Mother plants of the rootstock cultivar Rosa canina ‘Pfänder’ (clone internal code R02-6) were used in this study. These plants were grown in commercial peat substrate (Steckmedium 304: 20% black peat, 70% white peat, 10% perlite, Klasmann-Deilmann, Germany) fertilized with 100 g/m³ of Radigen (Terraflor, Germany) containing 2.0% Fe, 1.5% Cu, 1.0% Mn, 0.8% Mo, 0.6% B, 0.5% Zn, and 5.0% MgO. Stock plants were maintained at 20°C ± 2°C during the day and 16°C ± 2°C at night in the greenhouse at a relative humidity of 70% to 80%. Plants received additional light from sodium lamps (SON-T 400 W, Philips, The Netherlands) which produced an average light intensity of 630 µmol   m − 2   s − 1 (PPFD) for 8 h d^-1. Plant material was harvested at the end of October in 2022.

The laser marking process was implemented using a continuous wave CO₂ laser (BRM Pro 600, 80 W power, Netherlands) assisted by a rotary axis (CNC 80 mm Axis Rotary Chuck, LY. Group, China). The laser beam was focused through a 20 mm diameter ZnSe lens (Cloudray, USA), generating a Gaussian beam profile and a laser beam of 240 µm diameter. In order to evaluate the effect of wound position relative to the axillary bud and the leaf, three distinct sections were defined at the bottom part of the cutting: the section above the cutting base on the axillary bud or leaf-bud side (S_LS); the section above the cutting base opposite to the leaf-bud side (S_OS); and the section at the cutting base (S_CB) ( Figure 1 ). Two laser treatments were established depending on whether the additional laser wound was located at S_LS or S_OS. Laser treatment on the leaf-bud side is abbreviated in the following with L’_LS, while laser treatment on the opposite side with L’_OS. Cuttings without laser treatment (control) were named Ct. The geometry of the laser wound pattern for both treatments was a strip, 10 mm long and 1.0 mm wide, applied around 5 mm above the cutting base. The specific laser energy density required to expose the sclerenchyma, the tissue layer closest to the phloem, was based on the methodology by Morales-Orellana et al. (2022). The average cutting diameter was 2.5 mm ± 0.25 mm.

Rooting was carried out under aeroponic conditions, where each aeroponic growth box (measuring 60 cm x 15 cm x 40 cm) was equipped with an independent water pump (WP 25 W, Cadrim, China) containing four Gardena nozzle 360° sprinklers (Germany). The cuttings’ bases were secured using polystyrene rings and received continuous water irrigation through spraying. The pH of the water ranged from 5.5 to 6.5, with an electrical conductivity of 0.18 to 0.24 mS cm − 1 . No exogenous auxin was applied. In the growth chamber, temperature was maintained between 20°C and 23°C, an average light intensity of 630 µmol m^-2 s^-1 (PPFD) was provided by SON-T 400 W sodium lamps (Philips, The Netherlands) for 14 h per day, and relative humidity kept at 80% to 100% through a WHQ 12 head atomizer (DMWD, China) running for 10 min h^-1.

The experiment consists of three treatments ( Figure 1 ): laser treatment on leaf-bud side (L’_LS), laser treatment opposite to leaf-bud side (L’_OS) and cuttings without laser treatment - control (Ct). The experimental setup and timeline are described in detail in Table 1 . Long shoots were harvested from the stock plants using tree nursery scissors, put in a container with water and transferred to the laboratory, maintaining them wet via water spraying. They were divided into single-node cuttings and stored in distilled water before the laser treatment. After the laser application, cuttings were cultivated in aeroponic cultivation boxes. Tissue samples were taken for biochemical analysis, they were transferred immediately after preparation in dry ice and later stored at -80°C. Sampling for biochemical analyses, rooting evaluation and histological characterization were done in a period of 40 days after treatment (DAT). Selections of mother plants, long shoots, and cuttings were randomized for the three treatments.

On each next sampling day (1 DAT, 2 DAT, 4 DAT and 8 DAT) 15 cuttings per treatment were collected randomly from the boxes and partitioned in the three sections S_LS, S_OS and S_CB (3 cuttings were pooled, resulting in 5 samples per section and treatment or 15 samples per treatment or 45 samples per time point for the biochemical analyses, each comprising at least 70 mg). On sampling day 40 DAT, 45 cuttings per treatment had remained in the 9 boxes for rooting analysis. From these cuttings rooting percentage, absolute number of AR per treatment, number of roots per rooted cutting, average fresh mass per rooted cutting, and the position from which each root emerged was evaluated. The histological characterization at 0 DAT was done with additional 15 cuttings (5 replicates per treatment) and the histological characterization at 40 DAT was done with 12 cuttings (4 replicates per treatment) that had also been evaluated for their rooting response.

To evaluate the level of initial tissue penetration of the laser wounding, additional control cuttings and laser-treated cuttings were analyzed (5 cuttings per treatment) via histology as described previously (Morales-Orellana et al., 2022; Wamhoff et al., 2024). Moreover, histological characterization was done with the cuttings at 40 DAT (4 cuttings per treatment). Samples comprised of the whole stem base (S_LS and S_OS) and of the basal wounded section (S_CB) were fixed in AFE solution (76% ethanol, 1.8% formaldehyde, 20% water, and 5% glacial acetic acid) at 4°C for 24 hours. Dehydration involved a series of ethanol solutions (70%, 80%, 90%, and 96%) under vacuum. The samples were studied following the Technovit^® 7100 polymerization protocol (Kulzer, Germany). The pre-infiltration phase was carried out with a solution of basic Technovit 7100 and 96% ethanol in a 1:1 ratio under vacuum for 2 h. During the infiltration phase, the samples were left overnight in the basic Technovit 7100 solution with 1 g of hardener I per 100 ml of basic solution. The embedding was then performed in teflon molds (Histoform S, Technovit^®) after adding 1 ml of hardener II per 15 ml of the infiltration solution. Samples were sectioned using a microtome (Supercut 2065, Leica, Germany) to a thickness of 8 μm. Subsequently, all samples were stained with the FCA solution according to Etzold (New Fuchsine-Chrysoidine-Astra blue: 1000 ml H2O, 0.1 g New Fuchsine, 0.143 g chrysoidin, 1.25 g Astra blue, and 20 ml glacial acetic acid) (REF 11742, Morphisto, Germany) for approximately 4 min and rinsed with a series of ethanol solutions at 30%, 70%, and 30%. Bark penetration among laser treatments was measured in samples from 0 DAT. Additionally, the area of vascular tissue of stem cross sections in presence and absence of the laser wound was quantified in samples at 0 DAT and 40 DAT. Evaluation was done using a light microscope (BX61, Olympus, Japan) with a digital microscope camera (ColorView II, Olympus, Japan) and quantified with Cell* Imaging Software (Olympus, Japan).

The samples were pulverized in a mixer mill (MM 400 Mixer, Retsch, Germany) in 2 ml reaction tubes with two steel beads for 90 s at a frequency of 30 s^-1. Then, the frozen material was weighed with a precision scale (AB104SA, Mettler Toledo, United States) and divided into two pre-cooled tubes: one with approximately 20 mg for carbohydrate analysis and the second one with around 30 mg for plant hormone analysis. Special care was taken to avoid thawing.

All statistical analyses were performed using the software R 3.3.0 (R Core Team, 2021). AR formation was evaluated with the 135 remaining cuttings in a destructive final evaluation at 40 DAT. The determination of AR formation (yes/no), root number per rooted cutting and root fresh mass per rooted cutting were collected in addition to the number of roots classified by their position relative to the leaf-bud location. For rooting percentage or binomial data, generalized linear mixed models were used assuming a binomial distribution. For the root number per rooted cutting, a generalized linear mixed model was used, while the root fresh mass per rooted cutting was analyzed using linear mixed models. Data of root mass and length are presented as mean values with standard error (SE). The best fitting transformation was identified via the R package bestNormalize (Peterson, 2021). To test for significant differences in mean values between treatments, the nonparametric Kruskal−Wallis test of the R package agricolae was employed (Mendiburu, 2021) followed by Dunn’s post hoc test for pairwise comparisons.

In the histological section, data were subjected to linear model-based analysis, followed by an analysis of variance (ANOVA) of one factor. When significance was observed, mean values were compared using Tukey’s test (P ≤ 0.05) with the R package emmeans (Lenth, 2022). Data of histology is presented as mean values with standard deviation (SD).

For the biochemical analysis, data were subjected to two-factorial analysis of variance (ANOVA) taking treatment and section plus their interaction as factors. Mean values were compared using Tukey's test (Tukey, p < 0.05) with the R package emmeans (Lenth, 2022). This data is expressed as mean values with standard deviation (SD). The dynamics of each compound was presented for stem sections (S_LS, S_OS and S_CB) and adjusted as a heat map using the R package gplots (Warnes et al., 2023). ANOVA results are presented as a table below each heat map. Further graphs were created using the R package ggplot2 (Wickham, 2016).

Histological analyses at 0 DAT showed that both laser treatments reached the sclerenchyma layer and exposed the phloem proximities as intended ( Figures 2A, B ). The whole bark had an average thickness of 0.45 mm ( Figure 2C ). Laser wounding achieved an average depth between 0.25 mm and 0.35 mm for both laser treatments. Therefore, the laser energy applied was able to penetrate the layers of epidermis, collenchyma and parenchyma tissues, until stopping in the fibers of the sclerenchyma, thus, avoiding deeper damage to the vascular tissues (phloem, cambium and xylem). Based on our histological results at 40 DAT, the formation of callus, as well as the radial expansion of phloem and xylem, was necessary for tissue repair and reconnection in the laser wound. A pronounced development of wound-induced xylem and phloem took place directly in the wounded areas produced by the laser pattern ( Figure 2E versus Figure 2D ). Basal wound of laser-treated cuttings presented a robust root system directly connected to the sieve elements of the phloem, while control cuttings established AR connections mostly through callus ( Figures 2F, G ).

The application of both laser treatments resulted in an increase in rooting percentages compared to the control, as well as a significant increase in rooting parameters such as root fresh mass per rooted cutting, but not in the root number per rooted cutting ( Figure 3 ). In general, root formation was observed in the basal section S_CB of the cuttings in all treatments ( Figure 3A ). At the end of the cultivation period, a significantly higher percentage of 40% of the cuttings was rooted when the laser treatment was applied on the leaf-bud side (L’_LS) compared to untreated control cuttings. The rooting percentage of cuttings with the laser treatment L’_OS was 22%, but this value was not significantly higher than that of the control cuttings with 9% rooting ( Figure 3B ). The root number per planted cutting of 0.4 (18/45) roots for the cuttings of laser treatment L’_OS and 0.58 (26/45) for L’_LS was about 4 to 5 times higher than that of 0.11 (5/45) roots recorded for control cuttings ( Figure 3C ). The root distribution schemes showed a tendency for the ARs to be positioned preferentially on the leaf-bud side of the cutting, regardless of the treatment. Interestingly, the average number of ARs per rooted cutting was not altered by the presence or absence of additional wounds, being 1-2 roots per cutting regardless of the treatment ( Figure 3D ). However, one of the most notable features of laser-treated cuttings was the vigor of the root system, exhibiting a robust direct connection to the cutting and higher average fresh mass per rooted cutting compared to the control ( Figure 3E ), in which roots were less developed and accompanied by a greater amount of callus. Overall, the laser treatments stimulated rooting and also the position of the wound had an effect on the rooting response.

The rooting results confirmed our previous reports on cuttings of Rosa canina ‘Pfänder’, where laser-induced tissue damage led to an increase in the percentages of AR formation, even if no auxin was applied (Morales-Orellana et al., 2024). Therein, we showed that AR stimulation was positively and significantly correlated with an increase in vascular tissue, specifically with phloem enlargement. Furthermore, the first changes in the vasculature of the wounded tissue occurred within the first week, with callus growth being faster than vascular tissue growth. In the present study, the histological results showed that laser wounding increased the development of new vascular tissue directly in laser-treated sections as well ( Figure 2E ; Supplementary Figure S2 ). Similar to our results, studies by Matsuoka et al. (2021) reported an increase of wound-induced cambium and thickening of vascular tissue after wounding in flowering stems of Arabidopsis. These authors reported high expression of vascular tissue formation-related genes in these regions. According to Smetana et al. (2019) the regeneration of xylem is necessary to ensure the maintenance of the vascular cambium, making xylem a “stem-cell organizer” layer. Thus, xylem regeneration would open the possibility of extending the cambium layer, which in turn, stimulated phloem development in wounded regions. Consequently, the formation of novel vascular tissue led to a robust root system with ARs directly connected to the sieve elements of the phloem ( Figure 2G ). In microcuttings of Rhododendron, a more robust vascular system and enhanced rooting were related to a larger phloem:xylem width ratio (Dokane et al., 2014). Moreover, a recent report by Wamhoff et al. (2024) mentioned that lignin presence in rose cuttings may represent a mechanical barrier to AR outgrowth and reduce rooting capacity in some genotypes. Therefore, the stimulating effect of the laser treatment might be related to the reduction of mechanical strength by eliminating sclerenchyma fibers, thus, allowing an increase of vascular tissue.

According to Damodaran and Strader (2024), cells close to the basal wound are competent for AR formation, whereas in distal cells this capacity is reduced. The results of our study show that in cuttings containing two wounds at the same time, AR formation is still preserved in the basal wound. Furthermore, the positioning of ARs seems to be related to the vascular connection between the leaf-bud and the basal wound. This phenomenon seems to be related to a higher physiological activity in this cutting section as evidenced by the synthesis of new xylem on the leaf-bud side in control cuttings ( Figure 2 ), or the permanent higher levels of free IAA in the stem section below the axillary bud in all treatments ( Figure 6 ). These results suggest that additional laser wounds may impact long-distance signaling, while the vascular connection between the leaf-bud and the basal wound is permanently more active, and therefore, a site with greater competence for AR formation. Laser-treated cuttings displayed an increase in the total number of roots per treatment and root fresh mass per rooted cutting. Moreover, additional analyses showed that ARs from laser-treated cuttings were longer and had more lateral roots ( Supplementary Figure S1 ). These results correspond to those reported by Szajsner and Bąbelewski (2014) for laser treatment of Viburnum rhytidophyllum cuttings, and by Morales-Orellana et al. (2024) in leafy stem cuttings of rose under different laser pattern designs. Recent studies in wounded tissues of Arabidopsis mentioned that local water availability determined regeneration fates (Kareem et al., 2024). According to these authors, higher water availability regulated auxin response maxima which triggered de novo root formation, while low water availability gave rise to wound-induced callus, reduced spatial auxin response and inhibited root regeneration. Therefore, the stimulation of AR formation reported in our study may also be related to a higher local water availability through laser wounding.