Extracellular Self- and Non-Self DNA Involved in Damage Recognition in the Mistletoe Parasitism of Mesquite Trees

Psittacanthus calyculatus parasitizes mesquite trees through a specialized structure called a haustorium, which, in the intrusive process, can cause cellular damage in the host tree and release DAMPs, such as ATP, sugars, RNA, and DNA. These are highly conserved molecules that primarily function as signals that trigger and activate the defense responses. In the present study, we generate extracellular DNA (exDNA) from mesquite (P. laevigata) tree leaves (self-exDNA) and P. calyculatus (non-self exDNA) mistletoe as DAMP sources to examine mesquite trees’ capacity to identify specific self or non-self exDNA. We determined that mesquite trees perceive self- and non-self exDNA with the synthesis of O2•−, H2O2, flavonoids, ROS-enzymes system, MAPKs activation, spatial concentrations of JA, SA, ABA, and CKs, and auxins. Our data indicate that self and non-self exDNA application differs in oxidative burst, JA signaling, MAPK gene expression, and scavenger systems. This is the first study to examine the molecular biochemistry effects in a host tree using exDNA sources derived from a mistletoe.


Introduction
Prosopis laevigata (mesquite) is a Mexican endemic tree belonging to the Fabaceae family, predominately found in the center of Mexico and the south of the United States [1][2][3].The mesquite tree has ecological and economic significance, given that it promotes soil fertility by fixing N 2 and increasing PO 4 −3 , Fe 3+ , and C levels [4][5][6][7].Additionally, its biomass is used as biofuel, and its resin is utilized for the manufacture of footwear [1].Unfortunately, the mistletoe Psittacanthus calyculatus poses a threat to this tree [8] and is a photosynthetically active angiosperm that has lost its ability to anchor to the soil.It instead develops an endophytic root system (haustoria) that anchors it to the vascular bundle of mesquite branches to extract water, minerals, and other nutrients [9][10][11][12][13][14][15].
P. calyculatus mistletoe is an obligate hemiparasitic aerial plant whose nutrient consumption causes starvation, growth loss, disease predisposition, and, ultimately, death to the infected host tree [16,17].The prevalence of this parasitic plant is based on the behavioral patterns of frugivorous birds who eat fruits and contribute to their long-distance dispersal and the colonization of new hosts [16].While sufficient information exists regarding the pollination syndromes that P. calyculatus uses to attract and manipulate pollinators and visitors [18], the molecular mechanisms of the infection process remain unknown.Currently, it is well-known that seeds that birds deposit on branches can germinate, leading to the haustorium tissue formation, and the emergence of this intrusive organ is accompanied by high activity of cell-wall-degrading enzymes [19].The enzymatic degradation of the host bark causes a pit on the surface, that functions as an anchor for the haustorium penetration.When the haustorium begins to penetrate the host branches, it causes mechanical damage and breaks the host tissue, and the friction of the penetration also causes damage to the haustorium tissue, which further causes cellular damage that releases different damage-associated molecular patterns (DAMPs) [20][21][22].
Multicellular organisms suffer different types of cellular damage, including infectious processes.Janeway's classical model states that the immune system evolved to distinguish the infectious nonself from the noninfectious self [59].However, in most environments, injury to the outer layers of an organism (the skin or gut epithelia in the case of mammals and the epidermis of leaves and roots in the case of plants) inevitably leads to infection [45].Thus, organisms require an endogenous signaling pathway that enables them to perceive injury through DAMPs, mount adequate local and systemic responses to activate the immune system and achieve resistance [45,60].Therefore, the response to self-patterns is characterized by a rapid increase in reactive oxygen species (ROS) and cytosolic Ca 2+ levels, activation of mitogen-activated protein kinases (MAPKs) cascades, jasmonic acid (JA) signaling and antioxidant systems, which induce long-term resistance to pathogens and inhibits their growth (Table 1).Alternatively, the response to nonself plant patterns varies among species.For example, close species show similar responses to self-patterns, while distant species only induce innate immunity to pathogens and polyphenol accumulation (Table 1).In summary, the compilation of data in Table 1 derived from performed assays on simple photosynthetic organisms (algae) and higher plants is consistent with a high conservation of the response to DAMPS [20,61].
Table 1.Reports on the physiological responses during DAMP-mediated recognition in plant and microalgae species.

Vitis vinifera
Self-xyloglucan from cell walls
Many plants are vulnerable to attacks by parasitic plants because they lack the ability to perceive them, and relatively less attention has been paid to the host tree's ability to identify damaged compounds, including RNA, self exDNA, and cellular debris.In that sense, it has been recently investigated how DAMPs trigger defense-inducing signals, such as ROS production or MAPK signaling, minutes after harm is perceived.In the present study, we analyze mesquite's ability to distinguish its self-or non-self exDNA from that of a parasitic mistletoe, because it is one of the most researched DAMPs in plants that demonstrates species-specific responses.We obtained exDNA from P. laevigata and P. calyculatus by genomic DNA fragmentation.Then, we conducted field experiments in an endemic mesquite area where we sprayed mesquite leaves with self-or non-self exDNA solution at two different concentrations.Further, we quantified the of O 2•− , H 2 O 2 , polyphenols, and flavonoids concentrations, as well as the ROS-enzymes system that includes superoxide dismutase (SOD), peroxidase (PEX), and catalase (CAT) activities.We also quantified phytohormones content, and the relative expression levels of MAPKs as biomarkers of the activation of general stress against distinct exDNA sources.The results from the analysis of those biomarkers show that mesquite trees are able to distinguish between self-exDNA and non-self exDNA and suggest DAMPs as a potential biochemical signal that may be involved in the immune response against the progress of the parasitism.Furthermore, this is the first study to propose using exDNA as a tool for searching for molecular components that could be involved in mistletoe parasitism.

Study System
In order to evaluate mesquite's ability to respond to specific self or nonself patterns, we obtained exDNA from P. laevigata (self-exDNA) and P. calyculatus (non-self exDNA) by genomic DNA sonication extracted from field-collected leaves.To bring the data as close as possible to what happens in nature, mesquite leaves under field conditions were sprayed with deionized water (as a control) as well as with self-and non-self exDNA solutions, and and mesquite leaves had been mechanically damaged with a metal bristle brush (as a positive control for cell damage).Sonication produces exDNA fragments of approximately 200-1200 bp that were applied onto the leaf surfaces.exDNA solutions were prepared at two concentrations that are known to be efficient in inducing responses comparable to those generated by other plant species, with 100 µg/mL applied to Arabidopsis, Phaseolus vulgaris, Phaseolus lunatus, Lactuca sativa, and Capsicum annum [44,47,[56][57][58]67], and 200 µg/mL to Arabidopsis, Phaseolus vulgaris and Phaseolus lunatus [43][44][45]47,56].Leaf samples were collected at 0, 1, and 2 h after the treatments, a consensus time at which there is an accumulation of ROS and activation of JA signaling in different plant species as mentioned before (Table 1).

Effect of exDNA on the Oxidative Burst
The first line of defense of the innate immune system of plants coordinates defense responses through ROS, which are produced in the area where cellular damage has been detected due to the perception of DAMPs.In diverse reports, O 2 •− is rapidly accumulated at the apoplast [68,69], but due to its instability and its toxic effects in tissues, it is converted to H 2 O 2 [70,71].Cell damage usually increases the ROS level in an interval of 30 to 60 min (Table 1).For this reason, we detected O 2 •− and H 2 O 2 using the NBT and xylenol orange method, respectively, at 0, 1, and 2 h after different source treatments.
In that aspect, O 2 •− levels in mesquite leaves increased at different times (0, 1, and 2 h) with 100 or 200 µg/mL with self or non-self exDNA, respectively (Figure 1a).However, we detected O 2 •− after self exDNA treatments, which increased significantly with the application of 200 µg/mL of self exDNA, and the O 2 •− levels subsequently decreased at later times (1 and 2 h) (Figure 1a).In damage treatment, the O 2 •− increased at one hour of measurement and decreased at 2 h (Figure 1a).
Regarding the accumulation of H 2 O 2 levels with the exDNA source applications, in the control treatment, H 2 O 2 levels were not detected at early or later times.In contrast, the H 2 O 2 -level increased in a high concentration immediately after damage treatment and decreased at subsequent times.On the other hand, H 2 O 2 concentrations increased with the addition of 100 µg/mL of self exDNA at early times (0 h) and diminished dramatically at 1 and 2 h.In contrast, we observed that applications of self exDNA at 200 µg/mL, and non-self exDNA at 100 or 200 µg/mL did not raise H 2 O 2 concentrations at early time (0 h).The treatment with non-self exDNA at 200 µg/mL resulted in an increase of H 2 O 2 levels at later times (2 h) (Figure 1b).

Accumulation of Polyphenols and Flavonoids
The phenylpropanoid pathway (PPP) is positively regulated by JA [72], which is the precursor of many families of compounds, such as phenylpropanoids, flavonoids, lignins, monolignols, phenolic acids, stilbenes, and coumarins, which participate in the defense response [73].Particularly, mesquite stands out in the plant kingdom for producing many polyphenols and flavonoids [74].Therefore, we quantified the level of these molecules with the exDNA source application using spectrophotometric techniques.
We detected slight upward variations in the polyphenol content with the different treatments.For example, mechanical damage and application of 100 and 200 µg/mL of non-self exDNA presented the highest levels of polyphenols at time 0. In contrast, 100 µg/mL of self exDNA induced polyphenol levels at 1 h, while non-self exDNA treatment did not induce polyphenol levels.On the other hand, at a later time (2 h), the polyphenol content remained constant with exDNA application, and only damage treatment in-

Accumulation of Polyphenols and Flavonoids
The phenylpropanoid pathway (PPP) is positively regulated by JA [72], which is the precursor of many families of compounds, such as phenylpropanoids, flavonoids, lignins, monolignols, phenolic acids, stilbenes, and coumarins, which participate in the defense response [73].Particularly, mesquite stands out in the plant kingdom for producing many polyphenols and flavonoids [74].Therefore, we quantified the level of these molecules with the exDNA source application using spectrophotometric techniques.
We detected slight upward variations in the polyphenol content with the different treatments.For example, mechanical damage and application of 100 and 200 µg/mL of nonself exDNA presented the highest levels of polyphenols at time 0. In contrast, 100 µg/mL of self exDNA induced polyphenol levels at 1 h, while non-self exDNA treatment did not induce polyphenol levels.On the other hand, at a later time (2 h), the polyphenol content remained constant with exDNA application, and only damage treatment increased their polyphenol levels (Figure 2a).

Antioxidant Enzymes
An increase in the ROS rate, as shown in Figure 1, causes damage to host tissues [70].To avoid this damage and return to a redox balance, cells activate ROS scavenger systems [75,76].Thus, polyphenols act as nonenzymatic scavengers during damage or exDNA recognition.Therefore, to complement the ROS system, we analyzed the enzymatic anti-ROS machinery composed of the differential activities of SOD, PEX, and CAT, a conserved antioxidant system that regulates the level of ROS in the recognition of DAMPs (Figure 3).The O2 •− is reduced to H2O2 by SOD, and H2O2 is decomposed into H2O and O2 by PEX and CAT [54,70,71].SOD activity increases in wounded leaves at later times (2 h); however, the different exDNA applications only slightly increase in SOD levels (Figure 3a).PEX enzyme activity increased in response to damage and exDNA sources (Figure 3b); CAT activity increased when 100 and 200 µg/mL of non-self exDNA were applied at 0 h and decreased at 1 or 2 h.(Figure 3c).Contrary to what was observed with polyphenols, flavonoids have shown significant differences in their content between conditions.All treatments, including mechanical damage, self-exDNA, and non-self exDNA applications, showed increases in the flavonoid content compared to the control.Wounded leaves increased the flavonoid content at the early and later times (0, 1, and 2 h).Similarly, 100 or 200 µg/mL of self exDNA, induced the flavonoid levels, and their concentration remains constant at later times.In a similar manner, 100 µg/mL of nonself exDNA showed a flavonoid content greater than damage during the first hour after treatment, but the content decreased after 2 h, and 200 µg/mL of nonself exDNA showed contents similar to the damage treatment (Figure 2b).

Antioxidant Enzymes
An increase in the ROS rate, as shown in Figure 1, causes damage to host tissues [70].To avoid this damage and return to a redox balance, cells activate ROS scavenger systems [75,76].Thus, polyphenols act as nonenzymatic scavengers during damage or exDNA recognition.Therefore, to complement the ROS system, we analyzed the enzymatic anti-ROS machinery composed of the differential activities of SOD, PEX, and CAT, a conserved antioxidant system that regulates the level of ROS in the recognition of DAMPs (Figure 3).The O 2  [54,70,71].SOD activity increases in wounded leaves at later times (2 h); however, the different exDNA applications only slightly increase in SOD levels (Figure 3a).PEX enzyme activity increased in response to damage and exDNA sources (Figure 3b); CAT activity increased when 100 and 200 µg/mL of non-self exDNA were applied at 0 h and decreased at 1 or 2 h.(Figure 3c).

Activation of JA Signaling
The oxidative burst brought on by necrotic pathogen infection, herbivore attack, and mechanical damage causes JA biosynthesis to occur within minutes [77,78].The ROS found earlier and shown in Figure 1 may have increased the JA content.In our study system, the damage treatment is the only one that exhibits a statistically significant rise in JA levels within the first hour.Self-exDNA application at a concentration of 100 µg/mL did, however, show an increase within the same hour but not a statistically significant one.

Activation of JA Signaling
The oxidative burst brought on by necrotic pathogen infection, herbivore attack, and mechanical damage causes JA biosynthesis to occur within minutes [77,78].The ROS found earlier and shown in Figure 1 may have increased the JA content.In our study system, the damage treatment is the only one that exhibits a statistically significant rise in JA levels within the first hour.Self-exDNA application at a concentration of 100 µg/mL did, however, show an increase within the same hour but not a statistically significant one.Comparable values were observed for the remaining treatments and the control conditions (Figure 4).
Comparable values were observed for the remaining treatments and the control conditions (Figure 4).

Hormonal Accumulation during the Perception of Self-and Non-Self exDNA
The existence of crosstalk between the signaling pathways of defense hormones, such as abscisic acid (ABA), JA, and salicylic acid (SA), with growth regulatory hormones, such as auxins and cytokinins (CKs), is known to modulate and coordinate defense responses [79,80].In parallel to the quantification of JA, we quantified two other defense hormones (SA and ABA) and four canonical growth regulatory hormones by HPLC.We observed that the SA levels increased with 200 µg/mL self-exDNA at later times (2 h), and the nonself exDNA applications did not exert an increase in the concentration of this phytohormone (Figure 5a).The treatments did not show statistically significant differences for ABA levels at 0 h.However, after applying 100 µg/mL of self-exDNA for one hour, there was a noticeable rise in the amount of ABA (Figure 5b).
Additionally, we evaluated three cytokinins (CKs) that are related to leaf development.For example, zeatin was not found following various treatments, and iP did not exhibit statistically significant changes at any exDNA source application (0, 1, or 2 h), except for 100 µg/mL of non-self exDNA application, which caused an increase in iP at 2 h (Figure 5c).Conversely, Kinetin levels were detected (1 or 2 h) after applying 100 or 200 µg/mL of non-self exDNA and self-exDNA (Figure 5d).
To coordinate the many plant-development processes, auxins are generated in the shoot system and then transferred to the other organs via the phloem [81].In host and parasite plants, auxins mainly participate in the development of the root system and haustoria, respectively [82][83][84].Finally, we determined that the IAA phytohormone was present in the early stages of mechanical damage (0 or 1 h), but those levels were not detectable when exDNA treatments were applied.In contrast, after two hours of treatment, there was a rise in non-self exDNA at 100 or 200 µg/mL (Figure 5e).

Hormonal Accumulation during the Perception of Self-and Non-Self exDNA
The existence of crosstalk between the signaling pathways of defense hormones, such as abscisic acid (ABA), JA, and salicylic acid (SA), with growth regulatory hormones, such as auxins and cytokinins (CKs), is known to modulate and coordinate defense responses [79,80].In parallel to the quantification of JA, we quantified two other defense hormones (SA and ABA) and four canonical growth regulatory hormones by HPLC.We observed that the SA levels increased with 200 µg/mL self-exDNA at later times (2 h), and the non-self exDNA applications did not exert an increase in the concentration of this phytohormone (Figure 5a).The treatments did not show statistically significant differences for ABA levels at 0 h.However, after applying 100 µg/mL of self-exDNA for one hour, there was a noticeable rise in the amount of ABA (Figure 5b).
Additionally, we evaluated three cytokinins (CKs) that are related to leaf development.For example, zeatin was not found following various treatments, and iP did not exhibit statistically significant changes at any exDNA source application (0, 1, or 2 h), except for 100 µg/mL of non-self exDNA application, which caused an increase in iP at 2 h (Figure 5c).Conversely, Kinetin levels were detected (1 or 2 h) after applying 100 or 200 µg/mL of non-self exDNA and self-exDNA (Figure 5d).
To coordinate the many plant-development processes, auxins are generated in the shoot system and then transferred to the other organs via the phloem [81].In host and parasite plants, auxins mainly participate in the development of the root system and haustoria, respectively [82][83][84].Finally, we determined that the IAA phytohormone was present in the early stages of mechanical damage (0 or 1 h), but those levels were not detectable when exDNA treatments were applied.In contrast, after two hours of treatment, there was a rise in non-self exDNA at 100 or 200 µg/mL (Figure 5e).

Mapk Gene Expression Levels by Self and Non-Self exDNA
MAPK pathways are common and versatile signaling proteins that lie downstream of second messengers and play central roles in plant responses to various stresses [85] such as wounding, pathogen recognition, and general defenses [86].To determine whether the exDNA applications activate MAPK genes, we examined the relative expression of three MAPK gene family members using RT-PCR.MAPK2, MAPK44, and MAPK10 genes are associated with mechanical damage, immune defenses, and general stresses like plant pathogens.Accordingly, following mechanical damage, the activation of the MAPK2 gene expression was detectable at an early time (0 h), increased at 1 h, and gradually decreased at 2 h (Figure 6a).Furthermore, MAPK2 levels increase with the application of 200 µg/mL of self-exDNA and 100 and 200 µg/mL of non-self exDNA at 0 h.However, gene expression increases with any exDNA concentration at late times (1 or 2 h).Additionally, when 200 µg/mL of self-exDNA and 100 and 200 µg/mL of non-self exDNA are applied at 0 h, MAPK2 increases.Moreover, at late times (1 or 2 h), gene expression increases with any exDNA concentration.Regarding the MAPK4 gene, our findings show that 100 µg/mL of non-self exDNA application increased the relative expression at 0 h, and that this expression decreased at later times (1 or 2 h).Additionally, we observed that the gene expression always remained stable, except for the 100 µg/mL of self-exDNA application that decreased at 2 h (Figure 6b).However, early time (0 h), 100 and 200 µg/mL of self-exDNA increased MAPK10 gene expression significantly compared to 100 µg/mL of non-self exDNA.Nevertheless, those expression patterns decreased later (1 or 2 h) (Figure 6c).

Discussion
Based on the different perceptions of the DAMPs application (in particular exDNA), and the plant immune responses, we investigated distinct immunity-related traits that were similar to those reported in other plant species (Table 1) by using self-exDNA applications from P. laevigata (mesquite) and non-self exDNA from P. calyculaus (mistletoe), which are phylogenetically distant from each other and belong to different families (Fabaceae and Loranthaceae, respectively) [87,88].Data obtained in this study indicated that

Discussion
Based on the different perceptions of the DAMPs application (in particular exDNA), and the plant immune responses, we investigated distinct immunity-related traits that were similar to those reported in other plant species (Table 1) by using self-exDNA applications from P. laevigata (mesquite) and non-self exDNA from P. calyculaus (mistletoe), which are phylogenetically distant from each other and belong to different families (Fabaceae and Loranthaceae, respectively) [87,88].Data obtained in this study indicated that O 2 •− , H 2 O 2 , flavonoids, and ROS-stress enzymes were triggered; additionally, phytohormones and MAPK gene expressions were accumulated spatially, regardless of the self-exDNA or non-self exDNA applications.
Plants' innate immune systems use ROS to initiate defense responses when they detect cellular damage as a result of detecting foreign materials (DAMPs) [69,89].O 2 is accumulated inside the cell apoplast, but, due to its instability and its toxic effects on tissues, it transforms into H 2 O 2 [69,70].Our observations indicate that O 2 •− and H 2 O 2 levels were elevated at early times (0 or 1 h) in response to 100 or 200 µg/mL of self-and non-self exDNA.Biomarkers of oxidative burst with the exDNA source treatments suggest that the self-exDNA application could be an indicator of the stress levels similar to the damage suffered in mesquite tissues.In that aspect, similar effects were reported with the application of self-exDNA to root rice, which increased the production of O 2 •− and H 2 O 2 levels [55].Moreover, Vega-Muñoz et al. [58] and Zhou et al. [56] discovered that there were significant levels of H 2 O 2 in both self-and non-self exDNA applications in Arabidopsis ecotypes at early times.In addition, Solanum lycopersicum [42,56], Brassica napus [56], and Phaseolus vulgaris plants [45] showed high H 2 O 2 levels with the self-exDNA application.Taken together, self-and non-self exDNA can induce a species-specific response such as ROS in mesquite trees.
Specifically, large concentrations of metabolite accumulations, such as flavonoids and polyphenols were observed by the exDNA applications.For example, we saw that polyphenols were present at detectable quantities in all assessed conditions, regardless of whether the exDNA applications were made early or late.Flavonoid concentrations were produced in trees that underwent mechanical damage or non-self exDNA treatments.Prosopis leaves contain a significant amount of several metabolites, including flavonoids and polyphenols, and it has been determined that total polyphenols in P. glandulosa and P. juliflora trees exposed to metal stress are constitutively present [90,91].This could potentially account for the observed increase in polyphenol concentration following any kind of treatment, such as mechanical injury or exposure to exDNA sources.Recently it has been observed that the use of either self-or non-self exDNA in herbaceous plants has a significant effect on the amount of polyphenols and flavonoids.For example, in L. sativa and S. lycopersicum, the presence of self-exDNA or non-self exDNA increased the total polyphenols and flavonoids [44,52].To our knowledge, non-self exDNA applications strongly induce the synthesis of flavonoids, which serve as an auxiliary antioxidant system for host damage, with a specific function of diminishing ROS accumulation in plant cells [92].However, to avoid the stressors, the plants can use antioxidant enzyme systems to maintain and regulate their own homeostasis.In that aspect, SOD, PEX, and CAT enzymes revealed differential activities with non-self exDNA sources.SOD and PEX activities were induced with both exDNA sources; in contrast, CAT activities were induced only with non-self exDNA.These observations are similar to those obtained in tomato plants, where the application of self and non-self exDNA induced the enzymatic activities of SOD and CAT [52].In the same way, in L. Sativa, the applications of self and non-self exDNA induced the differential expression of genes-encoding SOD-and CAT-like enzymes [44].Those results might suggest that self and non-self-exDNA induce the activation of the enzymatic antioxidant system necessary to decrease the intensity of ROS.
It is well known that the biosynthetic pathways of plant phytohormones involve the interaction of growth-regulating hormones, such as auxins and cytokinins (CKs), and defense hormones, such as salicylic acid (SA), JA, and abscisic acid (ABA), which play important roles as primary messengers in signal transduction and cell-metabolism regulation [91].Additionally, they synchronize and influence defense responses through signaling pathways under different plant stresses [79,80].In this regard, plant research to detect phytohormones after applying exDNA sources has been based mainly on JA and SA analysis.Under mechanical damage and after applying 100 µg/mL of self-exDNA, we found a rise in JA levels associated with an oxidative burst; however, non-self exDNA treatment did not affect JA levels.Additionally, JA levels were found with the application of exDNA sources in a similar way to the different reports mentioned [25,32,36,55,57,87].Additionally, we quantified SA and ABA, and four canonical growth regulatory hormones such as Zeatin, Kinetin, iP, and IAA derived from the self and non-self exDNA applications.
SA is a signal molecule that is derived from the Shikimate-PPP pathway in plants and is recognized for its role in antioxidant defense systems and gene regulation [93].Moreover, the synthesis and metabolism of SA are necessary for the development of the stress symptoms and hypersensitive response, as well as the induction of gene-encoding pathogenesis-related proteins [94].We observed a pattern of SA contents in both self and non-self exDNA treatments, with high amounts at early and late times.In that aspect, many authors have reported that SA peaked minutes after or at 24 h after DAMP sources, similar to our results reported here [25,32,36,55,57,87].Notably, we detected an increase in ABA contents with non-self exDNA treatment at 2 h after treatment.ABA and JA signaling pathways converge at several regulatory points that are involved in ROS activation, MYC2 upregulation, and stomatal closure [78,[95][96][97][98][99].However, these pathways are erroneously considered to be antagonistic, and it is also speculated that biotic stresses like DAMP sources do not cause the ABA pathway to be activated.Chiusano et al. [48] analyzed ABA signaling with self and non-self exDNA applications, and they observed a downregulation of ABA-related genes with self-exDNA application at later times.Auxin and CKs have great potential to mitigate a variety of stressors and control plant development and stress tolerance (Raza et al., 2023).For example, CK (6-benzylaminopurine; BAP) and auxin (N6-(2-isopentenyl)adenine; iP) levels increased in A. thaliana at 16 and 24 h after treatment with self-exDNA [48].Additionally, an upregulation of CK-related genes was detected by the exDNA source applications [48].Similar effects were observed in Neochloris oleabundans microalgae with the self-exDNA applications, and differential responses in gibberellic acid, isopentenyladenine, and benzylaminopurine phytohormones compared with non-self exDNA application [54].In that aspect, we detected an increase in iP and Kinetin levels with self-exDNA and non-self exDNA treatment at 1 or 2 h.Subsequently, we observed a differential auxin content with 200 µg/mL of non-self exDNA at a late time.These data could indicate that the differential synthesis of CKs and auxin are regulated by the indistinct exDNA source application.
Plants require an endogenous signaling pathway that enables them to perceive danger, for example signals that can originate from the infection and injury through DAMPs to mount adequate local and systemic responses to activate the immune system and achieve resistance [100].DAMPs that trigger an immune response are ATP, extracellular matrix fragments, and exDNA, which activate signaling minutes after the perception of damage with events such as calcium fluxes in the cytosol, ROS, and MAPKs, which are considered conserved early responses [101].Hence, exDNA source application in P. laevigata activates stress-related responses.In that aspect, our observations indicate that mechanical damage, both self and non-self exDNA induce the expression of MAPK-related genes like MAPK2, MAPK4, and MAPK10.In particular, there was a slight increase of MAPK2 relative expression with exDNA sources but a significant increase with mechanical damage at one hour.Applying non-self exDNA can increase the relative expression of the MAPK4 gene at either early or late stages.Finally, the gene expression of MAPK10 increases its relative expression exclusively at early times when using self exDNA.These findings suggest that signaling and regulation of self and non-self exDNA, which the tree perceives as a stressor, is the reason for the increase in the relative expression of various MAPK genes.This aspect of plant-stress perception may be explained by the specific role that each encoding gene for the MAPK examined plays in its expression.For example, MAPK2 is a gene family member that is involved in mechanical plant wounding.After herbivory with Helycoverpa armigera and mechanical wounding treatments, the gene expression levels of MAPK2 changed in soybean cultivars [102], indicating that herbivores activate this gene.By the way, MAPK4 is a related gene in plant immune response against phytopathogens, due to mediate jasmonate-and salicylate-dependent defense responses, and acts with other MAPKs as a negative regulator of plant immune responses [103].Additionally, the gene MAPK10 is associated with resistance to plant pathogens [104].Previous studies support our findings.For example, single-stranded, and exDNA source applications from Arabidopsis induce the early expression of different MAPK genes involved in plant defenses [46,47].Moreover, the self exDNA application of Phytophthora capsici induces the relative expression of MAPK1, that are involved in the pathogenicity of the fungus in chili peppers [105].Duran-Flores and Heil [45] found MAPK accumulation in bean leaves treated with mechanical damage and exDNA sources.Further, in peach fruits, a complete MAPK cascade (MAPKKK1, MAPKK2, and MAPK1) was identified, which functions downstream of the FLS2-BAK1 receptor complex to exDNA [57].However, phosphorylation of MAPKs is one of the first reactions that cells have to DAMP sensing [106].exDNA is recognized by surface proteins known as pattern-recognition receptors (PRRs), which sets off immune responses such as Ca 2+ ion fluxes across the plasma membrane, the production of ROS, the activation of MAPKs, and a quicker regulation of defensive gene expression [19,24,41,43,47,58,99] to phosphorylate NADPH oxidases (NOX family) to produce O 2 •− in the apoplast, initiating the oxidative burst within a few minutes [107].Based on the above, the application of self-exDNA as a damage-specific DAMP strongly induces MAPK regulation in mesquite trees.
Mechanical damage, herbivory, and pathogen attack release DAMPs such as cell wall fragments, eATP, exDNA, and RNA, DAMPs recognition is a highly conserved mechanism across kingdoms [61,108].DAMPs can mediate the different plant immune responses.For example, they can cause an increase in the levels of H 2 O 2 and JA, activities of PPO, CAT, PEX, SOD, and hundreds or thousands of genes change their expressions.In particular, those that encode MAPKs, pathogenesis-related (PR) proteins, and enzymes of phenylpropanoid metabolism, as well as components of the JA and ethylene pathways in short or large period times induce resistance to pathogens through salicylic acid (SA) and ethylene (ET) signaling, downregulating photosynthesis genes, reducing cell division, and suppressing leaf and root growth [28,33,44,52,54,56,58,[109][110][111][112][113].
Relationship of exDNA in the P. calyculatus Establishment P. calyculatus has developed intelligent strategies to obtain nutrients from host trees through a structure called haustorium.The infective process takes place during seed germination, and the appearance of a prehaustorium occurs which activates the expression of genes encoding an arsenal of cell-wall-degrading enzymes that help to breach the host bark and the cortex [19].Through mechanical pressure, the haustorium penetrates the branch and will connect with the xylem host [19].The host experiences mechanical wounding during this penetration process, and the haustorium tissue may also experience mechanical wounding as a result of friction between the branch's woody walls due to the narrow process inside the branch.Then, the release of cellular compounds (self or nonself) from both species may occur, and the DAMP perception processes occurs, triggering different stress responses that activate the cascade of defense responses in the host tree.Thus, organisms require an endogenous signaling pathway that enables them to perceive injury through DAMPs, mount adequate local and systemic responses to activate the immune system and achieve resistance such as MAPKs [100].For example, in Striga hermonthica, a root parasite, the haustorium formation is induced by the haustoriuminducing factors secreted by host plants, such as Dimethoxybenzoquinone (DMBQ).DMBQ may induce the primary signaling cascade for haustorium induction through leucine-rich repeat-receptor-like kinase (CADL LRR-RLKs).It is known that DMBQ treatment of P. japonicum induces Ca 2+ signaling and MAP kinase (MAPK) phosphorylation.However, many elements of the signaling cascade are unknown.This includes the coreceptor for the CADL LRR-RLKs, the receptor-like cytoplasmic kinase (RLCK), the specific MEKKK, MEKK, and MAPK signaling cascade involved, and the Ca 2+ channel that facilitates Ca 2+ signaling, which subsequently induces pattern-triggered immunity [114].Tomato plants were able to respond to a small peptide factor found in the parasitic plant C. reflexa's extracts because they contain a cell-surface-receptor-like protein CUSCUTA RECEPTOR 1 (CuRe1), which is essential for the perception of this parasite-associated molecular pattern, as well as a rapid induction of ROS in a similar way to microbial pathogens [30].Moreover, C. campestris parasite extracts were injected into tomato plant hosts, and the host plants responded with reinforced cell walls at the infection sites against the parasite, thus, avoiding the parasitic infestation [62,63].These events may be related to lignocellulolytic degradation of the host's cell wall, and the released fragments accumulate by mechanical injury or enzymatic degradation that may serve as DAMPs and trigger defensive responses in the host [115].In addition, transcriptome analysis in the infective process of Psittacanthus schiedeanus mistletoe, an arsenal of cell-wall-degrading enzymes like glucosyl hydrolases is necessary for the host infection [116].Similarly, enzymatic analyses in P. calyculatus mistletoe revealed high activities of cell-wall-degrading enzymes such as cellulase and β-1,4-glucosidase that were primarily active in haustorium development, while xylanase and endo-glucanase were highly active in the haustorium penetration and xylem connection [19].Additionally, we found that the host reacts against P. calyculatus mistletoe by producing oxidative stress processes like H 2 O 2 , ROS-related enzymes, and different concentrations of phytohormones like JA and SA [117].These findings suggest that the host perceives self or non-self DAMPs.However, it is difficult to identify DAMPs as molecular effectors at particular stages of the infectious process, because they cannot be distinguished from the species from which they originated.The term "DAMPs" refers to self and non-self patterns that may originate from parasitic or host plants.These patterns are understood to be a possibility in the plant-host interaction, but further research must be conducted to rule out this theory.Hence, the approaches that we have taken in this research show us that exDNA is perceived as a DAMP.That is relevant in the mistletoe-mesquite interaction, which is an area where it has not been explored indepth.That requires the application of state-of-the-art tools such as "-Omics" to exploit in the future, with results that open a window of new knowledge, and those can be applied to other models of parasitic interaction, such as mistletoes that that infect host trees and root-parasitic plants.
Based on our results, we propose a working model of molecular interaction between the self and non-self exDNA host-tree perception (Figure 7).The application of self and parasite exDNA in mesquite leaves under field conditions allowed us to determine that mesquite recognizes and responds differentially to self-and non-self exDNA.exDNA is recognized as a DAMP by the cell host, activating MAPK signaling and increasing ROS, such as O 2 and H 2 O 2 .Subsequently, the antioxidative system, including enzymes such as SOD, PEX, and CAT, will help reduce this oxidative burst.Additionally, flavonoids detoxify H 2 O 2 by preventing signal amplification, and then JA levels increase.Surprisingly, there is an accumulation of flavonoids, which is independent of hormonal levels.After a period, H 2 O 2 levels return to their initial levels, flavonoids do not change, and kinetin and ABA phytohormone increase, which may cause flavonoid accumulation or prevent JA signaling.Lastly, levels for phytohormones such as kinetin and ABA return to initial levels after 2 h, but flavonoid levels peak.On the contrary side, independent of hormone accumulation, nonself exDNA perception induces H 2 O 2 , and CAT activity detoxifies these ROS, and flavonoid accumulation.Later, there is a decrease in CAT enzyme activity, a rise in H 2 O 2 , and a buildup of other phytohormones such as IAA.

exDNA Generation
Fresh mesquite (P.laevigata) (Figure S1) and mistletoe (P.calyculatus) leaves were collected near Irapuato, Guanajuato (20°42′42.18″N;101°19′22.66″O)and frozen with dry ice.To obtain genomic DNA, 5 g of ground tissue were homogenized with 20 mL of 100 mM Tris-HCl pH 8, 5 mM EDTA pH 8, 50 mM NaCl, and 10 mM β-mercaptoethanol [118].Genomic DNA concentrations were determined on a µDrop™ Duo plate and quantified in a microplate spectrophotometer (Multiskan SkyHigh, Thermo scientific, Waltham, MA, USA), then the extractions were treated with RNase to remove contaminating RNA.A solution of DNA (600 µg/mL) was resuspended in ultrapure and sterile water.We obtained short fragments of exDNA between 200-1200 pb after DNA sonication with the following parameters: 6:00 min at 55% of amplitude with a pulse of 1 s (active) and 1 s (not active).The exDNA concentration generated was adjusted with sterile distilled water to reach 100 or 200 µg/mL in the final concentrations that are reported to be effective in other model plants.Genomic exDNA, as well as fragmented exDNA, was verified on 3.0% agarose gel using ethidium bromide staining (Figure S2).

Study Site and exDNA Application
Field experiments were performed in a suburban area near Irapuato, Guanajuato.Young leaves and branches from mesquite trees were selected without visible infection by pathogens or herbivores, and, then, the branches selected were tagged.
Ten mL of self exDNA (P.laevigata) solution and nonself exDNA (P.calyculatus) at two concentrations (100 or 200 µg/mL with 0.1% of Tween 20) were sprayed onto the surface of the mesquite leaves from branches.Additionally, 10 mL of sterile deionized water

exDNA Generation
Fresh mesquite (P.laevigata) (Figure S1) and mistletoe (P.calyculatus) leaves were collected near Irapuato, Guanajuato (20 • 42 ′ 42.18 ′′ N; 101 • 19 ′ 22.66 ′′ O) and frozen with dry ice.To obtain genomic DNA, 5 g of ground tissue were homogenized with 20 mL of 100 mM Tris-HCl pH 8, 5 mM EDTA pH 8, 50 mM NaCl, and 10 mM β-mercaptoethanol [118].Genomic DNA concentrations were determined on a µDrop™ Duo plate and quantified in a microplate spectrophotometer (Multiskan SkyHigh, Thermo scientific, Waltham, MA, USA), then the extractions were treated with RNase to remove contaminating RNA.A solution of DNA (600 µg/mL) was resuspended in ultrapure and sterile water.We obtained short fragments of exDNA between 200-1200 pb after DNA sonication with the following parameters: 6:00 min at 55% of amplitude with a pulse of 1 s (active) and 1 s (not active).The exDNA concentration generated was adjusted with sterile distilled water to reach 100 or 200 µg/mL in the final concentrations that are reported to be effective in other model plants.Genomic exDNA, as well as fragmented exDNA, was verified on 3.0% agarose gel using ethidium bromide staining (Figure S2).

Study Site and exDNA Application
Field experiments were performed in a suburban area near Irapuato, Guanajuato.Young leaves and branches from mesquite trees were selected without visible infection by pathogens or herbivores, and, then, the branches selected were tagged.
Ten mL of self exDNA (P.laevigata) solution and nonself exDNA (P.calyculatus) at two concentrations (100 or 200 µg/mL with 0.1% of Tween 20) were sprayed onto the surface of the mesquite leaves from branches.Additionally, 10 mL of sterile deionized water was applied to the control branches.Simultaneously, other leaves from the branches were damaged with a steel-wire brush as a positive control for cell damage.Finally, the leaves were collected at 0, 1, and 2 h, and samples were flash-frozen with liquid nitrogen and stored at −80 • C until experimental use.

•−
The method can quantify O 2 •− radicals that are produced in plant tissues under stress and is based on the selective extraction of formazan produced after reduction with nitroblue tetrazolium (NBT) in histochemical staining [119].O 2 •− was measured according to Fonseca-García et al. [120], with some modifications.Briefly, 500 µL of 0.1% NBT (Sigma Aldrich, St. Louis, MO, USA) in 50 mM pH 7.5 NaH 2 PO 4 were added to a tube containing 10 pinnae (leaves).Samples were incubated at room temperature for 1 h in darkness.Then, NBT-stained leaves were bleached with 96% ethanol incubated for 1 h [121].NBT-stained leaves were ground with liquid nitrogen and dissolved in 2 M KOH-DMSO (1:1.6 v/v) and then centrifuged at 14,000 rpm for 10 min.Two hundred µL of samples were placed into a microplate, and the absorbance was measured at 630 nm in a microplate spectrophotometer (Multiskan SkyHigh.Thermo Scientific, Waltham, MA, USA).O 2 •− was indirectly quantified by a linear regression of the mean NBT absorbance values in samples compared to an H 2 O 2 standard curve (0, 3, 6, 9, 12, 18, 24, and 30 µM).

H 2 O 2 Quantification
We used the Peroxidase Assay Kit (Sigma Aldrich, MAK311).This assay utilizes the chromogenic Fe 3+ −xylenol orange reaction in the mixture.A purple complex is formed when Fe 2+ is oxidized to Fe 3+ by peroxides present in the sample that can be measured at 585 nm, and the absorbance is proportional to the peroxide level in the sample.The optimized formulation reduces interference by substances in the raw samples.A concentration of 1 µM H 2 O 2 equals 34 ng/mL, with a H 2 O 2 range detection between 0.2-30 µM (7-1020 ng/mL).Samples (50 mg) were ground with liquid nitrogen, and 0.5 mL of ultrapure water were added.Samples were mixed for 1 min and centrifuged at 14,000 rpm for 15 min at 4 • C. The supernatant was recovered for the assay, and a 40 µL sample was added to 200 µL of detection reagent in a 96-well plate.After 30 min of incubation, the absorbance was measured as before.The H 2 O 2 contents in the samples were calculated from a standard curve (0, 3, 6, 9, 12, 18, 24, and 30 µM).

Jasmonic Acid Quantification
Ground tissues (250 mg) were mixed with 1 mL of ethyl acetate and 10 µL of 0.1 mg/mL (±)-Dihydrojasmonic acid (internal standard; Cat: CDS022683, Sigma-Aldrich) and incubated at 4 • C in darkness overnight.The mixtures were centrifuged at 14,000 rpm for 15 min at 4 • C. Supernatants were collected and placed in a 1.5 mL tube.Additionally, plant residues were re-extracted with 1.0 mL of ethyl acetate, centrifugated as before, and the new supernatants were combined, evaporated, and concentrated entirely on a vacuum concentrator.Samples were derivatized with 100 µL of N ′ N ′ -diisopropylethylamine, 100 µL of chloroform, and 10 µL of pentafluorobenzyl at 65 • C in an extraction hood.The mixtures were cooled down on ice and were concentrated with gaseous nitrogen again.Finally, the residues were resuspended in methanol (70%) and centrifuged at 14,000 rpm for 5 min at 4 • C. Subsequently, the supernatants were used to analyze JA in a GC-MS (Agilent Technologies, Santa Clara" CA, USA) equipped with a DB-1MS column (approx.60 m length, 0.25 mm diameter, and 0.25 µm film) (Agilent Technologies, USA) with the following program 150 • C for 3 min and ramped up at 4 • C/min to 260 • C, with a final temperature after 25 min.The quantification of the JA was performed using the internal standard of dihydrojasmonic added to each sample, as indicated above.An average of the areas of the internal standard was obtained for all the samples, obtaining a correction factor by dividing the area of the internal standard by the area of each sample and multiplying it by the correction factor obtained for each sample.The average area of the internal standard corresponds to a concentration of 0.005 µg/µL [122].

RNA Extraction from Mesquite Leaves Treated with exDNA
Samples treated with exDNA were homogenized in liquid nitrogen with a mortar and pestle to obtain a fine powder.Next, 1 mL of Trizol ® reagent (Invitrogen, Waltham, MA, USA) was added before the samples thawed.Samples were mixed by immersion and incubated for 5 min at room temperature.After this time, 200 µL of chloroform were added and vortexed vigorously for 15 s, incubated for 3 min, and centrifuged at 10,000 rpm for 5 min.The aqueous phases were transferred to new 1.5 mL tubes, and 0.5 mL of isopropanol were added.The mixtures were incubated for 10 min at room temperature and centrifuged at 10,000× g for 10 min.The aqueous phase was removed from the samples, and the pellets were washed with 1 mL of 70% ethanol.Mixtures were then centrifuged at 10,000 rpm for 3 min, the supernatant was discarded, and pellets were air-dried for 5 min.RNA pellets were then dissolved in 50 µL DEPC-water.The integrity of the RNAs was verified at 230, 260, and 280 nm (the A260-A230 and A260-A280 ratios were calculated), as well as by electrophoresis on agarose gel (1.5%) [123].

cDNA Synthesis and Analysis of MAPKs Gene Expression
cDNA synthesis was performed using the oligo primer (dT)20 and the SuperScript III Reverse Transcriptase cDNA synthesis kit (Invitrogen ® ).Briefly, an aliquot of isolated RNA was mixed with 1 µL of Oligo dT20 + 1 µL of 10 mM dNTPs mix in a final volume of 10 µL with DEPC water.The mixtures were incubated at 65 • C for 5 min, and subsequently placed on ice for 1 min.The mixture for cDNA synthesis was prepared as follows: 2 µL of 10× reverse transcription buffer, 4 µL of 25 mM MgCl 2 , 2 µL of 0.1 M DTT, 1 µL of RNAase OUT, and 1 µL SuperScript III RT ® .The mixtures were incubated for 50 min at 50 • C, and the reaction was finished at 85 • C for 5 min and placed on ice.One µL of RNase H was added to the mixture and incubated for 30 min at 37 • C, followed by 1 µL of RNAase A for 30 min.The cDNA obtained was subjected to RT-PCR.

Quantification of Polyphenols and Flavonoids
Lyophilized samples (50 mg) were placed in a 2 mL microtube, mixed with 500 µL of 70% methanol, and homogenized at 1400 rpm in darkness at room temperature overnight.Samples were centrifuged at 14,000 rpm for 15 min to obtain the methanolic extract.Three µL of the sample extract were diluted with 237 µL of deionized water, 15 µL of 10% Folin-Ciocalteu, and 45 µL of 15% Na 2 CO 3 on a 96-well plate.The mixtures were incubated in darkness at room temperature for 30 min to read the absorbance at 760 nm (Multiskan SkyHigh, Thermo Scientific).Polyphenol concentrations were indirectly determined by linear regression of the mean absorbance values in samples compared to a standard curve.The standard curve was prepared with known concentrations of gallic acid dissolved in deionized water.Flavonoids were also determined according to [124] with some modifications; 12.5 µL of the methanolic extract were mixed with 7.5 µL of 5% NaNO 2 and incubated at 4 • C for 5 min.Fifteen µL of 10% AlCl 3 .6H 2 O were added, mixed, and incubated for 5 min.Finally, 50 µL of 1M NaOH and 165 µL of deionized water were added to read the absorbance at 510 nm (Multiskan SkyHigh Thermo Scientific).The flavonoid concentration was indirectly determined by linear regression of the mean absorbance values in the samples compared to a standard curve.The standard curve was prepared with known concentrations of quercetin dissolved in deionized water.

Antioxidant Enzyme Activities
Leaves from each treatment were ground with liquid nitrogen for protein extraction; ~100 mg of ground tissue were homogenized with 500 µL of 50 mM Na 2 PO 4 pH = 7, 1 mM EDTA, and 1% PVP.Next, the homogenates were centrifuged at 4 • C and 14,000 rpm for 15 min.The protein concentration extracted was estimated by Bradford prior to the enzymatic assays.SOD activity was determined according to Méndez-Gómez et al. [125] with some modifications.Briefly, in 1.5 mL microtubes, 10 µL of extract were mixed with 40 µL of 0.1 M EDTA, 20 µL 1.5 mM NBT, 600 µL of 0.0067 M KH 2 PO 4 pH = 7, and 50 µL of 12 mM of riboflavin.Reaction control is the same except for samples.The microtubes were illuminated with a 40 W lamp for 15 min.Posteriorly, absorbance at 560 nm was measured every 1 min for 10 min (Multiskan SkyHigh Thermo Scientific).For calculations, the blank absorbance was subtracted from each reading made at 560 nm on the samples that were tested.Activity was defined as the amount of enzyme that inhibits a change of 0.001 in absorbance caused by the photochemical reduction of NBT.For PEX, 140 µL of 100 mM Na 2 PO 4 pH = 6, 50 µL of 5 mM guaiacol, 10 µL of extracted, and 10 µL of 5 mM H 2 O 2 were mixed in a 96-well plate to read the absorbance at 480 nm every min for 30 min (Multiskan SkyHigh, Thermo scientific, Waltham, MA, USA).The activity was defined as the amount of enzyme that increases by 0.001 in absorbance by guaiacol oxidation.CAT activity was determined by a decrease in H 2 O 2 concentration [126]; 2.5 µL of extract were mixed with 150 µL of 50 mM Na 2 PO 4 pH = 7 and 20 µL of 50 mM H 2 O 2 in a 96-well plate to read the absorbance at 240 nm each min for 10 min (Multiskan SkyHigh, Thermo scientific, Waltham, MA, USA).The activity was defined as the amount of enzyme that disproportionates 1 µmol H 2 O 2 for 1 min.The H 2 O 2 extinction coefficient used for the calculation was ε = 45.2 /mM mL.

Phytohormone Quantification
The phytohormones were extracted and purified in accordance with previous reports by Zárate-López et al. [54].Samples were suspended in 250 µL of water of HPLC grade that were acidified at 0.1% with acetic acid (J.T. Baker, Phillipsburg, NJ, USA).Chromatographic analysis was performed on a UHPLC Ultimate 3000 (Thermo Scientific, San Jose, CA, USA) equipped with a quaternary pump, diode array detector (DAD), and a Zorbax RRHD C18 column (150 mm × 2.1 mm id, and 1.9 µm of particle size; Agilent Technologies, San Jose, CA, USA).The chromatography elution was done with ternary gradient elution combining methanol (B), acetonitrile (C), and water supplemented with 0.1% acetic acid (A) at a flow rate of 0.35 mL/min.The chromatography elution was done with ternary gradient elution combining methanol, acetonitrile, and water supplemented with 0.1% acetic acid at a flow rate of 0.35 mL/min.The following ternary gradient was used: 100% A (zero to one min), 60% A + 35% B + 5% C (one to 16 min), and 80% B + 20% C (16 to 21 min).A UV-DAD detector was set to record spectra between 210 nm and 270 nm.The quality control of the sample developed in a linear range between 0.98 to 35.91 uM of phytohormones.Calibration graphs were generated by plotting the peak areas against the corresponding concentrations injected into the HPLC column.Least-squares linear regression analysis was used to generate a linear calibration curve.The limit of detection (LOD) and limit of quantification (LOQ) were established as 3 times and 10 times, respectively, the signal-noise ratio.The HPLC chromatograms were analyzed using Chromeleon 7.0 software (Thermo Scientific, San Jose, CA, USA).

Statistics Analysis
For all experiments, the data were statistically analyzed using STATISTICA 8.0 Software (Dell StatSoft, Austin, TX, USA).Multivariate analyses with Tukey's post hoc test were used for testing differences in the analyzed parameters.In the graphs, the different letters indicate significant differences in comparison to the control (p ≤ 0.005).

Conclusions
In this study, we analyzed the stress responses to the exogenous application of self exDNA and nonself exDNA from the mistletoe P. calyculatus as a source of DAMPs in mesquite trees (P.laevigata).The study showed that the tree exhibits comparable reactions to both self and nonself exDNA applications, and both types of exDNA sources induced differential species-specific responses, such as ROS generation, elevated levels of defense metabolites like polyphenols and flavonoids, ROS-stress enzyme activities that reduce the oxidative burst, and the spatial synthesis of phytohormones.Furthermore, we observed differential expression in genes encoding MAPKs mainly when self exDNA was applied.Like other organisms, mesquite trees respond with self exDNA-inducing responses, such as oxidative burst and MAPKs activation.However, we believed that, for their ecological relation, mesquite trees respond to mistletoe exDNA with weaker responses compared to their self exDNA.

Figure 1 .
Figure 1.Quantitative determination of ROS in mesquite leaves treated with exDNA sources.Concentrations of O2 •− (a) and H2O2 (b).Mesquite leaves were wounded (damage) or sprayed with water for control, or self-exDNA (derived from mesquite leaves), and non-self exDNA (derived from P. calyculatus mistletoe leaves), and then collected at 0, 1, and 2 h.Bars represent means ± SE (n = 5; nmol per g fresh weight), and the different letters indicate a significant difference at p ≤ 0.05.

Figure 1 .
Figure 1.Quantitative determination of ROS in mesquite leaves treated with exDNA sources.Concentrations of O 2 •− (a) and H 2 O 2 (b).Mesquite leaves were wounded (damage) or sprayed with water for control, or self-exDNA (derived from mesquite leaves), and non-self exDNA (derived from P. calyculatus mistletoe leaves), and then collected at 0, 1, and 2 h.Bars represent means ± SE (n = 5; nmol per g fresh weight), and the different letters indicate a significant difference at p ≤ 0.05.

Figure 2 .
Figure 2. Polyphenol and flavonoid quantification in mesquite leaves treated with exDNA sources.Polyphenol (a), and flavonoid (b) levels.Mesquite leaves were wounded (damage) or sprayed with water for control, or self-exDNA (derived from mesquite leaves), and non-self xDNA (derived from P. calyculatus mistletoe leaves), and then collected at 0, 1, and 2 h.Bars represent means ± SE (n = 5; milligrams per g of dry weight: DW), and the different letters indicate a significant difference at p ≤ 0.05.

Figure 2 .
Figure 2. Polyphenol and flavonoid quantification in mesquite leaves treated with exDNA sources.Polyphenol (a), and flavonoid (b) levels.Mesquite leaves were wounded (damage) or sprayed with water for control, or self-exDNA (derived from mesquite leaves), and non-self xDNA (derived from P. calyculatus mistletoe leaves), and then collected at 0, 1, and 2 h.Bars represent means ± SE (n = 5; milligrams per g of dry weight: DW), and the different letters indicate a significant difference at p ≤ 0.05.

Figure 3 .
Figure 3. Antioxidant enzyme activities in mesquite leaves treated with exDNA sources.Superoxide dismutase (SOD) (a), peroxidase (PEX) (b), catalase (CAT) (c) activities.Mesquite leaves were wounded (damage) or sprayed with water for control, self-exDNA (derived from mesquite leaves) and non-self exDNA (derived from P. calyculatus mistletoe leaves), and then collected at 0, 1, and 2 h.Bars represent means ± SE (n = 5; units of activity per mg of protein), and the different letters indicate a significant difference at p ≤ 0.05.

Figure 3 .
Figure 3. Antioxidant enzyme activities in mesquite leaves treated with exDNA sources.Superoxide dismutase (SOD) (a), peroxidase (PEX) (b), catalase (CAT) (c) activities.Mesquite leaves were wounded (damage) or sprayed with water for control, self-exDNA (derived from mesquite leaves) and non-self exDNA (derived from P. calyculatus mistletoe leaves), and then collected at 0, 1, and 2 h.Bars represent means ± SE (n = 5; units of activity per mg of protein), and the different letters indicate a significant difference at p ≤ 0.05.

Figure 4 .
Figure 4. Quantitative JA determination in mesquite leaves treated with exDNA sources.Mesquite leaves were wounded (damage) or sprayed with water for control, self-exDNA (derived from mesquite leaves), and non-self exDNA (derived from P. calyculatus mistletoe leaves), and then collected at 0, 1, and 2 h.Bars represent means ± SE (n = 5; nanograms per mg of fresh weight), and the different letters indicate a significant difference at p ≤ 0.05.

Figure 4 .
Figure 4. Quantitative JA determination in mesquite leaves treated with exDNA sources.Mesquite leaves were wounded (damage) or sprayed with water for control, self-exDNA (derived from mesquite leaves), and non-self exDNA (derived from P. calyculatus mistletoe leaves), and then collected at 0, 1, and 2 h.Bars represent means ± SE (n = 5; nanograms per mg of fresh weight), and the different letters indicate a significant difference at p ≤ 0.05.

27 Figure 6 .
Figure 6.exDNA sources induce differentially relative gene expression of MAPKs in mesquite leaves.MAPK2 (a), MAPK4 (b), and MAPK10 (c) gene expressions.Mesquite leaves were wounded (damage) or sprayed with water for control, self-exDNA (derived from mesquite leaves) and non-selfexDNA (derived from P. calyculatus mistletoe leaves), and then collected at 0, 1, and 2 h.Bars represent means ± SE (n = 5), and the different letters indicate a significant difference at p ≤ 0.05.

Figure 6 .
Figure 6.exDNA sources induce differentially relative gene expression of MAPKs in mesquite leaves.MAPK2 (a), MAPK4 (b), and MAPK10 (c) gene expressions.Mesquite leaves were wounded (damage) or sprayed with water for control, self-exDNA (derived from mesquite leaves) and non-selfexDNA (derived from P. calyculatus mistletoe leaves), and then collected at 0, 1, and 2 h.Bars represent means ± SE (n = 5), and the different letters indicate a significant difference at p ≤ 0.05.

Figure 7 .
Figure 7. Responses of P. laevigata to the exogenous exDNA applications.O 2•− accumulation in the leaf represented by violet coloration, H2O2 accumulation in the leaf represented by red coloration, arrows = induction, perpendicular lines = inhibition, white squares = basal level of phytohormones and blue triangles = increase in phytohormone content (See details in text).

Figure 7 .
Figure 7. Responses of P. laevigata to the exogenous exDNA applications.O 2•− accumulation in the leaf represented by violet coloration, H 2 O 2 accumulation in the leaf represented by red coloration, arrows = induction, perpendicular lines = inhibition, white squares = basal level of phytohormones and blue triangles = increase in phytohormone content (See details in text).
•− is reduced to H 2 O 2 by SOD, and H 2 O 2 is decomposed into H 2 O and O 2 by PEX and CAT