THE MORPHOGENETIC EFFECTS OF THE HOODED · PDF filethe present study of the morphogenetic effects of the hooded ... in a drop of acetocarmine solution placed on ... epidermal cells

THE MORPHOGENETIC EFFECTS OF THE HOODED GENE IN BARLEY. I. THE COURSE OF DEVELOPMENT IN

HOODED AND AWNED GENOTYPES

G. LEDYARD STEBBINS AND EZRA YAGIL2

Department of Genetics, University of California, Davis

Received March 21, 1966

NTIL recently, the use of genes having pronounced effects on morphology as an aid to the solution of problems of morphogenesis has not been given

the attention which it deserves. In recent years, the value of such genes for this purpose has been greatly enhanced by our increased knowledge of the primary action of genes. There is now little doubt among geneticists that each gene pro- vides the “information” necessary for the specific structure of a polypeptide chain, and in this way determines specifically either the action of a particular enzyme or the properties of a structural protein. Single gene differences, therefore, determine primarily either the functioning of a particular enzyme, or some specific property of a cellular membrane or organelle of which a structural protein is an essential component. Their effects on visible morphological characteristics are always the secondary results of these primary activities at the biochemical and submicroscopic cellular level. The complete chain of events from primary gene activity to final expression of morphological differences is not known for any single gene. One step toward achieving this knowledge is to study intensively those events at the cellular and submicroscopic level which might provide a link between gene controlled processes and the form-determining activity of individual cells and groups of cells. It is with these goals in mind that the present study of the morphogenetic effects of the hooded gene has been undertaken.

The selection of hooded barley as an object of investigation was based primarily upon the fact that the effects of this gene upon the morphological and anatomical structure of the plant are as profound as are those of any single gene known to the writers, and have been variously interpreted by those botanists who have studied them previously. The principal developmental and anatomical characteristics of hooded barley have been described by MOSTOVOJ (1929) , BONNETT (1 938) , MYLER (1942) , TAKAHASHI et al. ( 1953) , and particularly by ARBER (1929,1934) and by HELM (1952). Their descriptions and interpretations will be reviewed and discussed critically at the end of the present paper.

The evidence that hooded behaves as a single Mendelian factor is extensive, and has been carefully reviewed by SMITH (1951 ) and NILAN (1964). Its action

‘This research was supported in part by National Science Foundation Grant No GB 227G Taken in part from a dissertation submitted to the Umverslty of California in partial fulfillment of the requirements for the Ph D degree

* Present address Uni5ersity of Tel Ariv, Israel

Genetics 54: 727-741 September 1966

728 G. L. STEBBINS A N D E. YAGIL

is generally regarded as dominant, although our observations, as well as those of MURTY and JAIN (1960), indicate that this dominance is incomplete. Under most conditions, the heterozygote ( K k ) can readily be distinguished from homo- zygous hooded ( K K ) . Linkage relationships indicate that it is located on the fourth chromosome (SMITH 1951; NILAN 1964). Mutations at the hooded locus have never been observed in barley lines cultivated by geneticists or breeders, even in the extensive experiments on mutagenesis by radiations and chemical agents which various geneticists have conducted ( GUSTAFSSON 1955; NOTZEL

1952). The absence of such mutations is remarkable, since nearly all of the other single gene differences commonly found in barley have been reproduced at least once in these experiments.

MATERIALS A N D METHODS

The barley varieties used predominantly in this investigation were the two lines Atlas 46 and Hooded Atlas 46. The former is a six-rowed, long-day requiring spring barley grown commer- cially in California. It was produced from Atlas barley as a variety resistant to scald and mildew. Hooded Atlas 46 was obtained through ten recurrent backcrosses of an original hooded variety, Colsess, to Atlas barley (SUNFSON and STEVENS 1957). Some additional observations have been made on awned and hooded lines of Atsel. This variety is day length neutral and flowers five to six weeks after sowing. I t arose as a mutant of Atlas (SUNESON and STEVENS 1957). Although these breeding precedures cannot be regarded as having produced completely isogenic lines, they ;an be regarded as isogenic in respect to genes affecting external morphology.

The plants were grown in 8-inch plastic pots either in the greenhouse or in a controlled environment chamber. The latter was adjusted to 16 hours light at 21 k 1°C and 8 hours dark at 17 i. 1°C. The light in the chamber was a combination of fluorescent and incandescent with an intensity of approximately 1500 ft.c. Plants for developmental studies grew in a 1:l mixture of Yo10 loam and sand fertilized by a commercial fertilizer mixture.

Observations on developing lemmas were made in two ways. Whole mounts were obtained either from fresh material or by fixing developing spikes in a mixture of 3 parts absolute alcohol: 1 part glacial acetic acid. After fixation, the material was stored in 70% alcohol. For observation, lemmas were dissected from the spikes with the help of a scalpel o r spear needle, then immersed in a drop of acetocarmine solution placed on a microscopic slide, covered with a cover glass, and gently heated for a few seconds over an alcohol lamp. After approximately 10 minutes the stain penetrated into all parts of the lemma. The lemmas were then transferred to a second glass slide on which was placed a drop of 46% acetic acid. It was then covered and sealed with wax. The lemma could be mounted with either the adaxial or the abaxial surface facing upwards, depending upon the observations to be made. This method permitted clear observations and photographs of the epidermal surface facing upwards, including cell walls and mitotic figures. The best photographs were those obtained immediately following the staining, since after a few hours the stain tended to diffuse.

In preparation for sectioning, apical meristems were fixed in either Carnoy’s or Navashin’s fixative. They were then dehydrated, using the tertiary butyl alcohol series, embedded in par- aftin, and sectioned longitudinally a t a thickness of 8 micra. In order to obtain the desired orientation of the specimens in the paraffin blocks, a method described by NICHOLS and MAY (1964) was used in which the specimens were threaded prior to embedding by a piece of thin wire inserted through the internode beneath the apex.

Following deparaffinization and hydration, the sections were stained with either the pyronin- methyl green stain for nucleic acids as developed by BRACHET and described by JENSEN (1962), or with the Feulgen stain for DNA. For the latter, the procedure described by BROOKS et al. (1950) was used with a modification developed and called to our attention by R. BAMMI (per-

MORPHOGENETIC EFFECTS O F HOODED 729

sonal communication), in which the basic fuchsin was decolorized for 20 minutes by means of SO, gas bubbling through the solution for 20 minutes.

In order to correlate the stages of development of the very different lemmas of the hooded and awned genotypes, anther length was measured with an ocular micrometer on each floret upon which observations were made. Since the development of the anthers and pollen in hooded and awned genotypes is identical, lemmas belonging to florets which had the same anther length o r stage of pollen development were regarded as being in the same stage of development.

RESULTS

Description of the mature hood: The following description of the mature lemma of the hooded genotype, based upon our observations, agrees with those of other authors (see introduction for references). The features described are illustrated in photographs (Figure 1 ) and a diagram (Figure 2). The lower portion of the lemma, which encloses the palea and the floret, is indistinguishable from the corresponding portion of the lemma in the awned genotype. Between this un- changed portion and the hood proper occur two lateral appendages, having es- sentially the same texture as the lemma proper. They may point either up (Figure 1 ) or down (Figure 2). Their shape vanes from triangular, which is the usual condition on the Atsel background, to elongate and attenuate, with awn-like tips. When they assume this latter shape, the lateral appendages are awn-like in both their macroscopic appearance and the shape and arrangement of their epidermal cells as seen under the microscope.

The hood itself is somewhat similar in shape to the lower part of the lemma proper, except for one important difference. It usually contains an extra palea which is inserted at its distal end. Enclosed between this palea and the hood proper is a rudimentary floret. In some instances, this floret contains the normal series of parts: two lodicules, three rudimentary anthers, and a rudimentary ovary with two small stigmas. Frequently, however, some of the parts are missing and some of them may be duplicated.

The fact that the extra palea and rudimentary floret are inserted at the distal end of the hood suggests that this structure may be inverted morphologically; i.e. its morphological “base” may be farther from the axis of the spikelet and spike as a whole than is its morphological “apex.” This impression of inversion is greatly strengthened by the appearance of the cellular pattern on the abaxial surface of the hood. Like the basal portion of the lemma proper, its epidermal cells are arranged in rows, in each of which elongate cells alternate with pairs of siliceous-suberous cells, as described originally by PRAT (1932) for leaves and other parts of various grasses. However, on the lemma proper, as well as on the upper leaf sheaths, culms, and other parts of the plants, the siliceous-suberous pairs have the siliceous cell in a distal and the suberous cell in a proximal position, while on the abaxial surface of the hood the reverse arrangement is found. So far as we are aware, this inverted condition is unique in the grass family.

The portion of the hooded lemma which lies distal to the hood proper is ex- tremely variable, depending upon both the genotype and its developmental environment. Its greatest development is shown in the diagram (Figure 2) which


2 FIGURE 1.-Photograph of three spikelets of hooded Atlas barley, showing the structures of

FIGURE 2.-Diag” showing the differences between awned and hooded lemmas, and the the hood.

direction of polarization in the latter.

MORPHOGENETIC EFFECTS O F HOODED 73 1

illustrates a short rachilla, distal to which is a rudimentary lemma-like structure enclosing a still more rudimentary palea and bearing a short, weak awn. Within this uppermost palea-like structure a small, rudimentary ovary sometimes occurs, but this second extra floret has been seen to possess rudimentary “stamens” only very rarely in our material. The commonest reductions from this condition are the reduction or absence of the awn, the weak development o r virtual absence of the rachilla connecting the two rudimentary florets, and the reduction of the uppermost “floret” to a tiny rudimentary palea. The development of the terminal awn is also very variable. In the Atsel hooded genotype, the entire upper portion of the hooded lemma is usually reduced to a blunt, triangular protuberance.

The nature of the hood, as well as the genetic evidence indicating that its formation is governed by a single Mendelian factor, raises the following problems, which are of basic morphogenetic significance. (1) Are all of the various structures found on the hooded lemma produced as secondary effects of a single primary gene action? (2) Are the extra florets found on the hood truly epiphyl- lous in position; i.e., is the axial tissue from which they arise completely sepa- rated from the remainder of the axial system of the plant and inserted on the distal portion of a foliar appendage? An alternative hypothesis is that these extra florets arose originally upon a separate pedicel, which later became completely adnate to the lower part of the lemma. (3) What is the basis of the remarkable inversion of polarity characteristic of the hood? Can it tell us anything about the factors which determine polarity in the organs of higher plants?

In the following description of development, those points will be emphasized which might contribute toward a solution of these three problems. They will be considered in detail in the DISCUSSION and in subsequent papers.

Early development of the hood: The development of the barley spike has been carefully described by BONNETT (1 935) with whose description our observations are in complete agreement. Up to and including the earliest stages of development of the lemma primordia, careful comparisons failed to reveal any differences between awned and hooded genotypes in either the shape or histological structure of any of the primordia formed. At this stage, the primordia are 700 to 800 micra long in the Atlas hooded and awned varieties, and 200 to 300 micra long in the Atsels.

At the macroscopic level, the initiation of the difference between the lemma primordia of the two genotypes can be seen as a divergence in the overall length of the organs. The allometric growth formula of HUXLEY (1932) was applied to the relative lengths of lemma and anther primordia in awned and hooded Atlas genotypes. The values obtained for these lengths were plotted to express the logarithmic derivation of the formula y = bzk, i.e., log y = log b f k log 2. Figure 3 shows the position of the values when plotted on a double-logarithmic scale. The values for the awned genotype fall along a single straight line, with k = 2.32. In the case of the hooded genotype, the data agree best with two straight lines, the upper of which has a gentler slope than the lower. Since the point of intersection of these two lines could be seen immediately to correspond with the stage of development at which histological differences between the lemmas of


FIGURE 3.-Linear regression lines showing growth correlations between anther and lemma length. Open circles are of awned genotypes, black circles represent hooded, LP = lemma primordia, C = cushion, HP = hood primordia. See text for further explanation.

2 0 0 400 6 0 0 ANTHER LENGTH le1

the two genotypes were becoming recognizable, the values for the hooded primordia were divided into two groups, those obtained prior to any visible sign of hood differentiation, and those made upon lemma primordia having some visible degree of hood differentiation. The two regression lines obtained are shown in Figure 3. The k value for predifferentiation stages is 2.06, which does not differ significantly from the value (2.32) obtained for the awned genotype. For the postdifferentiation stages of the hooded lemma primordia, k sinks to 0.91, and the difference between this value and the two others is highly significant.

The earliest histological differences between the lemma primordia of hooded and awned genotypes appear at the stage when the difference between them in overall growth rate, as shown in Figure 3, first becomes evident. This consists of a divergence in the size of the cells constituting the adaxial epidermis. In awned genotypes, these cells become gradually larger while in hooded they decline in size. This divergence, plotted against anther length, is represented in the graph, Figure 4. In spite of considerable variability within each genotype, this difference is clearcut, and soon reaches a point at which no overlapping exists between extreme values for the two genotypes.

200

I S 0

I O 0

so

MORPHOGENETIC EFFECTS OF HOODED 733

FIGURE 4.-Correlation between anther length and mean cell area in the adaxial epidermis of developing lemmas. Open circles, awned; black circles, hooded. The appendages of the circles represent the direction of cellular divisions.

This divergence in cell size is due to an increasing difference in the amount of cell elongation which takes place during the interphase between successive mitoses. In the awned genotype, this elongation is more than enough to make up for the loss in cell size which accompanies the division of one cell into two; in hooded, the amount of cell elongation during interphase is not enough to make up for this loss. Since cell width does not change in either genotype at this stage, the elongation is entirely parallel to the long axis of the awn. Evidence from experiments on the incorporation of tritiated thymidine, to be reported in a later paper, indicates that the reduction in the amount of elongation is the result of a shortening of the interphase between successive mitoses.

These differences in cell size are accompanied by equally conspicuous differences in cellular pattern, as shown in Figure 5 . In the awned genotype, the mitotic spindles are consistently oriented parallel to the long axis of the awn, and all divisions in the epidermal and subepidermal layers are anticlinal. In hooded, however, epidermal cells divide at various angles with respect to the long axis of the lemma. Although these divisions are all anticlinal, periclinal divisions occur in the subepidermal layer (Figure 6 B) . These divisions give rise to an elevated dome or cushion, from which the organs of the first extra floret are differentiated (Figure 7) . In its histological structure, this cushion bears a strong

734 G. L. STEBBINS AND E. YAGIL

L.". .lon..,., /

tl."yli." ,..I.. Cwhi.n

FIGURE 5.-Drawings of the adaxial epidermal surface of the lemma primordium of hooded at three stages of development (A, D, and E) and of 'awned (B, C ) at two later stages. These illustrate the cell pattern, and the relationship between cellular shape and the orientation of mitotic spindles.

resemblance to that portion of the primordial spike from which the spikelet primordia are differentiated. This structure is shown, for comparison, in Fig- ure 8. The significance of this resemblance will be discussed below.

Later development of the hood and the laterat appendages: The development of the rudimentary florets from the meristematic cushion has been well described and illustrated by MOSTOVOJ (1929) and HELM (195Q). The number of parts differentiated appears to be proportional to the size of the meristematic cushion from which they originate. One of the most significant facts about the development of this area is the persistence of the meristematic condition in many parts of it. At an anther length of 600 to 700 micra, with PMC's at the stage of meiotic prophase, the entire hood area is still completely meristematic. At a corresponding stage in the awned genotype, only the body of the lemma and the lower part of the awn are still in the meristematic condition, the upper third of the awn being either transitional or completely mature tissue. At an anther length of 1.4 mm, with anthers containing young pollen, the lateral appendages and the apical awn primordium are still completely meristematic, although the region around the base of the extra florets has reached a condition transitional toward maturity. At the same stage, meristematic tissue in the awned genotype is confined to the


~'IGIJIIFS 6-8: FIGURE 6.-hngitutlir1i11 niedian sections of lemma primordia of awned (6;)) And hooded (6b) lemmas stained with Feulgen. Figure 6b shows the effect of periclinal divisions in the subepidermal layer of the cushion. I h x " 7.-Median longitudinal section of a lemma primordium of hooded nt the stage when the organs of the extra floret are beginning to differentiate. FIGURE 8. An early stage in the formation of spikelet primordia from the spike primordium of Atlas barley. Awned and hooded genotypes are indistinguishable at this stage. Note the resem- Idancc in histological structure to Figure 6b.

central part of the body of the lemma and a restricted intercalary meristem at the base of the awn. When the anthers have reached a length of 2mm and contain 2-nucleate pollen grains, the tissue of the awn in the awned genotype is no longer meristematic, and only a few cell divisions persist in the central part of the lemma body. At this stage, the apical awn and a considerable portion of the hood in the hooded genotype are still meristematic. The termination of meristematic activity in the hood itself does not occur until shortly before anthesis.

The region which eventually forms the base of the hood proper remains in a meristematic condition for a longer time than any part of the hooded lemma. Until the anthers have reached a length of about 1.6 mm and contain uninucleate pollen grains, this region is completely meristematic, and the inverted pattern of cell differentiation, found in the mature hood, has not yet become evident. On the other hand, a normal uninverted cell pattern is found on the abaxial epidermis of the hood region at a relatively early developmental stage (anther length 0.8 to 1 mm, pollen mother cells in meiosis). As the hood grows, the progression of change in meristematic activity is similar to that in the body of the lemma proper, except that it is inverted. Asymmetrical mitoses as described by STEBBINS and SHAH (1959) but with inverted polarity appear first at the morphological

736 G. L . S T E B B I N S A N D E. YAGIL

base (topographically the distal region) of the hood, and soon afterward in the region nearest to and oriented toward the lateral appendages. The middle part of the hood resembles the corresponding portion of the lemma body in being the last to form asymmetrical mitoses and to reach maturity. In all of its developmental stages, the body of the hood resembles an inverted lemma, with its distal portions oriented toward the lateral appendages.

DISCUSSION

Single versus multiple primary gene action: BONNETT (1938) believed that the differences between hood and awn are too numerous and diverse to be produced by the action of a single gene. One of us (YAGIL) will present in a later paper evidence which shows that environmentally induced modifications of the expression of the hooded character affect all of the different structures of the hood complex simultaneously and to about the same degree. This suggests that alterations of a single physiological process can produce simultaneously all of the morphological differences between the hood complex and the awn. If this is so, then the process cencerned should affect principally or entirely the earliest of the histological divergences which can be observed, and this divergence should be responsible for the later changes. As pointed out in the descriptive section, this earliest change is the shortening of the interphase between mitoses on the adaxial epidermis and subepidermal layers in the distal region of the lemma primordium, which results in a greater proliferation of cells, a reduction in the amount of cellular elongation between mitoses, and a consequent reduction in mean cell size. Immediately following this change is the alteration in the orientation of the mitotic spindles from one plane to three. Are these phenomena causally connected?

The relationship between tempo of mitotic rhythm and orientation of mitoses which would account for a causal connection between the two is represented in

AWNED

0 - E - R-El-1 - - Division Elongation Division Elongation

HOODED

Gene acts 0-u- Elongation NO - H-H- Division Divieian Division

FIGURE 9.-A schematic diagram illustrating the relationship between changes in cell size, shape, and orientation of mitotic divisions in developing lemma primordia of awned and hooded genotypes.

MORPHOGENETIC EFFECTS O F HOODED 73 7

the diagram, Figure 9. If a developing structure such as an awn is elongating in a single direction, then a slow tempo of mitosis, which permits cells to become longer at each successive prophase, will produce cells with their long axes always oriented parallel to the long axis of the organ concerned. If some factor connected with cell shape determines that the orientation of the mitotic spindle shall be parallel to the long axis of the cell, mitoses will be consistently oriented in the same direction and will contribute to the elongation of the organ concerned. If, on the other hand, the tempo of mitosis exceeds that of cell enlargement, cell shape will change with each successive mitosis, and the orientation of mitotic spindles parallel to the long axis of the cell will produce divisions on many planes. A causal connection between alteration of mitotic tempo and orientation of mitoses would depend, therefore, on the existence of a similar connection between cell shape and orientation of the mitotic spindle.

The question as to whether such a connection exists has been carefully discussed by SINNOTT (1 960) , who has reviewed much of the older literature concerning it. HOFMEISTER, SACHS, and other botanists of the 19th century main- tained that new cross walls between cells are formed across the shortest distance possible in the preexisting cell. At the time when it was formulated, this was a reasonable supposition, but at present it is inadequate and misleading, since it focusses attention on wall formation, which is a late phenomenon of cell division, and is itself the consequence of the orientation of the mitotic spindle. As SINNOTT (1960) and others have pointed out, new cell walls are formed from a phragmo- plast which first appears across the equator of the mitotic spindle, so that the orientation of this structure is directly responsible for the position and orientation of the new cross wall.

The statement that the mitotic spindle is usually formed parallel to the long axis of the cell was first made by HERTWIG (1885) on the basis of observations on cleavage in invertebrate eggs. As a generalization, it is by no means univer- sally true in higher plants. Exceptions are well known in cambial tissue. In the epidermis of the barley lemma, the spindle is oriented at right angles to the long axis of the cell in many of the divisions taking place in the stomatal row prior to the formation of guard cell mother cells. Nevertheless, HERTWIG’S generalization does hold in many embryonic plant tissues, particularly the early stages of development of the embryo in both ferns and seed plants. Consequently, the question pertinent to the present problem is not the general validity of the prin- ciple, but whether it holds in the developing lemma primordium of hooded barley. Repeated observations indicate that it does, and that nearly all of the mitoses in these primordia are oriented with the axis of their spindles parallel to the long axis of the cell. This fact can be seen in Figure 5, D and E.

In general, the conclusion is justified that some factor related to cell shape has an orienting influence on the mitotic spindle, particularly in developing embryos and in rapidly dividing, evacuolate meristematic cells. In the developing lemma primordium of hooded barley, this factor produces a direct causal connection between the acceleration of cell proliferation and the shift from unidirectional and multidirectional orientation of cell division.

Given the shift in orientation of mitoses from one plane to three, and the con-

738 G . L. STEBBINS AND E. YAGIL

sequent elevation of the meristematic dome or cushion on the adaxial surface of the lemma primordium, the next question is: Does the existence of this cushion determine directly the differentiation from it of the various structures which constitute the hood and the rudimentary floret o r florets contained in it?

A complete answer to this question is at present impossible, since we do not know what factors determine the differentiation of appendages from undiff er- entiated meristem in any higher plant. Nevertheless, a comparison between the cushion and the spike meristem is enlightening. As shown by Figures 6 B and 8, both meristems consist of an outer tunica, in which all cells divide anticlinally, of three or four subepidermal layers in which both anticlinal and periclinal divisions are found, and of a lower layer of relatively large cells which are not dividing. The principal difference between the two meristems is that the tunica of the cushion on the hooded lemma is always one cell layer in thickness, while in the spike and spikelet primordia it may be either one or two cells in thickness.

In the developing spike of barley and other grasses, the differentiation of reproductive appendages follows so closely and consistently upon the formation of a histologically structured spikelet primordium, that a causal connection between the two may reasonably be postulated. In terms of modern developmental biology ( WADDINGTON 1962), they may be regarded as successive steps in an epigenetic sequence which is initiated by the shift from the vegetative to the reproductive meristem. The action of the hooded gene in initiating a second cycle of differentiation of reproductive parts becomes understandable on the basis of either one of two reasonable hypotheses. We may postulate either that the histological structure of the meristem is itself responsible in some way for the consistent differentiation of appendages from it, or that the histological structure and some unknown physiological conditions which are responsible for its differentiation are both determined by factors which are inherent in the building up of a three- dimensional, compact meristem possessing a tunica layer and consisting of small, rapidly dividing cells. On the basis of either postulate, we can make the hypothesis that the hooded gene initiates a new epigenetic sequence of reproductive organ differentiation by its acceleration of the mitotic rhythm, and the consequent organization of a three dimensional structure which has both the histological and the physiological properties of a floret primordium.

Possible factors concerned with the reversal of polarity: The later stages of hood development are characterized most strikingly by the reversal of polarity which is evident both in the mature hood and the way in which it develops. How is this reversal brought about?

A complete answer to this question will not be possible until the developmental sequence of the hooded lemma has been much more carefully studied. Neverthe- less, some facts are available which should provide a clue toward the better understanding of this problem.

The first important fact is that the mitotic divisions in the epidermis of the hood which have reversed polarization wcur very late in development. They do not begin until the rudimentary florets, which are distal to the main body of the hood and which at maturity are contained in its basal portion, are already well


developed. Furthermore, in developing lemmas which contain two extra florets, the rachilla between these florets is already differentiated at the time when growth with reversed polarization begins. The mitoses with reversed polarization have, therefore, one feature in common with the normally polarized mitoses which, much earlier in development, produce the main body of the lemma below the hood primordia. In both examples, the gradient of polarization radiates outward from a complex of tissues which represents a developing axis and its appendages. In the early lemma primordium, this axis is that of the spike itself and the much reduced secondary axis of the spikelet. In the hood primordium, polarization gradients radiate outward, both distally and proximally, from the already differentiated primordia of the extra florets, and of the isolated segment of rachilla which connects them. This similarity leads us to the hypothesis that in the developing spike of barley, polarization is somehow determined by the developing tissues of an axis-appendage system. In a normal barley spike, all parts of the axis are connected to each other, and polarization is always in a direction distal to the primary or secondary axis. In hooded, on the other hand, an extra axis-appendage system develops from the cushion which is located on the distal half of the lemma primordium, and this system functions as a secondary center for the radiation of polarity gradients.

The mechanism by which this radiation of polarity gradients is brought about is most probably associated with the formation of hormones or growth substances. Since a change in the tempo of the mitotic rhythm takes place, the action of a kinin or kinins as well as of indole acetic acid is probably involved. Furthermore, the distal o r nearly distal position of the development of the cushion on the lemma primordium would be most easily explained if this altered growth pattern was being affected by substances entering the primordium from below. In this connection, the observations of KULAEVA (1962) and KENDE (1965), that kinin-like substances are formed in the roots of plants and are transported upward through the stem, may be highly significant. Experiments to determine whether growth substances entering from below can affect the expression of the hooded gene are being initiated in the senior author’s laboratory. At present, therefore, it seems best to defer any further discussion of the primary action of the hooded gene until more information is available. A number of hypotheses can nevertheless be conceived which would explain all of the morphological characteristics of the hooded phenotype on the basis of a single alteration of the growth pattern at an early primordial stage of the lemma. Figure 10 is a summary of the probable epigenetic sequence of events which would make this secondary pleiotropic action possible.

SUM MARY

Previous studies, verified by the present authors, have shown that the dominant gene hooded in barley gives rise to a lemma which bears on its adaxial surface one or two extra florets, the proximal one of which shows inverted polarization. A pair of awn-like lateral appendages is found proximal to this floret. When growth is plotted on a logarithmic basis, that of the awned lemma

740

I

G. L. STEBBINS AND E. YAGIL

Early development o f lema primordium

(similar in kk and KK)

\ Action of kk -

J, . Gradual transition from meristematic to elongating and mature tissues in distal region. Restriction of meristem to proximal region.

I

Maturation of awn

b Action of KK

I .b

Increased cell proliferation and long persistent meristem in distal half of lemma primordium.

5 Formation of structured cushion L

Alteration in direction of movement and/or site of production of growth substances

,I Differentiation of / organs

J 1 Formation of rachilla- like base and connection between extra florets mitoses with inverted

Induction of intercalary growth and asymmetrical

1 polarization in hood

i 1 region

Formation of terminal awn

Formation of hood proper

FIGURE 10.-Diagram of the epigenetic sequence of events initiated by the action of the awned (kk) and the hooded ( K K ) alleles.

is linear and constant, while that of the hooded lemma undergoes an abrupt change at the time when the hood primordium is first differentiated. Histologi- cally, the first indication of hood differentiation is a more rapid proliferation of epidermal and subepidermal cells on the adaxial surface of the lemma. This is accompanied by a reduction in cell size and the orientation of mitotic spindles in several planes rather than strictly parallel to the long axis of the lemma. In this way, a meristematic dome or cushion is developed which bears a striking resemblance to the earlier stages of development of spikelet primordia. The parts of the extra florets are differentiated from this cushion.-In respect to its later development, the hooded lemma is distinctive in the long persistence of meristematic tissue in its distal regions. The lateral appendages emerge as localized con- tinuations of the marginal growth which begins in the middle part of the pn- mordium immediately after cushion formation. Their cells are continuous with those of both the main body of the lemma and the primordial hood. The inverted polarization in the hood region becomes manifest in the pattern of epidermal cells which are produced as the lateral appendages are maturing, and which are formed in rows continuous with those of the lateral appendages.-All of the morphological characteristics of the hooded lemma are probably produced by the

M O R P H O G E N E T I C E F F E C T S O F HOODED 741

action of a single gene, which alters the course of development at an early primordial stage, and initiates an entirely new epigenetic sequence of development. Hypotheses concerning the nature of this sequence are formulated. They indicate that a connection exists between rate of cell division and the orientation of divisions in meristems, and that inverted polarity is brought about first by the altered distribution of meristematic regions, and second by alterations in the distribution and/or the locations of formation of growth substances.

LITERATURE CITED

ARBER, A., 1929 Studies in the Graminae. VII. On Hordeum and Pariana, with notes on “Nepaul barley.” Ann. Botany 43: 507-533. - 1934 The Graminae. A study of cereal, bamboo and grasses. Cambridge, England.

Principles of Plant Physiology. Freeman, San Francisco. -

Hood and supernumerary spike development in barley. J. Agric. Res. 57: 371-377. Plant Microtechnique Manual.

BONNER, J., and A. W. GALSTON, 1952 BONNETT, 0. T., 1935

BROOKS, R. M.. M. V. BRADLEY, and T. I. ANDERSON, 1950

HELM, J., 1952

HERTWIG, O., 1885

HUXLEY, J. S., 1932 JENSEN, W. A., 1962 Botanical Histochemistry. Freeman, San Francisco. KENDE, H., 1965

KULAEVA, 0. N., 1962

MOSTOVOJ, K. I., 1929

MURTY, G. S., and K. B. L. JAIN, 1960

MYLER, J. L., 1942 NICHOLLS, P. B., and L. H. MAY, 1964

NILAN, R. A., I964

PRAT, H., 1932

SINNOTT, E. W., 1960 Plant Morphogenesis. McGraw-Hill, New York. SMITH, L., 1951 SUNESON, C. A., and H. STEVENS, 1957

barley. Agron. J. 49: 50-53.

TAKAHASHI, R., J. YAMAMOTO, S. YASUDA, and Y. ITANO, 1953 in barley. Ber. Ohara Instituts landwirtschaftliche Forschungen 10 : 29-52.

WADDINGTON, C. H., 1962 Press. New York.

The development of the barley spike. J. Agric. Res. 51 : 451-457. 1938

University of California, Davis.

Weizen. Flora 139: 96-147. Zur Ontogenese der Kapuze bei Kapuzengersten und der inflaten Spelzen beim

Welchen Einfluss iibt die Schwerkraft auf die Theilung der Zellen? Jenai-

Problems of Relative Growth. MacVeagh, N.Y.

sche Z. Naturwiss. 18: 175-205.

Kinetinlike factors in the root exudate of sunflowers. Proc. Natl. Acad. Sci. U.S. 53: 1302-1307.

on leaves. Soviet Plant Physiol. (Engl. Transl.) 9: 182-189.

Cesk. Akad. Zemedelske. Vest. 5 : 22-30.

the lateral florets and certain other characters. J. Indian Botan. Soc. 39: 281-308.

The effect of roots on leaf metabolism in relation to the action of kinetin

Ein Beitrag zur Kenntnis der Kapuzenentwicklung bei der Gabelgerste.

Genetic studies in barley. 11. Inheritance of fertility of

Awn inheritance in barley. J. Agric. Res. 65: 405-412. Studies on the growth of the barley apex. 11. On the

initiation of internode elongation in the inflorescence. Australian J. Biol. Sci. 17: 619-630.

The cytology and genetics of barley. Monographic Suppl. No. 3, Washing- ton State Univ. Vol. 32: No. 1.

L’Epiderme des Graminkes. Etude anatomique et systematique. Ann. Sci. Nat. Sei. 10, Bot. 14: 117-324.

Cytology and genetics of barley. Botan. Rev. 17: 1-51, 133-202, 285-355.

An evaluation of smooth-awned and Hooded-Atlas

Inheritance and linkage studies

N e w Patterns in Genetics and Deuelopment. Columbia University

THE MORPHOGENETIC EFFECTS OF THE HOODED · PDF filethe present study of the morphogenetic effects of the hooded ... in a drop of acetocarmine solution placed on ... epidermal cells

Documents