AOBPreview originally published online on May 21, 2004
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Annals of Botany 94: 179-186, 2004
© 2004 Annals of Botany Company
Recovery of Leaf Area through Accelerated Shoot Ontogeny in Thrips-damaged Cotton Seedlings
1 Cotton Research Unit, CSIRO Plant Industry, Locked Bag 59, Narrabri, NSW 2390, Australia
* For correspondence. Department of Environmental Solution Technology, Faculty of Science and Technology, Ryukoku University, 1-5 Yokoba, Seta-Oe, Otsu 520-2194, Japan. E-mail tomlei{at}rins.ryukoku.ac.jp
Received: 13 October 2003; Returned for revision: 4 December 2003; Accepted: 24 March 2004, Published electronically: 21 May 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSLITERATURE CITED |
|---|
• Background and Aims Leaf area of cotton seedlings (Gossypium hirsutum) can be reduced by as much as 50 % by early season thrips infestations, but it is well documented that plants can regain the difference in leaf area once infestation ceases. The processes involved in the recovery have not been identified. Hypotheses include enhancement of the photosynthetic rate of the damaged leaves, more efficient leaf construction (i.e. more leaf area per unit of dry matter invested in new leaves), and more branching.
• Methods This 2-year field study examined these hypotheses and found that thrips-affected plants recovered from a 30 % reduction in total leaf area. During the recovery period, repeated measurements of gas exchange, leaf morphology and individual leaf areas at all nodes were made to assess their contribution to the recovery.
• Key Results Recovery was not achieved through the previously proposed mechanisms. The pattern of nodal development indicated that the duration of leaf expansion of the smaller deformed leaves was shorter than that of control leaves, possibly because they had fewer cells. The production and expansion of healthy upper node leaves in thrips-affected plants could, therefore, begin sooner, about 1–2·5 nodes in advance of control plants. The proposed process of recovery was evident but weaker in the second year where thrips numbers were higher.
• Conclusions It is concluded that thrips-affected plants overcame the leaf area disparity through an accelerated ontogeny of main stem leaves. By completing the expansion of smaller but normally functioning lower node leaves earlier, resources were made available to the unfolding of larger upper node leaves in advance of control plants. The generality of this mode of plant resistance in pest damage remains to be determined.
Key words: Gossypium hirsutum, cotton, leaf ontogeny, thrips damage, defoliation, leaf area recovery.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSLITERATURE CITED |
|---|
Recovery from herbivory, or compensation, is an important process in plant–herbivore interactions in natural systems (Belsky et al., 1993). It is also a key component of integrated pest-management systems (IPM) for crops such as cotton, permitting the reduction of insecticide use (Fitt, 1994). In many cotton-producing regions, thrips are common pests (Hawkins et al., 1966; Quisenberry and Rummel, 1979; Wilson and Bauer, 1993; Atakan et al., 1996), and most crops receive some protection against thrips by seed treatment, insecticide application at planting, or foliar sprays. Once established, populations of thrips normally build up rapidly early in the growing season (Watts, 1937) leading to infestations that cause visually dramatic deformation of seedling leaves (Quisenberry and Rummel, 1979). The characteristically crinkled leaves are significantly smaller than normal leaves, and lower canopy leaf areas continue until thrips numbers drop, generally within 3–4 weeks of the initial increase (Sadras and Wilson, 1998). Affected plants then resume the production of normal leaves but development is accelerated so that plant leaf area equals that of unaffected plants within weeks (Hawkins et al., 1966; Sadras and Wilson, 1998). As is common in other crop species, cotton plants can often recovery fully from this type of early season defoliation without any economic consequences (Harp and Turner, 1976; Hay and Walker, 1989; Sadras and Wilson, 1998).
It is known that cotton seedlings can recover fully after losing as much as 70 % of total leaf area (Wilson et al., 2003), but the mechanism by which this is achieved has not been identified. Sadras and Wilson (1998) proposed four possible mechanisms for leaf area recovery: (1) increased photosynthetic capacity; (2) increased leaf area to mass ratio; (3) improved branching (Watts, 1937); and (4) the production of additional leaves. The first three mechanisms involve physiological or morphological adjustments, whereas the last requires an acceleration of shoot development. As there has been no evaluation of these potential mechanisms of recovery, the present study was carried out to examine the processes by which cotton plants could recover from reduced leaf area. Beyond quantifying the response of cotton to thrips, the findings from this study may have broader implications in understanding the physiological basis of aspects of plant tolerance to herbivory (Belsky et al., 1993).
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSLITERATURE CITED |
|---|
This study was conducted over two seasons (2001–2002 and 2002–2003) at the Australian Cotton Research Institute in Narrabri, NSW Australia (30·4°S, 149·8°E). Cotton (Gossypium hirsutum L. ‘Siokra V-16i’, transgenic cotton containing the Monsanto Cry 1Ac gene) was sown on 17 Oct. 2001 (year 1) and on 30 Sept. 2002 (year 2) at 10–12 plants per m2. Two treatments were imposed: a control where the systemic insecticide aldicarb (Temik, Aventis) was applied at sowing at 450 g ai ha–1 to protect seedlings against thrips species, and a thrips-damaged treatment where no aldicarb was used, allowing thrips to establish during early season growth. Plots (10 m x 4 rows at 1 m between rows) were laid out in a randomized block design with four replicates.
Thrips populations were monitored by weekly sampling of control and treatment plants from each plot. On each occasion, five plants were collected randomly from each treatment plot, placed in a ziplock bag and immediately brought back to the laboratory. The plants were thoroughly washed by adding water to the bags and agitating them vigorously for 1 min. The water containing dislodged thrips was poured through a fine mesh sieve. This process was repeated. The thrips retained in the sieve were then flushed on to a filter paper, which was placed in a Petri dish and stored frozen until they were counted using a stereomicroscope. Thrips populations were expressed as total numbers per plant in year 1, combining adults and larvae, but in year 2 larvae were separated from adults. The species identified were Thrips tabaci Lindeman and Frankliniella schultzei (Trybom). Wilson and Bauer (1993) also observed the predominance of T. tabaci in cotton at this time of the season at this location. The mean area of cotyledons and leaves at each node of the five plants was measured using a leaf area meter (LiCor, Nebraska, USA). Leaves and cotyledons were then placed in separate bags for each node and dried to constant mass at 70 °C, and the mean leaf mass to area ratio (LMA) was determined. Since leaf areas at individual nodes were recorded, it was possible to detect the change in nodal and whole plant leaf area during the phases of damage and recovery.
Gas exchange of individual leaves was measured in the field four times each year at 27, 35, 44 and 56 d after sowing (year 1) and at 31, 38, 46 and 56 d after sowing (year 2), using a portable photosynthesis system (Li-Cor 6400). The same parameter settings were used at all measurement dates: photon flux density 2000 µmol m–2 s–1 (Li-Cor light source); chamber block temperature 30 °C; vapour pressure deficit of the leaf maintained at less than 3 by adjusting the flow through the desiccant; and reference CO2 concentration 370 µmol m–2 s–1 (using a CO2 mixer). All measurements were made between 0830 and 1130 h Eastern Standard Time. The gas exchange of leaves at all nodes (including cotyledons) was measured for six plants of each treatment (3 plants x 2 replicates in year 1 and 2 plants x 3 replicates in year 2). If the leaf area was <6 cm2 (the leaf chamber opening), then the leaf was labelled and its area later determined using a leaf area meter. Gas exchange parameters were then recalculated using the correct leaf area. Given the small size of the cotton seedlings (<20 cm in length) and the spacing between seedlings (10 cm along row, 1 m between rows) during the measurement period, all leaves on both control and thrips infested plants were exposed to full sunlight for most of each day.
Plots were supplied with furrow irrigation as required but no additional pesticides were applied throughout the season. At the end of the season, all bolls were harvested from four randomly selected 1 m length rows of plants in each plot. Total boll numbers and seed cotton mass per metre were recorded.
As no block effect was detected using ANOVA (PROC GLM; SAS Institute, 1988) for any of the measured parameters, differences between control and thrips treatments were compared using a t-test (SAS Institute, 1988).
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSLITERATURE CITED |
|---|
Thrips-infested plants produced main stem leaves that were deformed (commonly circular in shape) and cupped (Fig. 1). Since the infestation occurred after the expansion of the cotyledons, and the thrips were concentrated in the shoot meristem, the cotyledons remained largely unaffected, as noted by Sadras and Wilson (1998). Thrips began to infest the seedlings shortly after germination and, in both years, thrips numbers in plots unprotected by aldicarb at sowing began to exceed those of controlled plots within 3 weeks after sowing. In year 1, by 33 d after sowing (DAS), the populations were 10·6 per plant, double those of control plants (P < 0·05, Fig. 2). In year 2, numbers of larvae were significantly higher on unprotected plants from 22 to 52 DAS, but numbers of adults did not differ between treatments. The reason for the peak in adult numbers at 38 DAS, observed in both control and unprotected plots, is almost certainly due to repeated immigration of adults. In a similar experiment, Sadras and Wilson (1998) found that immigrating adults fed enough to be killed by the Temik in protected plants but did little damage. On the basis of observations of control and unprotected plants during this period (Fig. 2), it was concluded that the influx of adults did not play a significant role in damaging leaves. Total numbers of thrips in year 2 were about twice the numbers in year 1; they declined naturally to a similar level in all treatment plots in each year but more quickly in year 1. The duration of infestation was about 20 d in year 1 and 40 d in year 2.
|
|
The reduction in leaf area caused by thrips began at about 3 weeks after sowing (Fig. 3), in close correspondence with the increase in thrips numbers (Fig. 2). Summing all leaves (including the cotyledons), the total reduction in leaf area of thrips-affected plants reached a maximum of 30 % relative to the control between 40 and 60 DAS in each year (Fig. 3). The more rapid reduction in leaf area in year 2 is attributed to the higher populations of thrips early in the infestation (Fig. 2). Sadras and Wilson (1998) found a comparable degree of leaf area reduction associated with peak populations of 10–30 thrips (larvae + adults) per plant. The total duration of leaf area reduction observed here, about 50 d (20–70 DAS), is similar to the 40 d (generally between 20 and 60 DAS) reported by Sadras and Wilson (1998). The recovery in leaf area per plant began at 50–60 DAS, reaching 89 and 83 % of the control by 68 and 67 DAS for the two years, respectively. The subsequent decline (after 70 DAS) in leaf area in year 2 was largely associated with the greater recruitment of non-main stem leaves located on vegetative and fruiting branches in control plants (Fig. 4B). Although Sadras and Wilson (1998) found a full recovery in leaf area after 60–80 d in most investigations, some showed a dip in leaf area, similar to that found here, about 80 d after sowing.
|
|
Study of the progression of leaf area development at each node (Fig. 4) shows that a significant reduction in the area of infested plants had begun by 23–24 DAS in each year, and it continued through to node 8 (68 DAS, year 1) and node 9 (67 DAS, year 2). Leaf area recovery began by 45 and 50 DAS, with the area of leaves of upper nodes beginning to exceed that of corresponding leaves in control plants. No vegetative branch growth was detected in either treatment during the measurement period (up to 68 DAS) in year 1. Some vegetative branch growth was observed in year 2 in both control and infested plants with additional leaf area on these branches contributing, respectively, 9 and 14 % to total leaf area on 59 DAS and 24 and 29 % to total leaf area on 74 DAS (insets of Fig. 4B). At the latter date, leaves from fruiting branches were also included. The additional leaf area contributed by non-main stem leaves was not significantly different (P > 0·05) between treatments at all sampling dates and therefore did not play a role in the recovery process.
One of the suggested mechanisms facilitating leaf area recovery (Sadras and Wilson, 1998) is an increased photosynthetic rate of damaged leaves: higher rates of assimilation might provide the necessary resource for the acceleration in leaf area expansion. The rate of net photosynthesis (A) of individual leaves (including cotyledons), measured at four dates spanning the infestation and recovery phases, revealed no significant enhancement in thrips-affected leaves (Table 1). In fact, there were several occasions where A was higher in the control than in damaged plants in year 2. Although A varied among nodes in accordance with the stage of leaf maturity, the pattern was the same for control and unprotected plants. Given that main stem leaves were smaller in unprotected plants, the similarity in A translates to a reduction in total carbon gain per plant comparable with that of leaf area (i.e. by up to 30 % between 40 and 60 DAS). It can, therefore, be concluded that compensatory photosynthesis is unlikely to contribute to recovery. It is, however, possible that damaged plants could have allocated their smaller pools of carbon preferentially to above-ground growth, reducing the supply to the root system. Under the conditions of the present experiments, soil moisture was maintained at non-stress levels, reduced root growth might have had no consequences for plant growth. This assumption requires further investigation.
|
Increased branching has also been suggested as a mechanism of recovery from thrips damage (Watts, 1937). Thrips damage to the apical meristem of cotton is known to promote vegetative branch growth (Sadras and Fitt, 1997; Jones and Wells, 1998; Lei and Gaff, 2003), but this occurs only at very high levels of infestation (L. J. Wilson, unpublished data). The type of damage found in the present study caused only deformation of the small unfolding leaves surrounding the meristem. A survey in year 1 found very low levels of tip damage in control (2·6 %) and thrips-affected plants (4·0 %). It can, therefore, be concluded that branching did not play a role in leaf area recovery; the degree of thrips damage resulted in leaf damage only and was insufficient to initiate lateral branch growth.
Wilson et al. (2003) have shown that it is possible for cotton crops to recover from up to 70 % loss of leaf area without affecting the yield of lint. This indicates that the supply of carbohydrate required to maintain normal shoot development during the recovery phase is relatively small; the 30 % reduction in leaf area found in this study may not have been sufficient to affect this requirement. Within 1–2 d after leaf unfolding, immature cotton leaves have already reached a positive carbon balance (i.e. carbon assimilation exceeds dark respiration; Constable and Rawson, 1980a, b). This means that the rate of leaf development in control plants was not carbon limited, but to achieve maximum A (at 75–90 % of final leaf area; Constable and Rawson, 1980a), an expanding leaf will require continued nitrogen imports for the maturation of chloroplasts. It is possible that the more rapid cessation in nutrient demand of damaged leaves on damaged plants contributed to the accelerated development of new leaves.
Decreased leaf mass to area ratio (LMA) has also been suggested as a mechanism of recovery. With a lower LMA, the damaged plant would need to invest less dry matter for a given leaf area. Table 2 shows little variation in LMA in corresponding upper node leaves between treatments; in fact, some damaged leaves (lower nodes) had significantly higher LMA than the control, particularly in year 2.
|
The remaining hypothesis is that leaf area recovery is achieved simply through a more rapid ontogeny. Although Sadras and Wilson (1998) suggested that the addition of new leaves can ‘partially compensate’ for the loss of leaf area, the present work suggests a mechanism by which this occurs. Initially, thrips feed in the apical bud region by piercing leaf primordia repeatedly before leaf unfolding. It is likely that clusters of laminal cells are destroyed in the process, leading to a reduction in cell number per leaf. The patchy destruction of cells throughout the lamina of the leaf primordia would also explain the cupped appearance of affected leaves (Fig. 1). Furthermore, deformed leaves have normal petiole development, suggesting that the turgor pressure required for normal cell expansion was not disrupted and that the remaining laminal cells are of normal size.
If thrips-affected leaves had fewer cells, then the duration of their expansion would be shorter, thus making resources available sooner for the production of upper leaves. Evidence for this is shown in Fig. 4A where by 50 DAS in year 1, all damaged leaves up to node 6 had reached full expansion (open arrow) while nodes 4–6 of control plants (solid arrow) were still expanding. By 58 DAS, thrips-affected node 7 had expanded fully compared with node 4 in control plants. A similar pattern was seen in year 2 where, on 67 DAS, node 6 and 7 were the uppermost fully expanded leaves for thrips-affected and control plants, respectively.
Since the leaf area of a plant is determined by the rate of leaf production/unfolding and the rate and duration of leaf expansion (Hay and Walker, 1989), results of this study indicate that the leaf area of infested plants recovered by both a shorter duration of the expansion of lower node leaves and a faster rate of unfolding of larger upper node leaves. The more rapid unfolding of upper node leaves is illustrated by the 2·5 node gain in infested plants over control plants (Fig. 4A, 68 DAS). The size of cotton leaves, as in other species (Hay and Walker, 1989), increases with each successive node, reaching a maximum at node 8–10 (Constable and Rawson, 1980a; Fig. 4). Thus, the recovery in leaf area of infested plants was mainly achieved through the earlier unfolding of larger leaves of the upper, unaffected nodes.
There was no residual effect of the early season damage on lint yield. At harvest, when all plots had reached >80 % open bolls, there was no significant difference (P > 0·05) between control and unprotected plants in total number of bolls and in seed cotton mass in each year (Fig. 5). Similar results were reported by Rummel and Quisenberry (1979) where a reduction in leaf area of up to 19 % had no effect on yield, but loss of 51 % of leaf area resulted in a yield loss of 9 %. In the same cropping area as this study, Sadras and Wilson (1998) reported a significant yield loss in only one of ten thrips-infested trials. The affected crop showed a relapse in leaf area recovery 60 d after sowing (i.e. at the beginning of fruit production), a pattern similar to that observed in year 2 of this study (Fig. 3). It is possible that higher infestations could delay the recovery of leaf area until after the start of fruit production. If the ontogenetic recovery in leaf area is incomplete at the initiation of fruiting branches, lower leaf area could be maintained, resulting in smaller plants with a lower assimilation capacity and reduced yields.
|
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSLITERATURE CITED |
|---|
Within a very short time period (<3 weeks), thrips feeding on leaf primordia led to a significant loss of leaf area at main stem nodes 1–8. Since the residency time of thrips was short, the affected cotton plants were able to recover completely. The transient loss and subsequent recovery of leaf area resulting from thrips feeding, at the densities experienced, was not achieved through physiological responses such as an improved carbon assimilation or increased allocation of biomass to leaf construction, but through changes in the pattern of development. This involved the earlier completion of expansion of lower, thrips-affected, main stem leaves which shortened the period during which these leaves competed for resources (Constable and Rawson, 1980b), making resources available for the earlier unfolding of upper, undamaged leaves. Since the upper node leaves are successively larger ontogenetically, their earlier unfolding accounted for the recovery in leaf area of thrips-affected plants.
|
ACKNOWLEDGEMENTS |
|---|
The technical assistance of Kellie Baguley, Deon Cameron, Simone Heimoana and Dee Hamilton is greatly appreciated. Our thanks to the farm staff at the Australian Cotton Research Institute for their help in establishing and maintaining the experimental fields. This study was supported in part by an Australian Cotton Research and Development Coorporation (CRDC) grants CSP124C and CSP103C.
|
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSLITERATURE CITED |
|---|
-
Atakan E, Coll M, Rosen D. 1996. Within-plant distribution of thrips and their predators: effects of cotton variety and developmental stage. Bulletin of Entomologic Research 86: 641–646.
Belsky AJ, Carson WP, Jensen CL, Fox GA. 1993. Overcompensation by plants: herbivory optimisation or red herring? Evolutionary Ecology 7: 109–121.
Constable GA, Rawson HM. 1980a. Effect of leaf position, expansion and age on photosynthesis, transpiration and water use efficiency of cotton. Australian Journal of Plant Physiology 7: 89–100.[ISI]
Constable GA, Rawson HM. 1980b. Carbon production and utilization in cotton: inferences from a carbon budget. Australian Journal of Plant Physiology 7: 539–553.
Fitt GP. 1994. Cotton pest management. 3. An Australian perspective. Annual Review of Entomology 39: 543–562.[CrossRef]
Harp SJ, Turner VV. 1976. Effects of thrips on cotton development in the Texas blacklands. Southwestern Entomologist 1: 40–45.
Hawkins BS, Peacock HA, Steele TE. 1966. Thrips injury to upland cotton (Gossypium hirsutum L.) varieties. Crop Science 6: 256–258.
Hay RKM, Walker AJ. 1989. An introduction to the physiology of crop yield. Harlow, Essex: Longman Scientific & Technical.
Jones MA, Wells R. 1998. Fiber yield and quality of cotton grown at two divergent population densities. Crop Science 38: 1190–1195.
Lei TT, Gaff N. 2003. Recovery from terminal and fruit damage by dry season cotton crops in tropical Australia. Journal of Economic Entomology 96: 730–736.[Medline]
Quisenberry JE, Rummel DR. 1979. Natural resistance to thrips injury in cotton as measured by differential leaf area reduction. Crop Science 19: 879–881.
Rummel DR, Quisenberry JE. 1979. Influence of thrips injury on leaf development and yield of various cotton genotypes. Journal of Economic Entomology 72: 706–709.
SAS Institute. 1988. SAS/STAT User’s Guide, Release 6.03 edition. Cary, NC: SAS Institute.
Sadras VO, Fitt GP. 1997. Apical dominance – variability among cotton genotypes and its association with resistance to insect herbivory. Environmental and Experimental Botany 38: 145–153.[CrossRef]
Sadras VO, Wilson LJ. 1998. Recovery of cotton crops after early season damage by thrips (Thysanoptera). Crop Science 38: 399–409.
Watts JG. 1937. Reduction of cotton yield by thrips. Journal of Economic Entomology 30: 860–863.
Wilson LJ, Bauer LR. 1993. Species composition and seasonal abundance of thrips (Thysanoptera) on cotton in the Namoi Valley. Journal of the Australian Entomological Society 32: 187–192.
Wilson LJ, Sadras VO, Heimoana SC, Gibb D. 2003. How to succeed by doing nothing: cotton compensation after simulated early-season pest damage. Crop Science 43: 2125–2134.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  A. A. Reddall, L. J. Wilson, P. C. Gregg, and V. O. Sadras Photosynthetic Response of Cotton to Spider Mite Damage: Interaction with Light and Compensatory Mechanisms Crop Sci., September 1, 2007; 47(5): 2047 - 2057. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. L. Grey, G. D. Buntin, P. M. Roberts, and D. C. Bridges Potential Interaction of Pendimethalin and Systemic Insecticides for Thrips Control in Cotton Agron. J., January 5, 2006; 98(1): 141 - 147. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||