REGULATION OF PATTERN OF GENE EXPRESSION
BY CONTROLLING ELEMENTS IN MAIZE

Barbara McClintock

The production of anthocyanin pigment
in maize requires the functioning of a
number of known genes distributed among

the chromosomes of the complement..

This pigment may be produced in most
parts of the plant. Its distribution and
intensity vary widely in different strains
of maize. Each pattern reflects the action
during development of genetic mecha-
nisms that regulate pigment production.
One of the gene loci participating in
anthocyanin formation is the C. locus in
chromosome 4. Two independent events
brought the action of the gene at this locus
under the control of the Spm (Suppressor-
mutator) system. The symbols co and
com? designate the modified loci resulting
from these events. In both instances, the
element of the Spm system that is inserted
at the C, locus contributes to regulation of
pattern of anthocyanin production in
plant and kernel. It can induce different
patterns, A change from one pattern to
another may be traced to modifications of
the gene locus initially produced by a
response of the element to an active Spm
element. Some of these responses May
occur in individual cells early in develop-
ment. The locus is thereby “set” at an
early stage to induce a particular pattern
of anthocyanin distribution in sporo-
phytic tissues formed by descendants of

such a cell. Some of the settings may sub-
sequently be “erased,” probably in those
progeny cells that are included in the
germ line. The evidence for such settings
and erasures is the subject of this report.

The C, gene locus was selected for these
studies because modifications of control
of its action may be detected in all parts
of the plant: root, stalk (including in-
terior cells), leaf parts, tassel parts (in-
cluding the anthers), husks, and portions
of the cob, including interior cells as well
as floral parts. Alterations of control of
C, action may also be detected in the
pericarp layer of the kernel, which is
derived from maternal cells, and in the
aleurone layer of the endosperm.

The studies to be reported required the
use of a recessive allele of Cs, initially
isolated and examined by E. H. Coe. It
will be referred to as the standard reces-
sive and symbolized as ¢2. When a plant
is homozygous for ¢:, anthoeyanin pig-
ment is produced in many parts of the
plant but its intensity is low. The aleurone
layer of kernels homozygous for c: 38
devoid of the pigment. When the standard
C; locus is present in a plant or kernel, on
the contrary, pigmentation is intense, pro-
vided all other genes involved in antho-
eyanin production are active within the
tissues.
GENETICS RESEARCH UNIT

The most revealing studies of altered
programming of action of the C. locus
were conducted with plants having a
genic constitution that would have al-
lowed intense pigment to be produced in
the various parts of the plant had the
standard C, gene been present rather than
c! or cx"-*. These plants were either
c2""\/cz or cy"*/cp in constitution, and
each had one or more Spm elements whose
components were fully active, at least
initially, in the young plant. The state of
ce" that was available for these tests is
one that responds to Spm late in develop-
ment, whereas one of the states of c,"?
can respond very early. Although the
types of modification produced by these
responses to bring about altered pigment
distributions and intensities were similar
in the two instances, those occurring at
co"? were more instructive, for the follow-
ing reason. Early-occurring alterations
that modify the programming of antho-
cyanin production are made evident in
the plant in large sectors, which include
within them various different tissues.
Distribution of anthocyanin to these tis-
sues, or its intensity in any one tissue or
tissue part, may be compared with that in
other sectors in the same plant.

Most of the examined plants produced
tillers (side branches) with fertile ears.
Observations were made of anthocyanin
pigment distribution in the tissues of the
main stalk and in those of the tillers. The
ears of each plant were pollinated by
plants that were homozygous for the
standard recessive c. (whose action is not
altered by Spm) and also for wz”-8; these
plants had no active Spm element. Wx
produces amylose starch in pollen grain
and endosperm; wz, the standard reces-
sive allele, does not produce this starch.
When wa”-8 is present, the action of the
gene at this locus is under the control of
the Spm system. Since the ear-bearing
ey" /e. and c:”—*/cz plants were also either
W2/wz or wx/wzx in constitution, the in-
corporation of wx-* in the testcross
allowed a determination of the presence
or absence of Spm in the cells producing

569

each ear, and also the state of Spm if
present. Ears are especially useful for
study of patterns of anthocyanin pigment
distribution, because the pigment is pre-
served in the cells of the dried ears, and
its distribution in parts of the cob as well
as in the pericarp and aleurone layers of
the kernel can be examined. Many of the
ears had sectors of various sizes that
exhibited altered patterns of distribution
of pigment in the cob and kernels. Each
sector arose from the progeny of a single
cell in which a modification had occurred
at the c:"—! or the c”~? locus. Comparison
of types of pigment distribution in the
various tissues within the sectors on a
single ear, or on different ears of the same
plant, gave evidence of the range of
altered programming that can be induced
by responses of the controlling element
to Spm. When intensity as well as distri-
bution of pigment was considered, the
range proved to be wide.

The tests were conducted with several
hundred plants carrying one state of
@”—, and with several hundred carrying
one or the other of two selected states of
c"—*, Because the results obtained with
the differently constituted plants are
comparable, discussion will be confined to

‘those obtained with one state of c—?.

Pigment Distribution in Parts of the Ear

A few illustrations will facilitate discus-
sion. Two ears produced on the main stalk
of one plant (8597C-4) that was o"-?/cg
in constitution are shown in Plate 1A.
The silks of these ears received pollen
from plants that were homozygous for
the standard recessive c:, and for wx™-%,
and had no active Spm. Thus at least half
the kernels on each ear should have been
homozygous for cz; and should have had no
pigment in the aleurone layer. The ear in
the lower part of the photograph had this
phenotype. On that ear two additional
classes of kernels appeared, one showing
spots of pigment in a colorless back-
ground, the other intensely and uniformly
pigmented. The spotted kernels received
a ce”? locus that responded to Spm in
570

development of the endosperm in the
manner expected of an unmodified c:"-?
locus. The fully pigmented kernels re-
ceived a c:”~? locus that had been modified
so that the gene functioned in the cells of
the aleurone layer in a manner resembling
that of the standard C2 gene. At the tip of
the ear there was a large sector in which
all the kernels that received a derivative
of c.”—* expressed this phenotype. In these
kernels the pericarp layer also was densely
pigmented, except at the region of the
crown. A discussion of pericarp pigmenta-
tion patterns will be deferred until later;
here it need only be stated that not all
the kernels on this ear had pigment in the
pericarp. Some had none; in others it was
confined to well-defined sectors within the
layer. A part of the ear shank, visible in
the photograph, was variegated for antho-
cyanin pigment, as were the parts of the
cob.

All kernels on the other ear produced
by the same plant were colorless. No pig-
ment appeared in the aleurone layer or the
pericarp layer, and the cob parts showed
no evidence of variegated patterns of pig-
mentation. The cells that gave rise to this
ear had more than one active Spm ele-
ment. Thus the uniformity of gene expres-
sion throughout the ear may not be
ascribed to somatic loss of Spm or of its
activity. A specific type of modification
at the locus of c"-? must have occurred
in the cell whose descendants produced
this ear.

Plate 1B shows the cobs of three ears
produced on another plant (8500-10)—
the two on the right by the main stalk,
the one on the left by the tiller. An active
Spm element was present in the cells that
gave rise to all these ears. The deep pig-
ment in the floral parts, visible as distinct
sectors in the two cobs produced by the
main stalk, resembled that which would
have appeared throughout the cob had the
standard C2 allele been present. In addi-
tion, the pericarp layer of the kernels
located within these sectors was deeply
pigmented. The aleurone layer of all
kernels on the right-hand cob was totally

CARNEGIE INSTITUTION

colorless, indicating that the ec”? locus
had been modified in the cell whose de-
scendants produced the ear. No early
modification occurred in the cells giving
rise to the middle ear, not even in the one
whose descendants produced the dark-
pigmented basal sector in the cob. On the
left-hand ear all kernels that received a
derivative of c”—? had deep and uniform
pigmentation in the aleurone and pericarp
layers. Thus the early modification of
c™—? in the cell whose descendants pro-
duced the ear induced a new pattern of
anthocyanin distribution in the plant
parts, allowing a C2 type of expression in
the maternally derived pericarp layer and
in the aleurone layer of the endosperm but
not in the floral parts of the cob.

Cobs of four ears produced by another
plant (8595B-1) are shown in Plate IC.
The cob at far right, from the ear of the
main stalk, had many small sectors show-
ing pigmentation of different intensities
in the floral parts. The other three cobs
came from ears produced by the three
tillers of the plant. The one pictured sec-
ond from right had no deeply pigmented
sectors. The adjacent cob on the left was
uniformly and deeply pigmented in all its
parts. The leftmost cob had both large
and small sectors with intense pigment in
the floral parts. Kernels on the two right-
hand cobs had pigmentation patterns in
the aleurone layer indicating that no early
modification of the cx"? locus had oc-
curred to alter gene action in that layer.
Such a modification did occur during de-
velopment of each of the two tillers whose
cobs are shown on the left. The event in
the cell whose descendants produced the
leftmost cob did not allow pigment to be
formed in the aleurone layer of its kernels.
In the tiller whose cob is adjacent, a very
early-occurring event at cy"? caused the
gene to act like the standard C2 The
altered expression was evident in all parts
of the tiller and also in the aleurone layer
of kernels on its ear. All those that re-
ceived the modified c”-? locus had deep,
uniform pigmentation in that layer.

The cobs and parts of the husks of two
GENETICS RESEARCH UNIT

ears from plant 8595C-9 are shown in
Plate 1D. The cob on the right was pro-
duced by the main stalk and the one on
the left by the tiller. In the right-hand
cob, pigment in the husks and floral parts
resembled that appearing when the stand-
ard Cy allele is present. The pericarp
layer of the kernels on this ear was also
deeply pigmented except at the very tip
of the ear, where only sectors of deep
pigment in a colorless background ap-
peared. The modification of co"? that
induced C, type of expression in the husks
and cob did not do so in the aleurone layer
of the kernels. Kernels that received both
cs? and Spm. had spots of pigment in a
colorless background in the aleurone layer,
as did kernels of the same constitution on
the tiller ear.

Plate 1Z shows cobs of two ears of plant
8595B-5. The right-hand cob, produced
by the main stalk, had many small sectors
with pigment of various intensities. The
one on the left, produced by the tiller,
had one large sector with deep pigment in
the floral parts and several smaller sectors
of this type. Control of aleurone pigmen-
tation pattern was not altered in kernels
of the right-hand cob. It was altered in
kernels of the left-hand cob: all those that
received the modified c,"-? locus had deep
pigment, uniformly distributed through-
out the aleurone layer. Though the peri-
carp layer of all kernels on this ear was
darkly pigmented, many kernels exhibited
sectors of even more intense pigmentation
in this layer. An active Spm was present
in the cells that gave rise to both of these
ears.

Plate 1F shows the cobs of two ears of
plant 8595B-7. The one on the right was
produced by the main stalk and that on
the left by the tiller. In the cells that gave
rise to these ears, no change had occurred
to alter the action of the gene in the
aleurone layer of the kernels, not even in
that part of the tiller ear where the floral
parts were deeply pigmented. All the
kernels within that sector, however, had
deep pigment in the pericarp layer.

The examples in Plate 1 give evidence

571

of altered patterns of distribution of
anthocyanin pigment, each referable to
a modification of the c:”~? locus in an in-
dividual cell during plant development as
a result of a response of its controlling
element to Spm. Some modifications ef-
fected a pattern resembling that pro-
duced by the standard C; allele. Others,
in combination with the standard reces-
sive C2, present in all cells of the examined
tissues, induced patterns resembling the
one that appears when this recessive is
homozygous. Many modifications, how-
ever, gave rise to pigment distributions
and intensities among the different tissues
that did not conform to either of these
phenotypes, as indicated in Table 2.

TABLE 2. Combinations of Phenotypic
Expression That May Appear in Cob and
Kernels as the Result of Different
“Settings” of the c"-? Locus*

 

In Layers of Kernels

 

 

In Floral
Parts of Pericarp Aleurone
Cob Layer Layer

C; C2 Cc,
C, Cr Cs
Ca C, C,
C2 Ca Ce
C2 C; unchanged
Cz unchanged unchanged

unchanged C2 C,
C2 unchanged unchanged
Cr Ce unchanged
Ce C2 Ca

 

*C,; phenotype resembling that produced by
the standard C; locus. ce: phenotype resembling
that produced by the standard recessive co.
Unchanged: ”-* locus unmodified in its expres-
sion.

Pigment Distribution in the Pericarp Layer
of the Kernel

The illustrations of altered patterns of
pigment distribution in kernels and cobs
(Plate 1) are supplemented by several
others that show pigment distributions in
the pericarp layer of the kernel. The peri-
carp is the outermost tissue layer of the
kernel. It represents the ovary wall and
thus the cells composing it are maternal
572

in origin. When pigment appeared in this
layer in mature kernels on ears of the
investigated plants, its intensity was often
low, particularly in the region of the crown
(Plates 1A and 2D). In contrast, the pig-
ment might be intense in all parts of the
pericarp layer in kernels that commenced
development after pollination but, for
some yet undetermined reason, ceased
development early. Many kernels of this
type appeared on ears that had been polli-
nated on the same morning in July 1965.
This was fortunate, for those kernels pro-
vided an opportunity to examine and
compare the many different patterns of
pigment distribution and intensity oc-
curring in the pericarp layer. Such kernels
are small and hollow when dry, as they
contain neither embryo nor endosperm
tissue. (See the hole in the second kernel
from right, lower row, Plate 2A.) They
will be referred to as abortive kernels.

The ovary wall is initiated by a ring of
cells about the base of a growing point
that will form the nucellus tissue in which
the megasporocyte will be differentiated.
The ring grows upward and over this
nucellus initial, gradually enclosing it by
a narrowing of the ring and a final coming
together of its parts at the position of the
future attachment of the silk. The silk
represents an extension of growth of one
section of the ring. This manner of de-
velopment of the ovary wall is well illus-
trated by the pigmentation patterns of
those abortive kernels in Plate 2A,B that
have deep-pigmented sectors in a colorless
background. The large pigmented sectors
are continuous and extend from the base
of the kernel to the point of silk attach-
ment. The silk, broken near its attach-
ment, is visible in the photographs in A.

All the abortive kernels shown in A and
B were present on the right-hand cob in
Plate 1C. Those in A, enlarged approxi-
mately <6, are viewed from above. Those
in B, enlarged approximately X23, are
viewed from the side—the germinal side
in the upper row and the abgerminal side
in the lower row.

The abortive kernels at each end of the

CARNEGIE INSTITUTION

upper row in Plate 24, and the 4 dark-
pigmented abortive kernels on the left in
2C, have been included to illustrate an-
other way in which the expression of a
phenotype may be modified in a very spe-
cial part of a tissue during its develop-
ment. In the region of the silk attachment
on the abgerminal side of the kernel, pig-
ment is either absent, or present in low
intensities, and basipetally directed forks
extend from this region. For development
of such patterns, pigment production
must be inhibited in some cells during the
later stages of ovary wall formation, or
else previously formed pigment must be
destroyed. All the cells derived from an
initial segment of the basal ring may be
subject to inhibition or destruction at
practically the same stage in develop-
ment, or the event may be initiated in
only one or a few such cells, the effect
spreading upward during subsequent
growth to include adjacent cells.

The abortive kernels shown in Plate 2C
were present on the tiller ear of plant
8597C-3. This ear had a large basal sector
in which all kernels receiving a derivative
of ce”? had intense and uniformly dis-
tributed pigment in the aleurone layer.
In Plate 2D, which shows a portion of the
ear, a small segment of this basal sector is
visible in two rows of kernels at the upper
left. The sector contained 31 abortive
kernels, and all had dark pigment in the
pericarp layer. Seven of them are shown
in Plate 2C. The white kernel on the left,
with pigmented streaks on one side only,
was located in a row adjacent to those
that formed the sector of dark-pigmented
kernels. The floral parts of the cob within
the sector were not uniformly pigmented:
subsectors displayed pigment intensities
ranging from dark to nearly colorless.

Presetting of the Controlling Element at the
Locus of c:™-*

The ear with the basal sector just de-
scribed was instructive in other ways. An
active Spm was present in the cells of that
sector and in those of several rows on each
side of it, but only in their basal part. The
GENETICS RESEARCH UNIT

em locus in the Spm-carrying cells of
these adjacent rows had not been modi-
fied. The response of the locus to Spm
resulted in pigmented spots in a colorless
background in the aleurone layer of ma-
ture kernels. No active Spm was present,
however, in the cells that produced the
remainder and larger part of the ear, and
no pigment appeared in the aleurone layer
of any kernels within that part. Neverthe-
less, control of patterns of pigmentation
was made evident by pigment distribu-
tions in the floral parts of the cob and in
the pericarp layer of the mature kernels.
In the middle of the photographed portion
of the ear was a sector, spanning two rows,
where the pericarp layer of the kernels
was pigmented throughout, although the
color intensity was greatly reduced in the
region of the crown. Below the right-hand
portion of this pigmented sector and ex-
tending to the right of it was another large
sector, in which none of the kernels
showed pigment in the pericarp layer. In
kernels to the left of these two sectors,
deep pigment appeared in the pericarp but
only as sectors in individual kernels.
Pigment distributions similar to those
illustrated in Plate 2D occurred on ears
of other plants, where there was no evi-
dence of the presence of Spm in the cells
producing the ears. Such phenotypes ap-
peared, however, only on ears of plants
that had commenced development with
an active Spm. This indicates that the
controlling element at c*-2 responded to
Spm early in plant development and was
thus conditioned to control patterns of
gene action in tissues produced very much
later, even though an active Spm was no
longer present. In other words, the initial
response to Spm ‘“preset’’ the element to
undergo specific types of change—par-
ticular “settings’—often many cell gen-
erations after the presetting event had oc-
curred. Each such setting then effected
a special expression of the gene in the
mature tissues subsequently formed by
descendants of the cell in which it had
occurred. Studies to be described in the
next section suggest that many of the sec-

573

tors observed on ears, in the floral parts
of the cob as well as the pericarp layer of
the kernels, arose through similar pre-
setting and subsequent setting events, and
that removal of an active Spm is not a
requirement for the subsequent settings.

Inheritance of Modified Pigmentation
Patterns
Studies were undertaken to determine
the inheritance of some of the modified
types of pigmentation patterns illustrated
in the ears and cobs of Plates 1 and 2. Un-

. til very recently, material for such studies

was limited. It was decided, therefore, to
explore in a preliminary manner each
available type in order to be able to select
in the future those that would provide the
most instructive information. The find-
ings of these preliminary tests can be
illustrated by describing the phenotypes
of plants derived from kernels of the three
ears whose cobs are shown in Plate 1B.
As was mentioned earlier, all three of
these ears were produced by cells that
carried an active Spm element.

Two classes of kernels, present in equal
numbers, were carried on the left-hand
cob in the photograph. One had intense
pigment throughout the aleurone layer;
the other had none. The pigmented ker-
nels received from the ear parent a deriva-
tive of the initial c,"—? locus that had been
modified in the somatic cell whose de-
scendants produced the ear. The colorless
kernels were homozygous for the standard
recessive C2. Both types had dense pig-
ment in the pericarp layer.

Plants derived from 13 of the pig-
mented kernels were intensely pigmented
throughout. The floral parts of their cobs,
unlike those of the parent cob, also were
deeply pigmented, the color intensity
resembling that of the dark sectors in the
other two cobs in Plate 1B. These plants
transmitted the modified c.”-? locus, now
performing as a stable C, allele, in a
strictly Mendelian manner. An active
Spm was present in each of 11 plants
tested for it, but the locus no longer re-
sponded to it.
574

Plants derived from 12 of the kernels
having a colorless aleurone layer also were
pigmented, but the color intensity was
very much lower than in the sister plants
derived from the colored kernels. An ac-
tive Spm was present in each of 11 plants
tested. Since these plants were homo-
zygous for Cz, no somatically occurring
change in gene expression was expected
and none was noted.

Plants were grown from selected kernels
on the middle cob in Plate 1B. The basal
sector of this cob where the floral parts
were deeply pigmented contained 6 ker-
nels, in which the pericarp layer also was
densely pigmented. This sector arose from
the descendants of a cell in which the
cs"? locus had been modified to allow a
C2 type of expression in the floral parts of
the cob and the pericarp layer of the
kernels. The aleurone layer of 3 of these
kernels had pigmented spots in @ non-
pigmented background—the expression
given by an unaltered cy? locus in the
presence of Spm. Obviously the modifica-
tion of the locus that allowed C2 expres-
sion in the sporophytic tissues did not
effect a similar expression in the endosperm
tissues. The aleurone layer of a fourth ker-
nel was deeply and uniformly pigmented,
indicating that a subsequent modification
of the c:-? locus had occurred before ferti-
lization. The remaining 2 kernels had a
colorless aleurone layer.

The plant derived from the kernel with
a darkly pigmented aleurone layer was
deeply pigmented in all its parts. The
floral parts of the cobs were densely and
_ uniformly pigmented, like the sector of

*the parent cob from which the kernel
came. The pericarp layer of all kernels
was darkly pigmented. Not only in the
plant parts, but also in the aleurone layer
of all kernels that received the modified
cy? locus, the phenotype resembled that
produced by the standard C2 locus. An
active Spm was present in the nuclei of
this plant, but the modified co™ locus no
longer responded to it in any detectable
manner.

In plants derived from the 3 kernels
with pigmented spots in a nonpigmented

CARNEGIE INSTITUTION

background, anthocyanin distributions
and intensities resembled in range of
types those of plants whose cobs and ker-
nels are illustrated in the plates. In that
regard, these 3 plants were indistinguish-
able from plants derived from 8 other
kernels on the same parent ear that had
pigmented spots in their aleurone layer.
The modification of the c:”-? locus that
was responsible for the deep-pigmented
basal sector in the parent cob was not
transmitted unaltered to the progeny.

The phenotype of the plants derived
from the 2 kernels with a colorless aleu-
rone layer resembled that of plants ho-
mozygous for the standard TECESSIVE C2.
Spm was present in each plant, but no
somatically occurring modifications affect-
ing anthocyanin distribution were noted.
Also, the aleurone layer of all kernels on
the testcross ears was colorless. These 2
plants probably were homozygous for Cs.

On the right-hand cob in Plate 1B, all
kernels had a colorless aleurone layer,
indicating that an early-occurring re-
sponse to Spm of the element at the cx”?
locus had nullified its capacity to con-
tribute to pigment production in this
layer. Nevertheless there were several
distinct sectors where the floral parts were
deeply pigmented, and kernels located
within these sectors had deep pigment in
the pericarp layer. The appearance of such
sectors indicated that the element at
co? had undergone subsequent modifica-
tions during development of the ear.
These changes allowed intense pigment
to be produced in sporophytic cells but
not in the aleurone layer of the kernel.
Because the aleurone layer was alike in
all kernels, those that received a modified
a? locus could not be distinguished
from those that received the standard
recessive Co.

Plants were grown, therefore, from the
5 kernels located within the dark-pig-
mented sector of the cob that is visible in
the photograph, and from 12 other kernels
located elsewhere on the cob (not in dark-
pigmented sectors). The 17 plants were
pigmented, but the intensity of pigmen-
tation resembled that in plants homozy-
GENETICS RESEARCH UNIT

gous for c;. None of them was variegated
for different intensities of pigmentation
within a tissue, as the ear parent had been,
even though Spm was present in 12 of the
13 plants tested for it. Also, the aleurone
layer of all kernels on ears of the plants,
produced by the testcross, was completely
colorless. None of the cobs, including
those of plants derived from kernels from
the dark-pigmented sector of the parent
cob, exhibited deep pigmentation in the
floral parts. Very probably some of the 12
plants known to have had an active Spm
element received from the ear parent the
modified derivative of c."-2, Further tests
will be required to determine whether
such plants may be distinguished from
those that are homozygous for the stand-
ard recessive. Present evidence suggests,
however, that the c"-? locus lost its
capacity to undergo further modifications
that would allow it to contribute to for-
mation of intense pigment in the sporo-
phytic tissues.

Other tests similar to those just de-
scribed were conducted. In addition, in-
dividual kernels were selected from some
ears and the plants derived from them
were examined for transmission of the
kernel phenotypes. Some of the selected
kernels had a fully pigmented aleurone
layer. Others came from ears having sec-
tors in which the aleurone layer was
colorless but the pericarp layer was either
colorless or variegated, or fully pigmented.
Still others were selected from ears of
plants that commenced development
without an active Spm element. None of
these last-mentioned plants showed evi-
dence of any somatically occurring change
in control of c:"? gene action, nor did
their progeny unless Spm had been intro-
duced in the cross of the ear, from which
kernels receiving Spm were selected. The
plants derived from such kernels did show
somatically occurring changes, whereas
the phenotype of sister plants derived
from kernels that did not receive Spm
resembled that of the ear parent.

Results of tests so far conducted with
plants carrying Spm and either c”— or
ce"? indicate that the modifications of

575

these loci which occur during plant de-
velopment and alter gene expression in
the aleurone layer of the kernel are herita-
ble in a-Mendelian manner, whereas most
changes affecting gene expression in the
sporophytic tissues are not heritable in
the same way. The modifications affecting
sporophytic tissues can occur after Spm
has been removed or inactivated, and
bring about the same range of alteration
in phenotypic expression as appears when
Spm remains active. a

This series of tests has led to the hy-
pothesis of “presetting”’ and subsequent
“setting” of the gene locus during de-
velopment of the sporophytic tissues,
mentioned earlier in this report. The non-
heritability of the settings suggests that
they are erased or modified, either in the
germ line or in the zygote or developing
embryo. There is some evidence, still in-
complete, to suggest that when a modifi-
cation occurs (instead of an erasure) it
is the same in all cells whose c,”—? locus
has undergone the same presetting and
setting process, and that its effect is ex-
pressed either in the presence or in the
absence of Spm.

Presetting and subsequent setting of a
locus, to control the action of its gene
many cell generations after the presetting
event has occurred, may be a general
attribute of control systems in maize.
First evidence of the process, obtained
with two states of ay"-*, was reported and
illustrated in Year Book 63 and Year
Book 64. In that material, contrary to the
findings reported here for c:”—! and c.”~?,
the effect of the presetting and setting
process is recorded in the aleurone layer
of the progeny kernels of plants carrying
Spm, but made evident only in those
kernels that do not receive Spm from
either the ear or the pollen parent. These
settings are nearly always erased in the
next plant generation at some period dur-
ing development, possibly in the germ
line. In a few cells, however, the locus
may escape the erasure process.

There is now a considerable body of
evidence that relates controlling elements
to the ordering of gene action during
576

development. A single system, such as
the Spm system, apparently accomplishes
this function in a number of distinctly
different ways. Each must reflect the
nature-of the elements themselves, their
modes of association with the gene loci,
and their individual actions and inter-
actions. The relation of the elements to
the gene loci has been investigated by
Dr. O. E. Nelson of Purdue University.
In studies (unpublished) of wx, wa,
and wx”-®, each of which arose through
the insertion of a controlling element at
the Wz locus, he has obtained evidence
that makes it possible to determine the
position of the element within the locus.
The site of insertion is different in each
of the three instances. The inserted
element does not noticeably reduce intra-
locus crossing over. It is not yet possible,
however, to relate controlling elements
to other known structures, as our knowl-
edge of chromosomal components is still
too limited. Nevertheless, the precision
with which the elements are able to regu-
late gene action reflects an orderly con-
trol mechanism, operating from the time
of zygote formation until the time of
maturation of tissues.

In eukaryotic organisms (plants and
animals having complex chromosomes
and well organized nuclei), the organiza-
tion of the chromatin is not the same in
all nuclei of an individual. Characteristic

CARNEGIE INSTITUTION

differences are observed in degree of
condensation of chromatin within nuclei
performing different functions. In some
types of cells, much of the chromatin
within a nucleus forms densely meshed
clusters, often connected with the nuclear
membrane. Biochemical studies have indi-
cated that the degree of activity of genes
within clusters is low by comparison with
that of genes in chromatin loosely dis-
persed in the nucleus. Thus some mecha-
nism must operate during development
to order the genes into clusters, and to
effect dissociation of particular genes
from such clusters so that they may
function at the proper times and to the
proper degrees. It is possible to consider
that in the regulation of this mechanism
the controlling elements may be involved.
Their initial positions and associations
within a cluster might result in some
modification that would alter their posi-
tions within the nuclei in subsequent cell
generations. Each altered association, in
turn, could affect the position of an ele-
ment in one or more succeeding cell
generations. Many of the observed com-
plexities of action of the controlling ele-
ments in maize could stem from such a
relatively simple sequence of events. The
“presettings” and “erasures” reported
here would appear less strange under this
interpretation.

BIBLIOGRAPHY

Burgi, E., A. D. Hershey, and L. Ingraham, Pre-
ferred breakage points in T5 DNA molecules
subjected to shear. Virology, 28, 11-14, 1966.

Burton, K., see Smith, M. G.

Hershey, A. D., The injection of DNA into cells
by phage, in Phage and the Origins of Molecular
Biology, J. Cairns, G. S. Stent, and J. D.
Watson, eds. Cold Spring Harbor Laboratory
of Quantitative Biology, pp. 100-108, 1966.

Hershey, A. D., see also Burgi, E:
Ingraham, L., see Burgi, E.

McClintock, B., The control of gene action in
maize, in Genetic Control of Differentiation.

Brookhaven Symposia in Biology, No. 18, pp-
162-184, 1965.

Skalka, A., Regional and temporal control of
genetic transcription in phage lambda. Proc.
Nail. Acad. Sci. U.S., 65, 1190-1195, 1966.

Skalka, A., see also Smith, M. G.

Smith, M. G., and K. Burton, Fractionation of
deoxyribonucleic acid from phage-infected
bacteria. Biochem. J., 98, 229-241, 1966.

Smith, M. G., and A. Skalka, Some properties of
DNA from phage-infected bacteria, in Sympo-
sium on Macromolecular Metabolism, New
York Heart Association, J. Gen. Physiol., 44,
No. 6, part 2, 127-142, 1966.
GENETICS RESEARCH UNIT

577

PERSONNEL

Year Ended June 80, 1966

Phyllis D. Bear, Carnegie Institution Fellow
Elizabeth M. Bocskay, Chief Clerk

Jennie S. Buchanan, Curator of Drosophila
Stocks

Elizabeth Burgi, Associate in Microbiology
Ruth Ehring, Carnegie Institution Fellow

Agnes C. Fisher, Secretary to Director;
Editor

Edward Goldberg, Carnegie Institution
Fellow

Alfred D. Hershey, Director
Laura J. Ingraham, Research Assistant

Barbara McClintock, Cytogeneticist
Gisela Mosig, Associate in Research

Anna Marie Skalka, Postdoctoral Fellow,
American Cancer Society

Mervyn G. Smith, Fellow, Damon Runyon
Memorial Fund

Rudolf Werner, Associate in Research _
Carole E. Thomason, Technical Assistant

Temporary and Part-Time
Anne K. Carhart, Technical Assistant
Frances C. Womack, Guest Investigator