Glechoma longituba (Nakai) Kuprian is a perennial herb widely distributed in the forest, on the roadside, and by creeks of tropical, subtropical, and temperate areas in China. The monopodial stolons are able to creep on the ground, and ramets can develop on each stolon node (Quan et al., 2021). Each ramet produces two petioles and leaves. The blades are heart-shaped, with margins bearing rounded teeth ( Figure 1 ). G. longituba has high phenotypic plasticity in response to light (Liu et al., 2015; Tie et al., 2022). In addition, for the rapid clonal propagation, G. longituba forms a complex ramet network; touch between ramets was inevitable.

The G. longituba in our experiment was collected from Jiwozi in the Qinling Mountains, Shaanxi, China. To ensure the uniformity of the genotypes and reduce the impact of the previous environment, the plant materials were collected from the same genet and then were vegetatively propagated for at least 4 months in a greenhouse at Northwest University in Xi’an (397 m a.s.l., 34.3°N, 108.9°E).

A total of 96 healthy and similarly sized (0.78 g ± 0.29 g) ramets were selected as parental ramets, and then were divided into 48 ramet pairs randomly. Each ramet pair was transplanted into a plastic pot (55 cm length × 28 cm width × 5.5 cm depth). The peat soil, perlite, and vermiculite (4:1:1, v/v/v) were mixed and utilized as the culture soil. The greenhouse conditions were a 24/20°C day/night temperature cycle, a 12/12-h light/dark cycle, and a 300 μmol m⁻² s⁻¹ photosynthetic photo flux density (PPFD) during the light cycle. PPFD levels were measured with a Quantum Metre (LQS-QM, Apogee Instruments Inc., USA). The relative humidity was maintained at 40%. Plants were watered into the soil directly every 3 days throughout the experiment and researchers avoided the droplets touching the leaves during watering. To minimize the effects of microenvironment variation in the greenhouse, the position of each pot was changed weekly.

After a week of transplanting, a 45-day experiment was conducted from 1 November 2022 to 15 December 2022. At first, 48 parental ramet pairs were divided into two groups randomly; half was placed under ambient light conditions, and the rest was placed in a shaded environment. Then, half of each group was subjected to touch treatment, and the other had no action. During the experiment, the newborn offspring was placed in an ambient light environment or under shade conditions ( Figure 2 ). Accordingly, the experiment included a total of eight treatments (two parental light conditions × two parental neighboring touch treatments × two offspring light conditions). Each treatment was designed with six replicates.

The shaded environment was achieved by hanging a shading net 20 cm above the top of ramets. The PPFD was about 80–100 μmol m⁻² s⁻¹ in the shaded environment. During the experiment, according to the experiment’s design, if the parent and offspring were in different light environments, the shading treatment was only performed above the parent or offspring.

The touch treatment was conducted between parental ramet pairs. Two petioles of the neighbor parental ramets were waved with a thin wooden stick gently to make the two leaf blades touch. During the treatment, we tried to minimize the bending angle of the petiole to avoid damage. The touch treatment was only performed on parental ramets and continued for 45 days. The touch was handled twice a day at 9:00 a.m. and 6:00 p.m., respectively; three contacts were made each time. To ensure the uniform touch intensity during repeated action, all the touch treatments were controlled by the same person. Furthermore, during the experiment, the distance between parental ramet pairs was maintained at 5 cm, a distance that is long enough for leaves of ramet pairs to not touch each other naturally during growth except for human-facilitated touch.

At the end of the experiment, we performed growth architecture and biomass measurements. For parents, the blade area and petiole length were first measured. Then, biomass indexes were measured, including total biomass, aboveground biomass (leaf biomass and node biomass), and root biomass. Finally, the following indicators of parental ramets were estimated: SLA (leaf area per leaf mass), specific petiole length (SPL; length of petiole per petiole biomass), leaf area ratio (LAR; blade area per total biomass), and leaf mass ratio (LMR; blade biomass per total biomass).

For offspring, the length of the longest stolon, total blade area, and the number of total ramets were recorded. After that, total biomass, leaf biomass, and stolon biomass were measured. The following indicators were estimated: specific stolon length (SSL; length of stolon per stolon biomass), stolon biomass/leaf biomass (S/L), LAR (blade area per total biomass), and LMR (blade biomass per total biomass). The comparison table of abbreviations is shown in Table 1.

Length of petiole and stolon was measured with a vernier caliper and tapeline, respectively. Blade area was calculated with Motic software (Motic Images Plus 2.0. Ink, Motic, China) after being scanned with a scanner (Perfection V19, EPSON, China). Biomass was measured after the samples were dried at 80°C for 48 h to a constant weight with an electronic balance (Sartorius BT25S, Beijing, China).

Before statistical analyses, to meet the assumptions of homoscedasticity and normality, the following data were subjected to logarithmic transformation: total biomass and root biomass of parent, and total biomass, stolon biomass, and S/L of offspring. We analyzed the integrated differences of phenotypic and growth among treatments of parental ramets and offspring, respectively, using a PERMANOVA (McArdle and Anderson, 2001), where parental neighboring touch treatment (NT; touch or not), parental light environment (PL; ambient or shade), offspring conditions (ambient or shade) and their interactions, and the covariates (initial fresh weight of parental ramet) were used as predictors, estimating the significance value by 999 permutations. Then, we conducted ANCOVAs to analyze the effects of PL, offspring light environment (OL), NT, and interaction regimes on all traits with parental fresh weight as a covariate. The LSD test was chosen as the method of multiple comparisons to test the significance among different treatments. Any effect of parental conditions, either direct (PL or NT) or in interaction with the offspring conditions (PL × OL, NT × OL, PL × NT × OL) that remained significant after removing the linear part of the maternal investment (initial fresh weight of parent), was considered a maternal effect (Bolker et al., 2009; Galloway et al., 2009; Puy et al., 2022). All data analyses and diagrams were carried out using R (v.3.2.3, R Core Team, 2016) with α = 0.05 as significance level. The images were further processed with Adobe Illustrator (v 28.1, Adobe Systems Incorporated, 2023).

For parental ramets, there were significant effects of PL, OL, and NT on biomass (including total, aboveground, and root biomass). Additionally, PL affected SLA, SPL, LAR, and R/S; OL showed an effect on the blade area, SPL, LAR, LMR, and R/S, while NT influenced SLA, LAR, and petiole length; the interaction of PL and OL affected the blade area, petiole length, LMR, LAR, and R/S significantly; PL × NT influenced the blade area, SLA, and LAR. Moreover, OL × NT and PL × OL × NT did not show an effect on parents ( Table 2 ).

When the genet (including parent and their offspring ramets) was in ambient light, neighboring touch (NT) mainly reduced the blade area and the petiole length of parental ramets but had no significant effect on other parental parameters. Additionally, compared with the ambient environment, the growth of genet in a shaded environment was depressed. Parental ramets displayed reduction in the blade area, biomass (aboveground, roots, and total biomass), and R/S, but increased SPL, LAR, and LMR. However, in the shaded environment, if touch between neighboring parental ramets happened, some traits induced by shade were changed. Upon recovery from the reduction in the parental blade area, SLA and LAR even became larger ( Figure 3 ).

For the offspring ramets, PL and OL displayed effects on almost all traits involved in our study, except for OL on stolon biomass. Neighboring touch treatment (NT) had effects on most traits except for stolon biomass/leaf biomass and LMR. Moreover, the PL × OL interaction affected the blade area, leaf biomass, and SSL of the longest stolon; PL × NT influenced total biomass, stolon biomass, blade area, and stolon/leaf biomass, but OL × NT only affected leaf biomass. Same as the parent, PL × OL × NT did not show an effect on the offspring ( Table 3 ).

When the genet was in ambient light, in offspring, parental neighboring touch decreased total leaf biomass and blade area, but enhanced S/L. When the genet was in a shaded environment, as compared with the ambient environment, shade leads to a reduction in biomass (stolon, leaf, and total biomass), the total blade area, and the length of the longest stolon. In contrast, there is an increase in SSL and S/L. However, if touch between neighboring parental ramets happened, it caused a greater decrease in the total biomass, stolon biomass, and length of the longest stolon than in those without touch ( Figure 4 ).

We compared the offspring performance in the same light environment as their parents or in a different light environment ( Figure 4 ). When the parental ramets were in ambient light, the offspring with a shaded environment did not show a significant difference from offspring with ambient light in the total biomass, stolon biomass, and length of longest stolon. However, they had a lower number of ramets, leaf biomass, blade area, LAR, and LMR, and higher SSL and stolon biomass/leaf biomass. Additionally, if the parental ramets are in a shaded environment, even if their offspring grew in an ambient environment, these offspring will still have a lower leaf, stolon, and total biomass, a smaller blade area, a shorter stolon length, but a larger SSL. In contrast to the offspring with ambient light, the offspring with a shaded environment will have the lowest total biomass, leaf biomass, blade area, LAR, LMR and stolon length, but the largest SSL.

When parents grew in ambient light, the following traits of offspring with ambient light were changed by parental neighboring touch: leaf biomass, leaf area, ramet number, LAR, and LMR were decreased, and stolon biomass/leaf biomass was increased. Total biomass, stolon biomass, SSL, and stolon length did not change significantly. For the offspring with a shaded environment, touch just reduced leaf area and LAR; others were unchanged ( Figure 4 ).

When parents grew in a shaded environment and offspring grew in ambient light, except for ramet number, stolon biomass/leaf biomass, LAR, and LMR without any changes, SSL was increased and all other traits decreased due to parental neighboring touch. For offspring with a shaded environment, however, leaf biomass, leaf area, ramet number, stolon biomass/leaf biomass, SSL, LAR, and LMR were not influenced by parental neighboring touch, with only total biomass and the length and biomass of stolon decreased ( Figure 4 ).

In the present study, the growth of the clonal plant G. longituba was affected by the touch between neighbor parents and not only parental ramets but also their offspring ramets. However, the effect varied with a light environment. For example, in an ambient environment, it seems that the effect of parental neighboring touch was mainly on leaves; the reduction in the blade area and petiole length of the parental ramets was the most intense phenotypic change observed after direct touch by a neighbor, which is also regarded as the core morphological change of thigmomorphogenesis (Brenya et al., 2022). The compression of leaf and petiole was associated with the increases in cell wall stiffness and decreases in cell elongation, which is correlated with ethylene-regulated pectin degradation induced by touch (Wu et al., 2020). Being a clonal herb with a long stolon, the petiole is the vertical support of G. longituba; a shorter petiole limits the bending moment, and together with a smaller blade, they reduce the risk of various mechanical strains, such as plastic deformation, uprooting, and buckling, and are likely adaptive in crowded vegetation (Langer et al., 2021; Telewski, 2021; Langer et al., 2022).

Parental environmental signals that can be perceived by clonal offspring via a connected spacer (stolon or rhizome) have been proven (Liu et al., 2015; Guo et al., 2022; Tie et al., 2022; Xue et al., 2022). Moreover, in this study, although touch only took place between parents, its impact was transferred to their offspring, causing the decrease of ramet number, leaf area, leaf biomass and leaf investment (decreased LAR and LMR), but relatively less reduction in stolon biomass (increased S/L). The performance of offspring ramets induced by parental neighboring touch owing to clonal integration was also proved with Leymus secalinus (Sui et al., 2011). The sharing of environmental signals among interconnected ramets was considered to favor young ramets (Wei et al., 2019). In this study, to some extent, the compression of phenotype and growth, and more allocation to the stolon, those induced by touch benefits plants in withstanding mechanical stress, which is also the form of plants in high mountain and grazed areas to protect plants from the stresses caused by high wind, rain, physical touch from other plants, or trampling (Braam, 2005; Sui et al., 2011).

When G. longituba was in a shaded environment, growth was obviously inhibited, as shown by the lower biomass accumulation and ramet production, which was due to the limitation of light resources. Shade induced a significant impact on both leaf and stolon. Despite the decreased biomass, some changes in parental ramets caused by shade were in favor of adapting to low-light conditions. For instance, notwithstanding the smaller leaf area, more investment to leaf in parental ramets showed increased LAR and LMR, which optimized light capture in the shade. Moreover, the longer SPL in parents and SSL in offspring represented expansion in the vertical and horizontal directions, which suggested an effort of clonal plants to escape from the shade ( Figures 3 , 4 ). Evidently, the clonal plant G. longituba exhibited both STA and SAS characters to tolerate the shaded environment.

Interestingly, if parental neighboring touch happened in the shade, it seems that some responses of G. longituba were more beneficial to survival in the shaded environments. For example, the reduction of parental leaf area in the shade was offset, and even more resources were allocated to the leaf area, causing larger SLA and more LAR ( Figure 3 ). Additionally, although the length and biomass of stolon decreased, the long SSL and SPL was maintained the same as in the shade ( Figure 4 ). In addition, the touch of the parent caused larger SLA in the shaded offspring, and these responses were not even represented in the shaded environment. Consequently, SAS or STS responses induced by shade were not restrained by parental neighboring touch and, instead, were promoted to some extent.

The connected stolon provides physical support for communication among ramet nets and guarantees the parental and ongoing effects on offspring ramets. The maternal effect and the light adaptation of clonal offspring induced by it have been discussed in many studies. Some documented that the phenotype induced by the maternal effect may facilitate offspring to pre-adapt to a light environment (Guo et al., 2022; Tie et al., 2022; Xue et al., 2022). The alternative view, however, stated that parental shading effects contributed little to the tolerance of clonal offspring to shading (Dong et al., 2019). Our results displayed that a parental high-light environment was in favor of offspring performance. If the parental ramets were in ambient light, their offspring always maintained a higher total biomass and ramet number regardless of OL. The assurance in total biomass of the shaded offspring was mainly related to stolons. By contrast, the offspring with shaded parents usually had a lower biomass even if they were under ambient light conditions. In brief, it seemed from our results that the offspring with a shaded environment did not benefit from the shading experience of their parent despite the matching parent–offspring environment. However, plants will avoid excessive growth or defense through a negative feedback–regulatory loop and achieve balance in response to adverse environments (Li et al., 2019), More precisely, the decrease in growth of G. longituba does not necessarily mean that they are completely unadaptable; resources may be more devoted to other traits not involved in the present study, such as defense and life span.

If parental ramets are in an ambient environment, the offspring with a shaded environment showed similar growth whether touch occured or not, and they both exhibited the same biomass (leaf, stolon, and total biomass), ramet number, LMR, SSL, and stolon biomass/leaf biomass. Instead, if touch happened when parents grew in a shaded environment, the biomass of offspring with a shaded environment was decreased, accompanied by a reduction in length and biomass of stolons. Of course, the longest SSL and a high S/L were still kept ( Figure 4 ). These traits responded in opposite directions, suggesting the potential for complex trade-offs among traits. Therefore, the effects of parental neighboring touch on offspring under shaded conditions depended on PL. If light resource in the parental environment was abundant, the impact of parental neighboring touch on the offspring with a shaded environment was not obvious, whereas when the parent was under limited light conditions, touch from a neighbor was more disadvantageous to the growth of the clonal plant G. longituba. The interaction of mechanical stimuli and shade has been reported by several studies; some documented that shading decreased or eliminated the thigmomorphogenesis (Henry and Thomas, 2002), while others showed opposite results (Feng et al., 2019). These results disclosed a complexity of interaction between shade and mechanical stimuli on plant morphology, growth, and allocation. Note that the mechanical stimuli in these studies were simulated with wind; the effect not only is a single mechanical stimulus, but also affects the exchange of heat, water vapor, and CO₂ around leaves (Feng et al., 2019), which made the findings more diverse. Our study focused on the effect of contact between neighboring leaves, which broadens our understanding of mechanical stimuli.