- Top of page
- Materials and Methods
As is the case for other arthropod-induced plant galls, the mechanism of ball gall formation by the larva of the tephritid fly Eurosta solidaginis on selected species of Solidago (goldenrod), is unknown. Extracts and secretions of selected species of gall-inducing insects have been shown to cause growth promotion and in some cases structures resembling galls (Martin, 1942; Boysen Jensen, 1948; Plumb, 1953; Leatherdale, 1955; McCalla et al., 1962). The growth promoting activity has been determined to be associated with salivary or accessory glands for some species of gall-inducing insects (Plumb, 1953; Hovanitz, 1959; McCalla et al., 1962), but it has not yet been determined which chemical or chemicals are responsible for insect-mediated gall induction and development.
Early studies suggested that applications of auxins to plant tissues resulted in structures similar to galls (Hamner & Kraus, 1937; Guiscafré-Arrillaga, 1949), while others were unsuccessful in their attempts to produce galls artificially (Plumb, 1953; Mandl, 1957). Applications of auxins to plant tissues have not induced the formation of galls of complex morphologies (Hough, 1953) and have not produced structures that exhibit the degree of hyperplasia found in most galls (Mani, 1964). A number of studies have provided bioassay evidence for the presence of indole-3-acetic acid in extracts of gall-forming aphid species (Link et al., 1940) or in their saliva (Nysterakis, 1948; Hopp, 1955). However Mani (1964) contends that the aphid saliva was contaminated by honey-dew, invalidating the results. Others have reported the presence of IAA in the saliva and salivary glands of selected species of homopterans (Hori & Endo, 1977; Hori, 1992). All of these studies were done using bioassays, which are of limited validity.
The goldenrod ball gall (Fig. 1) forms in response to the activities of a single larva of E. solidaginis in stems of selected species of Solidago, with the preferred host varying in different geographical regions (Novak & Foote, 1980). Ball gall formation in Ithaca, NY, USA, is on S. altissima. Adults emerge from overwintering galls in late May or early June in the Ithaca, NY, USA, area. After mating, adult females lay eggs in the folded leaves of the terminal bud of S. altissima. Once a larva has hatched, it travels down through the folded leaves of the terminal bud and burrows into stem tissue generally right below the apical dome. Gall growth continues over a 3–4-wk period. During that time, a 3-mm diameter stem enlarges to form a gall up to 30 mm in diameter (Uhler, 1951), through the processes of cell division, enlargement, and differentiation (Beck, 1947). During the course of gall development, the first instar larvae do not increase greatly in size. Larvae continue to grow after gall growth has ceased, reaching maximal size about 3.5 months after hatching. Larvae overwinter as third instars and pupate in the spring (Uhler, 1951).
As galls develop, bands of meristematic tissue form in the pith and then split into strands of meristematic tissue that radiate from the central pith region surrounding the larval chamber outward to the fasicular cambia of the vascular bundles. In a mature gall, some of the cells of the meristematic strands show reticular thickenings that stain more deeply and resemble the thickenings of the walls of some vessel cells (Beck, 1947). Cosens (1912) and Ross (1936) refer to these strands as ‘vascular strands’, and Blum (1952) presents evidence to support the vascular nature of the strands.
A few studies have focused on determining the mechanism of ball gall formation by E. solidaginis. Beck (1947) and Mills (1969) were unsuccessful in attempts to induce ball gall formation by injecting goldenrod stems with a variety of compounds. Gall formation did not occur in response to injections of goldenrod plants with IAA, p-chlorophenoxyacetic acid, kinetin, N-6-benzyladenine, ecdysterone, farnesyl methyl ether, or saline extracts of second or third instar larvae of E. solidaginis (Mills, 1969). As Mills provided insufficient details regarding the manner or location of the chemical injections, factors that are very important when attempting to simulate insect-induced gall formation, one cannot assess the significance of these negative results. Beck (1947) observed limited cell division and/or differentiation in response to applications of ammonium hydroxide, ammonium carbonate, trypsin, bacto-peptone, bacto-tryptone, tryptophan, IAA, and naphthalene acetic acid. Crude organic extracts of young galls and of larvae injected into goldenrod stem tips, in water, yielded no positive results, while distilled water extracts of the larvae and distilled water in which larvae had been kept, resulted in some cell division (Beck, 1947). Mills (1969) reported that butanol extracts of larvae resulted in galls in three out of 28 injections, but provided no information regarding the size, morphology, and/or anatomy of the galls that resulted.
In studies aimed at determining the changes in levels of free amino acids in E. solidaginis larvae over time, Uhler et al. (1971) detected tryptophan in third instar larvae only, at the end of August, after gall growth had ceased, while Heady et al. (1982) detected tryptophan in E. solidaginis larvae earlier in the season, in July, but at a point when most of the gall growth had ceased. However the reported absence of a particular amino acid at any stage cannot be considered conclusive in these two studies because both groups did not indicate the amounts of tissues extracted, did not present values on a per gram tissue basis, and did not account for losses or indicate the sensitivity of their assays.
Carango et al. (1988) detected hyper-induction of a native 58 kilodalton protein in ball gall tissues that was most evident during the peak period of gall growth. They suggested that ball gall formation is the result of an alteration of existing plant growth mechanisms in response to secretions of the larva. Abrahamson et al. (1991) detected high levels of phenolics in developing ball galls and suggested that phenolics may play a role in ball gall formation.
The objective of this study was to determine the levels of indole-3-acetic acid in a developmental series of galls, in normal goldenrod stem tissues, and in first instar larvae of E. solidaginis in an effort to gain insight into the mechanism of ball gall formation on S. altissima. In addition, experiments were conducted to determine whether ball galls would continue to grow in the absence of the apical bud and leaves, natural sources of auxin, and to determine whether the presence of terminal galls on goldenrod stems lacking apical buds would result in the inhibition of lateral bud release.
- Top of page
- Materials and Methods
The concentration of IAA (measured by GC-MS (SIM) with [13C6]-IAA as an internal standard) in goldenrod stem tissue on a weight/weight basis, was higher than the amount of IAA in developing ball galls found at equivalent locations along the stem (Table 2). Class A galls, the smallest galls in the developmental series, contained 119 ng g−1 f. wt of IAA, which was the highest concentration of IAA found in all gall classes, but was 43% of the IAA found in the corresponding stem tissue. IAA concentration declined with the developmental age of the galls, with the sharpest decline occurring in the early stages of gall development. The weight/weight IAA concentration in the stem tissue declined along the 7.0 cm length of the stem. The upper 1 cm of stem tissue contained 44% of the IAA, the middle 1.0–3.5 cm of stem tissue contained 31% of the IAA, while the lower 3.5–7.0 cm of stem tissue contained 25% of the IAA. The decline in concentration of IAA in the developing galls was much more pronounced than the decline observed in the stem tissues further from the apex. On a weight/weight basis, class G galls contained 7% of the IAA found in class A galls, while the corresponding stem tissue at the equivalent location contained 56% of the IAA found in the top 1 cm of stem tissue (Table 2).
Table 2. Levels of extractable indole-3-acetic acid (IAA) from seven developmental stages of ball galls, from the corresponding goldenrod (Solidago altissima) stem sections at which the stages are found, and from larvae of Eurosta solidaginis removed from Class A and B galls. IAA was detected and quantified by gas chromatography -mass spectrometry (GC-MS)
|Stem location: (distance from apical bud)||Size class of galls (diameter)||Gall [IAA] (ng g−1 f. wt)||Stem [IAA] (ng g−1 f. wt)||Gall [IAA] (ng cm−1)||Stem [IAA] (ng cm−1)||[IAA] in larvae|
|0 cm-1.0 cm||A (4–6 mm)||119.0 ± 17.0b||275.0 ± 24.0||14.5 ± 2.2b,c||11.6 ± 1.6||9139a ng g−1 f. wt|
| ||B (6–9 mm)|| 46.1 ± 2.7b|| ||14.2 ± 0.6b,c|| ||1.21a ng larva−1|
|1 cm-3.5 cm||C (9–12 mm)|| 33.1 ± 4.8||195.8 ± 15.0b||16.2 ± 2.3||11.2 ± 1.0b|| |
| ||D (12–15 mm)|| 20.7 ± 4.4|| ||17.0 ± 5.1|| || |
|3.5 cm-7.0 cm||E (15–18 mm)|| 14.9 ± 0.6||152.7 ± 16.0||18.1 ± 0.2||13.3 ± 2.2|| |
| ||F (18–21 mm)|| 10.0a|| ||15.4a|| || |
| ||G (21–24 mm)|| 8.3 ± 1.0|| ||19.6 ± 1.2|| || |
When the data were expressed on a weight/stem length basis (Table 2), it was found that the galls contained higher amounts of IAA compared with the stem tissues at the corresponding locations. The weight/stem length values for class A and B galls are equivalent to ng gall−1 as A and B galls are 1 cm in length.
First instar larvae removed from class A and B galls up to 9 mm in diameter, were shown to contain 9139 ng g−1 f. wt or 1.21 ng larva−1 (Table 2) as detected by GC-MS with [13C6]-IAA as an internal standard to account for losses during extraction. The concentration of IAA in the larvae was 33 times greater than that found in the upper 1 cm of goldenrod stem. At 1.21 ng larva−1, the presence of the larva contributes to the difference between the weight/stem length values for gall and stem tissues. Class A and B galls contained a mean of 14.2–14.5 ng gall−1 of IAA, while the corresponding control contained a mean of 11.6 ng for an equivalent length of stem (Table 2).
Growth of ball galls in the absence of the apical bud
Ball galls continued to grow in the absence of the apical bud (Figs 2, 3), as well as in the absence of the apical bud and absence of leaves on the gall. The mean final size of galls with or without the presence of the apical bud or leaves on the galls was not statistically different. (F2,25 = 2.22, P > 0.10). The presence of terminal galls on goldenrod stems lacking apical buds resulted in the inhibition of lateral bud release. Twenty-two percent of the stems showed evidence of lateral bud release below terminal galls on goldenrod stems lacking apical buds, and 22% of the stems showed evidence of lateral bud release below terminal galls on goldenrod stems lacking leaves on galls as well as apical buds (Table 3). Of the stems showing lateral bud release below galls, a mean of 1–1.5 nodes contained released buds. In contrast, 100% of the control stems lacking galls showed lateral bud release following removal of the apical bud, and a mean of 5.6 nodes were released per stem (Table 3). The number of nodes with released lateral buds differed significantly among the three treatment groups (Kruskal–Wallis T = 17.685, df = 2, P < 0.001).
Figure 2. Growth of the ball gall of Solidago altissima in the absence of the apex (no apex), in the absence of the apex and leaves of the gall (no apex or leaves), and growth of control galls. Each data point represents the mean diameter of galls with standard errors (open circles, no apex or leaves; closed squares, no apex; closed triangles, control). The mean final size of galls with or without the presence of the apical bud or leaves on the galls was not statistically different. (F2,25 = 2.22, P > 0.10).
Download figure to PowerPoint
Figure 3. One of the experimental ball galls of Solidago altissima that continued to grow despite the early removal of the apical bud above the gall. Lateral bud release below the terminal gall was inhibited despite the absence of the apical bud.
Download figure to PowerPoint
Table 3. The effect of apex removal on lateral bud release on goldenrod (Solidago altissima) stems in the presence or absence of ball galls. The percentage of stems showing lateral bud release below ball galls in terminal positions due to apex removal above the galls, compared to lateral bud release on normal stems after apex removal, as well as the mean number of nodes with released buds per stem. The leaves on the galls were also removed in one of the treatment categories. Data were obtained 53 d after apex removal
|Treatment||Percentage of stems with lateral bud release||Mean # of nodes with released buds per stem|
|Stem, no apex (n = 7)||100%||5.6 ± 0.5a|
|Terminal gall, no apex (n = 9)|| 22%||1.0 ± 0.0a|
|Terminal gall, no apex, no leaves on gall (n = 9)|| 22%||1.5 ± 0.5a|
- Top of page
- Materials and Methods
IAA concentrations on a weight/weight basis, were highest in the early stages of gall development and declined with the developmental age of the galls. The decline in IAA concentration found in the developmental series of galls was more pronounced than the decline in IAA levels found in stem tissues at the corresponding locations below the apex.
IAA concentrations were not elevated when analysed on a weight/weight basis in the early stages of ball gall development in comparison with control stem tissues. However the stem tissues already contained high levels of IAA as they were also in a state of active growth at the time of harvest. In addition, a marked change in hormone levels would not have been expected as the organized morphology of the ball gall is not suggestive of gross changes in hormonal levels as are the morphologies of crown galls and other tumours induced by bacterial pathogens. The results presented here differ from those of Kaldeway (1965) and Byers et al. (1976) who presented evidence for higher auxin bioactivity in the Avena coleoptile test for insect-induced gall tissues compared with control leaf tissues. Kaldeway (1965) found that oak apple gall tissues of Quercus robur had twice as much auxin activity as normal leaf tissues, while Byers et al. (1976) found that pinyon (Pinus edulis) needles with basal galls induced by larvae of the midge Janetiella sp. near J. colouradensis contained 3.7 times higher concentrations of auxin bioactivity compared with needles lacking galls on a fresh tissue weight basis, and 17 times more auxin activity per needle. Byers et al. (1976) reported that levels of auxin bioactivity dropped with the developmental age of the galls, with the highest levels of auxin activity per unit volume in the youngest galls, a result that was similar to the findings reported here.
When the data in this study were expressed on a weight/stem length basis, however, the galls were shown to contain higher amounts of IAA than the corresponding stem controls, indicating that the presence of the larva of E. solidaginis causes an increase in the level of IAA in the goldenrod stem segment that it inhabits. Additionally, the results of the apex removal experiments implicated ball galls as sources of auxin. Gall development was unaffected by the early removal of the apical bud, lateral buds, and leaves (natural sources of auxin) from developing galls. Removal of the apical bud above a developing gall did not result in the degree of lateral bud release that was consistently seen when the apical bud was removed on goldenrod stems lacking galls. In fact lateral bud release was greatly inhibited when galls were in a terminal position in the absence of the apical bud, suggesting that the galls were acting as a source of auxin and were replacing the apical bud in its role as the source of the auxin involved in the inhibition of lateral bud release (Tamas, 1995).
The larvae of E. solidaginis from developing ball galls were shown to contain very high levels of IAA at 9 µg g−1 f. wt or 1.21 ng larva−1, as detected and quantified by GC-MS. These results confirm preliminary findings of high levels of IAA in larvae of E. solidaginis from developing ball galls as detected by spectrofluorimetry following HPLC (Mapes, 1991). The results presented here are similar to those of Kaldeway (1965) who detected auxin bioactivity at a level of 0.5 ng larva−1 in Cynips quercusfolii, a cynipid wasp that forms oak apple galls on Q. robur. While Kaldeway (1965) found auxin bioactivity associated with nongall-forming inhabitants of the oak apple galls, the levels were much lower than those found associated with the gall former. In contrast, Byers et al. (1976) found no auxin activity in an extract of 200 of the gall-inducing larvae of the midge Janetiella sp. near J. colouradensis, despite providing evidence for high levels of auxin bioactivity in the needle galls on pinyon. There have been numerous reports of auxin bioactivity associated with gall-forming aphid species (Mani, 1964); however, no previous studies have provided evidence for the presence of indole-3-acetic acid in gall forming larvae with detection by GC-MS.
The higher concentration of IAA in gall tissues compared with stem tissues when the results were expressed on a tissue length basis, combined with the morphology of the ball gall, is consistent with a high concentration of IAA associated with the larva in the centre of the gall. A localized high concentration of IAA might exist in the tissue zone termed the inner nutritive-pith (Weis et al., 1989), which surrounds the larva. The nutritive pith tissue is presumably replenished by cell division in response to the presence of the larva (Beck, 1947). A low concentration of IAA in the three outer concentric zones surrounding the nutritive pith tissue would cause whole gall extracts to show relatively low concentrations of IAA. The data obtained in this study of extracts of entire galls would not reflect changes in IAA concentrations in the immediate vicinity of the larva.
Given the well-established role of auxin in vascular tissue differentiation (Aloni, 1995), the vascular tissue pattern of the gall is consistent with a central source of auxin associated with the larva. Vascular tissue radiates out from the nutritive-pith tissue surrounding the larva and connects with the stem vascular column in the periphery of the gall (Weis et al., 1989). Given that vascular strand formation has been shown to be induced along the path of auxin flow (Jacobs, 1952), the vascular tissue pattern in the ball gall may reflect diffusion of IAA from the larva. The lack of an increase in IAA concentrations in whole gall extracts could be reflective of the diffusion of IAA out of the gall. As IAA transport typically occurs at a rate of 5–20 mm h−1 (Rubery, 1987), tissue levels may not be increasing. IAA may not be accumulating in the tissues but diffusing out, causing vascular tissue differentiation along the paths of diffusion, and inhibiting lateral bud release below the galls.
Another explanation for the lack of an increase in free IAA concentrations in whole gall extracts could be rapid metabolism and/or conjugation of the IAA, without a concomitant increase in free IAA levels. Continuous uptake of exogenously applied radioactively labelled IAA by light grown stems of bean and pea has been shown to result in increasing tissue levels of free IAA primarily during the first six h of uptake only, with free IAA representing only 10–15% of radioactivity in the tissues after 24 h of uptake, and evidence that most of the IAA was present as conjugates at 24 h (Davies, 1973). Increased auxin biosynthetic activity as a result of transformation by the Ti plasmid of Agrobacterium tumefaciens has not resulted in an increase in the levels of free IAA (Wyndaele et al., 1985; Ishikawa et al., 1988). Similar results were reported by Sitbon et al. (1991) who found that transgenic tobacco plants which contained the two IAA biosynthesis genes of the Ti plasmid had only slightly higher concentrations of free IAA, but contained significantly higher levels of IAA conjugates compared with wild type plants. These studies provide evidence that the free level of IAA may not increase despite greater biosynthetic activities in the tissue.
We have demonstrated that the larva of E. solidaginis contains high levels of IAA and that the presence of an E. solidaginis larva causes an increase in the total amount of IAA present in the stem tissue that it inhabits. We have also demonstrated that ball galls act as sources of auxin. Given that the vascular tissue pattern of the gall is consistent with a centralized production of IAA, our results suggest that the larva of E. solidaginis acts as a source of IAA in developing ball galls on S. altissima.