RESEARCH NEWS
Supramolecular chirons
Ashwini Nangia
The most practical method for obtaining enantiomerically pure compounds
on a laboratory and industrial scale is
the optical resolution of racemates
through diastereomeric salt 1. However,
the procedure is tedious and capricious. The number of steps involved is
large, the maximum theoretical yield is
only 50%, and the undesired enantiomer has to be either recycled or dis carded.
The
exact
experimental
conditions have to be arrived at by
trial-and-error because the chemical
basis for the fractional crystallization
of a particular diastereomer is not
properly understood. The other
method for obtaining enantiomers is
mechanical separation of chiral crys tals. Discovered by Pasteur 2 a centuryand-a-half ago, success in this approach has been even slower. This
article discusses developments on the
former approach: how to obtain the
enantiopure molecule from a racemic
mixture of the compound? While there
have been studies to design new chiral
host compounds 3 to correlate the efficiency of resolution with crystal structures of less and more soluble salts 4
and to examine the role of the solvent 5,
there are as yet no general and globally applicable criteria for the selection
of a particular resolving agent, solvent,
and crystallization conditions for a
given substrate. The exact boundary
conditions for the fractional crystallization of a particular diastereomer are
guided by the experience of chemists.
Even so, the empirical rules have limited applicability and are valid only
within a family of structures. Many of
the thumb rules for efficient resolution
are part of the classified information
with pharmaceutical companies and
may never be disseminated in the public domain.
The recent paper of Hanessian et
al. 6 disscusses the rationale for enantiodifferentiation during self-assembly
through a detailed understanding of
molecular recognition between chiral
diamines and chiral diols. From an
374
analysis of the dozen or so crystal
structures of amine–alcohol complexes
(supraminols, Figure 1), the authors are
able to predict structures of the bestmatched pairs of diamines and diols
through the notion of supramolecular
chirons. Thus, supramolecular chiron
is ‘the minimal homo- or heterochiral
molecular unit or ensemble capable of
generating ordered superstructures by
self-assembly
through
hydrogen
bonding or other noncovalent forces,
and leading to topologically distinct
enantio- or diastereopure architectures’.
In what way is the Hanessian approach superior to related papers published recently? Saigo et al. 4 have
examined the resolution of 2-aryl-
alkanoic acids with (1R,2S)-2-amino1,2-diphenylethanol. Crystal structures
of the less soluble salts have a chiral
columnar hydrogen bond arrangement
of O–HLO and N–HLO bonds formed by
ammonium hydrogens and carboxylate
oxygens, reinforced by electrostatic
forces between the charged ions. The
structure includes a water molecule
which is bonded to the columnar structure through short OLO contacts.
There are no strong interactions between the polar, hydrogen bond columns except weak hydrophobic, van
der
Waals
inter-
Figure 1. Some vicinal diamines and vicinal diols comprising carbocyclic and alicyclic structures and their adducts.
Figure 2. Two CURRENT
types of SCIENCE,
supramolecular
the
VOL. 78,chirons
NO. 4, 25inFEBRUARY
2000
crystal structures of supraminol adducts. a, 1(R,R)·2(R,R); b,
1(R,R)·3(R,R). Motif a is pleated-sheet staircase and b is righthanded ribbon structure (Reproduced with permission of the
author, ref. 6).
RESEARCH NEWS
actions. The packing motif in the more
soluble complex is similar with comparable OLO and NLO distances except
that the water molecule is absent in the
polar domain. While the authors as cribe the additional hydrogen bonding
by the inclusion of water in the crystal
as a reason for the lower solubility,
this does not significantly improve our
understanding of selective crystallization. The phenomenon of hydration in
crystals is not properly understood in
terms of when water is included, for
what reasons, and its exact role on the
solubility of the crystal. Moreover,
comparison of the stability of a crystal
and its hydrate is non-trival because
the constituents are different. In another study, optical resolution of
trans-chrysanthemic
acid with (1R,2R)-1-(4-nitrophenyl)-2dimethylamino-propane-1,3-diol 5 was
found to be dramatically enhanced
when either pure methanol is used for
crystallization (93% optical purity) or
when methanol is added to ether-type
solvents (Et 2O, THF, MTBE, i-Pr2O;
91–99%). The authors postulate that
hydrogen bonding of the acid–diol
complex with methanol promotes nucleation and crystallization of the less
soluble diastereomer. It is not clear
from the paper why methanol works
and another alcoholic solvent, say
ethanol or iso-propanol gives no crys tals. This is a general problem in resolution chemistry – the results of one
system cannot be extended to a related
substrate.
In three papers during the last five
years, Hanessian et al. 6–8 have analysed the crystal structures of 1:1
complexes of chiral, C2-symmetric,
vicinal diamines and diols. In the crys talline
adduct of (1S,2S)-1,2-diaminocyclohexane and (1S,2S)-1,2-cyclohexanediol (1SS⋅2SS), the cyclohexane rings
align into four vertical columns and the
polar hydrogen bonding groups face
inward. The structure is a pleated
sheet-like array of eight-membered,
square planar hydrogen bonded units
in which the oxygen and nitrogen atoms are tetracoordinated. The remaining hydroxy and amino functional
groups are engaged in two symmetrical
side rows of tricoordinated zigzag hydrogen bond patterns which flank op-
posite sides of the central octagonal
staircase
core.
The crystal structure of the 1:1 adduct
of (1R,2R)-diamine and (1S,2S)-diol
(1RR⋅2SS) is gratifyingly predictable
and virtually identical to the structure
of the homochiral complex. Interestingly, chirality of the diol and diamine
components controls the tertiary structure of the complex, that is the (S,S)diamine and (S,S)-diol have a lefthanded helicate while the (R,R)diamine with the same diol produces a
right-handed helicate. The crystal
structures of other C2-symmetric vicinal diamines and diols in Figure 1 have
pleated-sheet staircase-like or righthanded
ribbon-helicate
hydrogen
bond architectures. A common theme
in this family of enantiomorphous
structures is the connection between
molecular functional groups and their
crystal packing motifs, thus opening
avenues for future design and optimization. A high degree of predictability
is ascribed to the supramolecular chirons displayed in Figure 2. These recurring hydrogen bond helicate motifs
in supraminol adducts could provide
the key to our understanding of enantio- and diastereodifferentiation during
crystallization.
What is the origin of supraminol
self-assembly? The ready formation of
amine–alcohol adducts may be understood from the mutual recognition
of complementary hydrogen bonding
donor–acceptor groups in the components: NH 2 has 2 donors and 1 acceptor while OH has 1 donor and 2
acceptor groups. Thus, in a 1:1 supraminol the tetracoordinated network is
saturated at both the heteroatoms
through one O–HLN and two N–HLO
hydrogen bonds 9. The authors propose that enantiodifferentiating recognition between a diamine and a
matching diol partner produces a thermodynamically stable architecture in a
single step, or alternatively, matching
diol (or diamine) molecules insert in the
lattice of diamine (or diol) to produce
the coordinatively saturated network.
This was verified through competition
experiments.
When (1R,2R)-cyclohexane diamine
is heated with racemic trans-1,2cyclohexane diol, the homochiral
(R,R)⋅(R,R)-complex (1RR⋅2RR) crys -
CURRENT SCIENCE, VOL. 78, NO. 4, 25 FEBRUARY 2000
tallizes out while the heterochiral
(R,R)⋅(S,S)-pair (1RR⋅2SS) remains in
the mother liquor. In practice, the
(R,R)-diol could be obtained from the
racemic trans-diol in 98% enantiomeric
excess. This and related experiments
show that the homochiral adducts
consist of the preferred (and therefore
stable) chiron combination given the
diol partner selected by a particular
diamine during crystallization. Further,
in the structure of achiral cis-1,2cyclohexane diol and enantiopure R,Rdiamine a markedly different hydrogen
bonding network (or supramolecular
chiron) is formed with two diamine and
two diol moieties in the asymmetric
unit. Notwithstanding the detailed
analysis of many closely related crys tal structures, it is not possible to say
definitively whether the observed
packing features are a result of kinetic
factors or thermodynamic stability. A major limitation in making such
predictions based on crystallization
properties is that though melting point
and crystal density generally correlate
with solubility, connection between
crystal energy and solubility can be
tenuous.
Hanessian et al.’s study is notable
for several reasons: (1) It is the first
systematic analysis of closely related
enantiomorphous crystal structures,
their crystallization behaviour, and
their hydrogen bonding patterns; (2)
The stereochemical information encoded in the molecule is reflected in
the recognition motifs of complementary functional groups; (3) Crystal
structures of the matched homochiral
supraminol adducts are analysed via
the newly introduced concept of s upramolecular chiron. In summary, the
authors provide a rational and improved interpretation of a classical
phenomenon using contemporary
ideas. Applications of these ideas for
enantiomer enrichment, resolution and
design of chiral auxiliaries should be
forthcoming in the near future.
The genesis of supramolecular chiron lies in its covalent sibling, chiron, a
term introduced by Hanessian more
than a decade ago to logically synthesize enantiomerically pure molecules
from the chiral pool 10. With a shift in
paradigm from the molecule to the s upermolecule in the nineties, the con375
RESEARCH NEWS
ceptual relationship between crystal
engineering and organic synthesis has
been proposed through another new
term ‘supramolecular synthon’ 11. Synthons are structure-directing motifs
involving non-covalent bonds and
contain the logic code for selfassembly
in
the
solid state. In effect, supramolecular
chirons are the chiral counterparts
of supramolecular synthons. Thus,
(supramolecular) synthons and chirons play the same focusing role in the
synthesis of target crystal structures
(supermolecules) that (molecular) synthons and chirons have in the synthesis of complex natural products
(molecules). All these exciting developments project that the challenge for
organic chemists in the new millennium
will lie in understanding and controlling another type of bond, the hydrogen bond.
Jacques, J., Collet, A. and Wilen, S. H., Enantiomers, Racemates and Resolutions,
Krieger Publishing, Malabar, Fl, 1994.
1. Pasteur, L., Ann. Chim. Phys., 1848,
24, 442.
2. Toda, F. and Shinyama, T., J. Chem.
Soc., Perkin Trans 1, 1997, 1759–1761.
3. Saigo, K., Kinbara, K. and Kobayashi,
Y., J. Chem. Soc., Perkin Trans 2,
1998, 1767–1775.
4. Kozsda-Kovács, É., Keserü, G. M., Böcskei, Z., Szilágyi, I., Simon, K., Bertók,
B. and Fogassy, E., J. Chem. Soc., Perkin Trans 2, 2000, 149–153.
5. Hanessian, S., Saladino, R., Margarita, R.
and Simard, M., Chem. Eur. J., 1999, 5,
2169–2183.
6. Hanessian, S., Simard, M. and Roelens,
S., J. Am. Chem. Soc., 1995, 117,
7630–7645.
7. Hanessian, S., Gomtsyan, A., Simard, M.
and Roelens, S., J. Am. Chem. Soc.,
1994, 116, 4495–4496.
8. Ermer, O. and Eling, A., J. Chem. Soc.,
Perkin Trans 2, 1994, 925–943.
9. Hanessian, S., Total Synthesis of
Natural Products: The Chiron Approach, Pergamon Press, New York,
1983.
10. Desiraju, G. R., Angew. Chem. Int. Ed.
Engl., 1995, 34, 2311–2327.
Ashwini Nangia is in the School of
Chemistry, University of Hyderabad,
Hyderabad 500 046, India.
Do our maternal and paternal genes pull us in different
directions?
Raghavendra Gadagkar
In all diploid organisms such as ourselves, each individual inherits one set
of chromosomes from the mother and
another set from the father. It is generally assumed that once these chromosomes reach our bodies, they lose any
‘memory’ of where they came from.
However there is evidence that chromosomes (and the genes they contain)
sometimes get differentially imprinted
as they pass through a male or female
body and this imprint may be retained
when the chromosomes are passed on
to the next generation 1–6. There is also
evidence that DNA methylation is a
mechanism by which chromosomes
may acquire such male-specific or female-specific imprints. Differential patterns of DNA methylation are known
tolead to different levels of gene expression 7–10. What all this means then
is that our paternally derived genes
and maternally derived genes may behave differently in our bodies even
though they may be otherwise identical. To the extent that genes influence
our behaviour it may well be that our
father’s genes and mother’s genes pull
us in different directions.
In 1992, David Haig11, then at the
University of Oxford, pointed out that
such a possibility has serious conse376
quences for the standard predictions
of sociobiological theory which is
based on the assumption that paternal
and maternal genes do not behave
differently. Let us consider two examples. In insects that belong to the order Hymenoptera (ants, bees, wasps)
females can lay both unfertilized, haploid eggs as well as fertilized, diploid
eggs. The fertilized diploid eggs develop into diploid adult females
whereas the unfertilized haploid eggs
develop into haploid adult males. Since
males are haploid, they produce sperm
that are clones of each other. The females, being diploid, produce haploid
eggs that receive a randomly chosen
50% of the maternal genome. In such
haplodiploid insects, two sisters would
be related to each other by a coefficient of genetic relatedness r of 0.75
but a female would be related to her
offspring by the usual 0.5 (as in diploid
species) (Figure 1). In 1964 W. D.
Hamilton12,13 pointed out that such
asymmetries in genetic relatedness
should select for altruistic behaviour
on the part of females to care for their
sisters rather than to produce their
own offspring. This is indeed what
workers (who are females) in many
social insect colonies do. In 1976 Triv-
ers and Hare 14 pointed out that although workers are more closely
related to their sisters (r = 0.75) they
are much less related to their brothers
(r = 0.25), as compared to their offspring (r = 0.5). They predicted therefore
Figure 1. Genetic relatedness
haplodiploidy (see text for details).
under
CURRENT SCIENCE, VOL. 78, NO. 4, 25 FEBRUARY 2000
RESEARCH NEWS
ing
Box 1. Implications of genomic
imprinting I (after Haig11)
Intra-genomic conflict between maternal and paternal genes over selfishness versus
altruism.
Hamilton’s haplodiploidy hypothesis showing preference for sisters over daughters
was based on average relatedness.
Hymenopteran colony with singly mated queen
Average
Maternal genes
Paternal genes
•
•
•
: Sisters > Daughters (1.5:1)
: Sisters = Daughters
: Sisters > Daughters (2:1)
Mild selection for rearing sisters instead of daughters (=
social evolution)
No social evolution if maternal genes are in control
Rapid social evolution if paternal genes are in control
Box 2. Implications of genomic imprinting II (after Haig11)
Intra-genomic conflict between maternal and paternal genes over sex investment ratio.
Trivers and Hare’s prediction that workers prefer a 3:1 investment in their sisters and
brothers was based on average relatedness.
Hymenopteran colony with singly mated queen
Average
Maternal genes
Paternal genes
•
•
•
: Sister : Brother
: Sister : Brother
: Sister : Brother
:: 3:1
:: 1:1
:: 1:0
Some queen–worker conflict if mean relatedness matters (as
queen prefers 1:1)
No queen–worker conflict if maternal genes are in control
Intense queen–worker conflict if paternal genes are in
control
that either workers should prefer their
own sons over their brothers or, if they
are forced to rear their sisters and
brothers, they should prefer to invest
in their sisters and brothers in the ratio
3:1 (0.75:0.25). A particularly fascinatCURRENT SCIENCE, VOL. 78, NO. 4, 25 FEBRUARY 2000
aspect of this prediction is that the
workers’ preferred ratio of investment
(3:1) is in conflict with the queen’s
preferred ratio of investment of 1:1 in
her daughters and sons. Hamilton’s
prediction and the prediction of Triversand Hare have since become the
cornerstones of sociobiology and both
have engendered an enormous body
of theoretical and empirical work.
Both these sets of predictions were
however based on the assumption that
maternal and paternal genes in the
bodies of the worker behave identically. Thus Hamilton as well as Trivers
and Hare computed relatedness values
(0.75, 0.5 and 0.25, discussed above)
by taking the average values for maternal and paternal genes. But if the
relatedness values are computed separately for the maternal and paternal
genes, they turn out to be quite different and so do the predictions. For example, from the point of view of the
maternal genes in a workers’ body,
sisters are as valuable as daughters, so
that altruistic rearing of sisters should
be favoured no more than selfish rearing of daughters (in both case, r = 0.5).
From the point of view of the paternal
genes on the other hand, sisters are
twice as valuable as daughters so that
altruistic rearing of sisters should be
even more strongly favoured than
selfish rearing of daughters (Box 1). A
similar situation occurs with the predicted sex investment ratios. From the
point of view of the maternal genes in
a workers’ body, sisters are as valuable as brothers so that a 1:1 sex investment ratio is favoured and thus
there should be no conflict between
queens and workers over sex investment. From the point of view of the
workers’ paternal genes however, all
the paternal genes are expected to be
found in sisters while none are expected to be found in her brothers.
Hence paternal genes should favour all
investment in sisters and none in
brothers. Therefore queen–worker conflict should now be even more severe
than what was predicted by a computation of average relatedness for maternal and paternal genes (Box 2).As
Haig 15 readily admits, whether a major
reappraisal of sociobiological theory is
required will depend on how common
genomic imprinting turns out to be in
377
RESEARCH NEWS
Figure 2. Life cycle of normal and PSR
strains of Nasonia vitripennis (after Werren et al.17) (see text for details). (after
Nur et al.18) (see text for details).social
insects. It is in this context that recent
evidence for the role of genomic imprinting in sex determination in a parasitic wasp assumes significance. The
wasp in question is Nasonia vitripennis and the study under consideration
is by S. L. Dobson and M. A. Tanouye 16 of the University of Caliat
Berkeley. Nasonia vitripennis is a
parasitoid wasp that is distributed
throughout the world. Female wasps
lay eggs in the pupae of flies that
breed in carcasses and in bird nests.
Like all Hymenoptera, N. vitripennis is
also haplodiploid and it is used as a
favourite laboratory model system in a
variety of genetic and evolutionary
378
studies. As it often happens with laboratory model systems, many unusual
mutants that cannot usually survive in
nature turn up in the laboratory cultures. Many strains of N. vitripennis
are now known that distort the sex
ratio
of
their
offspring –
variously called son killers and daughter killers! A rather famous one is
called PSR, for paternally transmitted
sex ratio factor 17. Unlike the wild type
strains, eggs fertilized by PSR males
also develop into haploid males but
these males do inherit the PSR factor
(Figure 2). It turns out that PSR is a
small, aberrant, unpaired chromosome
(such chromosomes are called B chromosomes) that enters the egg along
with the paternal chromosomes. Having done so it brings about the heterochromatization and hence the loss of
all paternal chromosomes 18. This
leaves the zygote only with the maternal chromosomes and the PSR itself
(Figure 3). Not surprisingly, such
zyotes develop into haploid, PSR containing males. PSR has thus been
dubbed the most selfish genetic element known 18 (see also Gadagkar 19). It
uses the male to reach the zygote and
having done so it destroys its cotravellers (the paternal chromosomes)
in order to ensure its future survival.
PSR cannot survive unless it converts
the diploid zygote that would normally
develop into a female, into a male because PSR cannot transmit through a
female body. PSR can only survive in a
male body and since male hymenopterans normally have no sons it has no
choice but to convert potential daughters into sons.
The aim of the Dobson and Tanouye 16 study was to understand the
mechanism of sex determination in the
Hymenoptera. Even though we know
that unfertilized eggs develop into
males and fertilized eggs develop into
females, the mechanism by which sex
is determined is far from clear. The
observation that unfertilized eggs develop into males and fertilized eggs
develop
into
female
is consistent with a variety of mechanisms – indeed there have been a variety of models proposed for sex
determination in the Hymenoptera.
1. Fertilization sex determination
(FSD): According to this model, the
very act of fertilization causes the
egg to develop into a female, quite
independent of the paternal genes
that fertilization may bring with it 20.
2. Single locus complementary sex
determination (SCSD): A single
sex determining locus is postulated
and individual homozygous or
hemizygous (as all haploid individuals are) are expected to develop
into males while those heterozygous are expected to develop into
females. Because the sex determining locus is believed to be highly
polymorphic, diploid homozygotes
are expected to be rare and the
usual way to get males is therefore
by the development of unfertilized
(hemizygous) eggs. As predicted by
the model, diploid, homozygous
males can be produced by inbreeding 21.
3. Multiple loci complementary sex
determination (MCSD): Because
the prediction of SCSD do not always fit the empirical data, multiple
sex determination loci have been
postulated for some species. The
prediction is that individuals homozygous or hemizygous at all of
these loci will develop into males
while those heterozygous for any
one of these loci will develop into
females 22,23.
4. Genic balance sex determination
(GBSD): According to this model,
sex is determined by a balance between male determining genes (M)
and female determining genes (F).
Because M is postulated to be more
powerful than F, haploid eggs with
CURRENT SCIENCE, VOL. 78, NO. 4, 25 FEBRUARY 2000
RESEARCH NEWS
one set of M and F each develop
into males (M > F). However fertilized eggs will have 2M and 2F. The
M genes are not expected to be additive in their effects while the F
genes
are expected to be additive. Thus
2F > M > F, so that fertilized eggs
develop into females 24.
5. Maternal effect sex determination
(MESD): This model proposes that
sex is determined by the ratio of nuclear and cytoplasmic factors. Haploid eggs, having one set of nuclear
and cytoplasmic factors each, develop into males. Fertilized eggs,
with one set of cytoplasmic factor
and two sets of nuclear factors (one
set received from the father), develop into females 25,26.
6. Genomic imprinting sex determination (GISD): This model proposes
that genes in the mother are so imprinted that they can only direct
male development. However genes
in the father are so imprinted that
they
can
Table 1. Sex determination in Nasonia vitripennis (after Haig15, and Dobson and Tanouye16)
Haploid eggs
Unfertilized
Fertilized by wild type male
Fertilized by PSR male
Diploid eggs
Maternal
chromosomes
Paternal
chromosomes
Sex
One set
One set
One set
Nil
One set
Only PSR factor
Male
Female
Male
direct female development despite the
presence of the maternal chromosomes. Only fertilized eggs contain
genes with the paternal imprint and
thus they develop into females 27,28.
It is fair to say that there is no satisfactory empirical support for any of these
models. Genomic imprinting sex determination is the most recent proposal
and the study of Dobson and Tanouye 16 makes it possible to exclude all
the previous five models and support
only GIST, for N. vitripennis. Several
other features of N. vitripennis permitted Dobson and Tanouye to design
experiments that are not usually possible with other organisms. I have already described the PSR factor. By
appropriate manipulations, one can
also produce triploid females (with
three sets of chromosomes) and diploid males. The triploid females produce haploid as well as diploid eggs
and the diploid males produce diploid
sperm! The basic experimental design
of Dobson and Tanouye involved fertilizing haploid and diploid eggs with
haploid and diploid sperm, with and
without the PSR factor. With the help
of three recessive eye colour markers
they were able to assess whether the
proportions of progeny of each sex
and eye colour were as expected by
different models for sex determination.
Their most important result can be
stated simply. When diploid eggs were
fertilized by PSR containing sperm, the
paternal chromosomes were lost as
expected, leaving an embryo with two
Maternal
chromosomes
Two sets
Two sets
Two sets
sets of maternal chromosomes and the
PSR factor. All previous models of sex
determination predict that these fertilized, diploid embryos should develop
into females. They are fertilized (satis fying FSD), they are diploid and heterozygous (satisfying SCSD and
MCSD) and diploid (satisfying GBSD
and MESD). Only the genomic model
predicts that these embryos should
develop into males. And Dobson and
Tanouye found that these embryos
indeed developed into males (Table 1).
As they admit, the possibility that PSR
itself has male determining genes cannot be completely ruled out. However
previous deletion analysis has failed to
separate the ability of PSR to convert
fertilized embryos into males and its
property of eliminating paternal chromosomes 29. Thus it appears that elimination of paternal chromosomes is the
mechanism by which PSR converts
fertilized eggs into males. Although
these eggs are fertilized and have a
heterozygous, diploid chromosome
composition, they only have maternally imprinted chromosomes. Lack of
paternally imprinted chromosomes can
thus be thought of the reason why
they do not develop into females.
Needless to say, this evidence in favour of imprinting as a mechanism of
sex determination in a Hymenopteran
insect (or indeed in favour of any
mechanism of sex determination!) is of
great interest to all of us who have
long been uncomfortable with our inability to come to grips with, as fun-
CURRENT SCIENCE, VOL. 78, NO. 4, 25 FEBRUARY 2000
Paternal
chromosomes
Nil
One set
Only PSR factor
Sex
Male
Female
Male
damental
a
problem
as
sex
determination, in the organisms we
study. Even more importantly, this
evidence in favour of genomic imprinting comes as a reminder that a
reappraisal of sociobiological theory
may be required sooner or later. The
best evidence for the role of genomic
imprinting followed by differential expression of maternal and paternal
genes, comes from mammalian sys tems 30,31. And it is in mammalian sys tems that the role of genomic
imprinting is also being vigorously
investigated in another area of sociobiological theory namely, inter-sexual
conflict 32,33. When females mate with a
different male each time they produce
an offspring, male–female conflict continues in the bodies of their offspring.
While the mother would like to distribute her resources nearly equally between her present and future offspring,
the father would like the present offspring (which is his) to get as much of
the maternal resources as possible,
unmindful of the health of future offspring (who are not likely to be his). It
has therefore been postulated that
genes which may be involved in
modulating the resource drawing abilities of offspring become differentially
imprinted, to express the conflicting
interests of the mother and the father.
The most famous example is the case,
or as Haig and Graham 34 call it, ‘the
strange case of the insulin-like growth
factor II’. Insulin-like growth factor II
(IGF II) is a polypeptide that helps
379
RESEARCH NEWS
rapid embryonic growth in mice. As
expected from the theory of genomic
imprinting mentioned above, the paternal copy of IGF II is well transcribed
while the maternal copy is almost silent 35. This is consistent with the idea
that the father’s genes are attempting
to enhance the resource drawing ability of the offspring while the mother’s
genes are not particularly encouraging
this. The ‘strange’ case concerns the
type 2 receptor for IGF II. While the
type 1 receptor appears to behave
normally, the type 2 receptor is unusual. First, it is transcribed mainly
from the maternal genome and not from
the paternal genome 36. Secondly, the
type 2 receptor is in other contexts a
cation-independent
mannose-6phosphate receptor which binds mannose-6-phosphate residues on lysosomal enzymes and transports them
into lysosomes. Haig and Graham have
theorized that the receptor which mediates the normal function of IGF II is
the type 1 receptor and that the type 2
receptor has been hijacked by the
mother to act as a sink for excess IGF II
and thus limit embryonic growth 34.
That the type 2 receptor gene is subject to imprinting of the opposite kind
as compared to IGF II is consistent
with this idea. Thus the father’s genes
appear to make plenty of growth factor
and promote embryonic growth while
the mother’s genes find a way of eating up this growth factor and limit embryonic growth.
There has been some effort at more
formal population genetic modeling of
these phenomena which, by and large,
lend support to the ideas of Haig and
his colleagues 37–39. Nevertheless, it
must be recognized that a great deal of
all of this speculation remains to be
tested, either by modeling or by experiments. But it is today’s speculation
that will guide tomorrow’s research.
6.
7.
8.
9.
Consequences, Cambridge University
Press, Cambridge, 1995.
Reik, W. and Surani, A. (eds), Genomic
Imprinting: Frontiers in Molecular Biology, Oxford University Press, Oxford,
1997.
Gold, J. D. and Pedersen, R. A., Curr.
Top. Dev. Biol., 1994, 29, 227–280.
Jaenisch, R., Trends Genet., 1997, 13,
323–329.
Kass, S. U. and Wolffe, A. P., in Epigenetics (Novartis Foundation Symposium
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
214), (eds Chadwick, D. J. and Cardew,
G.), John Wiley & Sons, New York,
1998, pp. 22–45.
Reik, W. and Walter, J., Curr. Opin.
Genet. Dev., 1998, 8, 154–164.
Haig, D., J. Theor. Biol., 1992, 156,
401–403.
Hamilton, W. D., J. Theor. Biol., 1964,
7, 1–16.
Hamilton, W. D., J. Theor. Biol., 1964,
7, 17–52.
Trivers, R. L. and Hare, H., Science,
1976, 191, 249–263.
Haig, D., Trends Ecol. Evol., 1998, 13,
380–381.
Dobson, S. L. and Tanouye, M. A.,
Genetics, 1998, 149, 233–242.
Werren, J. H., Nur, U. and Eickbush, D.,
Nature, 1987, 327, 75–76.
Nur, U., Werren, J. H., Eickbush, D. G.,
Burke, W. D. and Eickbush, T. H., Science, 1988, 240, 512–514.
Gadagkar, R., Survival Strategies – Cooperation and Conflict in Animal Societies, Harvard University Press,
Cambridge, Massachusetts, and Universities Press, Hyderabad, 1997.
Whiting, P. W., Genetics, 1960, 45,
949–970.
Whiting, P. W., Genetics, 1943, 28,
365–382.
Crozier, R. H., Am. Nat., 1971, 105,
399–412.
Snell, G. D., Proc. Natl. Acad. Sci. USA,
1935, 21, 446–453.
Da Cunha, A. B. and Kerr, W. E.,
Forma Functio, 1957, 1, 33–36.
Crozier, R. H., Annu. Rev. Entomol.,
1977, 22, 263–288.
Cook, J. M., Heredity, 1993, 71, 421–
435.
1. Peterson, K. and Sapienza, C., Annu.
Rev. Genet., 1993, 27, 7–31.
2. Chandra, H. S. and Brown, S. W.,
Nature, 1975, 253, 165–168.
3. Chandra, H. S. and Nanjundiah, V., Development, 1990, supplement, 47–53.
4. Ohlsson, R. (ed.), Genomic Imprinting:
An
Interdisciplinary
Approach,
Springer-Verlag, Berlin, 1999.
5. Ohlsson, R., Hall, K. and Ritzen, M.
(eds), Genomic Imprinting: Causes and
380
CURRENT SCIENCE, VOL. 78, NO. 4, 25 FEBRUARY 2000
RESEARCH NEWS
27. Poirié, M., Périquet, G. and Beukeboom,
L., Dev. Biol., 1992, 3, 357–361.
28. Beukeboom, L. W., BioEssays, 1995,
17, 813–817.
29. Beukeboom, L. W. and Werren, J. H.,
Genetics, 1993, 133, 637–648.
30. Barlow, D. P., Science, 1995, 270,
1610–1613.
31. Bartolomei, M. S. and Tilghman, S. M.,
Annu. Rev. Genet., 1997, 31, 493–525.
32. Haig, D. and Westoby, M., Am. Nat.,
1989, 134, 147–155.
33. Haig, D., Proc. R. Soc. London B.,
1997, 264, 1657–1662.
34. Haig, D. and Graham, C., Cell, 1991, 64,
1045–1046.
35. DeChiara, T. M., Robertson, E. J. and
Efstratiadis, A., Cell, 1991, 64, 849–
859.
36. Barlow, D. P., Stöger, R., Herrmann,
B. G., Saito, K. and Schweifer, N.,
Nature, 1991, 349, 84–87.
37. Mochizuki, A., Takeda, Y. and Iwasa,
Y., Genetics, 1996, 144, 1283–1295.
38. Spencer, H. G., Feldman, M. W. and
Clark, A. G., Genetics, 1998, 148, 893–
904.
39. Spencer, H. G., Clark, A. G. and Feldman, M. W., Trends Ecol. Evol., 1999,
14, 197–201.
CURRENT SCIENCE, VOL. 78, NO. 4, 25 FEBRUARY 2000
Raghavendra Gadagkar is in the Centre for Ecological Sciences, Indian
Institute of Science, Bangalore 560
012, India; and Evolutionary and
Organismal Biology Unit, Jawaharlal
Nehru Centre for Advanced Scientific
Research, Jakkur P.O., Bangalore
560
064,
India.
(e-mail:
ragh@ces.iisc.ernet.in;
internet:http://ces
.iisc.ernet.in/hpg/ragh)
381