**4.2 Positive or neutral effects of insect herbivory on photosynthesis**

Resistance and tolerance represent two general strategies of plant defense against herbivores, although interactions between these two strategies are assumed to occur under certain conditions, i.e. when the resources available for defense are limited or when both defensive strategies are physiologically costly (Leimu & Koricheva, 2006). Resistance involves the reduction of the amount of herbivore damage whereas tolerance leads to a reduction of the impact of herbivory on plant fitness (Rausher et al., 1993; Stowe et al., 2000). Resistance traits include mechanical and chemical characters that reduce herbivore performance (antibiosis) or preference (antixenosis). Conversely, proposed mechanisms for tolerance/compensation are re-growth stimulation, elevated rates of photosynthesis in remaining leaves of partially defoliated plants, increased branching through the release of apical dominance, alteration of phenology or plant architecture, production of new leaf area, utilization of high pre-herbivory stored carbon resources or the ability to reallocate them to less vulnerable tissues, resorption of nutrients from senescent/damaged leaves, especially nitrogen (N) and phosphorus (P), alteration of the external light environment and higher reproductive efficiency through increased percentage of fruit set (Mabry & Wayne, 1997; Hjalten et al., 1993; Strauss & Agrawal, 1999; Hochwender et al., 2000; Tiffin, 2000; Anten et al., 2003; Silla & Escudero, 2003; Leimu & Koricheva, 2006; Schwachtje et al., 2006). Compensatory ability in plants varies widely across species, and the degree in which it is manifested depends on the amount of leaf lost, with complete rather than fragmented defoliation usually being more conductive to an increased rate of net photosynthesis in the remaining or newly formed leaves (Welter, 1989). The mode of herbivore damage and herbivore type may also determine whether the overall effect on photosynthesis in the plant. This was elegantly evidenced in a recent report showing that herbivory on *N. attenuata* by *Tupiocoris notatus*, a cell-content feeder, (or by application on wounded plants of its salivary secretions), induced an elevated photosynthetic activity, and consequent CO2 assimilation, that appeared to compensate for lost tissue and for the fitness costs associated with the deployment of direct and indirect defenses. This compensatory effect was shown to be specific for this insect, since feeding by chewing *Manduca sexta* larvae resulted in a strong down-regulation of photosynthesis (Halitschke et al., 2011).

Environmental conditions and the timing of the herbivory event are also influential factors. Thus, compensation to damage in terms of timing of herbivory is usually more effective when required early in the growing season or before the reproductive phase has started. For example, a study performed in Lebanese cucumber (*Cucumis sativus*) to compare the ability to compensate for foliar herbivory at both the pre-flowering and flowering stages found that damage produced before flowering allowed plants to compensate more efficiently, in terms of vegetative biomass and fruit production, for leaf losses that sometimes reached 80% of the total leaf area in the plant groups examined. Higher compensation was correlated with a

Friend or Foe? Exploring the Factors that Determine the Difference

(Newingham et al., 2007).

on leaves (Kaiser et al., 2010).

Between Positive and Negative Effects on Photosynthesis in Response to Insect Herbivory 175

mechanism for insect herbivory. This welcome contribution to the rather unexplored area of plant root-insect interactions was designed to understand the high tolerance to root herbivory by bio-control agents shown by *Centaurea maculosa*, an invasive North American plant species. The use of 15N labeling indicated that infested plants were able to sustain growth and maintain a constant shoot N status under potentially devastating conditions characterized by a drastic reduction of whole plant and root N uptake as a result of herbivory, by shifting N allocation to the shoot, away from the reach of root herbivores

Compensatory photosynthesis is also deemed to play an important role in plants that utilize carbon-based defense strategies, by increasing the availability of carbohydrates that can potentially be allocated to defense. A recent report focused on the possible effects that diverse tritrophic interactions, involving browsing herbivores and several species of resident ants, could have on foliar photosynthetic rates, measured as net photosynthesis (Pn), transpiration and water use efficiency (WUE), and concomitant availability of carbon pools for metabolism and defense in *Acacia drepanolobium*. This species is an east-African, savannah-resident tree, that is known to exhibit carbon-based investments in direct defense (e.g. erection of physical barriers and accumulation of toxic chemicals), indirect defense (e.g. housing and feeding of beneficial ants that guard the plant from herbivores) and tolerance (e.g. stimulated rates of leaf and shoot growth) (King & Caylor, 2010). Their results, which represent the first evidence that indirect defenders of plants can also benefit plants by increasing their photosynthetic rates, indicated first, that *A. drepanolobium* trees exhibited elevated photosynthetic rates in response to browsing only when occupied by strongly mutualistic ants, and second, that this photosynthetic up-regulation mitigated the costs of herbivory by increasing pools of

photosynthate available for additional defense or for re-growth of lost tissue.

A unique example of positive manipulation of plant photosynthesis by insect herbivores is represented by the so-called green-island phenotype induced by leaf-miners in deciduous leaves in the autumn season, and persisting long after leaf abscission. These green-islands are characterized by photosynthetically active green patches in otherwise senescing leaves, and correspond to regions with an increased concentration in cytokinins, which are hormones involved in a variety of biological processes, many pertinent to the phenomenon in question, such as the inhibition of senescence, maintenance of chlorophyll and control of source-sink relationships for nutrient mobilization, and maintenance of enriched nutritional environments (Gan & Amasino, 1995; Balibrea Lara et al., 2004; Walters & McRoberts, 2008; Giron et al., 2007). The concentrated levels of nutrients that characterize green-islands in senescent leaves favor growth and reproduction of the leaf miners with only a limited consumption of leaf tissues. This, in turn, allows areas of uneaten tissue to be employed for thermal regulation and parasitoid avoidance (Djemaï et al., 2000; Giron et al., 2007). However, the origin of cytokinins in leaf-miner systems has not yet been determined, although several lines of evidence initially suggested that cytokinins were derived from the insect. Such a concept was questioned by recent findings suggesting that cytokinins might originate from bacterial endosymbionts known establish an intimate association with leaf mining insects. This was evidenced by the negative effects on insect fitness derived from curing leaf-miners of their symbiotic partner, which also abolished green-island formation

A number of selected examples in which insect herbivory has been shown to have a positive influence on photosynthesis, including many already described above, are shown in Table1.

higher photosynthetic efficiency and capacity, and with less dissipation of light energy as heat, leading to the proposal that herbivore-damaged plants may be induced to use a greater proportion of the absorbed light energy for photosynthesis as a result of altered carbohydrate source-sink relationships (Thomson et al., 2003). In contrast, an experimental setting designed to test the effects of partial de-budding on photosynthesis, stomatal conductance and nitrogen in *Picea jezoensis* seedlings led to the conclusion that the enhanced photosynthetic rate observed in de-budded seedlings was the result of an increased root/leaf ratio that reduced the stomatal limitation of photosynthetic rate, rather than of an altered sink-source relationship or increased leaf nitrogen content (Ozaki et al., 2004).

A pair of studies aimed at determining the carbon costs of herbivory by phloem-feeding scale insects on tress found that infested trees had a greater annual photosynthesis, as determined by measuring parameters such as *Vc* max, the maximum rate of Rubisco-catalysed carboxylation, *J* max, the rate of electron transport when irradiance is saturating and/or chlorophyll fluorescence (Retuerto et al., 2004; Dungan et al., 2007). The small negative effect on tree growth and reproduction and increased photosynthetic efficiency observed were taken as an indication that damaged trees were able to compensate fully for the relatively large loss of carbon to herbivory caused by the honeydew insects. According to these workers, the amelioration of carbon loss resulting from the additional sinks for photosynthates created by scale insect feeding was achieved by increased photosynthetic rates. These results were in agreement with previous data suggesting that defoliation, as well as removal of reproductive and other vegetative sinks, may improve photosynthesis in remaining leaf tissue by increasing carboxylation efficiency and the rate of Rubisco regeneration (Holman & Oosterhuis, 1999; Thomson et al., 2003; Ozaki et al., 2004; Turnbull et al., 2007). However, they were in contradiction with data generated from the meta-analysis of a collection of reports showing that sap feeding insects have an almost universal negative effect on growth, photosynthesis, and reproduction of woody plants (Zvereva et al., 2010). The discrepancy detected was adjudicated to experimental biases introduced by the utilization of improper controls (e.g selective assignation as controls to undamaged plant sections that were avoided by herbivores or herbivore preference for hosts with higher rates of photosynthesis). Other important findings of the above meta-analysis were the following: i) sap-feeders did not change the resource allocation in plants; ii) mesophyll and phloem feeders produced stronger effects than xylem feeders, whereas generalist sap-feeders reduced plant performance to a greater extent than did specialists; iii) methodology (e.g. greenhouse vs. field settings; natural vs. imposed herbivory and short-term vs. long-term feeding) was a significant factor influencing the outcome of the experiments, and iv) sap feeding was more detrimental at higher temperatures. Thus, sap-feeders were considered to exert a more severe overall negative impact on woody plant performance than defoliators, mostly due to the latter's lower ability to compensate for sap-feeders' damage in terms of both growth and photosynthesis.

Another study in which the effect of high and low soil nutrient levels on biomass re-growth and photosynthetic up-regulation, among genotypes of the Mediterranean annual grass *Avena barbata* subjected to simulated herbivory*,* obtained rather unexpected results. They showed that tolerance in this species was positively correlated only with pre-defoliation photosynthetic efficiency at high nutrients, since no evidence for photosynthetic upregulation in defoliated compared to control plants was observed regardless of nutrient treatment (Suwa & Maherali, 2008). In this context, a rather infrequent report describing compensatory responses to herbivory to the root system suggested a novel tolerance

higher photosynthetic efficiency and capacity, and with less dissipation of light energy as heat, leading to the proposal that herbivore-damaged plants may be induced to use a greater proportion of the absorbed light energy for photosynthesis as a result of altered carbohydrate source-sink relationships (Thomson et al., 2003). In contrast, an experimental setting designed to test the effects of partial de-budding on photosynthesis, stomatal conductance and nitrogen in *Picea jezoensis* seedlings led to the conclusion that the enhanced photosynthetic rate observed in de-budded seedlings was the result of an increased root/leaf ratio that reduced the stomatal limitation of photosynthetic rate, rather than of an altered sink-source relationship or increased leaf nitrogen content (Ozaki et al., 2004). A pair of studies aimed at determining the carbon costs of herbivory by phloem-feeding scale insects on tress found that infested trees had a greater annual photosynthesis, as determined by measuring parameters such as *Vc* max, the maximum rate of Rubisco-catalysed carboxylation, *J* max, the rate of electron transport when irradiance is saturating and/or chlorophyll fluorescence (Retuerto et al., 2004; Dungan et al., 2007). The small negative effect on tree growth and reproduction and increased photosynthetic efficiency observed were taken as an indication that damaged trees were able to compensate fully for the relatively large loss of carbon to herbivory caused by the honeydew insects. According to these workers, the amelioration of carbon loss resulting from the additional sinks for photosynthates created by scale insect feeding was achieved by increased photosynthetic rates. These results were in agreement with previous data suggesting that defoliation, as well as removal of reproductive and other vegetative sinks, may improve photosynthesis in remaining leaf tissue by increasing carboxylation efficiency and the rate of Rubisco regeneration (Holman & Oosterhuis, 1999; Thomson et al., 2003; Ozaki et al., 2004; Turnbull et al., 2007). However, they were in contradiction with data generated from the meta-analysis of a collection of reports showing that sap feeding insects have an almost universal negative effect on growth, photosynthesis, and reproduction of woody plants (Zvereva et al., 2010). The discrepancy detected was adjudicated to experimental biases introduced by the utilization of improper controls (e.g selective assignation as controls to undamaged plant sections that were avoided by herbivores or herbivore preference for hosts with higher rates of photosynthesis). Other important findings of the above meta-analysis were the following: i) sap-feeders did not change the resource allocation in plants; ii) mesophyll and phloem feeders produced stronger effects than xylem feeders, whereas generalist sap-feeders reduced plant performance to a greater extent than did specialists; iii) methodology (e.g. greenhouse vs. field settings; natural vs. imposed herbivory and short-term vs. long-term feeding) was a significant factor influencing the outcome of the experiments, and iv) sap feeding was more detrimental at higher temperatures. Thus, sap-feeders were considered to exert a more severe overall negative impact on woody plant performance than defoliators, mostly due to the latter's lower ability to compensate for

sap-feeders' damage in terms of both growth and photosynthesis.

Another study in which the effect of high and low soil nutrient levels on biomass re-growth and photosynthetic up-regulation, among genotypes of the Mediterranean annual grass *Avena barbata* subjected to simulated herbivory*,* obtained rather unexpected results. They showed that tolerance in this species was positively correlated only with pre-defoliation photosynthetic efficiency at high nutrients, since no evidence for photosynthetic upregulation in defoliated compared to control plants was observed regardless of nutrient treatment (Suwa & Maherali, 2008). In this context, a rather infrequent report describing compensatory responses to herbivory to the root system suggested a novel tolerance mechanism for insect herbivory. This welcome contribution to the rather unexplored area of plant root-insect interactions was designed to understand the high tolerance to root herbivory by bio-control agents shown by *Centaurea maculosa*, an invasive North American plant species. The use of 15N labeling indicated that infested plants were able to sustain growth and maintain a constant shoot N status under potentially devastating conditions characterized by a drastic reduction of whole plant and root N uptake as a result of herbivory, by shifting N allocation to the shoot, away from the reach of root herbivores (Newingham et al., 2007).

Compensatory photosynthesis is also deemed to play an important role in plants that utilize carbon-based defense strategies, by increasing the availability of carbohydrates that can potentially be allocated to defense. A recent report focused on the possible effects that diverse tritrophic interactions, involving browsing herbivores and several species of resident ants, could have on foliar photosynthetic rates, measured as net photosynthesis (Pn), transpiration and water use efficiency (WUE), and concomitant availability of carbon pools for metabolism and defense in *Acacia drepanolobium*. This species is an east-African, savannah-resident tree, that is known to exhibit carbon-based investments in direct defense (e.g. erection of physical barriers and accumulation of toxic chemicals), indirect defense (e.g. housing and feeding of beneficial ants that guard the plant from herbivores) and tolerance (e.g. stimulated rates of leaf and shoot growth) (King & Caylor, 2010). Their results, which represent the first evidence that indirect defenders of plants can also benefit plants by increasing their photosynthetic rates, indicated first, that *A. drepanolobium* trees exhibited elevated photosynthetic rates in response to browsing only when occupied by strongly mutualistic ants, and second, that this photosynthetic up-regulation mitigated the costs of herbivory by increasing pools of photosynthate available for additional defense or for re-growth of lost tissue.

A unique example of positive manipulation of plant photosynthesis by insect herbivores is represented by the so-called green-island phenotype induced by leaf-miners in deciduous leaves in the autumn season, and persisting long after leaf abscission. These green-islands are characterized by photosynthetically active green patches in otherwise senescing leaves, and correspond to regions with an increased concentration in cytokinins, which are hormones involved in a variety of biological processes, many pertinent to the phenomenon in question, such as the inhibition of senescence, maintenance of chlorophyll and control of source-sink relationships for nutrient mobilization, and maintenance of enriched nutritional environments (Gan & Amasino, 1995; Balibrea Lara et al., 2004; Walters & McRoberts, 2008; Giron et al., 2007). The concentrated levels of nutrients that characterize green-islands in senescent leaves favor growth and reproduction of the leaf miners with only a limited consumption of leaf tissues. This, in turn, allows areas of uneaten tissue to be employed for thermal regulation and parasitoid avoidance (Djemaï et al., 2000; Giron et al., 2007). However, the origin of cytokinins in leaf-miner systems has not yet been determined, although several lines of evidence initially suggested that cytokinins were derived from the insect. Such a concept was questioned by recent findings suggesting that cytokinins might originate from bacterial endosymbionts known establish an intimate association with leaf mining insects. This was evidenced by the negative effects on insect fitness derived from curing leaf-miners of their symbiotic partner, which also abolished green-island formation on leaves (Kaiser et al., 2010).

A number of selected examples in which insect herbivory has been shown to have a positive influence on photosynthesis, including many already described above, are shown in Table1.

Friend or Foe? Exploring the Factors that Determine the Difference

Damage type/ feeding guild

Foliagechewing feeder

Phloem feeder

Foliagechewing feeder

Plant species Herbivore

Soybean (*Glycine max*)

Beech trees *Nothofagus solandri* 

Wheat (*Triticum aestivum*)

*Acacia drepanolobium* species

Japanese beetles (*Popillia japonica*); corn earworm (*Helicoverpa zea*)

Scale insects (*Ultracoelostoma* 

Armyworm (*Spodoptera frugiperda*)

Resident ants (*Crematogaster mimosa; C. nigriceps; C. sjostedti; Tetraponera penzigi*)

*assimil)*

Between Positive and Negative Effects on Photosynthesis in Response to Insect Herbivory 177

Herbivory increased

transpiration without affecting carbon assimilation rates or photosynthetic efficiency. Reductions in net

photosynthesis and stomatal conductance occurred only when midvein was disrupted.

Infested trees had a greater annual photosynthesis measured as Vc max, J max and chlorophyll content. Consequently, annual canopy photosynthesis was 4% greater for infested trees.

Photosynthesis, intercellular CO2 and transpiration of injured leaves were not significantly affected; however, stomatal conductance values were

Spatial pattern of defoliation differentially affected photosynthesis; leaves defoliated at the basal portion had lower rates.

higher.

Trees exhibited elevated

mutualistic ants (*Crematogaster mimosa;* 

*C. nigriceps*).

of lost tissue.

photosynthetic rates in response to browsing only when occupied by strongly

Photosynthetic up-regulation mitigated the costs of herbivory by increasing pools for additional defense or for re-growth

Results Method Reference

GE, ChlF and thermal imaging (TI)

GE and ChlF

Aldea et al., 2005

GE Dungan

et al., 2007

Macedo et al., 2007

GE King and

Caylor, 2010


Friend or Foe? Exploring the Factors that Determine the Difference Between Positive and Negative Effects on Photosynthesis in Response to Insect Herbivory 177

176 Artificial Photosynthesis

only in birch.

Photosynthetic rates of grazed leaves increased following herbivory in alder; by contrast birch exhibited a decline in net photosynthesis. Differences related to the beetle's feeding behavior that often cut midribs

Photosynthetic rate and leaf nitrogen content were significantly raised by *T. salignus* feeding.

Higher compensation in terms of vegetative biomass and fruit production was correlated with an increase in

photosynthetic efficiency and capacity, and with less dissipation of light energy.

Insect infestation increased photosynthetic efficiency; effect enhanced by high temperature and light. Insects altered the photosynthesis of leaves not directly affected by

Partial de-budding enhanced photosynthetic rates in 1-yearold needles but not in currentyear needles. Greater photosynthetic rate was accompanied by increased stomatal conductance.

Photosynthesis, respiration rates or non-structural carbohydrates on leaves were not affected by short-term aphid feeding. No increase in net photosynthesis during thrips infestation or recovery phases revealed the lack of compensation in affected

the insects.

leaves.

Results Method Reference

Gas exchange (GE)

Oleksyn et al., 1998

et al., 2001

Thompson et al.,2003

et al., 2004

et al., 2004

Wilson 2004; Gomez et al., 2006

GE Collins

ChlF Retuerto

GE Ozaki

GE Lei &

Chlorophyll fluorescence (ChlF)

Damage type/ feeding guild

Foliagechewing feeder

Sap feeder (stemfeeding)

Foliagechewing feeder

Phloem feeder

Bud feeder

Phloem feeder

Plant species Herbivore

*Alder (Alnus incana*, *A. glutinosa*) and birch (*Betula pendula)*

Willow tree (*Salix viminalis)*

Cucumber (*Cucumis sativum*)

European holly

(*Ilex aquifolium*)

Ezo spruce *Picea jezoensis*

Cotton (*Gossypium hirsutum*)

trees

species

Aphids (*Tuberolachnus salignus*; *Pterocomma salicis*)

snail

Brown garden

(*Helix aspersa*)

Scale insects (*Coccus* sp.)

Manual debudding (in nature *Choristoneura jezoensis*; aphids

*Adelges japonicus*)

(*Thrips tabaci* and *Frankliniella schultzei*)

Cotton aphid (*Aphis gossypii*);

thrips

Alder beetle (*Agelastica alni*)


Friend or Foe? Exploring the Factors that Determine the Difference

Macedo, 2001; García & Ehrlén, 2002; Zvereva et al., 2010).

levels (Reich et al., 1999; Mediavilla et al., 2001).

The nitrogen status of the plant is also considered to influence the way photosynthesis is affected by herbivory in plants. This is because of the strong positive correlation that is usually observed between photosynthesis rate and nitrogen concentration in plants, predominantly sequestered in the Rubisco enzyme (Field & Mooney, 1986; Evans, 1989; see above). It is not surprising then, that one of the mechanisms offered to explain why the localized decrease in N content negatively affects photosynthesis, a circumstance that has been frequently reported in damaged leaves of woody plants subjected to insect herbivory, is precisely that N deficiency directly affects CO2 assimilation rates by lowering Rubisco

Another little studied aspect of plant-insect interations is the effect that insect oviposition might have on photosynthesis. Most of the available data suggest, however, that the effect is predominantly negative. A recent study reported that net photosynthetic rate, J max, and Vc max of pine needles laden with eggs of an herbivorous sawfly were lower than in egg-free

Between Positive and Negative Effects on Photosynthesis in Response to Insect Herbivory 179

damage by skeletonizing Mexican bean beetles (*Epilachna varivestis*) caused substantial losses of photosynthesis in the remaining leaf tissue (Peterson et al., 1998). It was hypothesized that the scraping and crushing of interveinal leaf tissue caused by feeding adults and larvae of Mexican bean beetles may have exacerbated localized water stress, ultimately causing tissue desiccation and photosynthesis repression. The timing of damage is also considered to be an important factor. In this regard, early season damage has been usually found to cause more pronounced changes in plants than late season damage, which is in accordance with the assumption that vigorously growing foliage has a greater capacity to respond to various stimuli, including damage. Also, seedlings are generally more susceptible to photosynthetic damage because of a shortage of reserves due to their smaller size or to limitations in nutrient acquisition (Nykänen & Koricheva, 2004; Hódar et al., 2008). A report recording the response of potted fruitless grapevines (*Vitis labrusca* var. Niagara) to early and late season mechanical and insect defoliation was in accordance with this concept by showing that growth, single leaf photosynthesis, and whole-vine photosynthesis were more tolerant to foliar injury late in the season than early in the season (Mercader & Isaacs, 2003). Similar results were obtained from a series of experiments performed to examine a possible trade-off between photosynthesis with defense or reproduction in the common milkweed *Asclepias syriaca*, which is a plant that accumulates toxic cardenolides in a constitutive or inducible manner and is also susceptible to insect damage during its relatively long reproduction period (Delaney et al., 2009). The results of this study showed that leaf Pn impairment after partial leaf defoliation had a seasonal pattern which correlated with *A. syriaca* reproductive phenology but not with cardenolide accumulation. In this regard, the small or absent Pn impairment occurring in leaves of pre-flowering or maturing seed pod plants, contrasted with the moderate to severe leaf Pn impairment detected in leaves of flowering and early seed pod formation plants. Such a behavior led the authors to suggest that a physiological 'cost of reproduction' might be an additional susceptibility factor leading to Pn impairment after herbivory injury on a leaf. Another important aspect to consider is that photosynthesis will be usually more affected when plants are attacked by generalist herbivores, against which they show a higher susceptibility. This is believed to be derived from the lack of a previous and selective co-evolutionary process leading to adaptation (Parker et al., 2006; however, see above). The plant's capacity to tolerate injury, its phenotypic plasticity and the type of environment with which the plant is interacting may be important factors too (Alward & Joern, 1993; Trumble et al., 1993; Delaney &


Table 1. Some examples of positive or neutral effects on photosynthesis after herbivory damage.
