Limits...
Fine ‐ scale frequency differentiation along a herbivory gradient in the trichome dimorphism of a wild Arabidopsis

View Article: PubMed Central - PubMed

ABSTRACT

Geographic variation is commonly observed in plant resistance traits, where plant species might experience different selection pressure across a heterogeneous landscape. Arabidopsis halleri subsp. gemmifera is dimorphic for trichome production, generating two morphs, trichome‐producing (hairy) and trichomeless (glabrous) plants. Trichomes of A. halleri are known to confer resistance against the white butterfly, cabbage sawfly, and brassica leaf beetle, but not against flea beetles. We combined leaf damage, microclimate, and microsatellite loci data of 26 A. halleri populations in central Japan, to explore factors responsible for fine‐scale geographic variation in the morph frequency. We found that hairy plants were less damaged than glabrous plants within populations, but the among‐site variation was the most significant source of variation in the individual‐level damage. Fixation index (Gst″) of a putative trichome locus exhibited a significant divergence along population‐level damage with an exception of an outlier population, inferring the local adaptation to herbivory. Notably, this outlier was a population wherein our previous study reported a balancing role of the brassica leaf beetle Phaedon brassicae on the morph frequency. This differentiation of the trichome locus was unrelated to neutral genetic differentiation (evaluated by Gst″ of microsatellite loci) and meteorological factors (including temperature and solar radiation). The present findings, combined with those of our previous work, provide suggestive evidence that herbivore‐driven divergence and occasional outbreak of a specific herbivore have jointly contributed to the ecogeographic pattern in the frequency of two morphs.

No MeSH data available.


Related in: MedlinePlus

Individual‐level damage (proportion of leaf area loss) of hairy (h; gray bars) and glabrous (g; white bars) plants in 26 populations of Arabidopsis halleri subsp. gemmifera. Data were pooled between 2014 and 2015. The site IDs (1–26; assigned from east to west) correspond to those in Figure 2 and the Tables S1 and S2. An inset shows differences in residual damage between hairy (h; white) and glabrous (g; black) plants when excluding (‐Site) or including (+Site) the site ID in multiple regressions. Error bars indicate SE of mean
© Copyright Policy - creativeCommonsBy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC5383478&req=5

ece32830-fig-0003: Individual‐level damage (proportion of leaf area loss) of hairy (h; gray bars) and glabrous (g; white bars) plants in 26 populations of Arabidopsis halleri subsp. gemmifera. Data were pooled between 2014 and 2015. The site IDs (1–26; assigned from east to west) correspond to those in Figure 2 and the Tables S1 and S2. An inset shows differences in residual damage between hairy (h; white) and glabrous (g; black) plants when excluding (‐Site) or including (+Site) the site ID in multiple regressions. Error bars indicate SE of mean

Mentions: The trichome phenotype, study sites, and study dates were significant sources of variation in leaf damage of A. halleri, whereas the study years and maximum leaf length did not explain the significant amount of variation in leaf damage (Table 1). The significant negative coefficient of trichome phenotype in the linear multiple regression showed less damage on hairy plants than on glabrous plants (coefficient of hairy phenotype ± SE = −0.047 ± 0.009, t = −5.086, p < 10−6). This decreased damage on hairy plants was visualized by a residual plot standardizing variation among the study sites (inset of Figure 3). However, the considerably larger amount of variation was attributable to the study sites rather than the other sources of variation (Table 1, Figure 3), indicating that among‐site variation was the most significant source of leaf damage. When damages were pooled among study sites, hairy plants were more damaged compared to glabrous plants (inset of Figure 3). The result that among‐site variation was the largest in the individual‐level damage was also supported by our complemental GLM (Appendix S1).


Fine ‐ scale frequency differentiation along a herbivory gradient in the trichome dimorphism of a wild Arabidopsis
Individual‐level damage (proportion of leaf area loss) of hairy (h; gray bars) and glabrous (g; white bars) plants in 26 populations of Arabidopsis halleri subsp. gemmifera. Data were pooled between 2014 and 2015. The site IDs (1–26; assigned from east to west) correspond to those in Figure 2 and the Tables S1 and S2. An inset shows differences in residual damage between hairy (h; white) and glabrous (g; black) plants when excluding (‐Site) or including (+Site) the site ID in multiple regressions. Error bars indicate SE of mean
© Copyright Policy - creativeCommonsBy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC5383478&req=5

ece32830-fig-0003: Individual‐level damage (proportion of leaf area loss) of hairy (h; gray bars) and glabrous (g; white bars) plants in 26 populations of Arabidopsis halleri subsp. gemmifera. Data were pooled between 2014 and 2015. The site IDs (1–26; assigned from east to west) correspond to those in Figure 2 and the Tables S1 and S2. An inset shows differences in residual damage between hairy (h; white) and glabrous (g; black) plants when excluding (‐Site) or including (+Site) the site ID in multiple regressions. Error bars indicate SE of mean
Mentions: The trichome phenotype, study sites, and study dates were significant sources of variation in leaf damage of A. halleri, whereas the study years and maximum leaf length did not explain the significant amount of variation in leaf damage (Table 1). The significant negative coefficient of trichome phenotype in the linear multiple regression showed less damage on hairy plants than on glabrous plants (coefficient of hairy phenotype ± SE = −0.047 ± 0.009, t = −5.086, p < 10−6). This decreased damage on hairy plants was visualized by a residual plot standardizing variation among the study sites (inset of Figure 3). However, the considerably larger amount of variation was attributable to the study sites rather than the other sources of variation (Table 1, Figure 3), indicating that among‐site variation was the most significant source of leaf damage. When damages were pooled among study sites, hairy plants were more damaged compared to glabrous plants (inset of Figure 3). The result that among‐site variation was the largest in the individual‐level damage was also supported by our complemental GLM (Appendix S1).

View Article: PubMed Central - PubMed

ABSTRACT

Geographic variation is commonly observed in plant resistance traits, where plant species might experience different selection pressure across a heterogeneous landscape. Arabidopsis halleri subsp. gemmifera is dimorphic for trichome production, generating two morphs, trichome&#8208;producing (hairy) and trichomeless (glabrous) plants. Trichomes of A.&nbsp;halleri are known to confer resistance against the white butterfly, cabbage sawfly, and brassica leaf beetle, but not against flea beetles. We combined leaf damage, microclimate, and microsatellite loci data of 26 A.&nbsp;halleri populations in central Japan, to explore factors responsible for fine&#8208;scale geographic variation in the morph frequency. We found that hairy plants were less damaged than glabrous plants within populations, but the among&#8208;site variation was the most significant source of variation in the individual&#8208;level damage. Fixation index (Gst&Prime;) of a putative trichome locus exhibited a significant divergence along population&#8208;level damage with an exception of an outlier population, inferring the local adaptation to herbivory. Notably, this outlier was a population wherein our previous study reported a balancing role of the brassica leaf beetle Phaedon brassicae on the morph frequency. This differentiation of the trichome locus was unrelated to neutral genetic differentiation (evaluated by Gst&Prime; of microsatellite loci) and meteorological factors (including temperature and solar radiation). The present findings, combined with those of our previous work, provide suggestive evidence that herbivore&#8208;driven divergence and occasional outbreak of a specific herbivore have jointly contributed to the ecogeographic pattern in the frequency of two morphs.

No MeSH data available.


Related in: MedlinePlus