Printer Friendly

Paternal effects in inheritance of a pathogen resistance trait in Ipomoea purpurea.

A major goal of quantitative genetics is to estimate genetic parameters that permit prediction of evolutionary response to selection. For continuously variable characters the response to selection depends upon the proportion of phenotypic variance that is caused by additive genetic variance, or narrow-sense heritability, [h.sup.2]. Heritability may have important ecological implications as well (Simms 1990). Recent quantitative genetic models demonstrate that the extent of heritable variance for a trait can influence the stability of ecological interactions in which it is important (Pease 1984; Soloniemi 1993). Thus, the heritability of a trait in a particular situation frequently is a fundamental parameter sought by evolutionary ecologists.

Several quantitative genetic methods are available with which to estimate heritability. The goal of these methods is to partition the phenotypic variance for a trait in a way that isolates additive variance from other causes. Unfortunately, additive variance is rarely completely isolated from other causal components. In particular, the parental contribution to offspring phenotype encompasses more than simply Mendelian genetic effects. The component of variance due to a particular type of parent (mother or father) therefore includes non-Mendelian as well as additive causal components (Cockerham and Weir 1977).

In addition to the nuclear genetic contribution that a parent makes to its offspring, a parent can influence offspring phenotype in two other ways (Hohenboken 1985). First, phenotype may be influenced by nonnuclear DNA transmitted through the cytoplasm, including plasmids and organelles such as chloroplasts and mitochondria. Second, parental care, biochemical or structural components of the parental cytoplasm, parent-specific gene expression, and aspects of the parental phenotype that were determined by its environment can also influence offspring phenotype. The maternal parent is thought to be the greatest source of these nonnuclear influences on phenotype, and a major goal of most crossing designs used by quantitative geneticists to estimate narrow-sense heritability is to prevent maternal effects from biasing estimates of additive variance. However, paternal effects are not unknown (e.g., Geisel 1988; Futuyma et al. 1993; Schmid and Dolt 1994; Fox et al. 1995; Watson and Hoffmann 1995; see also references in Harding et al. 1991), and as evidence for them is sought, we are becoming cognizant of their importance in an ever broader array of organisms.

Non-Mendelian parental effects are important because characters influenced by these effects can evolve in qualitatively different ways than do most traits. For independent trait exhibiting Mendelian inheritance, evolutionary response can be predicted by the breeder's equation, [Delta]z = [h.sup.2]s, in which [Delta]z is the change in the population mean of trait z and s is the selection differential, s = z - [z.sup.*], where z is the mean of the trait in the parental population and [z.sup.*] is the mean of the trait among selected parents (Falconer 1981). In particular, the breeder's equation states that only selection during the current generation can influence the evolutionary response. In contrast, for traits influenced by maternal or paternal inheritance, the response to selection depends not only on the force of selection in the current generation, but also on the evolutionary response in the previous generation (Falconer 1965; Kirkpatrick and Lande 1989). Thus, maternal and paternal modes of inheritance produce time lags in the response of a population to selection, which can cause unusual evolutionary dynamics (Kirkpatrick and Lande 1989). It has also been suggested that time lags introduced by maternal or paternal inheritance could influence the numerical dynamics of local populations exhibiting these modes of inheritance (Rossiter et al. 1993).

In this paper, we present evidence to suggest that paternal effects are important in the inheritance of resistance to a fungal pathogen in the tall morning glory, Ipomoea purpurea. Like other types of parasites, plant pathogens are presumed to exert a major influence on the structure and dynamics of host populations and communities. Consequently, to understand many aspects of plant ecology, we must understand the evolution of plant responses to pathogens. The mode of transmission of such responses in natural host populations is one important aspect of the biology of these traits.


Experimental Organisms and Study Site

Ipomoea purpurea Roth (Convolvulaceae) is a self-compatible annual vine with a mixed mating system (Brown and Clegg 1984; Schoen and Epperson 1985; Epperson and Clegg 1987). It is commonly found as an agricultural weed throughout the southern United States (Oliver et al. 1976), and in North Carolina it is pollinated by bumblebees (Bombus pennsylvanicus), (Rausher and Fry 1993). One of the major pathogens of I. purpurea is Colletotrichum dematium Pers. ex Fr. f ipomoea v. Arx (Fungi imperfecti), which causes anthracnose (Simms and Rausher 1993). The fungus overwinters on dried infected leaves in the soil (Agrios 1988) and infects I. purpurea via asexual spores (conidia) after warm summer rains, causing characteristic regular lesions on the leaves (Simms and Rausher 1993). Anthracnose is a localized disease that can spread to other parts of the plant only by new infection.

The study site was a newly plowed field on the Wake Forest University Campus in Forsyth County, North Carolina. Prior to plowing, the field had been a lawn for at least 30 yr and harbored neither I. purpurea nor C. dematium, ensuring that disease observed on inoculated plants was due to treatments and not to indigenous fungus. Seeds for this study were each obtained from different individuals of I. purpurea growing in a vegetable garden near Pfafftown in Forsyth County, North Carolina. Fungal isolates were collected from two other Forsyth County I. purpurea populations.

Experimental Design and Protocol

The experiment described here was part of a larger experiment that was previously used to examine trade-offs associated with resistance to and tolerance of fungal infection (Simms and Triplett 1995). Twelve seedlings were raised in the greenhouse at Wake Forest University during winter 1988-1989 and used as parents in three four-by-four diallel crosses (Falconer 1981; Becker 1984) to produce 36 full-sib families within 12 paternal and 12 maternal half-sib families. To eliminate potential effects of inbreeding depression, families consisting of selfed offspring were not included in the design. Although we could not completely randomize the timing of specific crosses, to reduce the effect of parental developmental stage and environmental variance on offspring phenotype, we attempted to complete one round of pollen donors for each recipient before initiating another round of donors.

In early July 1990, offspring from these diallels were germinated in the greenhouse. After two weeks, four seedlings from each family were inoculated with one of two fungal isolates (105-2ya and 106-24x, for comparison with results in Simms and Triplett [1995]). Inoculations were performed by spraying with an aqueous spore suspension until dripping (Simms and Triplett 1995). Plants were then placed into a 27 [degrees] C dew chamber (Suretemp) for 12 h to maintain 100% relative humidity. Immediately following inoculation, all seedlings were planted into the field in a completely randomized, split-plot design with two treatment plots randomly located within each of four spatial blocks. Within plots, individual offspring from the 36 full-sib families were randomly assigned to locations 0.7 m apart in a square grid. Plots were separated by 1.4-m aisles to reduce cross-contamination by rain splash. The 274 plants analyzed here were located within a larger experimental population consisting of nearly 1300 plants. Plants were watered by sprinkler with well water for one week following transplanting to ensure establishment. During the last two weeks of September, plants were harvested and resistance was measured as the complement of the proportion of leaves damaged by anthracnose (1 - number of damaged leaves/total number of leaves).

Statistical Analysis

Damage was first analyzed with a univariate, repeated measures, mixed-model analysis of variance (ANOVA) appropriate for a split-plot design (Neter et al. 1990, pp. 1066-1073) using the GLM procedure of SAS (SAS Institute 1989). Sires and dams were considered random effects nested within diallels; treatments (fungal genotypes) were considered fixed effects assigned to sections of spatial blocks. Proportion damage was arcsine square-root transformed to conform to the assumptions of the ANOVA.

In addition to the mixed model ANOVA, we also used the diallel analysis developed by Cockerham and Weir (1977) to detect nonnuclear parental effects on proportion of leaves damaged. For this analysis, we partitioned total phenotypic variance, [V.sub.P], using a modified form of the standard factorial ANOVA model

[Y.sub.ijk] = [Mu] + [M.sub.i] + [P.sub.j] + [(MP).sub.ij] + [[Epsilon].sub.ijk] (1)

where [Y.sub.ijk] is the resistance level of offspring k from dam i and sire j; [Mu] is the grand mean; [M.sub.i] and [P.sub.j] represent the total effects (nuclear, cytoplasmic, and environmental) of mother i and father j, respectively; [(MP).sub.ij] represents the interaction of dam i and sire j; and [[Epsilon].sub.ijk] is random variance due to the specific environment of offspring k. All effects in the model are random, with variances [Mathematical Expression Omitted], [Mathematical Expression Omitted], and [Mathematical Expression Omitted]. Furthermore, when parent i functions as both sire and dam in a diallel design, it contributes the same nuclear genes to all its offspring, regardless of whether through ovules or pollen, therefore creating a covariance among its offspring, [cov.sub.MP]. Also, [(MP).sub.ij] and [(MP).sub.ji] (sibs in reciprocal families) may share additional causal factors, with covariance, [cov.sub.MxP]. Table 1 shows how these variances and covariances are related to the factors in the factorial model (eq 1). In addition to these effects, our model also included block, treatment, and treatment-by-block effects. Furthermore, the sire, dam, and sire-by-dam interactions were all nested within diallel, as required by our particular experimental design.

We then specified the appropriate contrast matrices (Timm 1975) to repartition the variance into effects in the diallel model (Cockerham and Weir 1977):


[Y.sub.ijk] = [Mu] + [g.sub.i] + [g.sub.j] + [s.sub.ij] + [d.sub.i] - [d.sub.j] + [r.sub.ij] + [[Epsilon].sub.ijk](2a)


[s.sub.ij] = [s.sub.ji] and [r.sub.ij] = -[r.sub.ji], (2b)

where [g.sub.i] is the average effect of individual i as parent (the general nonreciprocal combining ability, GCA [Falconer 1981, p. 49]), [s.sub.ij] is the average effect of crossing individuals i and j, i [not equal to] j (the specific nonreciprocal combining ability, SCA [Falconer 1981, p. 49]), [d.sub.i] is the difference between individual i as sire and individual i as dam (the general reciprocal combining ability [Antonovics and Schmitt 1986]), and [r.sub.ij] is the difference between reciprocal full-sib families (from dam i and sire j vs. from sire i and dam j; the specific reciprocal combining ability [Antonovics and Schmitt 1986]). These effects can be obtained from the factorial model (eq 1) by the following transformations:

[g.sub.i] = ([M.sub.i] + [P.sub.i])/2

[d.sub.i] = ([M.sub.i] - [P.sub.i])/2

[s.sub.ij] = [[(MP).sub.ij] + [(MP).sub.ji]]/2

[r.sub.ij] = [[(MP).sub.ij] - [(MP).sub.ji]]/2, (3)

and their variances can also be obtained directly from the variances and covariances of the factorial model (eq 1):

[Mathematical Expression Omitted]

[Mathematical Expression Omitted]

[Mathematical Expression Omitted]

[Mathematical Expression Omitted]. (4)

To calculate the sums of squares associated with the effects in the diallel model (eq 2a,b), we specified contrasts for the general nonreciprocal combining ability, [g.sub.i], the total non-reciprocal effect, ([g.sub.i] + [s.sub.ij]), the general reciprocal combining ability, [d.sub.i], and the total reciprocal effect ([d.sub.i] + [r.sub.ij]). The sums of squares and degrees of freedom associated with the specific nonreciprocal combining ability, [s.sub.ij], and the specific reciprocal combining ability, [r.sub.ij], were obtained by subtraction.

Finally, the effects from the diallel model (eq 2a,b) can be transformed to fit the biological model:

[Y.sub.ijk] = [Mu] + [n.sub.i] + [n.sub.j] + [t.sub.ij] + [m.sub.i] + [p.sub.j] + [k.sub.ij] + [[Epsilon].sub.ijk], (5)

where [n.sub.i] and [n.sub.j] represent the additive nuclear contributions [TABULAR DATA FOR TABLE 2 OMITTED] of the parents; [t.sub.ij] represents the specific interactions between the two nuclear genomes (assuming [t.sub.ij] = [t.sub.ji]), which are equivalent to nonadditive nuclear effects; [m.sub.i] is the maternal extranuclear effect; [p.sub.j] is the paternal extranuclear effect; and all nuclear, extranuclear, and nuclear by extranuclear interactions are included in [k.sub.ij] = [(nm).sub.ii] + [(np).sub.ij] + [(nm).sub.ji] + [(np).sub.jj] + [(mp).sub.ij] + higher order interactions. The variances associated with these effects are related to the previous models and were obtained by the relationships provided in Table 2.

To estimate the maternal and paternal extranuclear effects ([m.sub.i] and [p.sub.i], respectively), we used contrast statements to estimate

[Mathematical Expression Omitted] (6a)


[Mathematical Expression Omitted] (6b)

for u offspring per cross in a diallel of N parents. The expectations of these variables are

[Mathematical Expression Omitted] (7a)


[Mathematical Expression Omitted]. (7b)

We then applied the method of symmetrical products (Koch 1967) as described in Cockerham and Weir (1977). Because this method is sensitive to imbalance in our experimental design, these estimates should be considered approximates.

The above analyses were performed with the general linear models procedure (GLM) of the SAS statistical analysis system, version 6.07 for Windows[TM] (SAS Institute 1989). In all analyses, F- and P-values are approximate because of the highly skewed distribution of damage values.

Finally, because missing values produced an unbalanced design, we also estimated variance components for the standard factorial model using the restricted maximum likelihood method in the VARCOMP procedure of SAS (SAS Institute 1989). This method constrains variance components to non-negative values. A separate analysis was performed on each diallel in each fungal treatment to yield six estimates for each variance component. Sire and dam components of variance were compared using a t-test modified to account for unequal variances.
TABLE 3. ANOVA table for complete factorial analysis of damage.
[Y.sub.ijklmn] denotes the nth plant in the ith block R, inoculated
with the jth isolate T split within block, from the lth sire S and
mth dam D, nested within the kth diallel [Delta]. Treating block,
sire, and dam as random effects, the model is [Y.sub.ijklmn] =
[Mu]..... + [[Rho].sub.i] + [t.sub.j] + [([Rho]t).sub.ij] +
[[Delta].sub.k] + [s.sub.l(k)] + [d.sub.m(k)] + s[d.sub.lm(k)] +
[Rho]t[s.sub.ijl(k)] + [[Epsilon].sub.ijlmn(k)] where [Mu]..... is
a constant is a constant and [[Epsilon].sub.ijlmn(k)] is independent
N(0, [[Sigma].sup.2]).

Source                 df         Type III SS         F         P

Block                   3            0.8336         5.93      0.001
Trt                     1            0.1502         1.04(*)   0.4
Trt*blk                 3            0.4317         3.07      0.03
Diallel                 2            0.1270         1.35      0.3
Sire(dial)              9            0.9439         2.24      0.03
Dam(dial)               9            0.1257         0.3       0.9
Sire*dam(dial)         15            0.3910         0.56      0.9
Trt*sire*dial*blk      63            2.6206         0.89      0.7
Trt*dam*dial*blk       63            3.0110         1.02      0.5
Error                  91            4.2651

* Satterthwaite approximate F-test, df = 1, 3.


Damage to the experimental plants was within the range (0-25%) observed for naturally infected plants in a Durham County, North Carolina, population but with a higher mean level of damage as compared to that population (0.2%; Simms 1993).

Resistance, defined as the complement of proportion damage, varied significantly among paternal half-sib families in the factorial ANOVA (Table 3, [ILLUSTRATION FOR FIGURE 1 OMITTED]), suggesting that the base population from which parent plants were obtained would possess heritable variation for damage in an environment with an equal probability of encountering either of the two fungal isolates used here. However, the absence of significant variation among maternal half-sib families suggests that variation among paternal half-sib families may be due to a nonnuclear paternal effect. In particular, the sire-nested-within-diallel component of variance was significantly larger than the dam-nested-within-diallel component ([F.sub.9,9] = 7.51, P = 0.003). The diallel analysis corroborates this interpretation by showing that most of the variance among paternal half-sib families was due to a nonnuclear paternal effect (Tables 4 and 5). Variance components associated with both the dam effect and the sire by dam interaction were negative, although only the former was significantly different from zero (P = 1-0.96 = 0.04 for dam [diallel] effect in Table 4). This result indicates that progeny in different maternal families resembled one another more than progeny of the same mother.

A significant difference between the magnitudes of the maternal and paternal components of variance should produce a large variance component associated with the general, reciprocal combining ability. However, the expected mean square of this effect is a linear combination of the variance components associated with nuclear and nonnuclear interactions ([Mathematical Expression Omitted] and maternal and paternal effects of [Mathematical Expression Omitted] and [Mathematical Expression Omitted], respectively) (Cockerham and Weir 1977). Two of these variance components were negative (Table 5), significantly reducing the mean square due to the general, reciprocal combining ability.

To control for the possibility that the negative variance components and paternal effect might be artifacts of an unbalanced crossing design, we used a maximum likelihood method to determine whether the paternal effect would persist when negative variance components were constrained to zero. In this analysis, the paternal component of variance was nearly an order of magnitude larger than the maternal component (Table 6). In only one of the combinations of treatment and diallel was the maternal component larger than zero; in that combination the maternal component was similar the paternal component.


The results of the different analyses are consistent. Damage varied significantly among paternal half-sib families, suggesting that resistance could evolve in response to selection. However, because damage varied among paternal half-sib families significantly more than it did among maternal half-sib families, the inheritance of disease resistance appears to be significantly influenced by non-Mendelian paternal effects. Furthermore, progeny from different maternal families resembled one another significantly more than progeny of the same mother, suggesting that genomic imprinting in the endosperm could be in part responsible for this pattern of variance components.

Maternal effects on offspring phenotype are commonly observed and several causal mechanisms have been identified. Offspring frequently spend some portion of their lives with their mother, sharing her environment and in many cases benefiting from her care and/or provisioning. There are also genetic and nongenetic effects associated with cytoplasmic transmission, which usually occurs through the mother (Hohenboken 1985). For example, disease resistance is sometimes transmitted in the nonnuclear genomic component. Resistance to southern corn blight (Cochliobolus heterostrophus) and yellow corn blight (Mycosphaerella zeae-maydis) are both associated with the maternally inherited T male-sterile cytoplasm in maize (Zea mays) (Pring and Lonsdale 1989). The gene coding for C. heterostrophus resistance has been identified on the mitochondrial genome (Dewey et al. 1988). In tobacco, susceptibility to tentoxin produced by the pathogenic [TABULAR DATA FOR TABLE 4 OMITTED] fungus Alternaria alternata (tenuis) segregates with the chloroplast (Aviv et al. 1980; Flick and Evans 1982).

Except in organisms with paternal care, non-Mendelian paternal effects are rarely considered likely candidates for influencing offspring phenotype. Nonetheless, there is now substantial evidence that many plant species exhibit biparental inheritance of plastid or mitochondrial genomes (Metzlaff et al. 1981; Medgyesy et al. 1986; Boblenz et al. 1990; Masoud et al. 1990; Wagner et al. 1991) and this phenomenon may be more common than is currently appreciated (Milligan 1992). Furthermore, in some cases one or both nonnuclear genomes are inherited strictly paternally (Neale et al. 1989; Neale and Sederoff 1989; Schumann and Hancock 1989; Harrison and Doyle 1990). Recently, paternal inheritance of tentoxin (pathogen) resistant chloroplasts was demonstrated in a Nicotiana tabacum line containing N. undulatum plastids (Avni and Edelman 1991).

Our results suggest that mitochondrial and chloroplast transmission should be further explored in I. purpurea. We have found no information on mitochondrial inheritance in Ipomoea. Cytological evidence suggests that I. nil transmits plastid DNA biparentally (Corriveau and Coleman 1988), although an earlier study found strictly maternal transmission of a chloroplast-encoded trait (Miyake and Imai 1935). Neither mode of inheritance could be responsible for strictly paternal effects, however. Nevertheless, there is some evidence in petunia that the capacity of the pollen parent to transmit plastids to progeny is heritable (Derepas and Dulieu 1992), which could produce the strictly paternal pattern of inheritance we observed.

A paternal effect could also be produced by influences of paternal environment on pollen traits that affect progeny phenotype. Environmental conditions can affect pollen characters such as size, germination, pollen-number, and pollen-tube growth rate, as well as number of seeds sired (Young and Stanton 1990; Lau and Stephenson 1993; 1994; Quesada et al. 1995). However, the low variance in damage among maternal half-sib families and the close spacing of parent plants within the greenhouse suggest that environmental variance in the parental generation had little influence on late expression of anthracnose resistance in the progeny.

Gametophytic selection due to pollen-tube competition is another process that could cause paternal effects (Shaw and Waser 1994). There are at most six ovules in each I. purpurea flower (Radford et al. 1968) and crosses for this experiment were performed by brushing the paternal anther onto the maternal stigma. Each stigma thus received a large pollen load, creating an opportunity for competition among pollen grains (Walsh and Charlesworth 1992). Furthermore, the style is at least 2 cm long (pers. obs.), allowing space for pollen tube [TABULAR DATA FOR TABLE 5 OMITTED] competition. Differential paternity has been observed in I. purpurea, although its mechanisms have not been elucidated (Epperson and Clegg 1987).

Mulcahy and Mulcahy (1975) outlined three modes by which competition among pollen grains might alter progeny phenotype and modify gene frequencies. These modes are distinguished by the timing of gene transcription. In particular, genes may be transcribed in (1) only the haploid gametophyte; (2) only the diploid sporophyte; or (3) both the haploid and diploid phases of the life cycle. They pointed out that only in the latter case, when genes are transcribed in both the haploid and diploid stages (pleiotropy), could gametophytic competition influence the phenotype of progeny sporophytes (Mulcahy and Mulcahy 1975). Significant overlap in gene transcription and expression between the sporophyte and gametophyte has been found in tomato, Lycopersicon esculentum (Tanksley et al. 1981), barley, Hordeum vulgate (Pedersen et al. 1987), and three species of Populus (Rajora and Zsuffa 1986). Further, genetic correlation between sporophytic and gametophytic traits has been demonstrated for pollen-tube growth and seed size (Mulcahy 1971) and pollen-tube growth and seedling height in Zea mays (Mulcahy 1974), as well as low temperature tolerance by [TABULAR DATA FOR TABLE 6 OMITTED] microgametophytes and sporophytes in hybrids of L. esculentum and L. hirsutum (Zamir et al. 1981; 1982). Embryolethal mutations in Arabidopsis thaliana also influence pollen-tube growth rates (Meinke 1982). The taxonomic breadth of these examples suggest that pleiotropic gene effects in pollen and sporophytes due to transcription in both the diploid and haploid phases of the life cycle may be widespread in angiosperms.

Additive genes that are expressed in both life history phases should also produce detectable additive nuclear effects via both the maternal and paternal contribution. Because we did not find such nuclear effects, we suggest instead that (1) some paternal plants were heterozygous at loci controlling pollen-tube growth rate; and (2) fast-growing recessive alleles at these loci may have been pleiotropic for either high or low fungal resistance. If so, then faster growing haplotypes would be more likely to fertilize ovules and produce exceptional resistance levels in these paternal families. In contrast, when the same individuals act as maternal plants, the average resistance level of their offspring should not be exceptional. By this mechanism, paternal effects would be due to recessive alleles that produce additive effects when expressed in the haploid gametophyte.

Gametophytic selection would have interesting evolutionary implications, as it could provide another mechanism for fitness costs of resistance. Fitness costs associated with resistance strongly influence the location and stability of equilibria in coevolutionary models of plants and pathogens (reviewed in Parker 1992). Thus, an important goal for further understanding both host-pathogen coevolution and the maintenance of genetic variation in natural populations is to determine how commonly costs are associated with resistance and virulence (Simms 1992). If alleles coding for fast pollen-tube growth also produced low disease resistance in the sporophyte, then highly resistant plants might experience a cost owing to their reduced likelihood of siring offspring when competing with the pollen of less resistant individuals in mixed pollen loads.

Gametophytic competition is likely to occur in natural populations of I. purpurea. Pollen-tube competition has been observed in mixed artificial pollinations from white-flowered and colored-flowered donors in I. purpurea (Epperson and Clegg 1987). Furthermore, because I. purpurea has at most six ovules in each flower, seed set in the field is unlikely to be pollen limited (Snow 1986). In our populations, flowers receive frequent visits from bumblebee pollinators that also exhibit high fidelity to I. purpurea (Brown and Clegg 1984; Epperson and Clegg 1987). Thus, flowers often receive multiple visits and accumulate heavy pollen loads (Epperson and Clegg 1987). Thus, if the paternal effect in anthracnose resistance were due to gametophytic competition, it could also be a plausible mechanism for fitness costs of resistance in this species. Clearly, further study will be necessary to determine whether paternal inheritance of resistance in I. purpurea is due to gametophytic selection.

A negative maternal component of variance indicates that offspring from different mothers are more similar than offspring from a single mother. This pattern could be produced by competition among progeny for resources provided by their mother. It has generally been argued that mothers should allocate resources equally among progeny (Smith and Fretwell 1974). However, two mechanisms might produce variance among progeny. First, the lengthy duration of flowering and seed maturation in I. purpurea provides the opportunity for establishment of rank hierarchies among sibling seeds, which could produce considerable variance in offspring quality within a single mother (Capinera 1979; Crump 1981). This phenomenon has been observed in birds whose nestlings hatch asynchronously (Hahn 1981; Greig-Smith 1985; Mock and Ploger 1987; Lessels and Avery 1989; Magrath 1989). Even under scramble competition, asynchronous development can produce variation in quality among offspring (Parker et al 1989). Interestingly, Seigel and Ford (1992) found that female snakes with greater access to food exhibited more within-female variance in offspring quality than those receiving less food.' Thus, obtaining seeds from well watered and fertilized greenhouse-grown plants might have inflated the variance among progeny within a mother.

Our results could also be explained by genomic imprinting, a phenomenon in which alleles behave differently depending upon the sex of the parent from which they are inherited (e.g., Tsai and Silver 1991; Ramesar et al. 1993). Imprinting is caused by differences in gametic DNA from male and female parents in methylation patterns and subsequent expression (Li et al. 1993; Razin and Cedar 1994; Chaillet et al. 1995). Genomic imprinting in angiosperms occurs mainly in the triploid endosperm rather than the diploid embryo (Haig and Westoby 1989; 1991). For example, maize endosperms that lack a paternal copy of one arm of chromosome 10 produce small kernels (Kermicle 1970; Lin 1984) and kernel size cannot be restored by adding extra maternal doses of that arm (Lin 1982). Evidence suggestive of genomic imprinting in the endosperm has been obtained with interspecific hybrid crosses among Solanum (Solanaceae) species (Johnston et al. 1980) and oats (Avena, Poaceae) (Nishiyama and Yabuno 1978). Haig and Westoby (1989) have suggested that enhanced expression of paternally derived alleles is a mechanism by which fathers can conflict with the mother over the optimal quantity of resources to allocate to a seed. There is an outside possibility that adult disease resistance is influenced by seed provisioning, which could, in turn, be influenced by paternal gene expression in the endosperm. Further work is necessary to test this hypothesis.

A growing number of reports of paternal effects from quantitative genetic studies suggests that "novel" genetic effects or selection pressures may sometimes bias the estimation of genetic parameters using kinship arrays. New models must be derived (e.g., Shaw and Waser 1994) that take these mechanisms into account.


Thanks to P. Becker and the other Wake Forest undergraduates who assisted us in the field. We also thank D. Pilson, M. Rausher, J. B. Walsh, M. C. Rossiter, and an anonymous reviewer for thoughtful comments on the manuscript. We are grateful for the statistical expertise of S. Darin, F. Vaida, and P. McCullagh of the University of Chicago Department of Statistics. J.K.T. was supported during the field work by the Howard Hughes Summer Undergraduate Research Program of the Department of Biology, Wake Forest University. This project was supported by National Science Foundation grants BSR 89-18030 and DEB 9196188-A01 to E.L.S.


AGRIOS, G. N. 1988. Plant pathology. 2d ed. Academic Press, New York.

ANTONOVICS, J., AND J. SCHMITT. 1986. Paternal and maternal effects on propagule size in Anthoxanthum odoratum. Oecologia (Berlin) 69:277-282.

AVIV, D., R. FLUHR, M. BLEICHMAN, AND E. GALUN. 1980. Progeny analysis of the interspecific somatic hybrids: Nicotiana tabacum (CMS) + Nicotiana sylvestris with respect to nuclear and chloroplast markers. Theor. Appl. Gen. 56:145-150.

AVNI, A., AND M. EDELMAN. 1991. Direct selection for paternal inheritance of chloroplasts in sexual progeny of Nicotiana. Mol. Gen. Genet. 225:273-277.

BECKER, W. A. 1984. Manual of quantitative genetics. 4th ed. Academic Enterprises, Pullman, WA.

BOBLENZ, K., T. NOTHNAGEL, AND M. METZLAFF. 1990. Paternal inheritance of plastids in the genus Daucus. Mol. Gen. Genet. 220:489-491.

BROWN, B. A., AND M. T. CLEGG. 1984. Influence of flower color polymorphism on genetic transmission in a natural population of the common morning glory, Ipomoea purpurea. Evolution 38: 796-803.

CAPINERA, J. L. 1979. Qualitative variation in plants and insects: effect of propagule size on ecological plasticity. Am. Nat. 114: 350-361.

CHAILLET, J. R., D. S. BADER, AND P. LEDER. 1995. Regulation of genomic imprinting by gametic and embryonic processes. Genes & Dev. 10:1177-1187.

COCKERHAM, C. C., AND B. S. WEIR. 1977. Quadratic analysis of reciprocal crosses. Biometrics 33:187-203.

CORRIVEAU, J. L., AND A. W. COLEMAN. 1988. Rapid screening method to detect potential biparental inheritance of plastid DNA and results for over 200 angiosperm species. Am. J. Bot. 75: 1443-1458.

CRUMP, M. L. 1981. Variation in propagule size as a function of environmental uncertainty for tree frogs. Am. Nat. 117:724-737.

DEREPAS, A., AND H. DULIEU. 1992. Inheritance of the capacity to transfer plastids to the pollen parent in Petunia hybrida Hort. J. Hered. 83:6-10.

DEWEY, R. E., J. N. SIEDOW, D. H. TIMOTHY, AND C. S. LEVINGS III. 1988. A 13 kD maize mitochondrial protein in E. coli confers sensitivity to Bipolaris maydis toxin. Science 239:293-295.

EPPERSON, M. K., AND M. T. CLEGG. 1987. First-pollination primacy and pollen selection in the morning glory, Ipomoea purpurea. Heredity 58:5-14.

FALCONER, D. S. 1965. Maternal effects and selection response. Pp. 763-774 in S. J.Geerts, ed. Genetics today; proceedings of the XI international congress on genetics. Vol. 3. Pergamon, Oxford.

-----. 1981. Introduction to quantitation genetics. 2d ed. Longman, London.

FLICK, C. E., AND D. A. EVANS. 1982. Evaluation of cytoplasmic segregation in somatic hybrids of Nicotiana: tentoxin sensitivity. J. Hered. 73:264-266.

FOX, C. W., K. J. WADDELL, AND T. A. MOUSSEAU. 1995. Parental host plant affects offspring life histories in a seed beetle. Ecology 76:402-411.

FUTUYMA, D. J., C. HERRMANN, S. MILSTEIN, AND M. C. KEESE. 1993. Apparent transgenerational effects of host plant in the leaf beetle Ophraella notulata (Coleoptera, Chrysomelidae). Oecologia 96:365-372.

GEISEL, J. T. 1988. Effects of parental photoperiod on development time and density sensitivity of progeny in Drosophila melano-Master. Evolution 42:1348-1350.

GREIG-SMITH, P. 1985. Weight differences, brood reduction, and sibling competition among nestling Stonechats Saxicola torquata (Aves: Turdidae). J. Zool. Soc. Lond. A 205:453-465.

HAHN, D. C. 1981. Asynchronous hatching in the laughing gull: Cutting losses and reducing rivalry. Anim. Behav. 29:421-427.

HAIG, D., AND M. WESTOBY. 1989. Parent-specific gene expression and the triploid endosperm. Am. Nat. 134:147-155.

-----. 1991. Genomic imprinting in endosperm: Its effect on seed development in crosses between species, and between different ploidies of the same species, and its implications for the evolution of apomixis. Phil. Trans. R. Soc. Lond. B Biol. Sci. 333: 1-13.

HARDING, J., H. HUANG, T. BYRNE. 1991. Maternal, paternal, additive, and dominance components of variance in Gerbera. Theor. Appl. Genet. 82:756-760.

HARRISON, R. G., AND J. J. DOYLE. 1990. Redwoods break the rules. Nature 344:295-296.

HOHENBOKEN, W. D. 1985. Maternal effects. Pp. 135-149 in A. B. Chapman, ed. General and quantitative genetics. Elsevier, Amsterdam, The Netherlands.

JOHNSTON, S. A., T. P. M. DEN NIJS, S. J. PELOQUIN, AND R. E. HANNEMAN JR. 1980. The significance of genic balance to endosperm development in interspecific crosses. Theor. Appl. Genet. 57:5-9.

KERMICLE, J. L. 1970. Dependence of the R-mottled aleurone phenotype in maize on mode of sexual transmission. Genetics 66: 69-85.

KIRKPATRICK, M., AND R. LANDE. 1989. The evolution of maternal characters. Evolution 43:489-503.

KOCH, G. G. 1967. A general approach to the estimation of variance components. Technometrics 9:93-118.

LAU, T C., AND A. G. STEPHENSON. 1993. Effects of soil nitrogen on pollen production, pollen grain size, and pollen performance in Cucurbita pepo (Cucurbitaceae). Am. J. Bot. 80:763-768.

-----. 1994. Effects of soil phosphorus on pollen production, pollen size, pollen phosphorus content, and the ability to sire seeds in Cucurbita pepo (Cucurbitaceae). Sex, Plant Repro. 7: 215-220.

LESSELLS, C. M., AND M. I. AVERY. 1989. Hatching asynchrony in European bee-eaters Merops apiaster. J. Anita. Ecol. 58:815-835.

LI, E., C. BEARD, AND R. JAENISCH. 1993. Role for DNA methylation in genomic imprinting. Nature (London) 366:362-365.

LIN, B.-Y. 1982. Association of endosperm reduction with parental imprinting in maize. Genetics 100:475-486.

-----. 1984. Ploidy barrier to endosperm development in maize. Genetics 107:103-115.

MAGRATH, R. D. 1989. Hatching asynchrony and reproductive success in the blackbird. Nature (London) 339:536-538.

MASOUD, S. A., L. B. JOHNSON, AND E. L. SORENSON. 1990. High transmission of paternal plastid DNA in alfalfa plants demonstrated by restriction fragment polymorphic analysis. Theor. Appl. Genet. 79:49-55.

MEDGYESY, P., A. PAY, AND L. MARTON. 1986. Transmission of paternal chloroplasts in Nicotiana. Mol. Gen. Genetics 204:195-198.

MEINKE, D. W. 1982. Embryo-lethal mutants of Arabidopsis thaliana: Evidence for gametophytic expression of the mutant genes. Theor. Appl. Genet. 63:381-386.

METZLAFF, M., T. BORNER, AND R. HAGEMANN. 1981. Variations of chloroplast DNAs in the genus Pelargonium and their biparental inheritance. Theor. Appl. Genet. 60:43-53.

MILLIGAN, B. G. 1992. Is organelle DNA strictly maternally inherited? Power analysis of a binomial distribution. Am. J. Bot. 79:1325-1328.

MIYAKE, K., AND Y. IMAI. 1935. Maternal transmission of mutated plastids in the Japanese morning glory. Bow. Gaz. 96:571-574.

MOCK, D. W., AND B. J. PLOGER. 1987. Parental manipulation of optimal hatch asynchrony in cattle egrets: An experimental study. Anim. Behav. 35:150-160.

MULCAHY, D. L. 1971. A correlation between gametophytic and sporophytic characteristics in Zea mays L. Science 171:1155-1156.

-----. 1974. Correlation between speed of pollen tube growth and seedling height in Zea mays L. Nature (London) 249:491-493.

MULCAHY, D. L., AND G. B. MULCAHY. 1975. The influence of gametophytic competition on sporophytic quality in Dianthus chinensis. Theor. Appl. Genet. 46:277-280.

NEALE, D. B., AND R. R. SEDEROFF. 1989. Paternal inheritance of mitochondrial DNA in incense-cedar (Calocedrus decurrens). Theor. Appl. Genet. 77:212-216.

NEALE, D. B., K. A. MARSHALL, AND R. R. SEDEROFF. 1989. Chloroplast and mitochondrial DNA are paternally inherited in Sequioa sempervirens D. Don Endl. Proc. Nat. Acad. Sci. USA 86: 9347-9349.

NETER, J., W. WASSERMAN, AND W. H. KUTNER. 1990. Applied linear statistical models: Regression, analysis of variance, and experimental designs. 3d ed. Irwin, Homewood, IL.

NISHIYAMA, I., AND T. YABUNO. 1978. Causal relationships between the polar nuclei in double fertilization and interspecific cross-incompatibility in Arena. Cytologia (Tokyo) 43:453-466.

OLIVER, L. R., R. E. FRANS, AND R. E. TALBERT. 1976. Field competition between tall morning glory and soybean. I. Growth analysis. Weed Sci. 24:482-488.

PARKER, G. A., D. W. MOCK, AND T. C. LAMEY. 1989. How selfish should stronger sibs be? Am. Nat. 133:846-868.

PARKER, M. A. 1992. Disease and plant population genetic structure. Pp. 345-362 in R. A. Fritz and E. L. Simms, ed. Plant resistance to herbivores and pathogens: Ecology, evolution, and genetics. Univ. of Chicago Press, Chicago.

PEASE, C. M. 1984. On the evolutionary reversal of competitive dominance. Evolution 38:1099-1115.

PEDERSEN, S., V. SIMONSEN, AND V. LOESCHCKE. 1987. Overlap of gametophytic and sporophytic gene expression in barley. Theor. Appl. Genet. 75:200-206.

PRING, D. R., AND D. M. LONSDALE. 1989. Cytoplasmic male sterility and maternal inheritance of disease susceptibility in maize. Annu. Rev. Phytopathol. 27:483-502.

QUESADA, M., K. BOLLMAN, AND A. G. STEPHENSON. 1995. Leaf damage decreases pollen production and hinders pollen performance in Cucurbita texana. Ecology 76:437-443.

RADFORD, A. E., H. E. AHLES, AND C. R. BELL. 1968. The manual of the vascular flora of the Carolinas. Univ. of North Carolina Press, Chapel Hill.

RAJORA, O. P., AND L. ZSUFFA. 1986. Sporophytic and gametophytic gene expression in Populus deltoides Marsh., P. nigra L., and P. maximowiczii Henry. Can. J. Genet. Cytol. 28:476-482.

RAMESAR, R., M. BABAYA, AND D. VILJOEN. 1993. Molecular investigation of familial Beckwith-Wiedemann syndrome: a model for paternal imprinting. Fur. J. Human Genet. 1:109-113.

RAUSHER, M. D., AND J. D. FRY. 1993. Effects of a locus affecting floral pigmentation in Ipomoea purpurea on female fitness components. Genetics 134:1237-1247.

RAZIN, A., AND H. CEDAR. 1994. DNA methylation and genomic imprinting. Cell 77:473-476.

ROSSITER, M. C., D. L. COX-FOSTER, AND M. A. BRIGGS. 1993. Initiation of maternal effects in Lymantria dispar: Genetic and ecological components of egg provisioning. J. Evol. Biol. 6:577-589.

SAS INSTITUTE. 1989. SAS/STAT user's guide. Vers. 6.4th ed. Vol. 2. SAS Institute Inc., Cary, NC.

SCHMID, B., AND C. DOLT. 1994. Effects of maternal and paternal environment and genotype on offspring phenotype in Solidago altissima L. Evolution 48:1525-1549.

SCHOEN, D. J., AND B. K. EPPERSON. 1985. The influence of flower color on outcrossing rate and male reproductive success in Ipomoea purpurea. Evolution 39:1242-1249.

SCHUMANN, C. M., AND J. F. HANCOCK. 1989. Paternal inheritance of plastids in Medicago sativa. Theor. Appl. Genet. 78:863-866.

SEIGEL, R. A., AND N. B. FORD. 1992. Effect of energy input on variation in clutch size and offspring size in a viviparous reptile. Funct. Ecol. 6:382-385.

SHAW, R. G., AND N. M. WASER. 1994. Quantitative genetic interpretations of postpollination reproductive traits in plants. Am. Nat. 143:617-635.

SIMMS, E. L. 1990. Examining selection on the multivariate phenotype: Plant resistance to herbivores. Evolution 44:1177-1188.

-----. 1992. Costs of plant resistance to herbivory. Pp. 392-425 in R. S. Fritz and E. L. Simms, ed. Plant resistance to herbivores and pathogens: Ecology, evolution, and genetics. Univ. of Chicago Press, Chicago.

-----. 1993. Genetic variation for pathogen resistance in tall morning glory. Plant Dis. 77:901-904.

SIMMS, E. L., AND M. D. RAUSHER. 1993. Patterns of selection on phytophage resistance in Ipomoea purpurea. Evolution 47:970-976.

SIMMS, E. L., AND J. K. TRIPLETT. 1995. Costs and benefits of plant responses to disease: Resistance and tolerance. Evolution 48: 1973-1985.

SMITH, C. C., AND S. D. FRETWELL. 1974. The optimal balance between the size and number of offspring. Am. Nat. 108:499-506.

SNOW, A. A. 1986. Pollination dynamics in Epilobium canum (On-agraceae): consequences for gametophytic selection. Am. J. Bot. 73:139-151.

SOLONIEMI, I. 1993. A coevolutionary predator-prey model with quantitative characters. Am. Nat. 141:880-896.

TANKSLEY, S. D., D. ZAMIR, AND C. M. RICK. 1981. Evidence for extensive overlap of sporophytic and gametophytic gene expression in Lycopersicon esculentum. Science 213:453-455.

TIMM, N. H. 1975. Multivariate analysis with applications in education and psychology. Brooks/Cole, Monterey, CA.

TSAI, J. Y., AND L. M. SILVER. 1991. Escape from genomic imprinting at the mouse T-associated maternal effect TME locus. Genetics 129:1159-1166.

WAGNER, D. B., J. DONG, M. R. CARLSON, AND A. D. YANCHUK. 1991. Paternal leakage of mitochondrial DNA in Pinus. Theor. Appl. Genet. 82:510-514.

WALSH, N. E., AND D. CHARLESWORTH. 1992. Evolutionary interpretations of differences in pollen tube growth rates. Q. Rev. Biol. 67:19-37.

WATSON, M. J. O., AND A. A. HOFFMANN. 1995. Cross-generational effects for cold resistance in tropical populations of Drosophila melanogaster and D. simulans. Aust. J. Zool. 43:51-58.

YOUNG, H. J., AND M. L. STANTON. 1990. Influence of environmental quality on pollen competitive ability in wild radish. Science 248:1631-1633.

ZAMIR, D., S. D. TANKSLEY, AND R. A. JONES. 1981. Low temperature effect on selective fertilization by pollen mixtures of wild and cultivated tomato species. Theor. Appl. Genet. 59:235-238.

-----. 1982. Haploid selection for low temperature tolerance of tomato pollen. Genetics 101:129-137.
COPYRIGHT 1996 Society for the Study of Evolution
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1996 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Simms, Ellen L.; Triplett, Jim K.
Date:Dec 1, 1996
Previous Article:Meta-analyses of the association between multilocus heterozygosity and fitness.
Next Article:Hierarchical analysis of inbreeding depression in Peromyscus polionotus.

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters