by screening a cDNA library. Dr. 
Gomer's laboratory is currently identifying addi- 
tional genes that, when their expression is 
disrupted, cause an abnormal ratio of cell types. 
Density Sensing 
During Dictyostelium development, the expres- 
sion of some genes depends on cell density. This 
effect is mediated by soluble factors referred to as 
conditioned medium factors (CMFs), which the de- 
veloping cells secrete at very low rates and simulta- 
neously sense. There are at least two classes of CMFs: 
one is an 80-kDa glycoprotein; the other is a set of its 
breakdown fragments, with molecular weights be- 
tween 6.5 and 0.65 kDa and higher specific activity 
than the 80-kDa CMF molecule. The two classes of 
molecules do not need to be combined for activity. 
Dr. Gomer's laboratory has purified CMF, se- 
CELL BIOLOGY AND REGULATION 63 
