574 The Plant Gell
LETTER TO THE EDITOR
Goethe, Sex, and Flower Genes
Although Iappreciatethe excellent series
of papers on plant reproduction in the
October 1993 special issue of The Plant
Cell, I would, nevertheless, take exception
to some of the basic comments on flower
morphologyby Coen and Carpenter(1993)
in their otherwise informative article.
The authors argue, with Goethe (1790),
that flowers and shoots are fundamentally
equivalent, that is, that the different parts
of a flower (sepals, petals, stamens, car-
pels) are equivalent to the leaves of a
shoot. Because internodes in the flower
are “so short as to be barely visible”
(p. 1175) or, in contrast to floricaula, “un-
detectable” (p.l176), a shortening of the
internodes is envisioned. The phyllotaxy
of floral organs is usually different from
that of leaves, and flower growth is deter-
minate (both apically and laterally) instead
of indeterminate. Thus, organ identity,
internodeIength, phyllotaxy, and determi-
nacy have all been changed. These
changes are thought to be “simply differ-
ent modifications of a common growth
plan,” produced by “different permutations
of a few key features of plant growth”
(p. 1175).
In positing this equivalence theory of
the flower and stem, the authors have,
however, not considered the most impor-
tant aspect of flower morphology:sexuality
or sexual reproduction, i.e., that aspect
of the flower for which it exists at all. In
fact, Goethe, seeing that his theory
(“everything is leaf”) did not correspond
to the sexual aspect of plant reproduction,
later denied sexuality in plants (Goethe,
1820),together with the botanists Schelver
(1812) and Henschel(1820),although the
sexual functions of the stamens and car-
pels had been shown conclusively by
Camerarius(1694), Linn6 (173!5), Kalreuter
(1761-1766), and, finally, Sprengel(1793).
One should also keep in mind that
Goethe’s statements were made in the tra-
dition of Platonic idealism and had no
phylogenetic implications (Goethe speaks
repeatedly of the “inner identity” of flow-
ers and shoots and the “identityof all plant
parts”).
However, there is no equivalenceof the
flower and shoot concerning sexual
reproductive functions. Kamalay and
Goldberg (1980, 1984) showed that “both
the anther and the ovary contain approxi-
mately 10,000 diverse mRNAs that are not
detectable in heterologousorgan system
mRNA or nuclear RNA populations”
(Drews and Goldberg, 1989, p. 259). Sev-
era1 other authors have obtained similar
results (Willing and Mascarenhas, 1984;
Smith et al., 1989; Koltunow et al., 1990).
And, beginningwith Sommer et al. (1990)
and Yanofsky et al. (1990), several flower
morphology genes have been cloned and
sequenced that are expressed only in the
flower and nowhere else.
Whatever the exact number of special
regulatory and target sequences that are
involved in and expressed during flower
development only, the independentresults
of several authors that a large number of
genes are expressed solely in the flower
clearly disproves the simple equivalence
of flowers and shoots claimed by Goethe
and his followers.
In contrast to the equivalence theory
(and bearing in mind that from the func-
tional point of view, all the flower structures
finally serve, or are produced for, the aim
of sexual reproduction and propagation),
one could argue that during flower devel-
opment most of the typical characters of
the vegetative shoot are more or less pro-
gressively lost, being replaced by the new
genetic programs and correspondingfea-
tures for sexual reproduction. In this
respect, the floweris not equivalent to, but
essentially different from, and definitely
more than, the shoot Sexuality is the rea-
son why the meristem structure, organ
identity, internode length, phyllotaxy, and
determinacyof theflower are all sodiffer-
ent from those of the shoot. Let it be
emphasized again that entirely different
genetic programs are switched on during
this stage of development. Or, speaking
allegorically, a new chapter with a new
topic is opened, a new movement of the
developmental symphony has begun.
This does not exclude, of course, the
possibility that the introductory phase of
flower development (sepals and petals)
consists largelyof the repetitionof already
known genetic programs (Drews et al.,
1992), transformed by a minority of new
regulatory gene functions for special de-
velopmental and morphological tasks. The
mainpart of flower development,however,
is made up of the organs for sexual
reproduction, i.e., the stamens and car-
pels, so that the large majority of the new
programs is expressed here.
Although many researchers have felt
that mutants transforming floral organs
into leaflike structures seem at first glance
to imply that flowers and leaves are in fact
equivalent, the following homeotic exam-
ples will reveal just the opposite. In the
well-known plena mutants(see, for exam-
ple, Masters, 1869), stamens are replaced
by petals, and carpels by sepals, followed
by further petaloid structures in the cen-
tral part of the flower. Because the
MADS-box gene PLENA in Antirrhinum
(and the homologous gene AGAMOUS in
Arabidopsis) is essential for turning on the
complex genetic programs for stamen and
carpel formation, its loss of function also
implies the loss of these sex organs, and
the simpler peta1 and sepal programs
are continued into the two inner flower
whorls-that is, the field where the PLENA
gene is normally expressed (Yanofsky et
al., 1990; Bradley et al., 1993).
Now, does this loss of PLENA gene
function, with the accompanying IOSS of
the thousands of additional diverse
mRNAs for sexorgan formation, imply that
the genetic programs for leaves, Sti“nS,
and carpels are all equivalent? Sattler
(1988, p. 1607) gives the answer for the
homeotic plena change by the following

analogous illustration: “lf a botanist in a
biology department is replaced by a zool-
ogist (as too often happens), the latter is
nota transformed botanist simply because
his predecessor was a botanist; he only
occupies the same position. From the
sameness of position, it does not follow
that members occupying it are also es-
sentially the same; they may be similar
or very different.” And in the case ofplena,
the members (floral organs) are, in fact,
very different: lnstead of sex organ
formation, we find the substitutional ex-
pression of the simpler petal and sepal
programs. On the other hand, ectopic ex-
pression of PLENA or AGAMOUS in the
two outer flower whorls and the corre-
sponding formation of carpelloid sepals
and staminoid petals (Mizukami and Ma,
1992; Bradley et al., 1993) means, first and
foremost, that the additional diverse
mRNAs are expressed earlier in develop-
ment and does not per se prove that the
programs for the vegetative structures are
equivalent to those of the sex organs.
Even for such similar nonreproductive
organs as leaves, sepals, and petals, the
results of molecular,genetics are not al-
ways as simple as might perhaps be
expected. In combiningGoethds ideas on
metamorphosis with Darwin’s theory,
green petal mutants have often been in-
terpreted as atavisms. However, one may
raise the question of whether changes in
the MADS-box gene DEFlClENS (DEF),
whose alleles cause different degrees of
greenish petals in Antiffhinum (Sommer
et al., 1990; Schwarz-Sommeret al., 1992),
corroborate this idea. The loss of fully
functioning binding sites of a transcrip-
tion factor that normally upregulates
expression of DEF in cooperation with
GLOBOSA (Trobner et al., 1992), as well
as all their target genes in petals and sta-
mens, shows only that in the mutants the
switch to the petal program is not very suc-
cessful and that (among other homeotic
deviations), the sepal program is now
more or less continued into the second
flower whorl.
The MADS-box sequences are con-
served from yeast to humans. As
expected, losses of functions in MADS-
box genes affect the different biological
systems very differently, and there is no
trend to classify the deviations as atavisms
(for a review, see Davies and Schwarz-
Sommer, 1994). Moreover, hardly anyone
would assert that the different (promoter,
MADS-box, and K-box) alleles of DEF
causing different degrees of greenish pe-
tals (including the temperature-sensitive
allele defA-101) are the atavistic gene se-
quences of flowering plants. The same
can probably be said of the alleles of the
homologousAPETALA3 gene in Arabidop-
sis (Jack et al., 1992). The substitution of
a late developmental program by an ear-
lier one does not necessarily prove the
equivalence of both. Although the bulk of
diverse mRNAs is equivalent in leaves and
petals, the different govefning fegulatory
systems and the corresponding qualita-
tive and quantitative mRNA differences
of these distinct organ classes are obvi-
ously not. Becauseof organ-specific gene
expression at the RNA and/or protein
levels (Jack et al., 1994), targeted gene
tagging can often concentrateon special
flower (or other) characteristics (see, for
example, Lonnig and Huijser, 1994).
Furthermore, the null alleles of DEFand
APETALA3 display not only sepals instead
of petals but also carpels in place of sta-
mens. As far as I am aware, it has never
been proposedfrom this that stamens are
“derived” from carpels.
In the often quoted case of Goethe’s “ex-
ample of a rose grown through” (Goethe,
1790; Weberling, 1989)-a malformation
in which flower development is stopped
shortly before, or at the beginning of, an-
ther development and shoot growth is
resumed-the flower programs are
switched off at this developmental stage
and the stem programs are turned on
again. Asserting the equivalence of both
programsfor this reasonwould be simply
illogical.
In short, the expression of specialshoot
features in abnormal flowers shows only
that such shoot programs (or parts of
them) were turned on et the wrong time
and at the wrongplace. They do not prove
the equivalence of the genetic programs
for stems and flowers any more than the
May 1994 575
LETTER TO THE EDITOR
accidental mixing of two different com-
puter text programson biology and poetry
proves their equivalence, although the
same letters, words, and basic grammar
may be involved (the words may be iden-
tical, but the message is not).
Beginning with systematic research by
Payer (1857), many other authors have ba-
sically come to the same conclusion
(involving also the telome theory, in which
both stamens and integumentsare thought
to derive not from leaves but from telomes,
as well as the sui generis theories of flow-
er structures and many further ideas and
arguments that cannot be discussed
here). Abridged histories of the different
views are given by Meeuse (1987) and
Leroy (1993).
Molecular genetics has definitely dis-
proved Goethe’s idea of the equivalence
of flowers and shoots by showing that an-
thers and ovaries express thousands of
diverse mRNAs that are not detectable in
heterologous organ systems and that cer-
tain specific cloned genes are expressed
only in the flower and not in the shoot (or
vice versa). Although “simply different
modifications of a common growth plan”
may be envisioned to be involved in some
of the changes between the stem and the
flower, the basic cause for a// the differ-
ences in meristem structure, floral axis
anatomy, organ identity, internodelength,
phyllotaxy, and determinacy is sexual
reproduction having noequivalencein the
stem.
Because misconceptions on this topic
are very widespread, I think that publica-
tion and discussion of this letter would
help plant science to clarify these points.
Wolf-Ekkehard Lijnnig
Max-Planck-lnstitut für
Züchtungsforschung
Carl-von-Linnb-Weg 10
50829 Koln
Germany
REFERENCES
Bradley, D., Carpenter, R., Sommer, H.,
Hartley, N., and Coen, E. (1993). Com-
576 The Plant Cell

LETTER TO THE EDITOR

plementaryfloral homeotic phenotypesresult Kamalay, J.C., and Goldberg, R.B. (1980). W.-E., Saedler, H., and Sommer, H. (1992).
from opposite orientations of a transposon Regulationof structural gene expressionin Characterization of the Antirrhinum floral
at the plena locus of Antirrhinum. Cell 72, tobacco. Cell 19, 935-946. homeotic MADS-box gene deficiens: Evi-
85-95. Kamalay, J.C., and Goldberg, R.B. (1984). dente for DNA binding and autoregulation
Camerarlus, R.J. (1694). Epistola ad M.B. Organ-specific nuclear RNAs in tobacco. of its persistentexpressionthroughout flower
Valentini de sexu plantarum. In Ostwalds Proc. Natl. Acad. Sci. USA 81, 2801-2804. development. EMBO J. 11, 251-261.
Klassiker der exakten Naturwissenschaften, Kolreuter, J.G. (1761-1766). VorlSiufige Smith, A.G., Gasser, C.S., Budelier-Sachs,
No. 105, 1899 (Leipzig: Verlag von Wilhelm Nachricht von einigen das Geschlecht der K.A., Hinchee, M.A., McCormick, S.,
Engelmann). Pflanzen betreffenden Versuche und Beo- Horsch, R.B., and Fraley, R.T. (1989).Struc-
Coen, E., and Carpenter, R. (1993). The bachtungen, nebst Fortsetzungen1,2, u. 3. ture and regulation of organ- and
metamorphosis of flowers. Plant Cell 5, In Ostwald’s Klassikerder exakten Wissen- tissue-specificgenes: Regulatedexpression
1175-1181. schaften, No. 41, 1893 (Leipzig: Verlag von of flower-specificgenes. In Cell Culture and
Davles, D., and Schwarz-Sommer, 2. (1994). Wilhelm Engelmann). Somatic Cell Genetics, J. Schell, and I.K.
Control of floral organ identity by homeotic Koltunow, A.M., ’Ruettner, J., Cox, K.H., Vasil, eds (San Diego:Academic Press), pp.
MADS-box transcription factors. In Plant Wallroth, M., and Goldberg, R.B. (1990). 197-214.
Promoters and Transcription Factors, L. Different temporaland spatialgene expres- Sommer, H., Beltran, J.-P., Huijser, P., Pape,
Nover, ed (Berlin: Springer-Verlag), pp. sion patterns occur during anther H., Wnnig, W.-E., Saedler, H., and
235-258. development. Plant Cell 2 , 1201-1224. Schwarz-Sommer, 2. (1990). Deficiens, a
Drews, G.N., and Goldberg, R.B. (1989). Leroy, J.-F. (1993). Origine et Bvolution des homeoticgeneinvolvedinthe controlof flower
Geneticcontrolof flowerdevelopment. Trends Plantes A fleurs (Paris: Masson). morphogenesis in Antirrhinum majus: The
Genet. 5, 256-261. Llnnb, C. (1735). SystemaNaturae (Leiden:The- proteinshows homologyto transcriptionfac-
Drews, G.N., Beals, T.B., Bui, A.Q., and odorus Haak). tors. EMBO J. 9, 605-613.
Goldberg, R.B. (1992). Regional and cell- Li)nnig, W.-E., and Huijser, P. (1994). Gene Sprengel, C. K. (1793). DasentdeckteGeheim-
specific geneexpressionpatternsduringpet- tagging by endogenoustransposons. In Plant nisder Naturim Bau undin der Befruchtung
al development. Plant Cell 4, 1383-1404. Molecular Biology Manual, S.B. Gelvin, der Blumen. In Ostwald‘s Klassiker der ex-
Goethe, J.W. (1790). Versuch die Metamor- R.A. Schilperoort, and D.P.S. Verma, eds akten Naturwissenschaften,Nos.48-51,1894
phoseder Pflanzenzu erkltiren. Gotha: Carl (Dordrecht: KluwerAcademic Publishers),in (Leipzig: Verlag von Wilhelm Engelmann).
Wilhelm Ettinger, 1984facsimile (Weinheim: press. Triibner, W., Ramirez, L., Motte, P., Hue, I.,
Acta humaniorader Verlag ChemieGmbH.) Masters, M.T. (1869). Vegetable Teratology (Lon- Huijser, P., Lhnig, W.-E., Saedler, H.,
Goethe, J.W. (1820). Verstiiubung, Verdun- don: Ray Society). Sommer, H., and Schwarz-Sommer, 2.
stung, Vertropfung. In Goethe, Samtliche Meeuse, A.D.J. (1987). All about Angiosperms (1992). GLOBOS: A homeotic gene which
Werke, Vol. 12,1989 (München:Carl-Hanser- (Delft: Eburon). interactswith D€F/C/€/VSin the controlof An-
Verlag), pp. 212-224. Mizukami, Y., and Ma, H. (1992). Ectopic ex- tirrhinumfloral organogenesis. EMBOJ. 11,
Henschel, A.W. (1820). Von der Sexualitatder pression of the floral homeotic gene 4693- 4704.
Pflanzen (Breslau: Wilhelm Gottlieb Korn). AGAMOUSintransgenicArabidopsisplants Weberllng, F. (1989). Morphology of Flowers
alters floral organ identity. Cell T1, 119-131. and lnflorescences(Cambridge:Cambridge
Jack, T., Brockmann, L.L., and Meyerowltz,
E.M. (1992). The homeotic geneAPETALA3 Payer, J.-B. (1857). TraitB d’Organog6nie Com- University Press).
of Arabidopsisthalianaencodesa MADSbox par& de Ia Fleur, reprint, 1966 (Lehre: J. Willing, R.P., and Mascarenhas, J.P. (1984).
andis expressedin petalsandstamens. Cell Cramer). Analysis of the complexity and diversity of
68, 683-697. Sattler, R. (1988). Homeosis in plants. Am. J. mRNAs from pollen and shoots of Trades-
BOt. 74, 1606-1617. cantia. Plant Physiol. 75, 865-868.
Jack, T., Fox, G.L., and Meyerowltz, E.M.
(1994). Arabidopsis homeotic gene Schelver, F. J. (1812). Kritikder Lehrevonden Yanofsky, M.F., Ma, H., Bowman, J.L., Drews,
APETALA3 ectopic expression: Transcrip- Geschlechtemder Pflanze(Heidelberg: Gott- G.N., Feldmann, K.A., and Meyerowltz,
tional and posttranscriptional regulation lieb Braun). E.M. (1990). The protein encoded by the
determine floral organ identity. Cell 76, Schwarz-Sommer, Z., Hui, I.,Huijser, P., Flor, Arabidopsis homeoticgene agamous resem-
703-716. P.4, Hansen, R., Tetens, E, Liinnig, blestranscriptionfactors. Nature346,35-39.
 Goethe, Sex, and Flower Genes
W. E. Lonnig
PLANT CELL 1994;6;574-576
DOI: 10.1105/tpc.6.5.574

This information is current as of October 15, 2010

Permissions https://www.copyright.com/ccc/openurl.do?sid=pd_hw1532298X&issn=1532298X&WT.mc_id=pd_hw1532298
X

eTOCs Sign up for eTOCs for THE PLANT CELL at:

http://www.plantcell.org/subscriptions/etoc.shtml
CiteTrack Alerts Sign up for CiteTrack Alerts for Plant Cell at:
http://www.plantcell.org/cgi/alerts/ctmain
Subscription Information Subscription information for The Plant Cell and Plant Physiology is available at:
http://www.aspb.org/publications/subscriptions.cfm

© American Society of Plant Biologists

ADVANCING THE SCIENCE OF PLANT BIOLOGY