Transformation of hard pollen into soft matter

Pollen’s practically-indestructible shell structure has long inspired the biomimetic design of organic materials. However, there is limited understanding of how the mechanical, chemical, and adhesion properties of pollen are biologically controlled and whether strategies can be devised to manipulate pollen beyond natural performance limits. Here, we report a facile approach to transform pollen grains into soft microgel by remodeling pollen shells. Marked alterations to the pollen substructures led to environmental stimuli responsiveness, which reveal how the interplay of substructure-specific material properties dictates microgel swelling behavior. Our investigation of pollen grains from across the plant kingdom further showed that microgel formation occurs with tested pollen species from eudicot plants. Collectively, our experimental and computational results offer fundamental insights into how tuning pollen structure can cause dramatic alterations to material properties, and inspire future investigation into understanding how the material science of pollen might influence plant reproductive success.

1) The main result of this work is the discovery that treatment of defatted pollen grains with KOH substitutes the methyl groups for hydroxyl groups in the pectins of the intine wall. I was left with several questions about this process and the data that were presented.  (20), 16533-16539). In the earlier work, sunflower pollen grains are treated with 6% KOH for 12hrs while in the current manuscript the treatment is with 10% KOH first for 2hrs and then for up to 12hrs. To my surprise, the conclusion of the 2016 paper, where a weaker solution of KOH was used, is: "The SECs isolated from alkaline lysis are completely damaged and lose their unique spiky microstructure". Scanning electron micrographs of the resulting material ( Fig. 1b in the 2016 paper) show an amorphous mass of material. In contrast, the conclusion of the current manuscript is: "Taken together, the findings reveal that incubating pollen in alkaline conditions for extended periods of time (replicating classical soapmaking22) results in the transformation of hard pollen grains into pliable, soft microgel particles". Although the authors talked about potential degradation of the pollen grains after long exposure to KOH, Fig. 1d of the manuscript shows that there are still pollen grains after 12hr since they have measured their diameter. Also, the so-called SEC shown in the current manuscript have the "spiky microstructure" of untreated pollen grains. I have trouble understanding why the results are so different and why the authors have not cited their own paper and explained the discrepancies.
(b) The experimental protocol used in this manuscript involves two successive treatments with KOH, one for 2hrs and a subsequent one for 0 to 12hrs. All other experimental conditions seem identical. Despite the similarity between the two treatments, the authors claim that their final outcomes are different. The first 2hr treatment would extract the shell (i.e. remove the pollen content and "expose pectin epitopes") while the second treatment of variable duration would substitute the methyl groups for OH groups ("extensive demethylation of exposed pectin molecules"). Supplementary Fig. 2 makes the same claim graphically. It is not clear why the authors make such as stark distinction between the net result of two identical and successive treatments with KOH. Why take such a categorical view of the experimental protocol? Is it not possible that substitution (demethylation) of pectins begins with the first KOH treatment? Is there not indication of that in Suppl.  Fig. 2c. At first, I interpreted Fig. 2c as statistics on MANY pollen grains that were treated with KOH in the same way and then exposed to rapid changes in pH. I also assumed that the area of pollen grains was normalized by the area at pH=2 which would explain why the standard deviation is zero for the Area Swelling Ratio of 1. However, I note now that the standard deviation also tends to zero a pH = 12 (Area Swelling Ratio of about 1.8). This result would certainly not be expected if MANY pollen grains are used to produce this curve and the normalization is done only with the area at pH=2. Specifically, it is extremely unlikely that many pollen grains would swell to exactly the same area swelling ratio. This curve must be explained better and ideally replaced with one that has proper statistics on a population of pollen grains.
(d) Suppl Fig. 3 shows JIM5 labelling of pollen grains treated with KOH for different time durations. Why is JIM5 labelling nearly absent at 6 and 12 hrs if the pollen grains are still able to swell after such exposure? Where are the de-esterified pectins in those treatments? Also note that the 0hr treatment shows extensive JIM5 fluorescence suggesting that the 2hr KOH pretreatment already started the de-esterification process.
(e) Finally, some controls seem necessary. For example, have the authors treated their pollen grains with PME to remove the methyl groups enzymatically? This simple control experiment would help solidify the claims made in the first part of the manuscript (Figs. 1 and 2). Also, I would have liked to see in Fig. 2c the effect of pH on defatted pollen grains that have NOT been treated with KOH. It seems the authors have done this experiment since they show pollen grains with this treatment in Fig. 3c ME/I = 3. The pollen grains do show substantial swelling with pH although not as much as the KOH-treated one. Fig. 2c must include proper statistics for the pH swelling of KOHtreated and non-treated pollen grains.
2) A mechanical analysis follows the swelling analysis of KOH-treated pollen grains. At this point, the manuscript becomes more obscure and I am concerned that the authors have not captured properly the mechanical behavior of the pollen grains.
(a) I am concerned about the validity of the protocol used to infer the elastic modulus of the exine and intine. According to the authors, intact pollen grains were used to measure the modulus of the exine. In this configuration, it is not clear how the specific modulus of the exine can be disentangled from the large scale deformation of the entire pollen grain under the force applied by the AFM. On the other hand, the modulus of the intine was measured on open fragments. Presumably the intine is still bound to the exine. How can the authors claim to have measured the elastic modulus of the intine if they are probing a thin layer of material that is still mechanically connected to the exine? Moreover, the intine wall is probably anisotropic in its structure.
(b) I could not find what was the tip radius of the AFM probe, nor the final displacement of the probe. In fact, the authors do not present the force-displacement curves that are normally used to infer the elastic properties of materials. The authors only mention that "the approach speed of the AFM cantilever was fixed to 0.8 μm/sec with a maximum loading force of 4.8 μN." Measuring material properties of soft elastic materials with an AFM is not a simple task when working with complex structures such as the pollen grain. The authors present their results as if they were routine measurements. A better description of the AFM measurements and some controls are necessary to validate the moduli reported in the manuscript.
(c) During harmomegathy, the volume of the pollen grain increases substantially but this volume increase is achieved in many species not by stretching the exine but by unfolding or unrolling it. In other words, swelling of pollen grains during harmomegathy does not necessarily involves stretching of the exine. I believe some unfolding or unrolling of the exine is seen in the pollen grains used to illustrate the simulations in Fig. 3c, as indicated by the change in curvature of the exine segments. The case of M_E/I = 1.6 in Fig. 3c is particularly striking because the normally convex exine segments become flat or even concave as the pollen swells. Thus the exine shell is unrolling to accommodate the increase in pollen grain volume. The finite element simulations are presented and implemented as if the exine is simply a hard shell that much be STRETCHED to allow swelling of the pollen grain. This is not a fair assumption. In fact, close inspection of the simulations for M_E/I = 1.6 reveals that the simulations are not reproducing faithfully the deformation of the pollen grains. If the swelling during harmomegathy is achieved without much softening or stretching of the exine, is it not possible that KOH-treated pollen grains can also swell, at least to some extent, without stretching the exine? If this is the case, the resistance offered by the exine is not a simple stretching deformation but one that involves also bending of the exine shell.
(d) The authors should define more clearly what they mean by stiffness. Are they talking about the Young's modulus of the material or do they also include the thickness of the exine and intine shells as part of their definition. Also, at some point in the manuscript, the authors say: "Moreover, atomic force microscopy (AFM) measurements revealed that prolonged alkaline treatment caused significant loss in mechanical strength of the exine layer (Extended Data Figs. [16][17]." The word "strength" in mechanics refers to the maximal stress that can be supported by a material. The authors have inferred, at best, the elastic modulus of the material not its strength.
Based on these comments, I believe the claim that "these findings reveal that a softened exine layer plays an important role in modulating the swelling behavior of the microgel particles by allowing greater swelling of a hydrated intine." has not been established.
3) The authors must be more careful with the claims they are making and the references or results they are citing in support of those claims. I'm listing some examples below.
(a) "This chemomechanical approach, inspired by classical soapmaking22, mimics key aspects of angiosperm reproduction pathways while achieving material properties not found in nature." It is not clear how the KOH treatment mimics key aspects of angiosperm reproduction. I suppose the authors are referring to the enzyme pectin methylesterase involved in remodeling the pectin structure of growing pollen tubes. If the mimicking is limited to this, the parallel is quite remote and their results, although interesting on their own, do not inform us in any concrete way about angiosperm reproduction. The papers do not support the claim made. Also, cellulose microfibrils do not form "tubules". Their representation in Fig. 1b and Fig. 2a,d has open tubes is misleading.
(d) "Biologically, this enzymatic activity is spatially controlled within the pollen wall structure, confounding prior efforts to engineer pollen structures with tunable material properties17, 18 Figure 2d does not establish this critical claim and, as far as I can tell, the reference cited does not either.
(f) A reference is needed for this statement "The pH-dependent swelling response is analogous to that of ionizable polymer networks whose time scales for de-swelling occur similarly within seconds." (g) "Our investigation of pollen grains and spores from across the plant kingdom reveals that microgel formation is restricted to pollen from eudicot plants." This statement is made on the basis of only three non-eudicot pollen grains or spores: Lycopodium, pine, and cattail. This is not enough to exclude ALL non-eudicots. Moreover, the cattail pollen shows significant swelling (Extended data Figure 30). The area increase would be 50% given that the diameter increases from 20µm to 25µm. ROM Fig. 1. Comparison of two alkaline processing methods and microscopic characterization of the processed pollen samples. Key differences are stirred vs. static incubation, incubation time, washing solvents, centrifugation, and drying conditions. For example, in additional control experiments, we verified that KOH incubation in static vs. stirred conditions had a significant effect on particle morphology, as indicated in ROM Fig. 2. Pollen incubation in 10% KOH at 80 °C for 12 h under stirred conditions yielded damaged, fragmented particles while pollen incubation in 10% KOH at 80 °C for 12 h under static conditions yielded intact particles without morphological damage. On a related note, we agree with the Reviewer that it is useful to cite our past work that tangentially describes another alkaline processing method (we overlooked its citation because the previous work was focused on SEC production and yielded fragmented particles) in order to provide context to the readership and also emphasize that the specific details of the processing protocol are important to control the processing outcome. In the revised manuscript (pg. 16, lines 349-353), these points have been added as follows: "While alkaline processing conditions can be used to convert hard pollen grains into soft microgel particles, not all alkaline processing protocols work and different results such as particle fragmentation (see, e.g., Ref. 32) can occur depending on processing parameters such as incubation time, static or stirred incubation conditions, washing solvents, and drying procedures." (b) The experimental protocol used in this manuscript involves two successive treatments with KOH, one for 2hrs and a subsequent one for 0 to 12hrs. All other experimental conditions seem identical. Despite the similarity between the two treatments, the authors claim that their final outcomes are different. The first 2hr treatment would extract the shell (i.e. remove the pollen content and "expose pectin epitopes") while the second treatment of variable duration would substitute the methyl groups for OH groups (" extensive demethylation of exposed pectin molecules"). Supplementary Fig. 2 makes the same claim graphically. It is not clear why the authors make such as stark distinction between the net result of two identical and successive treatments with KOH. Why take such a categorical view of the experimental protocol? Is it not possible that substitution (demethylation) of pectins begins with the first KOH treatment? Is there not indication of that in Suppl. Reply: We sincerely thank the Reviewer for this comment. First, we would like to address the Reviewer's comment that the "experimental conditions seem identical" aside from the incubation time format. As mentioned in our reply to the previous comment, there are other significant and important differences between the current protocol such as the static vs. stirred incubation which can affect the processing outcomes (cf. ROM Figure 2) along with washing steps and drying protocol. Thus, while the protocols may appear similar at first glance, they lead to different processing outcomes in a highly reproducible manner.
As for the Reviewer's critical remark about the "categorical view of the experimental protocol," we appreciate the Reviewer's suggestion and agree that the description could be rephrased, especially as it pertains to the two KOH incubation processing steps. As the Reviewer points out, we fully agree that substitution (demethylation) of pectin molecules can begin during the first KOH treatment step (as indicated in Supplementary Figure 3). We apologize for any confusion about this point and have clarified the schematic illustration in Supplementary Figure 2 to more clearly represent the two KOH incubation steps, including describing the intine layer as being constituted of 'weakly' and 'highly' de-esterified intine rather than giving the impression of fully esterified or de-esterified intine. The revised figure is presented below:

Supplementary Figure 2 | Schematic illustration of pollen microgel fabrication process.
Microgel fabrication process involves the following four processing steps (as indicated by arrows): (1) Treatment with organic solvent to remove excess lipid coating; (2) Pollen incubation in alkaline conditions for 2 h at 80 °C under stirring condition ("Pollen shell extraction"); (3) Extended incubation in alkaline conditions at 80 °C for up to 12 h ("Pollen microgel formation"); and (4) Gelation of pollen microgels induced by water rinsing.
Along this line, we verified that the JIM5 and JIM7 antibody labeling experiments support that deesterified pectin is already present in the pollen shells after the first 2 h KOH treatment (expressed as "0 h" results) while longer incubation times can result in more extensive de-esterification (see Supplementary Fig. 3 below from revised manuscript). More detailed analysis of the antibody labeling, especially with respect to the weakened intensity of the labeling markers with greater incubation time (from 6 h onward) are explained in our response to comment (d) below.
" (2) Fig. 2c. At first, I interpreted Fig. 2c as statistics on MANY pollen grains that were treated with KOH in the same way and then exposed to rapid changes in pH. I also assumed that the area of pollen grains was normalized by the area at pH=2 which would explain why the standard deviation is zero for the Area Swelling Ratio of 1. However, I note now that the standard deviation also tends to zero a pH = 12 (Area Swelling Ratio of about 1.8). This result would certainly not be expected if MANY pollen grains are used to produce this curve and the normalization is done only with the area at pH=2. Specifically, it is extremely unlikely that many pollen grains would swell to exactly the same area swelling ratio. This curve must be explained better and ideally replaced with one that has proper statistics on a population of pollen grains.

Reply:
We thank the Reviewer for this excellent comment and apologize for the lack of clarity. The data in Figure 2c are time-lapsed optical micrographs of individual pollen particles tethered onto a functionalized surface and these experiments were conducted to measure the time scale of microgel particle swelling behavior. For these experiments, the mean ± s.d. are reported from n = 5 particles, as described in the figure legend of the original article. This particle number corresponds to one field of view and we could only focus on one field of view per experiment because we used the highest imaging resolution on our instrument. We repeated the experiment numerous times for each condition and, qualitatively, the results across different repeats agree with the quantitative trend across all experiments. We focused our quantitative analysis on a single image frame (n = 5 particles) because the image processing is very time-intensive and the particle size must be determined manually for each particle across each image frame. Since even n = 5 particles yielded sufficiently clear data and the experiment was repeated many times, we felt that it was sufficient to present these data for understanding the kinetics of the swelling behavior. At the same time, we would like to emphasize that the time-lapsed optical microscopy experiments complemented the main DIPA experiments, which measured the pH-dependent swelling behavior of suspended pollen particles in bulk solution and were conducted on n = 500 particles per condition.
We also understand and appreciate the Reviewer's critique about the similar swelling ratio, however, we would like to emphasize that the pollen particles, as produced by nature, are remarkably monodisperse and have uniform material properties (viz. composition and architecture). For example, DIPA measurements showed that the defatted sunflower pollen grains without KOH treatment have a relatively narrow size distribution, i.e., diameters of 36.8 ± 1.7 and 39.5 ± 1.9 μm at pH 2 and pH 10 conditions, respectively. Also, the 6 h KOH-treated microgel particles have similar levels of uniformity, i.e., diameters of 28.3 ± 1.6 and 49.6 ± 3.4 μm at pH 2 and 10 conditions, respectively (ROM Table 1 in Part 2). As such, the relatively small (but still apparent) measurement errors reported in the time-lapsed optical microscopy experiments after pH equilibration (swelling/de-swelling, first and last 2 s in Fig. 2c) are consistent with the DIPA measurement results. Fig. 3 shows JIM5 labelling of pollen grains treated with KOH for different time durations. Why is JIM5 labelling nearly absent at 6 and 12 hrs if the pollen grains are still able to swell after such exposure? Where are the de-esterified pectins in those treatments? Also note that the 0hr treatment shows extensive JIM5 fluorescence suggesting that the 2hr KOH pretreatment already started the de-esterification process.

(d) Suppl
Reply: We thank the Reviewer for this excellent question. The JIM5 and JIM7 antibodies are commonly used to detect partially methyl-esterified residues, whereby JIM5 and JIM7 recognize weakly and highly esterified pectin molecules, respectively, and both antibodies cannot detect fully de-esterified samples 2,3 .
Thus, the JIM5 signal increases for samples up to around 3 h due to moderate de-esterification compared to the defatted control sample, but then becomes negligible at 6 h once the pectin molecules become fully de-esterified. By contrast, the JIM7 signal is seen for the defatted control in which case the intine is composed of highly esterified pectin molecules, but the signal diminishes due to increasing levels of pectin de-esterification in KOH-treated samples.
Along with the amendments in Supplementary Fig. 3, related statements in the manuscript have been also edited as follows: Pg. 3, lines 60-68 "To verify de-esterification, we conducted immunofluorescence microscopy experiments using the monoclonal JIM5 antibody that recognizes weakly esterified epitopes of pectin 21 (Fig. 1c). The results confirmed successful processing as indicated by increased de-esterification after 3 h incubation while longer incubation periods resulted in more extensive de-esterification of pectin molecules 22 (Supplementary Fig. 3a). Dynamic image particle analysis (DIPA) further revealed that the treated particles swelled from approximately 35 to 43 μm in diameter after 6 h incubation, which was judged to be the optimal condition based on the greatest extent of particle swelling (Fig.  1d). These results were complemented by immunolabeling experiments with JIM7 antibody, which recognizes highly esterified epitopes of pectin 21 (Supplementary Fig. 3b)." Reply: We thank the Reviewer for these excellent suggestions and have performed the requested control experiments: 1. We tested the effect of solution pH on defatted pollen grains that have NOT been treated with KOH. The DIPA measurements (n = 500 particles) and time-lapsed optical microscopy (n = 5 particles) confirmed that the defatted pollen grains are not sensitive to changes in solution pH. This information has been included in the revised manuscript as Supplementary Figure 9 (DIPA measurements) and incorporated in Figure 2c (time-lapsed optical microscopy measurements). The updated figures are presented below along with text revisions.
2. We have also enzymatically treated processed pollen microgel particles with pectinase -which catalyzes the removal of pectin molecules -in order to verify that pectin molecules play a key role in controlling the pollen's mechanical behavior. Upon pectinase treatment, the microparticles had negligible pH-dependent swelling. The data are presented below as ROM Fig. 3.

3.
We also considered performing pectin methylesterase (PME) experiments but decided that, among enzymatic processing control options, the pectinase option was more appropriate for our study design and data analysis strategy. In the pectinase experiments, the pollen particles could be treated using our exact protocol (including the initial 2 h KOH incubation step which is necessary for extracting the pollen shells) and then the pectinase enzyme treatment to remove pectin before conducting the pH-dependent swelling experiments. By contrast, the PME experiments would ideally be conducted whereby PME is the only component that is capable of causing pollen de-esterification but the initial 2 h KOH incubation step is necessary for extracting the pollen shells (thus, there would be pectin de-esterification from a combination of enzymatic and chemical factors) so we decided that the pectinase experiment was a more useful enzymatic control and it verified the importance of pectin molecules in dictating the stimuli-responsive microgel behavior.
Control experiments with defatted pollen grains Pg. 4, lines 87-90 "In marked contrast, when the solution pH was changed from 12 to 2, the microgel particles underwent relatively faster de-swelling or compression while defatted, unprocessed pollen grains exhibited negligible pH-responsive behavior in both directions (Fig. 2c)." Control experiment with enzymatically treated pollen grains ROM Figure 3. Swelling characterization of pectinase-treated pollen microgel particles. pHdependent swelling behavior of pollen microgel particles without or with pectinase treatment was characterized by DIPA measurements. Data are reported as mean ± s.d. (n = 500 particles per condition).
2) A mechanical analysis follows the swelling analysis of KOH-treated pollen grains. At this point, the manuscript becomes more obscure and I am concerned that the authors have not captured properly the mechanical behavior of the pollen grains.

Reply:
We thank the Reviewer for very thoughtful and valuable feedback about the mechanical analysis section of the paper. In the revised manuscript, we have carefully provided more details about the AFM measurement methods along with control experimental results for validation of our measurement approach. We also have clarified our findings on the mechanical behavior of the pollen grains, particularly regarding the stretching mechanism of pollen shells as part of the extensive swelling of pollen microgels.
(a) I am concerned about the validity of the protocol used to infer the elastic modulus of the exine and intine. According to the authors, intact pollen grains were used to measure the modulus of the exine. In this configuration, it is not clear how the specific modulus of the exine can be disentangled from the large scale deformation of the entire pollen grain under the force applied by the AFM. On the other hand, the modulus of the intine was measured on open fragments. Presumably the intine is still bound to the exine. How can the authors claim to have measured the elastic modulus of the intine if they are probing a thin layer of material that is still mechanically connected to the exine? Moreover, the intine wall is probably anisotropic in its structure.

Reply:
We thank the Reviewer for this valuable comment. Our AFM stiffness measurement protocol is based on a recently published literature reference 5 from a top nanomechanical group working on fundamental characterization of pollen grains, and we followed a similar approach while setting the testing parameters according to the following criteria: (1) 10% depth rule; and (2) assumption of isotropic exine/intine layers. We explain each point below: 1. In order to avoid the substrate effect, we followed the 10% depth rule whereby the indentation depth should be less than 1/10 th of the layer thickness. This rule has been well-known for various film stiffness measurements by AFM or nanoindentation 6 . In the dry condition, the average thickness of the exine and intine layers was 0.58 and 0.54 μm, respectively. In the wet condition, the pollen microgel particles swell, where only the intine layer was significantly increased to ~1.5 μm while the exine remained stable (ROM Fig. 4). Likewise, we carefully chose the indentation depth, which could only induce local deformation without large-scale deformation. Thus, we selected to analyze data where the indentation depths were in the range of 20-60 nm (within 10% of 0.58 μm) for dry and wet exine, <50 nm for dry intine (within 10% of 0.54 μm), and <100 nm for wet intine (within 10% of 1.5 μm). Figure 4. Schematic illustration of AFM measurements with pollen microgel particles. a, Nanoindentation experiments on the whole pollen shell consisting of two layers, exine and intine, for the stiffness measurement of exine with indentation depths of 20-60 nm under indentation loads of 3-6 µN. The pollen samples were prepared in both dry and wet conditions. b, The AFM tip can approach either the exine or intine layer depending on the orientational direction of the specimens, which were prepared from pollen shell fragments. The average thickness of the exine layer was ~0.6 µm for defatted and 0 h KOH-treated pollen specimens, and became thinner (~0.5 µm) with increasing KOH incubation time. Therefore, the indentation depths for measuring the exine of fragmented specimens were kept the same as those for the entire sample across both dry and wet conditions. By contrast, the thickness of the intine layer had significant variation between dry and wet conditions (0.5-1.5 µm), and thus the indentation depths were carefully controlled to be below 60 nm for dry samples and 50-100 nm for wet samples.

ROM
Additionally, we also checked all the data outside of those ranges, and confirmed that the retrieved stiffness, or modulus, of the exine or intine layers was very similar to the retrieved data from within the ranges of the indentation depths. This is because we set our indentation depth ranges following the 10% depth rule of the layer thickness in a very conservative manner. Biological samples typically have inhomogeneous thickness distributions, and thus sometimes larger indentation depths can provide valid data. We also tested rate-dependency and tip stiffness effects during the AFM measurements. Regardless of the different measurement conditions, all the values obtained from the tests showed no statistically significant difference (Supplementary Figure 31). Finally, we performed more than 20 indentations per sample for at least five pollen samples.
Supplementary Figure 31 | Effects of various AFM parameters on stiffness measurement of exine and intine layers of dry, defatted pollen grains. a, Representative force-displacement curves from AFM measurements using two types of AFM probes, an aluminum reflex-coated silicon cantilever PPP-NCHR (Nanosensors, Neuchâtel, Switzerland) with a typical spring constant of ~42 N/m and a tip end radius of 5 nm and a diamond cantilever TD26135 (Micro Star Technology, Texas, USA) with a spring constant of ~150 N/m and a tip end radius of ~5 nm. b and c, Young's moduli of exine (b) and intine (c) retrieved from force-displacement curves after AFM measurements using two AFM probes, NCHR and diamond probes at a maximum contact force of 20 µN with setpoint of 6 µN with zero contact time and an approach speed of 0.8 µN/s. AFM measurements were performed using the NCHR probe, varying the maximum contact force (10 to 30 µN), contact time (0 to 1, 3, and 5 s), and approach speed (0.5 to 1.2 µN/s). NS denotes non-significant differences among all measurement conditions. d, Representative forcedisplacement curves of exine layer measurements in the wet condition. e, Representative forcedisplacement curves of intine layer measurements in the wet condition.
In the revised manuscript, this information has been added as follows: Pg. 19-20, lines 500-532  Fig. 31). Thus, measurements were conducted at various positions (more than 16 data points) in a 5 μm × 5 μm area at an approach speed of 0.8 μm/s with the maximum loading force of 4.8 μN and zero contact time. Before experiments, the AFM cantilever was rinsed with water and ethanol, and treated with a UV light cleaner for 30 min in order to remove organic contaminants. The spring constant and sensitivity of the deflection signal were also calibrated by recourse to the thermal vibration of the AFM cantilever 38 by employing the commercial software (XEP, Park Systems). The force-versus-displacement curves were corrected by subtracting the deflection distance of the AFM probe from the total displacement." 2. We assumed that both the exine and intine layers are isotropic in this experiment since we focused on the differences in stiffness between the exine and intine. Also, we used an AFM probe with a tip end radius of 5 nm, and thus the indentation depths were significantly larger than this tip end radius. Thus, the classical Hertz model was used for data analysis, which is consistent with another recent nanomechanical report on pollen grains.
In the revised manuscript, more detailed information about these aspects has been added as follows: Pg. 20, lines 534-542 "For data analysis, we assumed that both the exine and intine layers were isotropic. Also, we used an AFM probe with a tip end radius of 5 nm, and thus the indentation depths (20- (Supplementary Fig. 31). We treated the geometry of the AFM tip as a parabolic model whereby it has a tip radius of Rc, so that the force ( can be expressed as: where E is Young's modulus, ν is the Poisson's ratio, and δ is the indentation depth. The Poisson's ratio ν was set at 0.5, which is typical for natural materials." (b) I could not find what was the tip radius of the AFM probe, nor the final displacement of the probe. In fact, the authors do not present the force-displacement curves that are normally used to infer the elastic properties of materials. The authors only mention that "the approach speed of the AFM cantilever was fixed to 0.8 μm/sec with a maximum loading force of 4.8 μN." Measuring material properties of soft elastic materials with an AFM is not a simple task when working with complex structures such as the pollen grain. The authors present their results as if they were routine measurements. A better description of the AFM measurements and some controls are necessary to validate the moduli reported in the manuscript.

Reply:
We thank the Reviewer for this suggestion. Since one paper was already published regarding the AFM measurement of pollen exine 5 , and our method was mostly consistent with those methods and results, we did not provide the details of our measurement methods and validations in our original manuscript. Nevertheless, we fully agree that we should provide a better description of the AFM measurements and have done so in the revised manuscript. Additional methods descriptions for experiments and data analysis were added too, as described in detail in our response to the previous comment.
) 3 1− 2 (c) During harmomegathy, the volume of the pollen grain increases substantially but this volume increase is achieved in many species not by stretching the exine but by unfolding or unrolling it. In other words, swelling of pollen grains during harmomegathy does not necessarily involves stretching of the exine. I believe some unfolding or unrolling of the exine is seen in the pollen grains used to illustrate the simulations in Fig. 3c, as indicated by the change in curvature of the exine segments. The case of M_E/I = 1.6 in Fig. 3c is particularly striking because the normally convex exine segments become flat or even concave as the pollen swells. Thus the exine shell is unrolling to accommodate the increase in pollen grain volume. The finite element simulations are presented and implemented as if the exine is simply a hard shell that much be STRETCHED to allow swelling of the pollen grain. This is not a fair assumption. In fact, close inspection of the simulations for M_E/I = 1.6 reveals that the simulations are not reproducing faithfully the deformation of the pollen grains. If the swelling during harmomegathy is achieved without much softening or stretching of the exine, is it not possible that KOH-treated pollen grains can also swell, at least to some extent, without stretching the exine? If this is the case, the resistance offered by the exine is not a simple stretching deformation but one that involves also bending of the exine shell.

Reply:
We thank the Reviewer for these valuable comments. We agree that, during harmomegathy, the volume decrease of the pollen shell should occur through closing its apertures or folding its shell structure. Also, during harmomegathy, hydration and dehydration occur depending on the environmental conditions around a pollen grain 10,11 . As Katifori et al. reported in their paper, the unfolding/folding of pollen shells can accommodate this harmomegathy process with minimal energy cost, with which we strongly agree 11 . However, in our current study, we focused on the subsequent "germination" process after harmomegathy is halted. During this germination process, lipid components from the outer layer of pollen grains are gradually dispersed, allowing increased water uptake 12,13 . Through this process, apertures are opened, which accelerates water uptake, leading to germination with significant swelling of pollen grains 13 . Meanwhile, in the exine layer, pollen cements, which occupy the cavity of the sporopollenin network in the exine, are released. Along these lines, we found that our processed pollen microgel samples exhibit key similarities to germinated pollen grains as follows: 1. During germination, the maximum swelling volume of pollen grains should be 1.5-3 times 13 , which cannot be reached by just opening the apertures of pollen grains based on the pollen structure. Particularly, sunflower pollen has spiky shells without any foldable shell design (e.g., wrinkles or folded structure) except for apertures. For instance, 0 h KOH-treated sunflower pollen showed a small change in the equatorial diameter (RpH 10 / RpH 2= ~1.2) from pH 2 to pH 10, whereas 6 h KOH-treated sunflower pollen showed a more significant change in the equatorial diameter (RpH 10 / RpH 2= ~1.8), as shown in ROM Table 1. Also, the spikes restrict extensive folding of the pollen shell during harmomegathy due to structural hindrance. As indicated in Supplementary Figure 9, the areal change in the pollen microgel particle increases with longer KOH incubation time. Thus, the increased volume of pollen grains can be achieved by further expansion of pollen through stretching of the pollen shell. 2. We also found that opening the apertures of pollen microgel particles is a very fast process, but increasing the swelling volume does require some additional time to reach its equilibrium. Thus, in the supplementary figures and videos, during the cyclical pH changes, initial expansion from pH 2 to pH 12 or final compression from pH 12 to pH 2 is relatively faster. We indicated that this aperture opening or unfolding process as reversible harmomegathy when the stigma condition is not sufficient and pollen grains need to restore harmomegathy. Then, after the germination process is initiated, a significant volume change accompanied by structural changes in the exine and intine layers will lead to pollen tube growth 13 .

ROM
3. Since pollen grains are already open to the external environment due to the defatting process and pollen shell extraction, the hydrated intine can facilitate hydration of pollen grains even though the exine is still rigid (Supplementary Figure 9 and ROM Table 1). In this case, the exine only allows aperture opening or unfolding of pollen with minimal volume change (e.g., 0 h KOH-treated pollen). The swelling of intine or water uptake of pollen is significantly constrained by the rigid exine layer. Thus, germination is efficiently ceased or suspended until pollen allows more hydration from the environment (Supplementary 4. Through the second KOH treatment, the exine layer exhibited significant structural changes, becoming more porous (Supplementary Figure 10). During germination, the exine also releases pollen cement or lipid components, increasing cavity size and hydrophilicity in order to accommodate fast water uptake 13 . Moreover, through the AFM measurements, we confirmed that the exine layer became less stiff (Supplementary Figure 31d,e).

Supplementary Figure 10 | Scanning electron microscopy (SEM) images of defatted pollen grains and pollen shells after treatment with 10 w/v% KOH for various incubation times.
Top: SEM image of entire pollen microgels after defatting without or followed by 10 w/v% KOH incubation (2 nd KOH treatment) for 0-12 h. Longer KOH treatment resulted in greater opening of the apertures. Bottom: Surface morphology of exine layers at higher magnification. Exines of the defatted pollen grain exhibited a dense and smooth surface morphology with a few microscale pores around the spikes, while the exine surfaces of KOH-treated pollen are rough and porous with exposed sporopollenin skeleton. The increased porosity of the exine surface was attributed to the release of pollen cement ("pollenkitt") due to KOH treatment 13,28 .
5. The Reviewer pointed out that "the normally convex exine segments become flat or even concave as the pollen swells." We agree that concave or flat pollen shell segments were clearly observed in some cases. We speculate that inhomogeneous material properties or thickness across the pollen shell may induce abnormal shapes during swelling. In future work, we intend to investigate this abnormal swelling mechanism of pollen microgel particles by characterizing the spatial distribution of structural and mechanical measurements of pollen shells from both normal and abnormal swelling pollen specimens. Reply: We thank the Reviewer for the thoughtful comment. We defined stiffness as the modulus value of the exine or intine layer. The thickness ratio of those two layers, exine and intine, is important, and in this study, we treated their apparent dry thicknesses as almost the same based on experimental observations. Also, we agree that "strength" is not a proper description in the sentence. Therefore, we reworded the description of the stiffnesses of the exine and intine layers to instead discuss the moduli of the exine and intine layers throughout the whole manuscript and supplementary information as follows: Pg. 5, lines 107-109

Figs. 11-12)."
(e) Based on these comments, I believe the claim that "these findings reveal that a softened exine layer plays an important role in modulating the swelling behavior of the microgel particles by allowing greater swelling of a hydrated intine." has not been established.

Reply:
We thank the Reviewer for this comment. As we discuss in our responses to previous comments, the results support that a rigid exine layer significantly constrained the swelling of the intine layer, thus suppressing the volume change of pollen microgel particles as indicated in Supplementary Figure 9 and ROM Table 1. We also provided experimental evidence showing the structural change of the exine layer, including increased porosity, in addition to AFM nanomechanical measurements on the exine layer (Supplementary Figure 10). Thus, we believe that our findings on the interplay between the exine and intine layers is still one of the key factors to modulate the swelling behavior of pollen microgel particles.
3) The authors must be more careful with the claims they are making and the references or results they are citing in support of those claims. I'm listing some examples below.
Reply: We thank the Reviewer for this important advice and we have carefully checked the references accordingly. In some cases, we tried to incorporate conceptually related references in order to emphasize what inspired a statement based on work in related fields since pollen materials science is a new and emerging field. We appreciate the Reviewer's efforts to keep the paper citations grounded and we have added more technical references to support specific facts and to complement the conceptually related references. Reply: We thank the Reviewer for his/her critical feedback and we were indeed referring to how the pectin methylesterase enzyme is involved in remodeling the pectin structure of growing pollen tubes. We appreciate that the Reviewer thinks our findings are interesting and also understand that these findings can stand on their own merit while the potential link to angiosperm reproduction is limited at present and not directly tested in this paper. As such, in the revised manuscript, we have decided to rephrase the sentence as follows:

Pg. 1, lines 21-24
"This chemomechanical approach, inspired by classical soapmaking 5   Reply: We agree with the Reviewer that the cellulose-hemicellulose structure is better described as a "microfibril" network rather than a "tubule" structure because the intine layer shows various nanoscale structural features that are entangled laterally to form ribbons, strands, or cables 14 . In the original manuscript, we represented the cellulose microfibril in a rod shape to distinguish it from other components, but we agree that the open tubes should be improved. Thus, in the revised manuscript, we replaced the "tubule" structures with "microfibril" network structures with appropriate literature citations, and modified the relevant figures (Figs. 1b, 2a, and 2d) as follows (additions in yellow highlight): Pg. 2, lines 45-48 "The outermost layer ("exine") is made up of sporopollenin, which is a strong, crosslinked biopolymer 10 , while the inner layer ("intine") is composed of elastic, load-bearing cellulose/hemicellulose microfibrils and pectin 16,17 (Fig. 1b)."  Reply: We agree with the Reviewer that this statement should be adjusted (especially the reference location) and, in the revised manuscript, have revised the statement as follows (additions in yellow highlight): Pg. 2, lines 50-52 "Biologically, this enzymatic activity is spatially controlled within the pollen wall structure 19,20 , and inspires biomimetic strategies to engineer pollen structures with tunable material properties."  Figure 2d does not establish this critical claim and, as far as I can tell, the reference cited does not either.

Reply:
We apologize for the confusion on this point. The reference we cited is about the divalent cation effect (and its relationship to charged functional groups), and in the revised manuscript, we added an additional reference about the apparent acid dissociation constant value of pectin (pH value at which there is 50% dissociation). In the revised manuscript, the sentence has been amended as follows: Pg. 4-5, lines 96-99 "Under neutral pH conditions, the carboxylic acid functional groups in the pectin molecules are predominately deprotonated 26 and thus sensitive to divalent cations 27 (Fig. 2d). Time-lapsed optical microscopy experiments on immobilized pollen particles provided direct evidence of the dynamic response to calcium ions (Fig. 2e)." Cited References (f) A reference is needed for this statement "The pH-dependent swelling response is analogous to that of ionizable polymer networks whose time scales for de-swelling occur similarly within seconds." Reply: We thank the Reviewer for this great suggestion. In the revised manuscript, we provided detailed references about the pH-responsive swelling/de-swelling behaviors observed in polymer networks as follows (additions in yellow highlight):

Pg. 4, lines 90-91
"The pH-dependent swelling response is analogous to that of ionizable polymer networks whose time scales for de-swelling occur similarly within seconds 24,25 ."  Figure 30). The area increase would be 50% given that the diameter increases from 20µm to 25µm.

Reply:
We thank the Reviewer for this comment and agree that the statement of "microgel formation is restricted to pollen from eudicot plants" should be restricted to "among the tested species" and we have rephrased our conclusions accordingly. As the Reviewer points out, for some other pollen species, there is modest swelling observed at the highest tested pH condition (pH 14) but pollen species in that category do not form microgels. This is reported in our study and also consistent with ongoing studies of additional pollen species (see ROM Fig. 6). Among the tested pollen species, the only species that formed microgel particles exhibited pH-dependent swelling/de-swelling behavior across the pH spectrum, including across pH 2 to pH 12. In the revised manuscript, we have revised our commentary as follows (additions in yellow highlight): Pg. 1, lines 26-29

"Our investigation of pollen grains and spores from across the plant kingdom further showed that microgel formation occurred with tested pollen species from eudicot plants while tested pollen species from non-eudicot plants did not form microgels."
Pg. 7, lines 146-154 "We also tested pollen grains from Baccharis and Camellia plants, which belong to the same eudicot clade as sunflower plants ( Fig. 4a and Supplementary Fig. 16). We discovered that pollen grains from tested eudicot plants are transformed into microgel particles (Fig. 4b " On the other hand, insufficient hydration of pollen can cease germination, leading to harmomegathy with minimal energy loss. Thus, our findings provide insight into the design principles of pollen grains in the context of harmomegathy (dehydration) and germination (hydration), while facilitating the development of a broadly effective strategy to prepare pollenderived microgel particles with high-performance material properties that go beyond the natural synthetic capabilities of plant reproductive machinery."

Reviewer 2 Comments
The authors reports a extensive study on production of a stimuli responsive (mono-and divalent cations, pH) sporopollenin microgels from various pollen grains. The manuscript is well-designed and it was written in a reader-friendly manner. The findings of the study was intersting and worth spreading. The discussions are relevant and very informative. The findings of the work should reach wider audience by publication. I think that the submission can be accepted for publication in the present form. My recommedation is ACCEPT for publication.

Reply:
We thank the Reviewer for positive endorsement of our work and are grateful that they recommend publication of the work. We have carefully revised the manuscript according to the suggestions of Reviewer 1 too and hope that the revised manuscript is deemed worthy of publication and will hopefully reach a wide audience and inspire future readers to explore this exciting field.

Reviewers' Comments:
Reviewer #1: Remarks to the Author: I thank the authors for their careful response to my comments. They have responded to the key concerns I had. I would only ask that they rephrase the legend of Suppl. Fig. 19. For the bottom part of the figure, the legend states: "Bottom: Pollen species from algae and green plants showed minimal response to changes in solution pH." The statement is factually wrong and, again, goes far beyond what authors have demonstrated. The words "algae" and "green plants" should be changed since algae do not have pollen or spores, and "green plants" is meaningless. In the case of Lycopodium, spores were tested not pollen. Finally, the authors DO NOT have the data to justify the sweeping statement that all no-eudicots pollen and spores cannot form gels.

Jacques Dumais
Reviewer #3: Remarks to the Author: The authors induce pectin de-esterification in pollen grains to "mimic the activity of key enzymes involved in pollen tube growth". The authors conclude that "subtle, chemically programmed variations in the mechanical properties of both the exine and intine layers can cause the swelling of pollen particles akin to the shape transformations that occur during the orchestrated sequence of pollen hydration, germination, and tube growth". The authors observed that as a result of changes in the cell wall, and specifically pectin de-esterification, the cell wall swells by pH and ionic treatment, a process that was seen to be reversible.
While there is in principle nothing wrong with a purely curiosity driven approach, the authors do suggest that they 1. address a biological knowledge gap with their study, and 2. provide the basics for a possible application (micro-gel particles). In my eyes, the present manuscript does not provide satisfactory information for either, and it fails to even properly identify the knowledge gap, which leaves the pollen biologist wanting.
As for the biochemical aspect of the manuscript, pectin gel swelling is not entirely novel since changes in mechanical properties and swelling of cell walls upon changes in pectin esterification is already well known. I am not an expert in biochemistry, however, and may have overlooked important aspects that warrant this study to be of high profile with relation to the biochemistry.
Unfortunately, I haven't been convinced yet by the computational model approach either. The main reason is that the justification for the conditions used are inadequately presented and consequently I may simply not have understood (despite having experience in FE modeling). Concretely, the modeling approach and the boundary conditions used would require additional explanations to reassure me that the interpretations are of relevance to the biological system. This is crucial since the authors proceed to draw important conclusions from their finite element model such as the one stated in the beginning of page 8: "…that the swelling of pollen particles occurs when the ratio of the mechanical strength of the chemically tuned exine and intine layers is within a certain, optimal range. This chemomechanical balance appears to be carefully regulated and suggests that there might be a chemorheological regulatory pathway at the individual particle level underpinning whether pollen hydration leads to successful germination or incipient drying".
In the following, detailed concerns are listed which the authors may wish to consider. (Annoyingly, the text didn't have line numbers and hence locations are approximate) Major concerns: 1. Finie element model: Even though some similarities between experimental and finite element data are drawn through matching a number of data points, I am not convinced that the working physics of the phenomena in model and experiments are actually the same. If a direct relation exists, is not clear enough in the present version of the manuscript.
The study aims at investigating the contribution of the mechanical properties of the exine and intine layers of pollen grain in relation to the swelling and de-swelling of the grain. In the finite element model used to simulate the swelling process, the authors implement an internal pressure in the lumen of the grain to simulate pollen grain 'swelling', similar to inflating a balloon. The authors mention that "hydration and swelling of the hyperelastic pollen shells associated with osmotic pressure effects were simulated based on a hypothetical internal pressure". The pollen grain's increase in size in the model does therefore seem to occur due to a pressure-induced stretching of both enveloping layers, the intine and the exine, and which essentially results in an increase of the volume of the internal cavity. In the biological sample this would presumably be achieved through osmotically driven water uptake into the cytoplasm and/or vacuole. The chemical treatment that is used in the experiments, on the other hand, is not cited to change the internal turgor pressure or volume of the internal cavity, but to cause a swelling of the wall material composing the inner layer of the envelope, the intine. It is unclear how the approach of inflating the structure by increasing the pressure in the cavity simulates the process of wall swelling in the inner of the two enveloping layers. Shouldn't this chemical process lead to a volume increase in the intine?
To justify their modeling approach, the authors refer to a publication by van der Sman (2015, reference 47). However, it is not exactly clear to me how this works and more detailed explanation is needed. van der Sman considered the hydration of the cellular tissue from two different sources: the vacuole water resulting in the turgor and the water retained in the hydrated cell wall (van der Sman considered the hydrated shell to consist of both the cell wall and the cytoplasm). According to the paper by van der Sman, "a theoretical analysis of the contributions of both the turgor and the hydrated cell wall to the total water holding capacity of cellular tissue" was performed. In this publication, the turgor was not a "hypothetical" element presumed but a real pressure applied in the cell cavity due to osmotic pressure of the solutes in the vacuole (i.e., turgor). As van der Sman mentions, membrane integrity is required and results in water retention inside the cell (symplastic water) and turgor. In addition, in formulation of the stress in the shell, a term for osmotic pressure is also added to account for the hydration of the cell wall material (e.g., eq. 3 of van der Sman). van der Sman further concludes that "This implies that even for thin cell walls, with RWC E 5%, it will still be important to include the water retention in the apoplast for correct prediction of mechanical behavior". Unless I overlooked it, no such term accounting for water content of the wall material is mentioned in the present manuscript. Instead, the authors seem to base their model on an inflation of the spherical shell by a ("hypothetical") turgor and seem to suggest that this is effectively the same as swelling of the shell wall. I am unable to confirm that this can be concluded from van der Sman's paper, or that it reflects the physics of the present experiments (swelling of the wall material).
One source of confusion may be the use of terminology. On page 22 the authors write "To simplify simulations, we disregarded ... volume changes in the intine due to swelling effects. Instead, intine swelling was imposed by tensile circumferential stresses ( , ) due to the osmotic pressure across its thin gel-like matrix." In this two-sentence passage, the term 'swelling' seems to refer to two different processes. In the first sentence the authors seem to refer to the increase in volume of the intine material due to the molecular repulsion (as in Figure 2), whereas in the second sentence the 'swelling' seems to refer to the increase in size of the entire pollen grain due to pressure from the internal cavity and resulting tension of the pollen envelope. If I interpret this correctly, I urge the authors to use two different terms for the two different phenomena, for example 'inflation' or 'turgor-driven expansion' for the latter. I could have this wrong and better explanations (and a clearer image in Supp Figure 13d) would be helpful. Further, when the authors mention that "intine swelling was imposed by tensile circumferential stresses", it initially gives the impression that these stresses were input as initial conditions in the model where in reality they seem to result from the stretch of the material due to turgor.
Crucially, the authors do not mention how the thickness of the wall is affected in this experimental process. As can be seen from Fig. 1 of van der Sman, the wall can get thinner when stretched by an internal turgor. It needs to be clarified how the changes in pollen wall thickness upon wall swelling are related between the model and the experiment and I strongly suggest that the authors add transmission electron micrographs of cross-sections of rapidly frozen and fixed pollen grains to demonstrate the actually volume and thickness change in the intine resulting from the chemical treatment. 2. In definition of the finite element boundary conditions: Why wasn't Abaqus/Standard used instead of /Explicit? Abaqus/Explicit is generally reserved for addressing problems of highly dynamic nature (e.g., impact). Here, the swelling process can be safely regarded as a quasi-static process and could be modeled as such using a Static step in Abaqus/Standard. While Abaqus/Explicit can also be used to solve such problems, the choice should really be justified. The reason is that ensuring the correctness of the values and patterns resulting from Explicit Dynamics can be challenging. In the same vein, how does use of such small time steps (in order of microseconds) correspond to the quasi static process of swelling? 3. Page 3: How can FTIR confirm that 'structural integrity' of the wall is maintained? What does structural integrity mean in this context? That polymers remain polymers or that they maintain a given network structure (i.e. linkages)? What exactly does FTIR allow to conclude? Please provide a bit of information to the non-expert. 4. Page 4: "can result in the transformation of hard pollen grains into pliable, soft microgel particles". It would be helpful to define microgel particle. The authors seem to repeatedly suggest that, upon treatment, pollen grains from some species form "microgel" particles while others do not. No clear definition is provided what properties are required for a particle to count as a microgel particle. The pollen grains clearly still have a lumen and a stiff outer wall (exine). What makes them microgel particles that might be useful for applications in any manner? 5. Methods section, page 16: In the "microgel formation" section it is mentioned that the sample was left at 80Celsius. Heat is known to alter the plant cell wall material. Did the authors verify how temperature affects their results? Minor comments 6. Page 2, line 4: 'transfers viable cellular material' should be more specific. The transfer consists in the gametes, or sperm cells, from the male gametophyte (=pollen) to the female gametophyte.
7. Page 2, bottom half. 'hollow' does not seem to express that the pollen grain is filled by a liquid (not a gas). Maybe simply leave 'hollow' out of the sentence? Similarly, further down, 'holes' may leave the wrong impression of actual empty spaces which is not the case since an aperture still has an intine. I suggest replacing by 'gap in the exine'. 13. Page 7: "we discovered that this distinction alone is insufficient to explain hydration-induced pollen swelling." This claim requires further discussion.
15. Figure 1C and E: the legend must indicate which primary antibody was used. What is "hydrostatic tensile stress"? Is it principal stress due to hydrostatic pressure?
16. Page 19: Was AFM used or another indentation device? I am asking since the text mentions "instrumented indentation". This term is generally used for cantilever-less indentation systems. 17. What was the AFM tip shape?
18. Page 19: Was the immunolabel not preceded by a chemical fixation step? The protocol speaks of 'labeled sections' which would mean there would be a resin embedding step as well. There doesn't seem to be any mention of that (unless it is elsewhere in the methods). Please provide exhaustive description of the method. 19. Page 19: To compare fluorescence intensity between samples it is crucial that they were prepared in parallel, that the microscope settings were identical for acquisition and that image processing was done in identical manner. Please confirm that this was the case. 20. Legend of Figure 1 and elsewhere: Please specify that the SEM images are pseudo-colored. 21. Why does particle size decrease between 6-12 hours ( Figure 1D)? 22. Figure 2: I realize that the drawings are conceptual and simplified. However, pectin is a highly branched molecule (rather than a single strain as indicated). Maybe this could be symbolized at least conceptually?
23. Figure 2a: The figure legend should specify whether the repulsion is made due to increase or decrease in pH. 24. Figure 2c: How is "area swelling ratio" determined? 25. Figure 2d: The legend is not descriptive of the events shown in figure. 26. Figure 3 and FEA model: Did the presence of spikes actually influence the modeling results? Did they serve any other purpose than looking prettier when put side-by-side with micrographs? (This is fine, but it should be mentioned somewhere whether or not they have an effect or are purely meant to make the model look similar to the biological sample). Figure 2: I understand that these are conceptual drawings, but filling the entire pollen grain lumen with 'genetic material' simply ignores the cellular structure. Not the entire protoplast consists of genetic material! To not change the drawing, the light green material would need to be identified as 'Protoplast of the vegetative cell and generative cell (or sperm cellsdepending on species)'. To draw this more accurately, at least the nucleus of the vegetative cell as well as the generative cell or sperm cells should be drawn. Reply: We sincerely thank the Reviewer for taking the time and expert effort to carefully review our manuscript and for providing many excellent comments to help us revise and improve the manuscript. In the revised manuscript, we have added/edited the relevant remarks to fully incorporate the Reviewer's suggestions. 2) Finally, the authors DO NOT have the data to justify the sweeping statement that all no-eudicots pollen and spores cannot form gels.

Reply:
We appreciate the Reviewer's thoughtful comment. We added more constrained information on the related remarks as follows:

Pg. 1, lines 25-28
"Our investigation of pollen grains and spores from across the plant kingdom further showed that microgel formation occurs with tested pollen species from eudicot plants while tested pollen species from non-eudicot plants did not form microgels under the conditions used in this manuscript."

Pg. 7, lines 159-161
"To extend our findings, we tested pollen grains and spores from other clades and discovered that those from flowering monocots (Cattail) and gymnosperms (Pine) as well as spore-bearing lycophytes (Lycopodium) did not form microgel particles (Fig. 4)." "Taken together, these results support that pollen grains from the tested eudicot plants have suitable material properties to facilitate microgel conversion using our natureinspired strategy."

Reviewer 3 Comments
The authors induce pectin de-esterification in pollen grains to "mimic the activity of key enzymes involved in pollen tube growth". The authors conclude that "subtle, chemically programmed variations in the mechanical properties of both the exine and intine layers can cause the swelling of pollen particles akin to the shape transformations that occur during the orchestrated sequence of pollen hydration, germination, and tube growth". The authors observed that as a result of changes in the cell wall, and specifically pectin de-esterification, the cell wall swells by pH and ionic treatment, a process that was seen to be reversible.

Reply:
We thank the Reviewer for taking the time to carefully review our manuscript and for providing many excellent comments to help us revise and improve the manuscript. Upon reading the Reviewer's comments, we identified one important point that could have led to misunderstanding of some of the key points in our manuscript in general. This possibility appears to have resulted from a fundamental misunderstanding of plant cell walls versus pollen shells, which represent distinct features of two very different systems.
Below, we provide detailed responses to each point raised by Reviewer #3 and explain how we have improved the revised manuscript in our attempt to respond to each of his/her comments. Where applicable, we have also provided additional Review-Only Material (ROM) to explain our approach and findings.
While there is in principle nothing wrong with a purely curiosity driven approach, the authors do suggest that they 1) address a biological knowledge gap with their study, and 2) provide the basics for a possible application (micro-gel particles). In my eyes, the present manuscript does not provide satisfactory information for either, and it fails to even properly identify the knowledge gap, which leaves the pollen biologist wanting.

Reply:
We appreciate the Reviewer's remark and concern. As addressed in the abstract, the biological knowledge gap that we address in the current manuscript is focused on how tuning pollen structure (through a facile chemical process) can alter pollen's mechanical, chemical, and adhesion properties. The ability of pollen to control its microstructure depending on the environment during harmomegathy, reverseharmomegathy, or germination 1 , is guided by an organized sequence of enzymatically controlled reactions to modify structural components within certain parts of the pollen wall 2,3 . Although this enzymatic process might be a better-known phenomenon (albeit still the subject of intense, ongoing investigations in labs worldwide) from a biological perspective, engineering pollen-inspired materials by recourse to and expanding on these remodeling processes is highly novel and results in high-performance material capabilities that exceed natural limits, as identified in this manuscript for the first time.
To better explain the significance of our work, we have revised the introduction of manuscript as follows:

Pg. 2, lines 43-55
"Common features of pollen grains across various plant species include a microcapsule structure, function-driven shape, and ornamental architecture 15 . We selected pollen grains from sunflower plants (Helianthus annuus), which have spiky appendages and a tripartite structure (Fig. 1a and Supplementary Fig. 1). The outermost layer ("exine") is made up of sporopollenin, which is a strong, crosslinked biopolymer 10 , while the inner layer ("intine") is composed of elastic, load-bearing cellulose/hemicellulose microfibrils and pectin 16,17 (Fig. 1b). The aperture gap in the exine layer is integral to pollen tube growth and is neighbored by pollen wall regions with distinct material properties. As part of remodeling processes, pectin methylesterase (PME) enzyme plays a key role in controlling wall elasticity by converting pectin into pectate 18 , exposing carboxylic acid functional groups and imparting greater surface charge density that modulates the intine structural arrangement 18 . Biologically, this enzymatic activity is spatially controlled within the pollen wall structure 19,20 , and inspires biomimetic strategies to engineer pollen structures with tunable material properties." We have carefully examined the existing references and our approach to better explain the concerns of the referee. As for the possible application of microgel particles, we have now included detailed information about our ongoing research in our reply to the Reviewer's major comment 4, and also updated appropriate sections in our revised manuscript.
As for the biochemical aspect of the manuscript, pectin gel swelling is not entirely novel since changes in mechanical properties and swelling of cell walls upon changes in pectin esterification is already well known. I am not an expert in biochemistry, however, and may have overlooked important aspects that warrant this study to be of high profile with relation to the biochemistry.

Reply:
We would like to clarify here how our work is distinct from plant cell walls that are typically studied by plant tissue swelling and what is significant about our work.
Plant cells always have a strong cell wall surrounding them. When they take up water by osmosis, they start to swell but the cell wall prevents them from bursting. Plant cells become "turgid" when they are put in dilute solutions. Turgid means swollen and hard. The pressure inside the cell rises, and eventually the internal pressure of the cell is so high that no more water can enter the cell. This liquid or hydrostatic pressure works against osmosis. Turgidity is very important to plants because this is what make the green parts of the plant "stand up" in the sunlight.
As noted above, the plant cell wall prevents plant cells from bursting due to osmosisinduced swelling. In the case of animal cells which are not surrounded by tough cell walls, they swell and eventually burst under osmotic pressure. The swelling of this tough cell wall, accommodated by pectin gels, is one of the mechanical strategies to reduce the 4 stress concentration throughout the cell wall by softening the cell wall for improved flexibility and improved resilience. By contrast, the pollen wall performs two major roles in order to maximize the success rate of reproduction. First, it should protect the generative or sperm cells in various environmental conditions, until the right stigma environments are settled. Thus, the hydrophobic exine layer blocks water penetration into the pollen wall, going through harmomegathy. Second, the pollen wall should accommodate germination which requires significant water intake and tubular growth to facilitate the reproduction process. Due to the stiff and hydrophobic exine layer, water intake is significantly constrained at the beginning. As a result, mechanical and chemical degradation of the exine by releasing lipid and pollen cements associated with increased porosity should be a key change occurring during germination. In the meantime, pectin gel of the intine layer increases the inflation capability of the pollen grain, allowing significant water intake and rapid tubular growth. Thus, in fact, both exine and intine layers play important roles in this germination process. In our study, we mimicked this germination process, inducing chemical and structural changes of the exine and intine layers, for transformation of pollen grains into inflatable soft microgels.
First, we would like to address that the plant cell wall from tissue and plant pollen are fundamentally different, and thus the biological implication of revealing each pectininvolving mechanism is also very different. The composition of plant cell walls mainly consists of cellulose, pectic and hemicellulosic derivatives, which are similar to pollen's inner layer, intine. However, pollen includes an inner intine layer as well as the outer exine layer, which is structurally much tougher and more complex than any plant cell wall. Also, plant cell walls form tissues, whereas pollen particles exist discretely (ROM Fig. 1). Along this line, broadly studied topics cover the growth of the plant cell wall or softening of ripe fruit mediated by pectin, which is directly relevant to the life cycle of plant tissue 4 . Second, the definition of 'gel' that the Reviewer is referring to and 'microgel' from our manuscript are conceptually different. A general term of 'pectin gel' refers to the highmethoxy pectin under acidic condition or low-methoxy pectin forming egg-box model, whereby divalent ions crosslinks or bridges the polysaccharide chains, thus showing the increased viscosity 5,6 . These types of gel are well-known as pectin is widely used in food as a gelling agent or stabilizer. However, 'microgel' refers to a gel formed from a network of microscopic polymer or macromolecular colloids that swell in a good solvent, whereby sometimes feature stimuli (e.g., temperature, pH, ions)-responsive behaviors 7,8 .
To the best of our knowledge, the current manuscript is the first paper reporting that the chemical treatment controls the de-esterification of pectin in order to produce a soft material including the exine layer.
Unfortunately, I haven't been convinced yet by the computational model approach either. The main reason is that the justification for the conditions used are inadequately presented and consequently I may simply not have understood (despite having experience in FE modeling). Concretely, the modeling approach and the boundary conditions used would require additional explanations to reassure me that the interpretations are of relevance to the biological system. This is crucial since the authors proceed to draw important conclusions from their finite element model such as the one stated in the beginning of page 8: "…that the swelling of pollen particles occurs when the ratio of the mechanical strength of the chemically tuned exine and intine layers is within a certain, optimal range. This chemomechanical balance appears to be carefully regulated and suggests that there might be a chemorheological regulatory pathway at the individual particle level underpinning whether pollen hydration leads to successful germination or incipient drying".
In the following, detailed concerns are listed which the authors may wish to consider. (Annoyingly, the text didn't have line numbers and hence locations are approximate).

Reply:
We appreciate your attention to our computational model approach and are able to confidently answer and clarify the following points.

1) Finite element model: Even though some similarities between experimental and finite element data are drawn through matching a number of data points, I am not convinced that the working physics of the phenomena in model and experiments are actually the same. If a direct relation exists, is not clear enough in the present version of the manuscript. The study aims at investigating the contribution of the mechanical properties of the exine and intine layers of pollen grain in relation to the swelling and de-swelling of the grain.
In the finite element model used to simulate the swelling process, the authors implement an internal pressure in the lumen of the grain to simulate pollen grain 'swelling', similar to inflating a balloon. The authors mention that "hydration and swelling of the hyperelastic pollen shells associated with osmotic pressure effects were simulated based on a hypothetical internal pressure". The pollen grain's increase in size in the model does therefore seem to occur due to a pressure-induced stretching of both enveloping layers, the intine and the exine, and which essentially results in an increase of the volume of the internal cavity. In the biological sample this would presumably be achieved through osmotically driven water uptake into the cytoplasm and/or vacuole. The chemical treatment that is used in the experiments, on the other hand, is not cited to change the internal turgor pressure or volume of the internal cavity, but to cause a swelling of the wall material composing the inner layer of the envelope, the intine. It is unclear how the approach of inflating the structure by increasing the pressure in the cavity simulates the process of wall swelling in the inner of the two enveloping layers. Shouldn't this chemical process lead to a volume increase in the intine?
To justify their modeling approach, the authors refer to a publication by van der Sman (2015, reference 47). However, it is not exactly clear to me how this works and more detailed explanation is needed. van der Sman considered the hydration of the cellular tissue from two different sources: the vacuole water resulting in the turgor and the water retained in the hydrated cell wall (van der Sman considered the hydrated shell to consist of both the cell wall and the cytoplasm). According to the paper by van der Sman, "a theoretical analysis of the contributions of both the turgor and the hydrated cell wall to the total water holding capacity of cellular tissue" was performed. In this publication, the turgor was not a "hypothetical" element presumed but a real pressure applied in the cell cavity due to osmotic pressure of the solutes in the vacuole (i.e., turgor). As van der Sman mentions, membrane integrity is required and results in water retention inside the cell (symplastic water) and turgor. In addition, in formulation of the stress in the shell, a term for osmotic pressure is also added to account for the hydration of the cell wall material (e.g., eq. 3 of van der Sman). van der Sman further concludes that "This implies that even for thin cell walls, with RWC E 5%, it will still be important to include the water retention in the apoplast for correct prediction of mechanical behavior". Unless I overlooked it, no such term accounting for water content of the wall material is mentioned in the present manuscript. Instead, the authors seem to base their model on an inflation of the spherical shell by a ("hypothetical") turgor and seem to suggest that this is effectively the same as swelling of the shell wall. I am unable to confirm that this can be concluded from van der Sman's paper, or that it reflects the physics of the present experiments (swelling of the wall material).
Reply: We sincerely appreciate the Reviewer's comments and the opportunity to explain our results here in greater detail. In this work, chemomechanical boundary conditions of particle swelling were reasonably simplified through a conventional solid mechanics approach, with continuum FE models that simulate the hydrogel-like intine layer as hyper-elastic solid. This approach has been well-accepted for modelling of smart hydrogel systems (e.g, hydraulic hydrogel actuators or natural seed capsule actuator from Delosperma nakurense) in a simpler manner 9,10 . Furthermore, the purpose of our modelling in the current manuscript is to understand how the interplay of exine and intine layers influences inflation of pollen microgels, mimicking germination of pollen grains. Thus, we would like to clarify in the following response: 1) what happens to pollen grains after chemical treatment, 2) how pollen microgels are inflated, and 3) then how we mimicked the inflation of pollen microgels using the continuum FE models, ignoring hydration and volume change of the intine layer. 1) For pollen grains, the intine layer is originally hydrophobic due to highly esterified pectin, whereas the exine layer consists of sporopollenin covered or filled with lipid and pollenkitt (pollen cement). Thus, the pollen shell is inherently hydrophobic and not swellable when it is intact. 2) After the chemical treatment, the highly esterified pectin becomes de-esterified and hydrophilic while the exine layer becomes porous, losing pollen cement and lipid. In the meantime, the cytoplasm is also fully removed, thus the internal cavity is fully empty (ROM Fig. 2). Also, since the cytoplasm with cells was released through three apertures, the continuous intine layer was ruptured. Thus the bi-layered pollen shell was fully discretized through the entire shell thickness at the apertures.
ROM Fig. 2. Schematic illustration of pollen grains and pollen microgels before/after chemical treatment.
3) As a result, a pollen microgel has a hollow shell structure with three apertures, consisting of the outer exine and inner intine hydrogel layers. Thus, the volume increase of this pollen microgel system is caused by the swollen intine layer interplaying with the relatively stiffer exine. Moreover, three apertures are open to the environment, allowing rapid water intake into internal cavity of the hollow microgel shell due to the inflation of the pollen shell (ROM Fig. 3). ROM Fig. 3. Schematic illustration of a pollen microgel before and after inflation from pH 2 to pH 7 4) In this study, we aimed to understand how the hollow pollen shell, consisting of stiff exine layer and swellable intine layer, could deform under pH change depending on the duration of the second KOH treatment (from 0 h to 12 h). Since we used a conventional solid mechanics approach using nonlinear continuum FE models, chemical boundary conditions (e.g., solute concentrations, water contents, charge balance, mass transfer, etc.) in the real case require assumptions in order to make this chemical/mechanical coupling problem solvable with reasonably simplified mechanical boundary conditions (e.g., internal pressure) as shown in Supplementary Figure 13,  Real case: since the intine layer is swellable depending on its osmotic pressure induced by ionization or de-ionization of pectin chains in response to pH changes, the induced tensile circumferential stresses ( , ) of the swollen intine layer with a resultant volume change of the intine layer inflate the pollen microgel. Subsequently, the increased cavity volume of the pollen microgel is filled with water coming from the opened three apertures very rapidly to reach equilibrium. However, in this case, typical continuum FE models cannot deal with those chemical boundary conditions for intine where hydration and swelling occur through mass transfer under driving force by induced chemical potential depending on charge balance and solute concentration.  Equivalent model case: instead of setting the chemical boundary conditions associated with swelling of the intine layer, we introduced the hypothetical hydrostatic pressure which can be directly applied to the internal surface of the intine layer. By using this hypothetical internal pressure, we could capture the inflation of pollen microgel using our continuum FE models. In order to make this boundary condition reasonable, we assumed that the deformation of pollen microgel is significantly large (>> 10%) and highly non-linear due to hyperelastic behavior and non-linear pollen geometry, ignoring hydration and volume increase of the intine layer associated with swelling. Thus, our models could predict the inflation of pollen microgels, showing good agreement with experimental observation.

Supplementary Figure 13 | Geometrical configuration of pollen shells used in the simulations, including an inner intine shell, outer exine shell, and spikes. d,
Schematic illustration of pollen shell expansion that is induced by an osmotic pressure ( ) associated with hydration and swelling of the intine shell (top) and an equivalent internal pressure ( − ) that is applied to the internal surface of the pollen shell (bottom). The inflation of the hyperelastic pollen shells due to hydrated and swollen intine was simulated using a hypothetical internal pressure, as proposed by van der Sman 47 .

5) As the reviewer pointed out, van der Sman investigated the inflation and
hydration of the elastic hydrogel shells due to internal pressure, p int (van der Sman et al. 2015, 11 ). In this context, we referred to a publication by his work on the internal osmotic pressure which induces the inflated hydrogel shell. However, since the pollen microgel system is a hollow hydrogel shell, it is significantly different from a typical plant cell fully enclosed with the cell wall and is different from the case modeled by van der Sman. Those cellular tissues can swell by increasing the osmotic pressure along with increase of solute content of the vacuole. On the other hand, our pollen microgel system can be inflated by pH responsive swelling behavior of the intine layer. 6) Also, in our follow-up work, we have been developing multi-phase models of the intine layers in conjunction with FE for a more accurate constitutive law of pollen microgels, considering hydration and swelling of pollen microgel in much more detail. We will address hydration, osmotic pressure, ionic concentration effects, etc. in our future works extensively, mainly focusing the improved modelling of pollen microgels.

One source of confusion may be the use of terminology. On page 22 the authors write "
To simplify simulations, we disregarded ... volume changes in the intine due to swelling effects. Instead, intine swelling was imposed by tensile circumferential stresses ( , ) due to the osmotic pressure across its thin gel-like matrix." In this two-sentence passage, the term 'swelling' seems to refer to two different processes. In the first sentence the authors seem to refer to the increase in volume of the intine material due to the molecular repulsion (as in Figure 2), whereas in the second sentence the 'swelling' seems to refer to the increase in size of the entire pollen grain due to pressure from the internal cavity and resulting tension of the pollen envelope. Figure 13d) would be helpful. Further, when the authors mention that "intine swelling was imposed by tensile circumferential stresses", it initially gives the impression that these stresses were input as initial conditions in the model where in reality they seem to result from the stretch of the material due to turgor.

If I interpret this correctly, I urge the authors to use two different terms for the two different phenomena, for example 'inflation' or 'turgor-driven expansion' for the latter. I could have this wrong and better explanations (and a clearer image in Supp
Reply: We agree that we used the term, "swelling", for both the intine layer and the microgel system, which caused the confusion. As the Reviewer pointed out, the term, "swelling" in the first sentence indicate the volume increase of intine due to osmotic pressure, whereas the swelling in the second sentence is the inflation of the pollen microgel as a result of swollen intine. Moreover, we agree that the sentence, "intine swelling was imposed by tensile circumferential stresses", caused some confusion about the initial boundary conditions of our models. We have updated those in the method section of our revised manuscript as follows:

"Boundary conditions for modeling pollen swelling/de-swelling To investigate pollen swelling and de-swelling behaviors as well as related deformation mechanisms, we employed a three-dimensional model based on finite element analysis (FEA) using a commercial software package ABAQUS 2017 (Dassault Systèmes SIMULIA, Johnston, RI). The model was generated by Python scripts and then run in parallel in 32 cores.
To simplify simulations, we disregarded solvent flow into and out of the pollen shell walls during swelling, as well as volumetric changes in the intine due to swelling effects. Instead, intine swelling was imposed by tensile circumferential stresses ( , ) due to the osmotic pressure exerted across its thin gel-like matrix. Hydration and swelling of the hyperelastic pollen shells associated with osmotic pressure effects were simulated based on a hypothetical internal pressure in our FE models, as illustrated in Supplementary  Fig. 13d." Crucially, the authors do not mention how the thickness of the wall is affected in this experimental process. As can be seen from Fig. 1 of van der Sman, the wall can get thinner when stretched by an internal turgor. It needs to be clarified how the changes in pollen wall thickness upon wall swelling are related between the model and the experiment and I strongly suggest that the authors add transmission electron micrographs of cross-sections of rapidly frozen and fixed pollen grains to demonstrate the actually volume and thickness change in the intine resulting from the chemical treatment.
Reply: We appreciate the Reviewer's suggestion. We observed the thickness of the exine and intine layers in dry condition after the chemical treatment using scanning electron microscope. Since pollen cement was removed from the exine layer, the reduced thickness and increased nanopores were clearly observed as the chemical treatment was prolonged (Supplementary Figure 10). In the meantime, through optical microscope, we confirmed that the intine layer was significantly hydrated and swollen in wet condition, increasing its total thickness. We fully agree that the precise thickness measurement of the hydrated intine layer can be done by transmission electron microscope as the reviewer suggested. However, this method allows us to observe the local thickness of a selected pollen microgel fragment in 2D. Since biological samples have some degree of structural inhomogeneity, we should do the thickness measurement more carefully.
Instead, we are working on nanoCT imaging of pollen microgels, under various pH and ionic conditions. Through this experiment, we hope we can clarify the real wall thickness of the hollow pollen shell in 3D. As the reviewer pointed out, we expect that we can capture the swelling behavior and underlying fundamentals of pollen microgels in a better way through this more accurate structural observation.
Moreover, during through our FE simulations, we also considered the thickness effect of the exine and intine layers as parametric simulations. The deformation mechanism was essentially independent of the thickness of the intine layer (e.g., for 1x, 2x and 4x intine thickness), whereas the magnitude of applied hypothetical internal pressure was increased to achieve the same level of inflation of pollen microgels with significantly increased simulation time, due to more prominent bulk behavior of the intine layer along with the increased thickness. In reality, along with swelling, the stiffness of hydrogel should be significantly reduced due to the increased water content within the hydrogel. Our new multi-phase models will capture the thickness effect of the intine layer in this context more accurately. Reply: We thank the Reviewer's attention to these references. As the Reviewer pointed out, we found that our processed pollen microgel samples exhibit key similarities to germinated pollen grains as follows: 1) During this germination process, lipid components from the outer layer of pollen grains are gradually dispersed, allowing increased water uptake 12,13 . Through this process, apertures are opened, which accelerates water uptake, leading to germination with significant swelling of pollen grains 13 . Meanwhile, in the exine layer, pollen cements which occupy the cavity of sporopollenin network in exine are continuously released. 2) Since pollen grains are already open to the external environment due to the defatting process and pollen shell extraction, the hydrated intine can facilitate hydration of pollen grains even though the exine is still rigid. In this case, the exine only allows aperture opening or unfolding of pollen with minimal volume change (e.g., 0 h KOH-treated pollen). The swelling of intine or water uptake of pollen is significantly constrained by the rigid exine layer. Thus, germination is efficiently ceased or suspended until pollen allows more hydration from the environment.

Supplementary Figure 29 | Pollen hydration/dehydration cycle of angiosperms with
corresponding FEA simulation images. The top sequence shows a dry pollen grain undergoing the water uptake process during germination. In situations where complete hydration is not possible, germination might be aborted, and the apertures close again as shown in the bottom sequence 13 .

13
3) Through the second KOH treatment, the exine layer exhibited significant structural changes, becoming more porous in the following figure. During germination, the exine also releases pollen cement or lipid components, increasing cavity size and hydrophilicity in order to accommodate fast water uptake 13 . Moreover, through the AFM measurements, we confirmed that the exine layer became less stiff.

Supplementary Figure 10 | Scanning electron microscope (SEM) images of defatted sunflower pollen grains and pollen shells before and after 10 w/v% KOH treatment.
Top: SEM image of entire pollen microgels after defatting without or followed by 10 w/v% KOH incubation (2 nd KOH treatment) for 0-12 h. Longer KOH treatment resulted in greater opening of the apertures. Bottom: Surface morphology of exine layers at higher magnification. Exines of the defatted pollen grain exhibited a dense and smooth surface morphology with a few microscale pores around the spikes, while the exine surfaces of KOH-treated pollen were rough and porous with exposed sporopollenin skeleton. The increased porosity of the exine surface was attributed to the release of pollen cement ("pollenkitt") due to KOH treatment 13,28 .
We also have added the information into the revised manuscript as follows:

Pg. 9, lines 199-209
"Understanding and tuning the biomechanics of natural materials holds promise for materials design and application. These findings are relevant to understanding how the morphological evolution of microgel particles might play a role in ensuring plant reproduction in nature 13 . The ease of aperture opening in pollen particles induces rapid water uptake from the stigma surface. Meanwhile, the subsequent stretching and expansion of pollen grains require significant water uptake and considerable structural change with high energy consumption, to promote germination. As a result, insufficient hydration of pollen due to the inappropriate stigma conditions ceases germination, leading to harmomegathy with minimum energy loss (Fig. S29). Thus, our findings provide insight into the design principles of pollen grains in the context of harmomegathy (dehydration) and germination (hydration), while explaining how our microgel formation and tuning approach can exceed the performance limits of these natural processes."

2) In definition of the finite element boundary conditions: Why wasn't Abaqus/Standard used instead of /Explicit? Abaqus/Explicit is generally reserved
for addressing problems of highly dynamic nature (e.g., impact). Here, the swelling process can be safely regarded as a quasi-static process and could be modeled as such using a Static step in Abaqus/Standard. While Abaqus/Explicit can also be used to solve such problems, the choice should really be justified. The reason is that ensuring the correctness of the values and patterns resulting from Explicit Dynamics can be challenging. In the same vein, how does use of such small time steps (in order of microseconds) correspond to the quasi static process of swelling?
Reply: We sincerely appreciate the Reviewer's great expertise in the FE modeling and fully agree with the Reviewer. Abaqus/Standard should be used for a static or quasi-static process since it provides more reliable solutions. As the reviewer pointed out, the swelling process of pollen microgel happened experimentally within a few seconds, implying that it definitely should be a quasi-static process rather than a dynamic process. However, due to the following reasons, we concluded that Abaqus/explicit should be more proper for our simulations than Abaqus/Standard.
1. We started our simulations using Abaqus/Standard. Unfortunately, the simulation failed when the swelling ratio of pollen microgel reached up to ~1.4, where local deformation exceeded ε>2. We realized that even if our pollen microgel swelling is a quasi-static process, it induces significantly large deformation (ε>2) due to both material and geometric nonlinearity, particularly around the aperture regions. Therefore, mesh refining around the apertures was essential, unreasonably increasing the total number of elements and consequently extending the total simulation time up to a few days. Since the maximum swelling ratio of pollen microgel was ~1.8, we concluded that Abaqus/Standard was not feasible to deal with this kind of large and non-linear deformation 9 . If we want to further model the swelling ratio > 1.4 using Abaqus/ Standard, we need to add a considerably greater number of elements around apertures to avoid convergence problems, which should be beyond our current computing limit as well as Abaqus's capability. 2. The quasi-static solution by Abaqus/explicit can improve the convergence of the model with a large deformation. Also, the quasi-static analysis with Abaqus/explicit is often used to solve large models and/or complicated contact problem which require fewer system resources than the implicit Abaqus/standards procedure [14][15][16] . Thus, we chose Abaqus/Explicit to solve our models. 3. As the reviewer commented, the quasi-static solutions by the Abaqus/Explicit should be carefully evaluated in terms of their correctness. The difference between the fully static solution (Abaqus/Standard) and quasi-static solution (Abaqus/Explicit) originates from kinematic effects.  Fig. 4, we carefully monitored the kinematic energy density as well as strain energy density of pollen microgel during its swelling simulation. We clearly found that the kinematic energy occupies a very small portion compared with the strain energy, indicating the external work was mainly transformed into the static deformation strain energy of pollens. Thus, we ensured that the obtained quasi-static solutions were reasonable. 4. Regarding the time steps, it is a pseudo time rather than the real physical time of the swelling event. Although in Abaqus/Explicit a longer step time reduces the kinematic energy, making quasi-static solutions closer to static, it requires too considerable computing resource. Hence, we carefully chose the step time of our simulations in times of the order of few microseconds, considering our large models (a total of 145,938 elements). As shown in ROM Fig. 4, with this microscale time step, the kinematic effects were almost negligible. Thus, we confirmed that we firmly obtained reasonable quasi-static solutions using Abaqus/Explicit under feasible model set-up conditions.

3) Page 3: How can FTIR confirm that 'structural integrity' of the wall is maintained? What does structural integrity mean in this context? That polymers remain polymers or that they maintain a given network structure (i.e. linkages)? What exactly does FTIR allow to conclude? Please provide a bit of information to the non-expert.
Reply: We agree with the Reviewer that the structural integrity might be misleading. In FTIR analysis, the structural characteristics of pollens depending on the treatment conditions were investigated. First, due to the existence of cytoplasm, the FTIR absorption spectrum of defatted pollen was more complicated and convoluted than those of KOH-treated pollen specimens. On the other hand, regardless of the different treatment time (0-12 h), the characteristic absorbance peaks of all KOH-treated pollen grains appeared almost identical. Therefore, we presumed that one of the key components, pectin should be a distinctive feature through the FTIR spectra, depending on its de-esterification. The two important characteristic peaks of pectin and de-esterified pectin are ~1740 and ~1620 cm -1 due to C=O stretching of esterified carboxyl groups (-COOCH 3 ) and free carboxyl groups (-COOH), respectively, along with the increase in ~1410 cm -1 peak when the pectin is de-esterified 17-19 20 . The peak shift may occur due to the existence of potassium ions and the influence of other cell wall components [21][22][23] . All pectin peaks were more clearly observed for KOH-treated pollen grains than defatted pollen, especially the peak at ~1620 cm -1 for defatted pollen was not prominent. However, up to 12 h incubation with KOH, any significant changes of chemical structures of pollen grains weren't observed through FT-IR even though the existence of de-esterified carboxyl group for pectin was clearly confirmed. In order to make this point clearer, we edited the FTIR description in the revised manuscript as follows:

"Fourier transform infrared (FTIR) spectroscopy experiments also verified that all KOH-treated pollen grains appear nearly identical, irrespectively of the treatment time,
whereas only the defatted sample exhibited more complex and convoluted spectral features along with peaks corresponding to pectin molecules due to residual cytoplasmic contents (Supplementary Fig. 4)."

4) Page 4: "can result in the transformation of hard pollen grains into pliable, soft
microgel particles". It would be helpful to define microgel particle. The authors seem to repeatedly suggest that, upon treatment, pollen grains from some species form "microgel" particles while others do not. No clear definition is provided what properties are required for a particle to count as a microgel particle. The pollen grains clearly still have a lumen and a stiff outer wall (exine). What makes them microgel particles that might be useful for applications in any manner?
Reply: We thank the Reviewer for this excellent question, and agree that the definition of microgel particles should be stated more clearly in the manuscript. A microgel is a gel formed from a network of microscopic polymer or macromolecular colloids that swell in a good solvent, whereby sometimes feature stimuli (e.g., temperature, pH, ions)responsive behaviors 7,8 . In our study, the KOH-treated, tested pollen species from eudicot plants formed gel (as observed in the reversed vials in Supplementary Figure 3c) and were highly responsive to pH changes, which is the similar attribute to conventional microgels, although the unique morphology of the pollen grains remained after KOH treatment.
As the pollen grains are known to be practically indestructible along with demonstrating excellent biocompatibility, these microgel particles can be applied to various conventional material processes. Those individual microgel particles can be used as smart drug carriers which are responsive to pH or ion strength changes. More importantly, those soft pollen microgel particles can be processed for fabrication of thin films or sponges as shown in ROM Fig. 5. As the pollen microgels are responsive to external stimuli including pH and ionic strength, we believe these systems can be utilized as biosensors, purification as chelating agents or smart tissue scaffolds with tunable porous structures. We are extensively expanding our fundamental research on pollen microgel systems into various applications as our future works. Moreover, natural polymer-based microgels are not thermally stable and have irregular particle shapes and heterogeneous particle sizes [24][25][26] . This limitation hampers the utility of existing natural printing materials used for freeform 3D printing. As the pollen microgel particles are uniform, thermally and mechanically stable, and stimuli-responsive material with strong material interfaces, they can be excellent source of natural materials for 3D/4D printing. [REDACTED] ROM Fig. 5. Our preliminary studies supporting the competitive advantages of plant pollen as a next-generation material. We have completed proof-of-concept experiments demonstrating that pollen-based microparticles from certain plant species are useful materials for various 3D printing strategies, paper, and sponge.
We have indicated this information in the revised manuscript as follows:

Pg. 9, lines 210-213
"Furthermore, the process offers new pathways to produce highly uniform microgel particles from pollen sources. Such particles could be useful in a wide range of applications where excellent quality control is essential as, for example, in highperformance sensors and actuators as well as in vivo drug delivery."

5) Methods section, page 16: In the "microgel formation" section it is mentioned that the sample was left at 80 degree Celsius. Heat is known to alter the plant cell wall material. Did the authors verify how temperature affects their results?
Reply: We thank the Reviewer for this excellent question and have studied the effect of the temperature during the KOH incubation stages. Indeed, the temperature is one of the important factors for the microgel formation. In this study, we intentionally induce a biomimetic germination process by 1) removing pollen cements from the exine layer and 2) de-esterify pectin molecules of the intine layer in order to hydrate and inflate hydrophobic pollen grains. In particular, to de-esterify pectin molecules, high temperature (>50 o C) and strong alkali conditions are often required. We also found that the processing time were extended up to a few weeks at lower temperature (such as 25 o C) to obtain the same swelling behavior of produced pollen microgel particles. We also found that high temperature degraded pectin significantly after 24 h treatment, when pollen microgels couldn't maintain their structures with significantly reduced mechanical stability. Thus, we limited the processing time up to 12 h.
Minor comments: 6. Page 2, line 4: 'transfers viable cellular material' should be more specific. The transfer consists in the gametes, or sperm cells, from the male gametophyte (=pollen) to the female gametophyte.
Reply: In the revised manuscript, the sentence is revised to be more specific as follows:

Pg. 2, lines 33-35
"Pollen is a remarkable natural material that plays a critical role in plant reproduction and transfers viable cellular material (i.e., male gametes or sperm cells) between different reproductive parts of plants 6,7 ." filled by a liquid (not a gas). Maybe simply leave 'hollow' out of the sentence? Similarly, further down, 'holes' may leave the wrong impression of actual empty spaces which is not the case since an aperture still has an intine. I suggest replacing by 'gap in the exine'.

Page 2, bottom half. 'hollow' does not seem to express that the pollen grain is
Reply: As explained above in major point 1), the pollen microgel has a hollow shell structure with three apertures entirely opened to the outer environment (also can refer to Fig. 1e). Therefore, we think that the terms of "hollow" and "holes" in this context are correct.

Page 3: "To address this challenge": what challenge? No challenge is specified as of yet except for an allusion to potential of getting inspired by design of pollen grain.
Reply: We thank the Reviewer for the comment and agree that the sentence starts with vague transition. In the revised manuscript, the sentences have been edited as follows:

Pg. 3, lines 56-57
"Motivated by these intricate biological features, we developed a nature-inspired strategy to de-esterify pectin molecules throughout the entire pollen wall structure." 9. Page 3, bottom half. 'Coinciding with an increase in gel-like properties...'. Until here, the manuscript has not mentioned any proof for gel-like properties to actually arise, hence it is somewhat misplaced to casually mention them here.

Reply:
We thank the Reviewer for this suggestion and agree that the gel-like properties should be addressed before CLSM experiments in the context. In the revised manuscript, we have added a Supplementary Figure 3c to show the response of pollen gels to gravity, and incorporated it as follows:

Pg. 3, lines 71-77
"These results were complemented by immunolabeling experiments with JIM7 antibody, which recognizes highly esterified epitopes of pectin 21 (Supplementary Fig. 3b). In addition, the pollen dispersions exhibited more gel-like character, as indicated by increased resistance to gravity due to de-esterification of pectin (Supplementary Fig. 3c).
Coinciding with an increase in gel-like properties, confocal laser scanning microscopy (CLSM) experiments revealed that chemical processing caused the pollen particles to swell and join together while individual particles remained structurally intact 23 (Fig. 1e)." Reply: We thank the Reviewer for this comment and along with the comment 9 above, we have added the additional information to show the gel-like character of pollen microgel particles in Supplementary Figure 3c.

Supplementary
As for the second point, it is correct. In fact, the inflation of pollen microgel particles consist of two deformation modes. The first deformation mode occurs by opening three apertures. From the evolution of elastic strain energy, we confirmed that this deformation did not require much strain energy, localizing the deformation around the tips of apertures. On the other hand, the second deformation mode was induced by stretching the pollen shell along with the swelling of the intine layer. This process requires high strain energy, leading to the large global deformation throughout the pollen shell. Therefore, we have clarified this sentence in the revised manuscript as follows:

Pg. 7, lines 148-155
"The aperture opening does not require significant strain energy with highly localized deformation around the apertures, whereas subsequent stretching is an energyconsuming process that involves large-scale global deformation throughout the pollen shell. Stress and strain contours of 6 h KOH-treated pollen microgel particles ( / = 1.6) indicate that the three apertures are initially opened, accompanied by a small volume change until the critical swelling ratio is reached, followed by a dramatic change in microgel particle diameter at the maximum swelling ratio, (Supplementary Videos 8 and 10)." 11. Page 6: What is the "initial swelling ratio, open"? Please define.

Reply:
The initial swelling ratio, open indicates the inflation of pollens mainly due to the opening of three apertures. When the intine layer is hydrated and swollen, the first deformation mode is opening those three apertures. This deformation mode doesn't require too much strain energy, and occur very rapidly, allowing water intake into the internal pollen cavity.
We have added this definition in the revised manuscript as follows:

Pg. 6-7, lines 139-148
"Numerical simulations also revealed the morphological evolution of a microgel particle during its structural expansion (Figs. 3b,c and Supplementary Figs. 14-15). We defined three key swelling ratios of pollen-derived microgel particles: , , and . is the initial swelling ratio ( =1) when the total volume of the microgel particle is minimized. Once pectin hydration begins, rapid aperture opening occurs and intine swelling-induced pressure and resulting strain are highly localized at the tips of the three apertures.
is the critical swelling ratio of a pollen microgel until its inflation is mainly accommodated by opening of three apertures. When a critical swelling ratio is reached, the deformation mode of a microgel particle transfers from aperture opening to stretching. Further expansion of the microgel particle induces large deformation of both the intine and exine layers until the maximum swelling ratio, , is reached." Reply: We thank the Reviewer for this valuable comment and in the revised manuscript, the relevant remarks are added as follows:

Pg. 1, lines 25-28
"Our investigation of pollen grains and spores from across the plant kingdom further showed that microgel formation occurs with tested pollen species from eudicot plants while tested pollen species from non-eudicot plants did not form microgels."

Pg. 7, lines 159-161
"To extend our findings, we tested pollen grains and spores from other clades and discovered that those from flowering monocots (Cattail) and gymnosperms (Pine) as well as spore-bearing lycophytes (Lycopodium) did not form microgel particles (Fig. 4)." 13. Page 7: "we discovered that this distinction alone is insufficient to explain hydration-induced pollen swelling." This claim requires further discussion.

Reply:
The underlying folding mechanisms of pollens have been investigated during harmomegathy where hydration and dehydration occur depending on the environmental conditions around a pollen grain 27,28 . As Katifori et al. reported in their paper, the unfolding/folding of pollen shells can accommodate this harmomegathy process with minimal energy cost 28 . However, in our current study, we focused on the subsequent "germination" process after harmomegathy is halted. During this germination process, lipid components from the outer layer of pollen grains are gradually dispersed, allowing increased water uptake 12,13 . Through this process, apertures are opened, which accelerates water uptake, leading to germination with significant swelling of pollen grains 13 . It is well-known that de-esterification of pectin in the intine layer is one of the key biochemical reactions during germination. However, none of studies have reported the role of the exine layer for germination. Through this study, we found that "interplay" of exine and intine layers only could induce inflation of pollen microgels, mimicking germination of pollen. We have added this point in the revised manuscript as follows.

Pg. 8, lines 176-187
"While experimental and computational studies of dehydration-induced harmomegathy have long recognized that the exine and intine layers of natural pollen grains have distinct mechanical properties 11 , we discovered that this distinction alone is insufficient to explain hydration-induced pollen swelling. Strikingly, through direct experimental investigation with SEM and AFM and supporting computational modeling based on a bilayered pollen shell structure, our findings reveal that the interplay of exine and intine layers plays a key role in dictating pollen swelling. The ratio of the stiffness values of the chemically tuned exine and intine layers must be within a certain, optimal range to trigger inflation of pollen-derived microgel particles in our experiments. This chemomechanical balance appears to be carefully regulated and suggests there is a chemorheological regulatory pathway at the individual particle level underpinning whether pollen hydration leads to successful germination or incipient drying 13 (Supplementary Fig. 29)."

Page 7, last line: "mechanical strength" or stiffness?
Reply: We thank the Reviewer for pointing this out as it should be stiffness. We have corrected the term in the revised manuscript as follows:

Pg. 8, lines 182-184
"The ratio of the stiffness values of the chemically tuned exine and intine layers must be within a certain, optimal range to trigger inflation of pollen-derived microgel particles in our experiments." 15. Figure 1C and E: the legend must indicate which primary antibody was used. What is "hydrostatic tensile stress"? Is it principal stress due to hydrostatic pressure?
Reply: We thank the Reviewer for the comment. The monoclonal JIM5 antibody was used as the primary antibody in Figure 1c (as well as Supplementary Fig. 3a), and Figure  1e is without any antibody treatment as the pollen is autofluorescence. In the revised manuscript, we added the antibody information in the figure legend as follows:

Pg. 10, lines 223-224
" Figure 1 | c, Immunofluorescence microscopy detection of de-esterified pectin within pollen shells using JIM5 as the primary antibody. d,…" As for the second point, hydrostatic tensile stress indicates induced isotropic stress of the pollen shell due to applied hydrostatic pressure. Through this stress field, we could investigate where applied pressure was localized. Due to the symmetric geometry of the spherical pollen shell, the stress contour shows strong correlation with the max principal strain contour, explaining why large deformation occurs at a particular region of the pollen shell.

Page 19: Was AFM used or another indentation device? I am asking since the text mentions "instrumented indentation". This term is generally used for cantilever-less indentation systems.
Reply: We apologize for the lack of clarity in the description. We used AFM for the stiffness measurement of pollen grains or microgel particles. In the revised manuscript, we have corrected the term to avoid this confusion as follows:

Pg. 19, lines 515-516
"This depth-sensing AFM indentation approach enables the quantitative determination of the Young's modulus of the shell material 35,36 ."

What was the AFM tip shape?
Reply: The AFM tips that we used were shaped as polygon-based pyramid with a halfcone angle of 20°. In the revised manuscript, we have added this information as follows: Reply: In the immunolabeling study, we did not use a chemical fixation step and resin embedding step. To avoid the confusion, we would like to revise the manuscript as follows:

Pg. 19, lines 507-510
"Antibody-labeled pollen particles were examined immediately with a confocal laser scanning microscope (Zeiss LSM 710) and without any antifade reagents. Pollen grains without primary and/or secondary antibodies were used as controls."

Page 19:
To compare fluorescence intensity between samples it is crucial that they were prepared in parallel, that the microscope settings were identical for acquisition and that image processing was done in identical manner. Please confirm that this was the case.

Reply:
We indeed confirm that the identical imaging acquisition settings (i.e., fluorescence light exposure time, strength) were used in the respective immunolabeling studies (JIM5 or JIM7). In the revised manuscript, we have added the information as follows:  Figure 1d, the diameter of the pollen gel was determined in pH 7 when pectin was in deprotonation state. Therefore, the pollens were in the swollen state due to the repulsive electrostatic force between -COOgroups. It is reasonable that the degradation of pectin will result in a lower swelling ratio. Figure 2: I realize that the drawings are conceptual and simplified. However, pectin is a highly branched molecule (rather than a single strain as indicated). Maybe this could be symbolized at least conceptually?

22.
Reply: We thank the Reviewer for the constructive comment and in the revised manuscript, the branched pectin structures are represented as follows in Fig. 1 and 2:  Reply: We thank the Reviewer for the interesting question and comment. We tried to mimic the pollen shell structure with spikes as close as possible for our modelling. However, there is another reason for us to introduce spikes on the sunflower pollen shell surface. As Katifori et al. 32 reported in their paper, the unfolding/folding of pollen shells can accommodate the harmomegathy process with minimal energy cost. In case of sunflower pollen, it has spiky shells without any foldable shell design (e.g., wrinkles or folded structure) except for apertures. Also, the spikes restrict extensive folding of the pollen shell during harmomegathy due to structural hindrance. Thus, we could set the minimal volume of those sunflower pollen microgel particles when their apertures were fully closed.

Reply:
We apologize for the missing information in the figure. The same confocal laser scanning microscopy described in the Methods section has been used, and the imaging conditions were identical for all samples after the first adjustment. Also, all presented images are from the middle optical sections while most of the pollens display the largest 2D diameter. We understand that this is an important information for the readers to follow; thus, in the revised manuscript, following sentences were added to the Methods section:

Reviewers' Comments:
Reviewer #3: Remarks to the Author: The authors have taken great care to provide additional explanations and many of the previously highlighted points have been clarified. Thile this has improved the manuscript incrementally, my main concern remains: the validity of the finite element model. The additional explanations provided by the authors make it clear that my initial interpretation was accurate: Deformation of the outer envelope is accomplished by applying a hypothetical turgor pressure from the inside. While it could be argued that all modeling approaches are based on both educated assumptions and simplifications, in this case this simplification is such that the result becomes meaningless in terms of the outcome. Even if a swelling of the virtual pollen grain is achieved through this pressure application, the simulation can not be used to validate any of the structural concepts. It does not not simulate the material swelling of the intine, nor the change in volume or the actual pressure that this process exerts on the exine.
The inconsistencies are expressed in the authors own words as they state "In the meantime, through optical microscope, we confirmed that the intine layer was significantly hydrated and swollen in wet condition, increasing its total thickness." and "Thus, the volume increase of this pollen microgel system is caused by the swollen intine layer interplaying with the relatively stiffer exine." but then they state: "To simplify simulations, we disregarded solvent flow into and out of the pollen shell walls during swelling, as well as volumetric changes in the intine due to swelling effects. Instead, intine swelling was imposed by tensile circumferential stresses…due to the osmotic pressure exerted across its thin gel-like matrix. Hydration and swelling of the hyperelastic pollen shells associated with osmotic pressure effects were simulated based on a hypothetical internal pressure in our FE models, as illustrated in Supplementary  Fig. 13d." In a nutshell, the model simply does not have any predictive power that helps us understand the biological, physical or chemical processes at hand. I can therefore not support publication of that aspect of the manuscript.
That said, the experimental and chemical part of the manuscript is actually very nice and novel and that part alone could represent an full manuscript, although probably better suited for a different journal.
Reviewer #4: Remarks to the Author: I have been asked to look specifically at the modelling aspect of this paper, and determine whether the framework is valid. I agree with Reviewer 3; the model as set out does not sufficiently capture the underlying mechanisms required to have predictive power. A critical element is the swelling of the intine which has been neglected -imposing a hypothetical internal turgor pressure is not equivalent. I also do not think the reference used to justify this assumption (van der Sman) is doing the same thing -there the internal pressure is representing the inflation of the internal cell vacuole. The van der Sman paper does include the effect of swelling of the gel, but I think that is via a different element in the model.
Editorial Note: Based on the advice from reviewers an additional reviewer with specific expertise was added; this reviewer is numbered Reviewer #4

Reviewer 3 Comments
The authors have taken great care to provide additional explanations and many of the previously highlighted points have been clarified. While this has improved the manuscript incrementally, my main concern remains: the validity of the finite element model. The additional explanations provided by the authors make it clear that my initial interpretation was accurate: Deformation of the outer envelope is accomplished by applying a hypothetical turgor pressure from the inside. While it could be argued that all modeling approaches are based on both educated assumptions and simplifications, in this case this simplification is such that the result becomes meaningless in terms of the outcome. Even if a swelling of the virtual pollen grain is achieved through this pressure application, the simulation can not be used to validate any of the structural concepts. It does not not simulate the material swelling of the intine, nor the change in volume or the actual pressure that this process exerts on the exine. In a nutshell, the model simply does not have any predictive power that helps us understand the biological, physical or chemical processes at hand. I can therefore not support publication of that aspect of the manuscript.
That said, the experimental and chemical part of the manuscript is actually very nice and novel and that part alone could represent a full manuscript, although probably better suited for a different journal.
Reply: We sincerely thank the Reviewer for carefully reviewing our manuscript and for providing many excellent comments to help us revise and improve the manuscript. We were very encouraged by the Reviewer's positive comments that the experimental and chemical results of the manuscript are significant. We also agree that the modeling aspects can be improved and we have made significant efforts in this direction by removing the simplified hypothetical hydrostatic pressure and replacing it with a more detailed accounting of the swelling behavior of the intine layer.
To this end, we have developed a multiphysics model of intine layer swelling as a function of environmental conditions (e.g., ion effects) that works in conjunction with the FEA modeling. This development has been part of our ongoing research in the field and we decided to include it in the revised manuscript for a more thorough treatment of this swelling aspect. The following points briefly summarize why we added these elements and how they improve our modeling overall.
1) As the Reviewer pointed out, the hyperelastic exine layer is inflated by swelling of the intine layer. In our previous version of the manuscript, we captured the exine inflation by applying a hypothetical hydrostatic pressure to the inner surface of the intine layer. This approach captured the basic concept of swelling-induced inflation, however, we agree that it can be further refined to better describe intine swelling based on actual changes in the system such as ionic changes.
2) Thus, in our revised manuscript, we introduced a multiphysics model of the intine layer, capturing its swelling behavior by accounting for chemo-electro-mechanical coupled field effects. Through this multiphysics model of the intine layer swelling, the resulting swelling-induced mechanical pressure was analytically calculated depending on factors such as solution pH and ion type/concentration. Since the intine swelling pressure equals the equilibrium pressure for inflating the exine layer, we used the intine swelling pressure as the boundary condition for exine inflation in our FEA modeling. Below, we compare the previous and new modeling approaches used in the previous versions of the manuscript and in the revised manuscript (ROM Table 1). 3) Thus, two key factors that affect the swelling/deswelling behavior of the pollen microgel particle system were addressed through our extended numerical modeling: 1) intine swelling by the multiphysics model: the intine uptakes ionic solution from the surrounding medium to generate a considerable osmotic pressure, which is the driving force for inflating the outer exine; and 2) exine inflation by the FEA model: the stiffer exine along with the three apertures exerts an inhomogeneous constraint on intine swelling. The interplay between exine and intine layers plays a key role in controlling swelling behavior of pollen microgel particles, and our findings show strong agreement between experiment and simulation.
To further evaluate how changes in the chemomechanical properties of pollen substructure layers affect the swelling/de-swelling behavior of pollen microgel particles, we also conducted computational simulations based on a multiphysics model that incorporated finite element analysis (FEA) [29][30][31] . Specifically, the model simulated the extent of pollen intine swelling in different ionic solutions in line with the aforementioned experiments and accounted for the effects of chemo-electro-mechanical coupled fields on inflation/deflation of the exine layer (Supplementary Fig. 13). The pollen intine swelling pressure ( ) was analytically calculated as a function of changes in environmental conditions (i.e., ion types and concentrations) and the computed values used as boundary conditions for analyzing inflation of the exine layer (i.e., exine inflation pressure, ) to evaluate the swelling/deswelling behavior of pollen microgels, and we systematically studied / ratios from 0.15 to 8 (Supplementary Table 1). A stiffer exine ( / > 2) would be expected to impose a rigid boundary condition upon the hydrated intine, thereby constraining the potential for intine swelling (Fig. 3a). On the other hand, when / < 2, both experiments and simulations demonstrated a steep increase in particle diameter ( / = 1.6, Supplementary Video 7). Moreover, the de-swelling of pollen microgel particles that was experimentally observed in the presence of multivalent cations (cf. Fig. 2) was also captured in the simulations for / < 1, and arises from the intine becoming stiffer than the exine due to the chelation reaction as shown in Fig. 2d  ( / = 0.7, Supplementary Video 8). Together with the gelation of de-esterified pectin molecules within the intine layer, these findings reveal that a softened exine layer plays an important role in modulating the mechanical properties of microgel particles by allowing greater swelling of the hydrated intine. Thus, the interplay of mechanical responses in the exine and intine layers dictates the morphological behavior of the microgel particles.
Taking into account the mechanical behavior of an individual microgel particle, its representative volume element (RVE) in the FEA was defined as a structure that consisted of a hyperelastic exine layer only with one aperture and one-third symmetry (Supplementary Fig. 13). We captured the large elastic deformation and strain energy density of a microgel particle by using FEA to simulate pollen exine inflation. Numerical simulations also revealed the morphological evolution of a microgel particle during its structural expansion (Figs. 3b,c and Supplementary Figs. 14-15) Consequently, Supplementary Fig. 13, Supplementary Table 1, and Fig. 3 have been revised as follows: Supplementary Figure 13 | Multiphysics model of swelling of pollen microgel particles. a, Schematic illustration of pollen microgel swelling that is composed of the interplay of intine (hydrogel) swelling and exine (rubber) inflation. The swelling-induced mechanical pressure exerted on the intine layer ( ) due to osmotic pressure effects (Π) decreases as intine swelling proceeds. The intine swelling pressure ( ) is equal to the equilibrium pressure required to inflate the exine layer ( ). The swelling ratio of the intine layer, λ i , is equal to the inflation ratio of the exine layer, λ e , due to the intine layer being tightly bound to the exine layer. The hydration and swelling of pollen microgels due to osmotic pressure-related effects were predicted based on the multiphysics model of the intine layer ( = , and the pressure for exine inflation was applied to the internal surface of the exine layer using the boundary condition = , as proposed in Ref. 31. b, Schematics of the intine shell placed in an ionic solution with the computational domain and boundary conditions for numerical simulations. c, Geometrical configuration of pollen microgel particle shells used in the simulations, including exine layer and spikes with mesh grid concentrated on two tips of the aperture while avoiding excessive mesh 5 , , , , ) distortion during the aperture opening process. Note that the pollen microgel particle is modeled using one-third symmetry. d, Deformed contour of a pollen grain simulated with one-third symmetry to effectively use computational resources. e, Measurement setup for the swollen diameter of a deformed pollen microgel particle structure. Note that the deformation is irregular and the red circle marks the maximum deformed diameter.  ratio values ( / = 0.15, 1.6 and 3). c, Hydrostatic tensile stress and maximum principal strain contours of the pollen microgel particles ( / = 0.15, 1.6 and 3) for three critical swelling ratios ( , and ) (labeled FEA), along with representative optical micrographs of pollen microgel particles in various chemical environments (i.e., ionic changes) that triggered similar morphological evolutions. / = 0.15 corresponds to pollen microgel particles immersed in 100 mM CaCl2; / = 1.6 to 6 h KOH-treated pollen microgels incubated in pH 10 solution; / = 3 to defatted pollen grains. Source data are provided as a Source Data file.

Supplementary
Also, detailed descriptions of the extended modeling approach have been added in the revised Methods section as follows:

Numerical Modeling for Swelling/Deswelling of Pollen Microgel Particles
For pollen grains, the intine layer is naturally hydrophobic due to highly esterified pectin, whereas the exine layer consists of sporopollenin that contains lipids and pollenkitt (pollen cement). Thus, the pollen shell is inherently hydrophobic and not swellable in its natural form. After the chemical treatment, the highly esterified pectin becomes de-esterified and hydrophilic while the exine layer becomes porous (due to removal of pollen cement). In addition, due to chemical treatment, the cytoplasm is also fully removed and thus the internal cavity is fully empty (Supplementary Fig. 2). Also, since the cytoplasmic contents, including cells and cellular debris, have to be released through the three apertures, the once-continuous intine layer is also ruptured. Thus, in the modeling, the two-layer pollen shell was fully discretized through the entire layer thickness at the apertures as shown in Supplementary Fig. 2. As a result, a pollen microgel particle has a hollow shell structure with three apertures, consisting of the outer exine and inner intine layers. Thus, the volumetric increase in a pollen microgel particle is caused by the interplay of the swollen intine layer and relatively stiffer exine layer. In particular, we carefully defined the swelling/deswelling behavior of the hydrogel-like intine layer due to water absorption and desorption depending on the osmotic pressure and inflation/deflation of the hyperelastic exine due to swelling-induced mechanical pressure of the underlying intine layer. Moreover, the three apertures are open, which allows rapid water intake into the internal cavity of the hollow particle that is concomitant with inflation of the pollen exine. Taken together, two key factors affect the swelling/deswelling behavior of the pollen microgel particle system: 1) intine uptakes ionic solution from the surrounding medium to generate a considerable osmotic pressure, which is the driving force for inflating the outer exine; and 2) the stiffer exine along with the three apertures exerts an inhomogeneous constraint on intine swelling. Therefore, the swelling-induced mechanical pressure exerted at the outer surface of the intine layer ( ) due to osmotic pressure decreases as swelling proceeds. The intine swelling pressure ( ) equals the equilibrium pressure that is required to inflate the exine layer ( ). The swelling ratio of the intine, is equal to the inflation ratio of exine, ince the intine layer is tightly bound to the exine layer (Supplementary Fig. 13a). Details of the overall approach behind the multiphysics and computational models are described in Refs. 29-31.

Constitutive law for intine layer based on a multiphysics model
Based on the experimental results, the intine layer of pollen microgel particles is shown to behave as a stimuli-responsive hydrogel, which provides the driving force for inflation and deflation of the outer exine layer in pollen microgel particles. For simplification, the pollen intine immersed in bath solution with mobile ions are represented approximately by spherical shell structures without apertures, as shown in Supplementary Fig. 13b. The equilibrium of pollen intine swelling was modeled in the radial direction due to spherical symmetry. Therefore, the present computational domain consists of (i) the pollen intine layer represented by a shell-like structure, (ii) the surrounding buffer medium, including both internal and external solutions, and (iii) the interface over the pollen intine and solution domains.
For theoretical formulation of the model in a Lagrangian scheme, several assumptions were made as follows: (i) the pollen intine system is maintained under isothermal conditions, such that the dissociation constant K is independent of temperature; (ii) no chemical reaction occurs that generates extra ions(ion concentration is constant); and (iii) the pollen intine is placed in an unstirred solution and thus the effect of convection on ionic diffusion is negligible. In line with the experiments, all four diffusive species in the system, including the pollen intine and surrounding solution, are considered in the present multiphysics model, namely hydrogen ions (H + ), hydroxide ions (OH -), chloride ions (Cl -), and respective cations (i.e., K + or Ca 2+ ). According to the law of mass conservation, the Nernst-Planck equation is used for characterization of ionic diffusion, as given below, where X is the material coordinate, and (mM) are the valence number and molar concentration of the ionic species k, (V) is the electric potential, and , T and re the universal gas constant (8.314 J/(mol•K)), the room temperature (298 K), and Faraday constant (9.6487×10 4 C/mol), respectively. In addition, C = F T F , where F = ∂x ∂X = (λ , λ , λ ) is the deformation gradient.
For the distributive electric potential, the Poisson equation is given as follows, k where J = det F is the change in volume of the pollen intine layer due to swelling, is the vacuum dielectric permittivity (8.854×10 -12 C 2 /N • m 2 ), is the relative dielectric permittivity, is the valence number of the fixed charge, and is the fixed charge density given below 43 , where is the initial molar concentration of ionizable groups within the pollen intine in the dry state, K (mM) is the dissociation constant, and is the molar concentration of hydronium ions.
For the effect of mechanical equilibrium on pollen intine swelling, the mechanical governing equation with nonlinear deformation is employed as 30 , where G (GPa is the pollen intine shear modulus associated with the Young's modulus E and the Poisson's ratio υ via G = E/[2(1 + υ)], and Π is the osmotic pressure given by 29 , where the overbar ( denotes the ion concentration in the solution, is the temperature in the unit of energy, is the volume per solvent species, is the number of solvent species divided by the volume of the dry polymer, and is the Flory-Huggins parameter to characterize the interactions between solvent molecules and the polymer network inside the pollen intine. From Equation (6), it is found that the osmotic pressure inside the pollen intine Π is due to mixing of the polymer and solvent molecules, as well as due to the imbalance of ions in the external solution and the pollen intine, including mobile ions and fixed charges.
The stress in the radial direction is then obtained by substituting the deformation gradient and osmotic pressure into Equation (7), as follows: Supplementary Fig. 13b, for the spherical symmetric problem, the Neumann type of the electrochemical boundary conditions are imposed at the center of the pollen, located at = 0 , namely, A free swelling process is conducted to identify the ionic concentrations at equilibrium, and an equilibrium state over the spherical shell-solution interface is given as The driving pressure is then achieved by inputting the equilibrium ionic concentrations into Equation (8) 31 . In order to perform numerical simulations of pollen intine swelling in the ionic solution, all the inputs required by the multiphysics model are tabulated in Supplementary Table 2. Among the inputs required in the present model, only the initial fixed charge density in the dry state was not directly obtained in the experiment. Thus, by an approach to the inverse problem, = 1.7848 × 10 4 mM was identified through the present model associated with the experimental data, where the swelling of the deformed intine layer ( = 250 MPa) was λ =1.5946 at pH = 7. Moreover, the constant = 1.7848 × 10 4 mM was further validated with the FEA modeling of the exine layer where intine swelling was constrained by the external exine layer, thus free swelling of the intine layer became swelling-induced mechanical pressure exerted to the internal surface of the exine layer. Since is an initial material property of the pollen intine, and is independent of the pollen intine swelling process, the well-identified value of could provide consistent modeling prediction which showed good agreement with experimental observations.

Constitutive law for exine layer
The maximum swelling diameter of pollen microgel particles reached up to 1.8 times the size of their original diameter, as seen for the case of 6 h KOH-treated pollen samples. Large deformation around the tips of the three apertures can occur due to the high-stress concentration factor at the tip of a crack 44 . Moreover, the swelling/de-swelling behavior of pollen microgel particles was reversible. Thus, the particles clearly showed rubber-like hyperelastic behavior. Therefore, we built threedimensional computational models for simulating the micromechanics of the exine layer in pollen particles based on the classic hyperelastic neo-Hookean model 45 . In the context of pollen microgel particles, about the paucity of reliable data on material properties of exine layer and the lack of detailed studies of its constitutive response led us to choose the simple but efficient neo-Hookean model, rather than complex models with many adjustable parameters.
The strain energy density of the neo-Hookean model is given by where U is the elastic strain energy per unit reference volume; , , and are the three principal stretches; and s the total volume ratio, defined as J = 45,46 . Two material constants, and 1 , were chosen to simulate the inflation and deflation behavior of the exine layer under large elastic strain.
is related to the initial shear modulus G as follows, where is the bulk modulus, and G and are related to the Young's modulus E, which retrieved from AFM force-based nanoindentation tests performed on the pollen particles (Supplementary Figs. 11-12), by the formulas, G = E/2(1 + υ) and = E/3(1 − 2υ) with a Poisson's ratio, υ. Thus, we can express the two material constants, and 1 , in terms of exine material parameters, E and υ, as follows, Considering the presence of pores in the exine layer, a value of 0.4 for the Poisson's ratio was selected to account for exine compressibility.

FEA modeling for inflation/deflation of exine layer
The geometry of pollen microgel particles was determined by direct experimental measurements using a scanning electron microscope (SEM). The average thickness of the exine layer was determined to be 0.58 μm, and the mean outer diameter of the particles was 28 μm. Spikes associated with the exine layer were defined as cones with a height of 5 μm and a base diameter of 3 μm. The aperture length was described relative to a central angle of 90°, and the tip radius of curvature of the aperture was set as 1.0 μm. This latter value was selected in order to avoid convergence issues related to large deformation around the aperture during finite element simulations of inflation of pollen exine (Supplementary Fig. 13c). A one-third symmetry structure with only one aperture and 38 spikes was constructed as the exine for one microgel particle, where symmetric boundary conditions were applied at the perimeter (Supplementary Fig. 13d). Furthermore, the density of exine was taken as 1 g/cm 3 . Considering the small size and low density of the pollen grains in solvents, gravitational effects were not considered in the model. The outer surface of the exine layer was tightly attached to the spikes, and an inflation pressure ( ) generated by osmotic pressure of the intine ( ) was exerted on the inner surface of the exine layer. The effects of mesh size on the deformation capacity were studied by an adaptive mesh strategy, where the largest mesh size was 0.5 μm. Therefore, various mesh sizes were employed for small deformation at regions distant from the aperture and were decreased to a minimum value of 0.02 μm for large deformation (ε > 2) around the tip of the aperture. Meanwhile, the artificial strain energy of the whole model associated with hourglass was less than 2% of the total internal energy. It should be noted that a severely inhomogeneous deformation was observed during the swelling process so the swelling diameter was measured by a red circle in the direction of maximum deformation, as indicated in Supplementary Fig. 13e. To investigate the inflation/deflation processes as well as related deformation mechanisms, we employed a three-dimensional model of the exine layer based on finite element analysis (FEA) using a commercial software package ABAQUS 2017 ® (Dassault Systèmes SIMULIA, Johnston, RI). The exine model was generated by Python scripts and then ran in parallel in 32 cores. Quasi-static solutions were calculated using ABAQUS/Explicit 46,47 to simulate the inflation process of the pollen exine layer with different mechanical properties under an internal pressure while kinematic effects were ignored. The finite element mesh entailed solid eight-node brick elements with reduced integration (C3D8R), while the spike on the pollen exine surface was treated as a rigid surface without deformation throughout the entire simulation. A total of 75,938 elements were used in the pollen exine model. A pseudo step time of 30 μs for inflating and 30 μs for deflating the 12 exine structure was set in order to minimize the kinetic energy. The inflation pressure ( = ) was predicted by the intine swelling model ( = based on the relationship, = ; thus, the boundary condition applied to the internal surface of the exine layer was = . A maximum internal pressure ( , ) of 8.3 MPa, induced by the swollen intine layer, was predicted by the multiphysics model that corresponded to 6 h KOH-treated pollen microgels incubated in pH 10 solution; this was the treatment condition which exhibited maximum swelling among all pollen microgel particle specimens. The reference pressure was, therefore, set as 8.3 MPa ( , ) and 0 MPa ( , ) for conditions corresponding to pH 10 and pH 2 conditions, respectively, for all of the simulations, and the Young's modulus values and the predicted internal pressure values were systematically varied in the parametric studies, as indicated in Supplementary Table 1. The pressure boundary condition on the inner surface of the exine layer was applied with a defined load profile gradually increasing up to 8.3 MPa and then decreasing to zero to improve the convergence of the model in large deformation simulations. Table 2 has been added in the revised manuscript and describes the input parameters for the numerical simulations as follows:  1. Flowchart of the modeling process by the dual approach combining multiphysics and FEA modeling. A. Flowchart of the modeling process for the determination of the initial fixed charge for the intine multiphysics model. B. Flowchart of the modeling process for stress/strain analyses of pollen microgels, after obtaining the constant value of the initial fixed charge by repeating the process. material to be = √ ⁄ ≈ 600 / according to its Young's moduli, , of 400MPa and density of 1g/ 3 . At the same time, we calculated the impact velocity, = ⁄ = 0.8 × 28 30 ⁄ ≈ 0.7 / , according to i) the step time t =30 and ii) global deflection displacement, = 0 ( − 1), where the original pollen diameter ( 0 ) is 28 μm and the maximum swelling ratio ( ) of pollen is about 1.8. Therefore, the quasistatic requirement is fully satisfied since ⁄ ≪ 1%. It is noteworthy that the scale of deformation is on the order of micrometers so the required time steps for quasi-static analysis in ABAQUS/ Explicit should be on the order of micrometers as well. 2. In fact, the time steps in the FEA simulation represent a pseudo time rather than real physical time of the swelling event. Although, in Abaqus/Explicit, a longer step time reduces the kinematic energy, making quasi-static solutions closer to static, it requires considerable computing resources. Hence, we carefully chose the step time of our simulations in times on the order of microseconds. As shown in ROM Fig. 2, with this microscale time step, the kinematic effects were almost negligible. Thus, we confirmed that we firmly obtained reasonable quasi-static solutions using Abaqus/Explicit under feasible model set-up conditions.