Osmosensor-mediated control of Ca2+ spiking in pollen germination

Pei, Songyu; Tao, Qi; Li, Wenke; Qi, Guoning; Wang, Borong; Wang, Yan; Dai, Shiwen; Shen, Qiujing; Wang, Xi; Wu, Xiaomei; Xu, Shijian; Theprungsirikul, Lynn; Zhang, Jingyuan; Liang, Liang; Liu, Yuantao; Chen, Kena; Shen, Yang; Crawford, Bridget M.; Cheng, Mengjia; Zhang, Qi; Wang, Yiqi; Liu, Hongli; Yang, Benguang; Krichilsky, Bryan; Pei, Jessica; Song, Karen; Johnson, Douglas M.; Jiang, Zhonghao; Wu, Feihua; Swift, Gary B.; Yang, Huanghe; Liu, Zhonghua; Zou, Xuexiao; Vo-Dinh, Tuan; Liu, Feng; Pei, Zhen-Ming; Yuan, Fang

doi:10.1038/s41586-024-07445-6

Download PDF

Article
Open access
Published: 22 May 2024

Osmosensor-mediated control of Ca²⁺ spiking in pollen germination

Songyu Pei^1,2,3,4,5^na1,
Qi Tao^4,5^na1,
Wenke Li^4,5^na1,
Guoning Qi⁵^na1,
Borong Wang⁵,
Yan Wang¹,
Shiwen Dai¹,
Qiujing Shen⁵,
Xi Wang^4,5,
Xiaomei Wu⁵,
Shijian Xu²,
Lynn Theprungsirikul²,
Jingyuan Zhang²,
Liang Liang⁵,
Yuantao Liu^4,5,
Kena Chen⁵,
Yang Shen⁵,
Bridget M. Crawford³,
Mengjia Cheng⁵,
Qi Zhang^4,5,
Yiqi Wang⁵,
Hongli Liu⁵,
Benguang Yang⁵,
Bryan Krichilsky²,
Jessica Pei^2,6,
Karen Song ORCID: orcid.org/0000-0002-0591-8393²,
Douglas M. Johnson⁷,
Zhonghao Jiang²,
Feihua Wu²,
Gary B. Swift⁷,
Huanghe Yang ORCID: orcid.org/0000-0001-9521-9328⁸,
Zhonghua Liu¹,
Xuexiao Zou¹,
Tuan Vo-Dinh ORCID: orcid.org/0000-0003-3701-3326³,
Feng Liu ORCID: orcid.org/0000-0002-9781-4048^1,2,
Zhen-Ming Pei ORCID: orcid.org/0000-0003-1473-696X^2,3^na2 &
…
Fang Yuan ORCID: orcid.org/0000-0001-6498-8079^1,2,5^na2

Nature volume 629, pages 1118–1125 (2024)Cite this article

24k Accesses
16 Citations
14 Altmetric
Metrics details

Subjects

Abstract

Higher plants survive terrestrial water deficiency and fluctuation by arresting cellular activities (dehydration) and resuscitating processes (rehydration). However, how plants monitor water availability during rehydration is unknown. Although increases in hypo-osmolarity-induced cytosolic Ca²⁺ concentration (HOSCA) have long been postulated to be the mechanism for sensing hypo-osmolarity in rehydration^1,2, the molecular basis remains unknown. Because osmolarity triggers membrane tension and the osmosensing specificity of osmosensing channels can only be determined in vivo^3,4,5, these channels have been classified as a subtype of mechanosensors. Here we identify bona fide cell surface hypo-osmosensors in Arabidopsis and find that pollen Ca²⁺ spiking is controlled directly by water through these hypo-osmosensors—that is, Ca²⁺ spiking is the second messenger for water status. We developed a functional expression screen in Escherichia coli for hypo-osmosensitive channels and identified OSCA2.1, a member of the hyperosmolarity-gated calcium-permeable channel (OSCA) family of proteins⁶. We screened single and high-order OSCA mutants, and observed that the osca2.1/osca2.2 double-knockout mutant was impaired in pollen germination and HOSCA. OSCA2.1 and OSCA2.2 function as hypo-osmosensitive Ca²⁺-permeable channels in planta and in HEK293 cells. Decreasing osmolarity of the medium enhanced pollen Ca²⁺ oscillations, which were mediated by OSCA2.1 and OSCA2.2 and required for germination. OSCA2.1 and OSCA2.2 convert extracellular water status into Ca²⁺ spiking in pollen and may serve as essential hypo-osmosensors for tracking rehydration in plants.

Development of a robust hydroponic method for screening of sunflower (Helianthus annuus L.) accessions for tolerance to heat and osmotic stress

Article Open access 18 January 2021

Hydrogen peroxide signal photosynthetic acclimation of Solanum lycopersicum L. cv Micro-Tom under water deficit

Article Open access 11 August 2023

Changes in physio-biochemical parameters and expression of metallothioneins in Avena sativa L. in response to drought

Article Open access 12 February 2023

Main

All living organisms, particularly sessile land plants, must monitor water in their environment to programme growth and development^3,7. Land plants evolved from aquatic ancestors and adapted to the terrestrial environment by overcoming two seemingly insurmountable obstacles: water deficiency and fluctuation⁸. To survive dry environments and harsh seasons, ancestral land plants not only acquired specific structures for water acquisition, transport and management, but also improved developmental strategies. One of their most effective strategies is to arrest cellular activities via water loss to form drying or desiccation-tolerant structures, such as spores, pollen and seeds^8,9,10. Rehydration is likely to occur via hypo-osmosensing processes that resemble cell surface hypo-osmosensors in other organisms^4,11,12 and involve second messenger systems (Extended Data Fig. 1a)—that is, the lower osmolarity in extracellular spaces relative to cytosol is tracked and converted to a second messenger to initiate biological activities. However, the plant hypo-osmosensor remains unidentified^13,14.

More than 35 years ago, Tazawa et al. observed a Ca²⁺ requirement for cell turgor regulation in algae², and recorded HOSCA using injected Ca²⁺-sensitive bioluminescent aequorin¹⁵, which was later confirmed in algae using injected fluorescence dyes^16,17 and in aequorin-expressing tobacco¹⁸. HOSCA has long been hypothesized to be a hypo-osmolarity perceiving mechanism, as Ca²⁺ acts as a second messenger for various stimuli^19,20,21 (Extended Data Fig. 1a–c). Indeed in animals, some transient receptor potential (TRP) channels—a family of around 30 sensory Ca²⁺-permeable channels for diverse stimuli—function as hypo-osmosensors^5,12,22,23. Nevertheless HOSCA-associated molecular components remain unidentified in plants, partially owing to technical challenges of isolating them (Extended Data Fig. 1b–d).

To demonstrate a function for HOSCA in sensing of hypo-osmolarity (hypo-osmosensing), several criteria must be met. First, the HOSCA mutant must exhibit defective HOSCA. Second, the physiological processes in response to hypo-osmotic treatment must be compromised in HOSCA mutants. Third, if the HOSCA encodes an ion channel, the channel must be gated by hypo-osmolarity in planta. We previously used aequorin Ca²⁺ imaging-based genetic screens to identify the cell surface sensors for hyper-osmotic stress (OSCA1 (ref. ⁶)), salt stress (GIPC²⁴) and H₂O₂ (HPCA1 (ref. ²⁵)). Similarly, the receptors for ATP, lipopolysaccharide and quinone have also been indentified^26,27,28. In contrast to these stimuli, for HOSCA measurements, vegetative tissues must be treated first with hyper-osmotic stress, and then with hypo-osmotic shock, which causes enormous variations of Ca²⁺ increases (Extended Data Fig. 1d), precluding Ca²⁺-dependent genetic screens. Conversely, given that several OSCAs have been identified as hyper-osmolarity-gated Ca²⁺-permeable channels^{6,29,30,31,32}—that is, hyper-osmosensors—it is possible that some of the 15 OSCA family members in Arabidopsis could function as hypo-osmosensors. In eukaryotes, mechanosensing channels include TRPs, epithelial sodium channels, two-pore ___domain potassium channels, MscS-like channels (MSLs), piezo channels and OSCAs^{4,5,6,23,33,34,35}. There are no plant homologues of TRP, epithelial sodium channels or two-pore ___domain potassium channels, whereas they do express MSLs, piezo channels and OSCAs^13,20,35. Most MSLs are anion channels that are located in endomembranes and regulate organelle size³³, and piezo is also localized to endomembranes^36,37. In addition, MCA1 and MCA2 function as Ca²⁺-permeable mechanosensitive channels in roots³⁸. Structural analyses of OSCAs have shown the importance of lipid interactions and conformational changes in their activation, and some OSCAs can be activated by hyper-osmotic treatments and mechanical poke or negative pressure to the plasma membrane when analysed in heterologous expression systems^{6,29,30,31,32}, but they have not been demonstrated to function as hypo-osmosensors in planta. However, the expression systems that have been used to analyse other hypo-osmosensors, such as animal cells for TRPV4 and E. coli for MscS and MscL mechanosensitive channels^22,39, have not been used to screen OSCAs stringently for their hypo-osmosensing activities.

During our initial cloning of OSCAs, we found that they were toxic to E. coli and that the toxicity was suppressed by increasing the osmolarity of the medium, suggesting that the toxicity might be caused by hypo-osmosensitive activities. Here we developed a robust growth screen in E. coli to systematically analyse this subtle phenotype, identified OSCAs as hypo-osmosensors, revealed their activities responsible for HOSCA in Arabidopsis pollen, and identified pollen Ca²⁺ oscillations as the second messenger that enables apoplastic osmolarity to initiate germination.

Screening for hypo-osmosensitive OSCAs

The lack of molecular information on HOSCA prompted us to use established approaches to identify hypo-osmosensing channels, but all attempts unfortunately failed. As TRPV4 is a hypo-osmosensor²², we used expression cloning in human embryonic kidney 293 (HEK293) cells. We sought to develop Ca²⁺-genetic screens for HOSCA mutants. We expressed 15 OSCAs in HEK293 cells but observed mixed channel specificities that did not distinguish between hypo-osmotic shock and mechanical stress. We expressed 15 OSCAs in the E. coli 7-MscL/MscS-knockout strain³⁹. We also tested individual transfer DNA (T-DNA) mutants of 15 OSCAs, but could not identify phenotypes with reduced HOSCA. Eventually, we followed the assumption that OSCAs could encode both hyper- and hypo-osmosensing channels. During our initial cloning of OSCAs into E. coli vectors, we encountered difficulties, possibly owing to their toxicity⁴⁰. Because excess Na⁺ ions might permeate through OSCA channels, we grew the E. coli cells with reduced NaCl in the medium to clone several OSCAs into the vectors, but some of these OSCAs could not be cloned until the osmolarity was increased. We speculated that these OSCAs may be active and toxic under low osmolarity, which motivated us to conduct thorough analyses on these channels.

In the low-salt hypo-osmotic medium, E. coli clones encoding 15 different OSCAs grew normally without isopropyl β-d-1-thiogalactopyranoside (IPTG) induction (Fig. 1a and Extended Data Fig. 2a). In the presence of IPTG, cells expressing OSCA2.1, OSCA2.5 and OSCA1.3 grew more slowly (Fig. 1b), and the Ca²⁺ channel inhibitor La³⁺ restored growth (Fig. 1c). Increases in osmolarity of the medium rescued growth for cells expressing OSCA2.1 and OSCA1.3, but not those expressing OSCA2.5 (Fig. 1d). Direct comparison of these cells confirmed that OSCA2.1, OSCA1.3 and OSCA2.5 were toxic under hypo-osmotic conditions and that the Ca²⁺ channel inhibitors La³⁺ and Gd³⁺ blocked the toxicity of all three channels, whereas sorbitol suppressed the toxicity of OSCA2.1 and OSCA1.3 only (Fig. 1e–g and Extended Data Fig. 2b–f). These results suggest that OSCA2.1 and OSCA1.3 may be hypo-osmosensitive channels, whereas OSCA2.5 in likely to have a distinct physiological activation mechanism.

Fig. 1: Identification of hypo-osmosensitive OSCA2.1 by a functional expression screen in E. *coli.*

OSCAs are essential for pollen germination

Receptors convert an external signal into a second messenger (Extended Data Fig. 1a) to regulate downstream physiological processes. We therefore tested whether hypo-osmosensitive OSCAs are required for processes that are known to be regulated by water availability. Pollen germination and seed germination are essential processes for fitness and survival under extreme water regimes^7,8,10,14, in which cells emerge from a desiccated state to initiate biological activities in response to increases in water availability (Extended Data Fig. 1c). Single-cell haploid pollen provide a useful model for studying plant responses to external signals^9,10,41, whereas seed germination is involved in the coordination of multiple cell types.

We first screened Arabidopsis lines mutagenized by T-DNA insertion of 15 OSCAs using an in vitro pollen germination assay, but found no significant phenotypes (Fig. 2a). On the basis of the expression of OSCA2.1, OSCA1.3 and OSCA2.5 in E. coli, we generated double and triple mutants, and observed lower germination rates in the osca2.1/osca2.2 (hereafter osca2.1/2.2) double-knockout mutant (Fig. 2b and Extended Data Fig. 3a,b). This phenotype was verified by direct comparison of osca2.1 and osca2.2 single and double mutants (Fig. 2c and Extended Data Fig. 3c). The osca2.1/2.2 pollen exhibited less germination on stigmas than wild type, as shown by aniline blue staining and GFP fluorescence imaging (Fig. 2d,e and Extended Data Fig. 3d–h), showing that OSCA2.1 and OSCA2.2 (OSCA2.1/2.2) have physiologically important roles. OSCA2.1 and OSCA2.2 expression through their endogenous promoters complemented osca2.1/2.2 germination defects almost completely (Extended Data Fig. 3c,i), demonstrating that OSCA2.1 and OSCA2.2 functioned redundantly during pollen rehydration without altering pollen viability (Extended Data Fig. 3j,k).

Fig. 2: OSCA2.1 and OSCA2.2 are required for water-initiated and Ca²⁺-dependent pollen germination in *Arabidopsis.*

Regulation of pollen germination via OSCAs

It has long been known that Ca²⁺ is a requirement for pollen germination⁴². We reasoned that given that OSCA2.1 and OSCA2.2 might convey extracellular hypo-osmotic information to Ca²⁺ signalling, if osca2.1/2.2 cells were unable to track water status, decreases in medium osmolarity could overcome their germination defects. Given that sucrose serves as both a carbon source and an osmolyte, we prepared germination medium with a constant 300 mM sucrose—resulting in osmolarity lower than that of standard medium^9,10,41 (535 mOsm)—and varied the osmolarity using sorbitol. At high osmolarity, osca2.1/2.2 pollen germinated at lower rates than the wild type (Fig. 2f, left), and at low osmolarity, osca2.1/2.2 germinated at similar rates to the wild type (Fig. 2f, middle). However, osca2.1/2.2 pollen did not germinate when the Ca²⁺ concentration was decreased from the standard 5 mM to 0.5 mM (Fig. 2f, right). The synergistic effect of osmolarity and Ca²⁺ showed that osca2.1/2.2 mutant pollen is less sensitive to decreases in osmolarity than wild type (Fig. 2g), and lowering Ca²⁺ concentration exacerbated this phenotype (Fig. 2h). Increased Ca²⁺ alone did not rescue the mutant phenotype (Extended Data Fig. 4a,b). These data show that OSCA2.1 and OSCA2.2 have an essential role in the regulation of pollen germination by medium Ca²⁺ calcium concentration^41,42,43.

OSCAs are vital for hypo-osmotic Ca²⁺ spiking

To examine HOSCA in planta^16,17,18 (Extended Data Fig. 1c,d), we generated transgenic lines expressing the GFP-based Ca²⁺ indicator GCaMP6 (ref. ⁴⁴). We placed pollen grains from these lines on agarose medium containing 300 mM sorbitol for 1 h, and then treated them with hypo-osmotic solutions. Hypo-osmotic shock did not induce increases in intracellular Ca²⁺ concentration ([Ca²⁺]_i) in osca2.1/2.2 pollen (Fig. 3a–c and Supplementary Video 1). HOSCA was not affected in osca2.1 or osca2.2 single mutants, and the HOSCA phenotypes of osca2.1/2.2 mutant pollen could be rescued by OSCA2.1 or OSCA2.2 (Extended Data Fig. 4c–e). Increasing the osmolarity of the medium used to pretreat the pollen intensified HOSCA in both osca2.1/2.2 and wild type, but the effect was much weaker in osca2.1/2.2 pollen (Fig. 3d).

**Fig. 3: OSCA2.1 and OSCA2.2 function as hypo-osmosensitive Ca²⁺-permeable channels in pollen grains and HEK293 cells.**

To further pinpoint molecular mechanisms and physiological functions of OSCA2.1 and OSCA2.2, we determined their expression patterns and subcellular localization. In GUS-reporter transgenic plants, OSCA2.1 and OSCA2.2 were expressed in whole seedlings, flowers and pollen (Fig. 3e) and expression was increased during pollen hydration (Extended Data Fig. 4f). OSCA2.1–GFP and OSCA2.2–GFP accumulated at the cell surface before hydration and at germination, and also distributed in the cytosol, possibly in preparation to form new plasma membrane (Fig. 3f), similar to MSL8 (ref. ⁴⁵). OSCA2.1–GFP and OSCA2.2–GFP were also targeted to the cell surface and appeared to undergo exocytosis and endocytosis in newly developed pollen tubes, whereas GFP remained in the cytosol (Extended Data Fig. 4g,h).

OSCAs form hypo-osmosensing Ca²⁺ channels

To determine whether OSCA2.1 and OSCA2.2 can facilitate Ca²⁺ influx in response to hypo-osmotic treatments, we analysed OSCA2.1 and OSCA2.2 in HEK293 cells^6,22. In response to decreases in osmolarity, cells expressing OSCA2.1 or OSCA2.2 showed larger [Ca²⁺]_i increases than those harbouring empty vector (Fig. 3g–j). OSCA2.1–GFP and OSCA2.2–GFP were localized to the vicinity of the cell surface (images in Fig. 3i,j and Extended Data Fig. 5a). Similar to TRPV4, OSCA2.1 and OSCA2.2 enabled HOSCA in HEK293 cells, and their activities were blocked by La³⁺ (Fig. 3k and Extended Data Fig. 5b–i). HOSCA decayed similarly in cells expressing OSCA2.1 or OSCA2.2, and differed in cells expressing empty vector or TRPV4 (Fig. 3l). Similar to TRP desensitization^23,34,46, in response to successive hypo-osmolarity challenges, robust desensitization was recorded in cells expressing OSCA2.1 or OSCA2.2 but not those with empty vector, whereas TRPV4-expressing cells displayed only mild desensitization (Extended Data Fig. 6). These results show that OSCA2.1 and OSCA2.2 have a distinct hypo-osmosensor activity.

OSCAs control pollen Ca²⁺ oscillations

Pollen dehydration occurs before anthesis, which is vital for resistance to environmental stress during dispersal^9,10. Although Ca²⁺ gradients and oscillations in pollen tubes were initially studied using microinjected Ca²⁺-sensitive dyes^41,43, Ca²⁺ levels in pollen grains can only be detected using transgenic Ca²⁺ indicators^47,48. Ca²⁺ signatures in pollen grains and their regulatory mechanisms as well as the causal relationship with germination thus remain poorly understood.

To analyse whether and how OSCA2.1 and OSCA2.2 act as hypo-osmosensors to control pollen Ca²⁺ signalling, we used GCaMP6m, which has high sensitivity and balanced kinetics⁴⁴. We established a GCaMP6-based assay in wild-type pollen in standard medium, and observed [Ca²⁺]_i oscillations over a 300-min period (Supplementary Video 2). To take into consideration the variation of GCaMP6 expression and fluorescence bleaching associated with long recordings, we first analysed fluorescence intensities over a longer timescale to determine the broad trend in Ca²⁺ signalling during germination. Pollen [Ca²⁺]_i increased from the early stage of germination to about the germinating stage in both genotypes, but was weaker in osca2.1/2.2 pollen, leading to lower total [Ca²⁺]_i over 300 min in osca2.1/2.2 (Extended Data Fig. 7).

To quantify systematically and statistically the timecourse of [Ca²⁺]_i oscillations, we recorded patterns of [Ca²⁺]_i. We observed a first resting phase (RePh1), followed by [Ca²⁺]_i oscillations with small amplitudes (CaOsc^S), then a second resting phase (RePh2), followed by [Ca²⁺]_i oscillations with large amplitudes (CaOsc^L) that led to tube protrusion, and finally a third resting phase (RePh3) (Fig. 4a and Supplementary Video 2). Dozens of CaOsc^S appeared evenly distributed in the cytosol, and several CaOsc^L gradually propagated towards the germination aperture and finally triggered tube protrusion.

Fig. 4: [Ca²⁺]_i oscillations in pollen grains prior to germination are impaired in *osca2.1/2.2.*

Typical osca2.1/2.2 pollen had markedly reduced [Ca²⁺]_i oscillations, including but not limited to largely extended RePh1, shortened CaOsc^S, abolished RePh2 and diminished CaOsc^L with fewer irregular spikes, which often did not trigger germination within 300 min (Fig. 4b). If pollen did not germinate, several CaOsc^S → RePh2 → CaOsc^L → RePh3-like modules re-occurred afterwards until germination was triggered in both genotypes, but much more so in osca2.1/2.2 pollen. Given that OSCA2.1 and OSCA2.2 are hypo-osmosensors, the difference in Ca²⁺ spiking in osca2.1/2.2 pollen suggests that Ca²⁺ spiking may function as a second messenger for extracellular water status.

OSCAs couple external osmolarity to Ca²⁺ spiking

To further assess whether [Ca²⁺]_i oscillates faithfully according to water availability in the medium, we used hypo-osmotic medium, which resulted in osca2.1/2.2 germination defects being largely restored, and hyper-osmotic medium, which resulted in prominent osca2.1/2.2 germination defects (Fig. 2g). At 350 mOsm, the whole set of [Ca²⁺]_i signatures seen in Fig. 4a occurred in wild-type pollen, whereas at 680 mOsm, CaOsc^S were overwhelmingly extended over RePh2, CaOsc^L and RePh3 (Fig. 5a,b, Extended Data Fig. 8 and Supplementary Videos 3 and 4). [Ca²⁺]_i patterns were substantially altered under the hyper-osmotic condition, including but not limited to extended CaOsc^S, delayed and/or diminished CaOsc^L, and reduced amplitudes of CaOsc^S and CaOsc^L (Fig. 5c,d).

**Fig. 5: Induction of pollen [Ca²⁺]_i oscillations by gradual increases in water availability that mimic the rehydration process is disrupted in *osca2.1/2.2* pollen grains.**

In 350 mOsm medium, [Ca²⁺]_i patterns in osca2.1/2.2 pollen resembled those seen in the wild-type pollen at 535 mOsm, however, [Ca²⁺]_i spiking was almost abolished at 680 mOsm (Fig. 5e–h). The CaOsc^S and CaOsc^L had reduced amplitudes and delayed initiation times in osca2.1/2.2 pollen. Results similar to these at 350 mOsm were also observed in the 420 mOsm hypo-osmotic medium (Extended Data Fig. 9), consistent with the germination phenotypes. For comparison, we quantified the period and the duration of individual spikes, and found that CaOsc^S and CaOsc^L in osca2.1/2.2 pollen had longer periods and longer durations (Extended Data Fig. 10a–d). Together, these results show that Ca²⁺ spiking in pollen was tightly controlled by the osmolarity of the medium, and that this link was weakened in osca2.1/2.2 pollen. Of note, prolonged pollen Ca²⁺ spiking and intensified Ca²⁺ spiking at the germination aperture have been observed previously, but were not quantified⁴⁸.

We assessed the number of [Ca²⁺]_i spikes in pollen placed on media with osmolarity reduction from 680 mOsm to 350 mOsm to mimic increases in water availability during rehydration (Fig. 2g). Decreases in osmolarity increased the total number of spikes in both genotypes, but much less in osca2.1/2.2 (Fig. 5i,j). The periods of CaOsc^S and CaOsc^L were shortened and their amplitudes were increased in an osmolarity-dependent manner (Fig. 5k,l and Extended Data Fig. 10e,f), illustrating that Ca²⁺ spiking was enhanced during rehydration, whereas osca2.1/2.2 pollen exhibited much weaker Ca²⁺ spiking. To further verify the causal relationship between [Ca²⁺]_i oscillations and OSCA2.1 and OSCA2.2, we analysed osca2.1, osca2.2 and osca2.1/2.2 complementation lines. [Ca²⁺]_i oscillations were not affected in osca2.1 or osca2.2 pollen, and OSCA2.1 and OSCA2.2 rescued osca2.1/2.2 defects in respect to the periods and amplitudes of CaOsc^S and CaOsc^L (Extended Data Fig. 10g–j). These findings demonstrate that the coupling of medium osmolarity to [Ca²⁺]_i oscillations is mediated by OSCA2.1 and OSCA2.2, serving as a molecular mechanism for sensing water availability during rehydration.

Discussion

Here we answer two long-standing questions regarding the nature of the molecular nature of HOSCA and whether HOSCA functions as a hypo-osmosensing mechanism. We also reveal that Ca²⁺ oscillations serve as a second messenger for a primary stimulus in pollen grains. Since second messengers are intracellular small molecules that transfer the extracellular signal received by cell surface receptors to the cytosol, our results reveal that OSCA2.1 and OSCA2.2 perceive extracellular osmolarity and convert to Ca²⁺ spiking in pollen. Note that although the exact gating mechanisms remain unknown, OSCA family members may be hypo- or hyper-osmosensitive, similar to TRP family members sensing hot and cold temperatures^23,34,46.

A central question is whether Ca²⁺ spiking could serve specifically as a second messenger for osmolarity in extracellular spaces, even though other cellular processes that occur during rehydration are also induced by water. Here we present several lines of evidence that this is likely. osca2.1/2.2 pollen did not germinate in normal 535 mOsm medium, suggesting that the initiation switch controlled by OSCA2.1 and OSCA2.2 is required even when there is sufficient water for germination. osca2.1/2.2 pollen also did not develop normal Ca²⁺ oscillations at 535 mOsm, suggesting that OSCA2.1 and OSCA2.2 are required for these signals. In addition, osca2.1/2.2 pollen did not display HOSCA, indicating that OSCA2.1 and OSCA2.2 sense hypo-osmotic shock. OSCA2.1 and OSCA2.2 also formed Ca²⁺-permeable channels, indicating that they may directly convert hypo-osmotic stimulus into Ca²⁺ increases. Together, these results closely link Ca²⁺ spiking to medium osmolarity via OSCA2.1 and OSCA2.2, establishing that water in the medium, OSCA2.1 and OSCA2.2, and Ca²⁺ spiking form a novel sequential signalling cascade.

The water potential (Ψ_w) controls water movements across cell membranes^7,8,12,14. Ψ_w is composed of solute potential (Ψ_s), hydrostatic pressure potential (Ψ_p) and matrix potential (Ψ_m) (Extended Data Fig. 10k). In freshwater land plants, Ψ_w in the extracellular solution (\({\psi }_{{{\rm{w}}}_{o}}\)) varies from −0.03 to −4.5 MPa (equivalent to a 150-fold difference in osmolarity from 12 to 1,800 mOsm). By contrast, in mammals there is a limited range of osmolarity fluctuations from −0.66 to −0.82 MPa, about 10% around the set-point of approximately 300 mOsm (Extended Data Fig. 10l). The large detection ranges of OSCA2.1 and OSCA2.2 in plants compared with hypo-osmosensors in animals is likely to be owing to the presence of a cell wall in plants. Desiccated seeds and pollen grains^7,8 can have extremely low Ψ_w of less than −100 MPa. During rehydration, \({\psi }_{{{\rm{w}}}_{o}}\) increases gradually, and when the cell membrane is established, \({\psi }_{{{\rm{w}}}_{o}}\) exceeds the Ψ_w in the cytosol (\({\psi }_{{{\rm{w}}}_{i}}\)), which activates hypo-osmosensors, leading to Ca²⁺ influx.

Ca²⁺ spiking with distinct signatures occurs in various signalling processes in plants, including nodulation and mycorrhizal symbiotic establishment in root hairs, abscisic acid-induced stomatal closure, circadian oscillations, diatom osmoregulation and pollen tube tip growth^9,19,21,41. Ca²⁺ spiking in pollen grains resembles Ca²⁺ oscillations in pollen tube tips^41,48, although pollen Ca²⁺ spiking occurred less evenly and with greater fluctuations in amplitudes and periods, and with periods longer than those in pollen tubes⁴³ (4–10 min versus around 20 s). Whether OSCA2.1 and OSCA2.2 have a role in Ca²⁺ oscillations in pollen tubes remains to be determined. Notably, several Ca²⁺ channels have been associated with Ca²⁺ oscillations in pollen tube tips, but the gating mechanisms and roles remain poorly understood^9,10. For instance, cyclic nucleotide-gated channels and glutamate-like receptors are known to be involved in Ca²⁺ oscillations in pollen tubes^9,49,50. In addition, cell surface MSL8 permeates anions and regulates pollen volume and integrity but inhibits pollen germination⁴⁵. A key open question is how OSCA2.1 and OSCA2.2 work in concert with such channels to monitor water availability to generate Ca²⁺ spikes essential for pollen germination, tube growth and egg fertilization. It is possible that at a given osmolarity, many osmosensors, including hyper- and hypo-osmosensors, might function together to maintain general Ca²⁺ homeostasis in pollen. Our identification of the water → OSCA2.1–OSCA2.2 → Ca²⁺ spiking pathway in pollen grains will open new avenues for dissecting Ca²⁺ oscillations in pollen tubes. Nevertheless, although water is the most important driving force for cell growth^3,7,14,43, it remains to be determined whether this water–hypo-osmosensor–Ca²⁺ spiking pathway exists in vegetative tissues. The OSCA family originated during the evolution of protists³⁵, implying a role in maintaining cell shape and turgor^5,6. Moreover, plant OSCAs expended greatly and evolved coincidently with the plant transition from water to land³⁵, suggesting an essential role in plant responses to water fluctuations. In conclusion, we have identified and characterized OSCA2.1 and OSCA2.2 as the long-sought hypo-osmosensitive Ca²⁺ channels in plants, and identified the role of Ca²⁺ spiking as the second messenger for water availability during pollen germination.

Methods

Screen based on functional expression of OSCA in E. coli

The bacterial growth assay was designed with the consideration of either the complementary effect or the toxicity effect caused by ion channels as described^{39,40,51,52,53}. DNA sequences encoding 15 Arabidopsis OSCA family members⁶ were cloned into the Gateway entry plasmid pENTR/D-TOPO, and then into the destination plasmid pDEST14 for T7 promoter-based expression. Normal lysogeny broth (LB) medium contains 1% tryptone, 0.5% yeast extract and 171 mM NaCl (417 mOsm). Low-salt hypo-osmotic lysogeny broth (LSHypo-LB) medium was modified from LB medium, and contained 1% tryptone, 0.5% yeast extract, 5 mM NaCl and 3 mM KCl (70 mOsm). The osmolarity of LSHypo-LB medium was further adjusted by adding sorbitol as indicated. For spot growth assays, OSCA plasmids were transferred into the BL21(DE3) pLysS strain, which provides a tight control for expression of toxic proteins, and the freshly transformed strains were spread on an ampicillin and chloramphenicol plate and incubated overnight at 37 °C. Cultures were centrifuged, and precipitations were re-suspended using the LSHypo-LB medium. Cells were adjusted to optical density at 600 nm (OD₆₀₀) of 0.2, and serially diluted at 10×, 10³× and 10⁵×. These dilutions were spotted at the volume of 1.5 μl per spot onto LSHypo-LB plates. IPTG (1 mM), sorbitol, LaCl₃ and GdCl₃ at the indicated concentrations were added. Bacterial spot cultures were photographed and the bacterial growth rate was analysed using ImageJ⁵⁴. Osmolarity was measured with a vapour pressure osmometer (VAPRO 5600, Wescor).

Plant materials and growth conditions

Arabidopsis thaliana (Col-0) was used as the wild type. A. thaliana (Col-0) stably expressing GCaMP6m under the control of pollen-specific LAT52 promoter were generated and also used as the wild-type. The osca1.1-1 mutant was from our previous study⁶. Arabidopsis T-DNA insertion lines (Col-0) for 15 OSCAs were obtained from the ABRC and the GABI-Kat project (Supplementary Table 1), and the homozygous T-DNA insertion lines were verified as described previously⁶. Arabidopsis plants were grown on soil (Sungro, Professional growing mix), or in Petri dishes in half-strength Murashige and Skoog salts (0.5× MS; Sigma), 1% (w/v) sucrose (Sigma), and 0.8% (w/v) agarose (Sigma) in controlled environmental chambers or rooms at 21 ± 2°C and 65% relative humidity. The fluency rate of white light was ~110 μmol m⁻² s⁻¹. The photoperiods were 16 h light/8 h dark cycles. Arabidopsis seeds were sown on soil or 0.5× MS medium, placed at 4 °C for 4 days in the dark, and then transferred to growth rooms.

In vitro pollen germination assay

Pollen grains from freshly opened mature flowers (stage 12–13) of 8–12 week-old Arabidopsis plants were dispersed onto standard solid pollen germination medium containing 0.5% agarose, 300 mM (9.7% w/v) sucrose, 225 mM sorbitol, 1.6 mM boric acid, 1 mM CaCl₂, 1 mM Ca(NO₃)₂ and 1 mM MgSO₄ (adjusted to pH 6.3 with KOH; adjusted to osmolarity 535 mOsm with sorbitol), and placed at room temperature (22–24 °C) for 6 h for the in vitro germination assay^{55,56,57,58,59,60,61}. Pollen grains were photographed using inverted fluorescence microscopes (Axio Observer 3; Zeiss) equipped cooled CCD/CMOS cameras (CoolSNAP HQ2/Prime 95B; Teledyne Photometrics)^6,62 and MetaMorph 7.7 and MetaFluor 7.7 (Molecular Devices), and the germination rate was analysed using ImageJ. Osmolarity was measured with a Wescor vapour pressure osmometer.

In vivo pollen germination assays

In vivo pollen germination analyses were carried out by two approaches: imaging pollen constitutively expressing GFP driven by a pollen-specific LAT52 promoter^58,63, and aniline blue staining^64,65. The osca2.1/2.2 mutant was crossed into wild-type Arabidopsis expressing pLAT52-driven GFP (pLAT52::GFP) and homozygous lines were obtained. Flowers from the wild-type plant (not expressing GFP) were emasculated 24 h prior to pollination and left in the growth chamber or room until in vivo pollen germination assay. Pollen grains with or without pLAT52-driven GFP expression from newly opened flowers were dabbed onto the surfaces of pre-emasculated wild-type stigmas. For GFP-visualization assay, after 30 min of pollination, stigmas with pLAT52-driven GFP-expressing pollen grains were removed from the plants, placed on the cover glass, and imaged using a fluorescence stereo microscope (Axio Zoom.V16; Zeiss) equipped with a CCD camera (Axiocam MR R3, Zeiss). Excitation was provided at 488 (20) nm, and GFP fluorescence images at 509 (10) nm emission were collected using Zen 2012 software (Zeiss). For aniline blue staining, after 2 h of pollination using pollen grains without GFP expression, pistils were excised and fixed in Carnoy’s fixative (75% ethanol and 25% acetic acid), softened in 10 M NaOH, and stained in 0.1% aniline blue^64,65. Stained pistils were observed under the Axio V16 microscope, and fluorescence images were collected with excitation at 359 (20) nm and emission at 457 (20) nm. Because non-germinated grains were washed out for the aniline blue staining, the number of germinated grains that adhered to the stigma were counted. These in vivo germination analyses for wild type and osca2.1/2.2 were carried out side-by-side and with switched order between these genotypes to minimize variations, and the data represent more than five independent experiments.

Imaging of [Ca²⁺]_i in HEK293 cells

Analyses of ion channel activities in HEK293 cells were carried out as described previously^{6,62,66,67,68}. HEK293 cells were cultured and maintained in DMEM supplemented with 10% fetal bovine serum, 1% penicillin and streptomycin in CO₂ incubators at 37°C. For transfection, cells were seeded onto poly-lysine-coated eight-well chambered cover glasses (Nunc) overnight and transfected with plasmid DNA using Lipofectamine 3000 (Invitrogen). Cells were loaded with Fura-2AM (5 μM; Sigma), and a Fura-2-based Ca²⁺ imaging assay was performed in cells 18 to 24 h after transfection using an inverted fluorescence microscope (Axio Observer 3) equipped with two filter wheels (Lambda 10-3; Sutter Instruments), and cooled CCD and CMOS cameras (CoolSNAP HQ2/Prime 95B; Teledyne Photometrics)^6,62. Emission ratiometric images (F_{340 nm}/F_{380 nm}) were collected using MetaFluor software or Micro-Manger software (https://micro-manager.org/). Cells were incubated in a standard buffer containing 50 mM NaCl, 3 mM KCl, 0.6 mM MgCl₂, 10 mM glucose, 0.1 mM CaCl₂, 10 mM HEPES, and 160 mM mannitol (adjusted to pH7.4 with NaOH and osmolarity 300 mOsm with mannitol) for 30 min. For hypo-osmotic treatment, the bath was perfused using a peristaltic pump (Dynamax RP-1; Rainin) with hypotonic solution (140 mOsm; standard buffer without 160 mM mannitol), and HOSCA were recorded. For HOSCA desensitization analysis, cells were incubated in standard isosmotic solution (300 mOsm), and then treated with standard hypo-osmotic solution (140 mOsm) 3 times as illustrated in Extended Data Fig. 6a. Osmolarity was measured with a Wescor vapour pressure osmometer.

Analysis of OSCA–GFP/YFP expression

For analysis of OSCA–GFP in HEK293 cells⁶, cells were transfected transiently with pCMV::OSCA2.1-GFP, pCMV::OSCA2.2-GFP or pCMV::GFP, and about 18 to 24 h after transfection, GFP fluorescence imaging was carried out using a confocal microscope (LSM710 or LSM880 with Airyscan; Zeiss). For analysis of OSCA–GFP or OSCA–YFP in Arabidopsis, both OSCA endogenous promoter OSCA–YFP and pLAT52-driven OSCA–GFP transgenic single-insertion homozygous lines (pOSCA2.1::OSCA2.1-YFP, pOSCA2.2::OSCA2.2-YFP, pLAT52::OSCA2.1-GFP and pLAT52::OSCA2.2-GFP) were generated as described^{6,25,61,69,70}. The pLAT52::GFP transgenic plants were used as controls. Whole seedlings were imaged with the stereo microscope (Axio Zoom.V16) or the confocal microscope. The fluorescence in pollen grains and pollen tubes was analysed using a LSM880 confocal microscope with Airyscan. The plasma membrane was labelled with the FM4-64 dye and analysed as described previously^43,71,72,73 using the confocal microscope. For each transgenic line, more than ten independent transformants were analysed and similar results were obtained. The plasma membrane localization is also consistent with the prediction by SUBA4 (http://suba.plantenergy.uwa.edu.au/)⁷⁴.

Histochemical GUS activity analysis

Histochemical staining for GUS activity using the OSCA endogenous promoter-driven OSCA full-length genomic DNA–GUS (pOSCA2.1::genomic_OSCA2.1-GUS, pOSCA2.2::genomic_OSCA2.1-GUS) transgenic lines as described^6,25. Seedlings grown in 0.5× MS medium or soil were used for the histochemical staining⁷⁵. Data represent more than five independent lines, which displayed similar staining patterns. Flowers and seedlings were placed into GUS reaction solution, and imaged microscopically (SteREO Discover V20, Zeiss).

Pollen grain viability assay

Pollen grains from mature flowers at stage 12–13 were used for viability assay^45,76,77. Pollen grains were released on slides containing the solid germination medium (420 mOsm), the slides were placed upside down and incubated for 30 min in a chamber with ∼ 95% relative humidity. Then 100-μl drops of solution containing 1 μg ml⁻¹ fluorescein diacetate (FDA) and 0.5 μg ml⁻¹ propidium iodide (PI) were added. FDA and PI fluorescence was recorded in the GFP and dsRED fluorescence channels, respectively, using the confocal microscopy. FDA stains live grains, while PI stains the edge of live grains as well as compromised grains.

Imaging of [Ca²⁺]_i in pollen grains

Wild-type (Col-0) expressing GCaMP6 (ref. ⁴⁴) driven by the pollen-specific LAT52 promoter^48,49,72 (pLAT52::GCaMP6m) was used as the wild type. The osca2.1 and osca2.2 single mutants and the osca2.1/2.2 double mutant expressing GCaMP6, osca2.1 GCaMP6, osca2.2 GCaMP6 and osca2.1/2.2 GCaMP6, respectively, were used as osca2.1 and osca2.2 and osca2.1/2.2 mutants. OSCA2.1 osca2.1/2.2 GCaMP6 and OSCA2.2 osca2.1/2.2 GCaMP6 were generated using lines 9 and 10 described in Extended Data Fig. 3f, and used as complementation lines. Imaging and analysis [Ca²⁺]_i in pollen grains were carried out as described in previous studies with modifications^{6,25,56,72,78,79,80,81,82,83}. Pollen grains from freshly opened mature (stage 12–13) flowers were dispersed on the solid germination medium with indicated osmolarity. For GCaMP6-based [Ca²⁺]_i imaging, excitation was provided at 485 nm, and 510 nm emission images were taken with an exposure time of 100 ms and collected at the indicated intervals using the Axio Observer 3 microscope. More than 50 grains were imaged in one view each time, and GCaMP6 fluorescence intensity for each individual pollen grain was obtained for further analysis. For the analysis of relative [Ca²⁺]_i, GCaMP fluorescence ΔF/F₀ was calculated as (F − F₀)/F₀, where F₀ is the baseline fluorescence signal averaged over the first ten stable frames immediately before the start of treatments, and analysed using ImageJ.

For the hypo-osmotic shock treatment in plants, hyper-osmotic conditions were applied for a relatively long time (hours) first to generate a stable high osmotic status in plants, and then hypo-osmotic solutions were applied quickly for analysis of hypo-osmotic signalling processes that occur in minutes, which have been well described for various studies, including Ca²⁺ signalling^16,17,18,84. For analysis of hypo-osmotic shock-induced cytosolic free Ca²⁺ concentration ([Ca²⁺]_i) increases (HOSCA), pollen grains from freshly opened mature (stage 12–13) flowers were dispersed on high-osmolarity media modified from standard solid germination medium (535 mOsm) by adding sorbitol at the indicated concentrations for 1 h, and then perfused with the low-osmolarity solution modified from the standard solid germination solution without addition of sorbitol and agarose. Standard solid pollen germination medium contained 0.5% agarose, 300 mM (9.7% w/v) sucrose, 225 mM sorbitol, 1.6 mM boric acid, 1 mM CaCl₂, 1 mM Ca(NO₃)₂, and 1 mM MgSO₄ (adjusted to pH 6.3 with KOH; adjusted to osmolarity 535 mOsm with sorbitol). GCaMP florescence images were collected every 2 s with an exposure time of 100 ms for 350 s. Osmolarity was measured with a Wescor vapour pressure osmometer. Hypo-osmotic solution, which was prepared as the standard solution without the addition of 225 mM sorbitol, was added into the bath at the indicated time during continues imaging.

For the analysis of [Ca²⁺]_i oscillations in pollen grains, a two-step approach was used: visual analysis of the GCaMP6 fluorescence video, and quantification of the GCaMP6 fluorescence, both of which were then matched and adjusted for accuracy. First, the GCaMP6 fluorescence video, which was composed of images taken for pollen grains placed on germination medium every 30 s for 300 min, such as Supplementary Videos 2–4, was played using a 55 inch QLED 4 K TV monitor (Q80, Samsung) in a darkroom, and individual Ca²⁺-spiking events of [Ca²⁺]_i oscillations with small amplitudes (CaOsc^S) and large amplitudes (CaOsc^L) and their corresponding image number and time for each pollen gain were manually identified and counted. CaOsc^S were eye-detectable ‘flashes’ with not only increases in fluorescence intensity (about 2–3 times the baseline intensity), but also increases in the area of fluorescence. CaOsc^L were intensive flashes with increases in fluorescence intensity to 5 times or more the baseline intensity. Second, individual grains in the fluorescence images at the initial stage (taken within 5 min) were circled or drawn manually as regions of interest, and florescence intensities of regions of interest throughout the germination process for 300 min were calculated. Note that increases in pollen fluorescence area were not incorporated into the quantitative analyses. The fluorescence baseline was adjusted with fluorescence photobleaching correction. Oscillations of fluorescence intensity were matched to these CaOsc^S and CaOsc^L events, which were obtained by visual analysis in the first step, and classified into the simplest categorizations of relatively quiet resting phases (RePh1, RePh2 and RePh3) and CaOsc^S and CaOsc^L as described in detail in the main text.

Water potential

Ψ_w is composed of Ψ_s, Ψ_p and Ψ_m (interaction with matrices of solids and of macromolecules, such as cell walls) in plants⁷: Ψ_w = Ψ_s + Ψ_p + Ψ_m. The osmotic potential of pure water is zero, and the osmotic potential for a solution is always negative. The osmotic potential of a solution (in molarity) is calculated by using the following formula: Ψ_s = −iCRT, where i is ionization constant, C is molar concentration, R is the pressure constant, and T is the absolute temperature. The water potentials in plant cells and mammalian cells illustrated in Extended Data Fig. 10k,l are largely estimated using information from the literature^{7,8,12,85,86,87}—actual water potentials for a given species may vary greatly.

DNA constructs and transgenic lines

For E. coli growth assay, cDNAs encoding 15 Arabidopsis OSCA family members⁶ were amplified by PCR using primers listed in Supplementary Table 2. The PCR products were cloned into the Gateway entry vector pENTR/D-TOPO vector in host cell TOP10, and the verified pENTR/D-TOPO vectors containing interesting gene were subsequently recombined into the destination vector pDEST14 for OSCA gene expression in E. coli BL21(DE3) pLysS cells.

The osca2.1 osca2.2 double mutant (osca2.1/2.2) was generated by crossing osca2.1 and osca2.2 (Extended Data Fig. 3a,b). For complementation, clones containing OSCA2.1 promoter driving OSCA2.1 genomic DNA or OSCA2.2 promoter driving OSCA2.2 genomic DNA were amplified by PCR using primers (Supplementary Table 2). The transgenic lines of osca2.1/2.2 expressing pOSCA2.1::OSCA2.1 (OSCA2.1 osca2.1/2.2) or pOSCA2.2::OSCA2.2 (OSCA2.2 osca2.1/2.2) were generated via floral dip transformation as described previously^6,69.

For HEK293 Ca²⁺ imaging analysis, cDNAs encoding Arabidopsis OSCA2.1 and OSCA2.2, and mouse TRPV4 were amplified by PCR using primers (Supplementary Table 2). The PCR products were cloned into the Gateway entry vector and subsequently the destination vector pcDNA3.2 for gene expression in HEK293 cells.

For OSCA–GFP analysis in HEK293 cells, cDNAs encoding OSCA2.1 or OSCA2.2 were cloned into the pEGFP-N1 vector (Clontech), and these plasmids were used to transfect HEK293 cells.

For OSCA–GFP or OSCA–YFP analysis in Arabidopsis, clones containing OSCA2.1 promoter driving OSCA2.1 genomic DNA or OSCA2.2 promoter driving OSCA2.2 genomic DNA were amplified by PCR using primers listed in Supplementary Table 2. The promoter and the DNA fragments were fused by PCR and the products were cloned into the Gateway pENTR-1A vector, which were subsequently recombined into the destination vector pGWB540 to express GFP fusions with OSCA2.1 or OSCA2.2 in Arabidopsis via transformation^6,69.

For GUS activity analysis, the genomic DNA containing the promoter and coding sequence for OSCA2.1 or OSCA2.2 were amplified by PCR using primers listed in Supplementary Table 2, cloned into the Gateway pENTR-1A vector, and subsequently recombined into the destination vector pGWB533 to express tissue-specific promoter-driven GUS in Arabidopsis via transformation^6,69.

For imaging of [Ca²⁺]_i in pollen grains, wild-type (Col-0) expressing GFP fluorescence-based Ca²⁺ indicator GCaMP6m (GCaMP6)⁴⁴ driven by the pollen-specific LAT52 promoter^48,49,72 (pLAT52::GCaMP6m) was generated by transformation^6,69. The osca2.1 and osca2.2 single mutants and the osca2.1/2.2 double mutant were crossed into the wild-type expressing GCaMP6, and over five independent homozygous lines of osca2.1 GCaMP6, osca2.2 GCaMP6, and osca2.1/2.2 GCaMP6 were obtained and analysed with similar results. GCaMP6-expressing lines of OSCA2.1 osca2.1/2.2 (OSCA2.1 osca2.1/2.2 GCaMP6) and OSCA2.2 osca2.1/2.2 (OSCA2.2 osca2.1/2.2 GCaMP6) were generated accordingly by transformation^6,69.

OSCA mRNA analysis

OSCA2.1 and OSCA2.2 RNA were isolated from leaves using a RNA extraction kit and reverse transcription kit (Bioline). The abundance of mRNAs from the wild-type and osca2.1/2.2 seedlings was analysed by RT–PCR, and UBQ was used as a loading control as described^6,88. Gene-specific primers (Supplementary Table 2) were used for OSCA2.1, OSCA2.2 and UBQ expansion by standard PCR.

Statistics and reproducibility

To minimize the system variations, wild-type and osca mutants as well as transgenic lines were grown side-by-side in the agarose medium in Petri dishes or in soil in trays, and the Petri dishes and trays were rotated every other day in positions in the growth chambers or rooms to establish relatively even temperature and light. Independent experiments were performed at least three times with similar results, unless indicated otherwise. The statistical analysis was performed using EXCEL 2016 software (Microsoft), and P values were calculated via unpaired (two components with equal variance) or two-tailed t-test for bar graphs, unless stated otherwise. Data are presented as mean ± s.d. or mean ± s.e.m. To analyse the difference between genotypes in curve or line graphs, two-way ANOVA was carried out using SAS software (SAS Institute). Values of P < 0.05 were considered statistically significant.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

Information on OSCA T-DNA insertion lines and PCR primers and vectors are provided in Supplementary Tables 1 and 2. DNA and RNA gel images presented in Extended Data Fig. 3a,b are included in Supplementary Fig. 1. All other data and materials supporting the findings of this study are available in the main text or the supplementary materials. Source data are provided with this paper.

References

Hetherington, A. M. & Brownlee, C. The generation of Ca²⁺ signals in plants. Annu. Rev. Plant Biol. 55, 401–427 (2004).
CAS PubMed Google Scholar
Tazawa, M., Shimmen, T. & Mimura, T. Membrane control in the Characeae. Annu. Rev. Plant Physiol. 38, 95–117 (1987).
CAS Google Scholar
Kung, C. A possible unifying principle for mechanosensation. Nature 436, 647–654 (2005).
CAS PubMed ADS Google Scholar
Arnadottir, J. & Chalfie, M. Eukaryotic mechanosensitive channels. Annu. Rev. Biophys. 39, 111–137 (2010).
CAS PubMed Google Scholar
Kefauver, J. M., Ward, A. B. & Patapoutian, A. Discoveries in structure and physiology of mechanically activated ion channels. Nature 587, 567–576 (2020).
CAS PubMed PubMed Central ADS Google Scholar
Yuan, F. et al. OSCA1 mediates osmotic-stress-evoked Ca²⁺ increases vital for osmosensing in Arabidopsis. Nature 514, 367–371 (2014).
CAS PubMed ADS Google Scholar
Hsiao, T. C. Plant responses to water stress. Annu. Rev. Plant Physiol. 24, 519–570 (1973).
CAS Google Scholar
Oliver, M. J. et al. Desiccation tolerance: avoiding cellular damage during drying and rehydration. Annu. Rev. Plant Biol. 71, 435–460 (2020).
CAS PubMed Google Scholar
Johnson, M. A., Harper, J. F. & Palanivelu, R. A fruitful journey: pollen tube navigation from germination to fertilization. Annu. Rev. Plant Biol. 70, 809–837 (2019).
CAS PubMed Google Scholar
Kim, Y. J., Zhang, D. B. & Jung, K. H. Molecular basis of pollen germination in cereals. Trends Plant Sci. 24, 1126–1136 (2019).
CAS PubMed Google Scholar
Bremer, E. & Kramer, R. Responses of microorganisms to osmotic stress. Annu. Rev. Microbiol. 73, 313–334 (2019).
CAS PubMed Google Scholar
Bourque, C. W. Central mechanisms of osmosensation and systemic osmoregulation. Nat. Rev. Neurosci. 9, 519–531 (2008).
CAS PubMed Google Scholar
Waadt, R. et al. Plant hormone regulation of abiotic stress responses. Nat. Rev. Mol. Cell Biol. 23, 680–694 (2022).
CAS PubMed PubMed Central Google Scholar
Zhang, H., Zhu, J., Gong, Z. & Zhu, J.-K. Abiotic stress responses in plants. Nat. Rev. Genet. 23, 104–119 (2022).
PubMed Google Scholar
Okazaki, Y., Yoshimoto, Y., Hiramoto, Y. & Tazawa, M. Turgor regulation and cytoplasmic free Ca²⁺ in the alga Lamprothamnium. Protoplasma 140, 67–71 (1987).
Google Scholar
Taylor, A. R., Manison, N. F. H., Fernandez, C., Wood, J. & Brownlee, C. Spatial organization of calcium signaling involved in cell volume control in the fucus rhizoid. Plant Cell 8, 2015–2031 (1996).
CAS PubMed PubMed Central Google Scholar
Bickerton, P., Sello, S., Brownlee, C., Pittman, J. K. & Wheeler, G. L. Spatial and temporal specificity of Ca²⁺ signalling in Chlarnydomonas reinhardtii in response to osmotic stress. New Phytol. 212, 920–933 (2016).
CAS PubMed PubMed Central Google Scholar
Takahashi, K., Isobe, M., Knight, M. R., Trewavas, A. J. & Muto, S. hypo-osmotic shock induces increases in cytosolic Ca²⁺ in tobacco suspension-culture cells. Plant Physiol. 113, 587–594 (1997).
CAS PubMed PubMed Central Google Scholar
Kudla, J. et al. Advances and current challenges in calcium signaling. New Phytol. 218, 414–431 (2018).
PubMed Google Scholar
Hedrich, R. Ion channels in plants. Physiol. Rev. 92, 1777–1811 (2012).
CAS PubMed ADS Google Scholar
Zipfel, C. & Oldroyd, G. E. D. Plant signalling in symbiosis and immunity. Nature 543, 328–336 (2017).
CAS PubMed ADS Google Scholar
Liedtke, W. et al. Vanilloid receptor-related osmotically activated channel (VR-OAC), a candidate vertebrate osmoreceptor. Cell 103, 525–535 (2000).
CAS PubMed PubMed Central Google Scholar
Julius, D. TRP channels and pain. Annu. Rev. Cell Dev. Biol. 29, 355–384 (2013).
CAS PubMed Google Scholar
Jiang, Z. et al. Plant cell-surface GIPC sphingolipids sense salt to trigger Ca²⁺ influx. Nature 572, 341–346 (2019).
CAS PubMed ADS Google Scholar
Wu, F. H. et al. Hydrogen peroxide sensor HPCA1 is an LRR receptor kinase in Arabidopsis. Nature 578, 577–581 (2020).
CAS PubMed ADS Google Scholar
Choi, J. et al. Identification of a plant receptor for extracellular ATP. Science 343, 290–294 (2014).
CAS PubMed ADS Google Scholar
Laohavisit, A. et al. Quinone perception in plants via leucine-rich-repeat receptor-like kinases. Nature 587, 92–97 (2020).
CAS PubMed ADS Google Scholar
Kutschera, A. et al. Bacterial media-chain 3-hydroxy fatty acid metabolites trigger immunity in Arabidopsis plants. Science 364, 178–181 (2019).
CAS PubMed ADS Google Scholar
Zhang, M. F. et al. Structure of the mechanosensitive OSCA channels. Nat. Struct. Mol. Biol. 25, 850–858 (2018).
CAS PubMed Google Scholar
Thor, K. et al. The calcium-permeable channel OSCA1.3 regulates plant stomatal immunity. Nature 585, 569–573 (2020).
CAS PubMed PubMed Central ADS Google Scholar
Liu, X., Wang, J. W. & Sun, L. F. Structure of the hyperosmolality-gated calcium-permeable channel OSCA1.2. Nat. Commun. 9, 9 (2018).
ADS Google Scholar
Jojoa-Cruz, S. et al. Cryo-EM structure of the mechanically activated ion channel OSCA1.2. eLife 7, e41845 (2018).
PubMed PubMed Central Google Scholar
Hamilton, E. S., Schlegel, A. M. & Haswell, E. S. United in diversity: mechanosensitive ion channels in plants. Annu. Rev. Plant Biol. 66, 113–137 (2015).
CAS PubMed Google Scholar
Venkatachalam, K. & Montell, C. TRP channels. Annu. Rev. Biochem. 76, 387–417 (2007).
CAS PubMed PubMed Central Google Scholar
Wu, X. M., Yuan, F., Wang, X. W., Zhu, S. & Pei, Z.-M. Evolution of osmosensing OSCA1 Ca²⁺ channel family coincident with plant transition from water to land. Plant Genome 15, e20198 (2022).
CAS PubMed Google Scholar
Radin, I. et al. Plant PIEZO homologs modulate vacuole morphology during tip growth. Science 373, 586–590 (2021).
CAS PubMed ADS Google Scholar
Mousavi, S. A. R. et al. PIEZO ion channel is required for root mechanotransduction in Arabidopsis thaliana. Proc. Natl Acad. Sci. USA 118, e2102188118 (2021).
CAS PubMed PubMed Central Google Scholar
Yoshimura, K., Iida, K. & Iida, H. MCAs in Arabidopsis are Ca²⁺-permeable mechanosensitive channels inherently sensitive to membrane tension. Nat. Commun. 12, 6074 (2021).
CAS PubMed PubMed Central ADS Google Scholar
Levina, N. et al. Protection of Escherichia coli cells against extreme turgor by activation of MscS and MscL mechanosensitive channels: identification of genes required for MscS activity. EMBO J. 18, 1730–1737 (1999).
CAS PubMed PubMed Central Google Scholar
Rosano, G. L. & Ceccarelli, E. A. Recombinant protein expression in Escherichia coli: advances and challenges. Front. Microbiol. 5, 172 (2014).
PubMed PubMed Central Google Scholar
Hepler, P. K., Kunkel, J. G., Rounds, C. M. & Winship, L. J. Calcium entry into pollen tubes. Trends Plant Sci. 17, 32–38 (2012).
CAS PubMed Google Scholar
Brewbake, J. L. & Kwack, B. H. Essential role of calcium ion in pollen germination and pollen tube growth. Am. J. Bot. 50, 859–865 (1963).
Google Scholar
Cheung, A. Y. & Wu, H. M. Structural and signaling networks for the polar cell growth machinery in pollen tubes. Annu. Rev. Plant Biol. 59, 547–572 (2008).
CAS PubMed Google Scholar
Chen, T.-W. et al. Ultrasensitive fluorescent proteins for imaging neuronal activity. Nature 499, 295–300 (2013).
CAS PubMed PubMed Central ADS Google Scholar
Hamilton, E. S. et al. Mechanosensitive channel MSL8 regulates osmotic forces during pollen hydration and germination. Science 350, 438–441 (2015).
CAS PubMed PubMed Central ADS Google Scholar
Clapham, D. E. TRP channels as cellular sensors. Nature 426, 517–524 (2003).
CAS PubMed ADS Google Scholar
Iwano, M. et al. Ca²⁺ dynamics in a pollen grain and papilla cell during pollination of Arabidopsis. Plant Physiol. 136, 3562–3571 (2004).
CAS PubMed PubMed Central Google Scholar
Diao, M., Qu, X. L. & Huang, S. J. Calcium imaging in Arabidopsis pollen cells using G-CaMP5. J. Integr. Plant Biol. 60, 897–906 (2018).
CAS PubMed Google Scholar
Wudick, M. M. et al. CORNICHON sorting and regulation of GLR channels underlie pollen tube Ca²⁺ homeostasis. Science 360, 533–536 (2018).
CAS PubMed ADS Google Scholar
Frietsch, S. et al. A cyclic nucleotide-gated channel is essential for polarized tip growth of pollen. Proc. Natl Acad. Sci. USA 104, 14531–14536 (2007).
CAS PubMed PubMed Central ADS Google Scholar
Nichols, R. J. et al. Phenotypic landscape of a bacterial cell. Cell 144, 143–156 (2011).
CAS PubMed Google Scholar
Ali, R., Zielinski, R. E. & Berkowitz, G. A. Expression of plant cyclic nucleotide-gated cation channels in yeast. J. Exp. Bot. 57, 125–138 (2006).
CAS PubMed Google Scholar
Kung, C., Martinac, B. & Sukharev, S. Mechanosensitive channels in microbes. Annu. Rev. Microbiol. 64, 313–329 (2010).
CAS PubMed Google Scholar
Schneider, C. A., Rasband, W. S. & Eliceiri, K. W. NIH Image to ImageJ: 25 years of image analysis. Nat. Methods 9, 671–675 (2012).
CAS PubMed PubMed Central Google Scholar
Zonia, L. & Munnik, T. Osmotically induced cell swelling versus cell shrinking elicits specific changes in phospholipid signals in tobacco pollen tubes. Plant Physiol. 134, 813–823 (2004).
CAS PubMed PubMed Central Google Scholar
Michard, E. et al. Glutamate receptor–like genes form Ca²⁺ channels in pollen tubes and are regulated by pistil d-serine. Science 332, 434–437 (2011).
CAS PubMed ADS Google Scholar
Zhang, J. et al. Sperm cells are passive cargo of the pollen tube in plant fertilization. Nat. Plants 3, 17079 (2017).
PubMed PubMed Central Google Scholar
Steinhorst, L. et al. Vacuolar CBL-CIPK12 Ca²⁺-sensor-kinase complexes are required for polarized pollen tube growth. Curr. Biol. 25, 1475–1482 (2015).
CAS PubMed Google Scholar
Zonia, L., Müller, M. & Munnik, T. Hydrodynamics and cell volume oscillations in the pollen tube apical region are integral components of the biomechanics of Nicotiana tabacum pollen tube growth. Cell Biochem. Biophys. 46, 209–232 (2006).
CAS PubMed Google Scholar
Li, H. J., Meng, J. G. & Yang, W. C. Multilayered signaling pathways for pollen tube growth and guidance. Plant Reprod. 31, 31–41 (2018).
CAS PubMed Google Scholar
Li, H., Lin, Y., Heath, R. M., Zhu, M. X. & Yang, Z. Control of pollen tube tip growth by a Rop GTPase–dependent pathway that leads to tip-localized calcium influx. Plant Cell 11, 1731–1742 (1999).
CAS PubMed PubMed Central Google Scholar
Jacob, P. et al. Plant “helper” immune receptors are Ca²⁺-permeable nonselective cation channels. Science 373, 420–425 (2021).
CAS PubMed PubMed Central ADS Google Scholar
Cheung, A. Y., Boavida, L. C., Aggarwal, M., Wu, H. M. & Feijo, J. A. The pollen tube journey in the pistil and imaging the in vivo process by two-photon microscopy. J. Exp. Bot. 61, 1907–1915 (2010).
CAS PubMed Google Scholar
Huang, J. B. et al. Stigma receptors control intraspecies and interspecies barriers in Brassicaceae. Nature 614, 303–308 (2023).
CAS PubMed PubMed Central ADS Google Scholar
Zhang, L. L. et al. FERONIA receptor kinase-regulated reactive oxygen species mediate self-incompatibility in Brassica rapa. Curr. Biol. 31, 3004–U3072 (2021).
CAS PubMed Google Scholar
Caterina, M. J. et al. The capsaicin receptor: a heat-activated ion channel in the pain pathway. Nature 389, 816–824 (1997).
CAS PubMed ADS Google Scholar
Han, S. C., Tang, R. H., Anderson, L. K., Woerner, T. E. & Pei, Z.-M. A cell surface receptor mediates extracellular Ca²⁺ sensing in guard cells. Nature 425, 196–200 (2003).
CAS PubMed ADS Google Scholar
Coste, B. et al. Piezo1 and Piezo2 are essential components of distinct mechanically activated cation channels. Science 330, 55–60 (2010).
CAS PubMed PubMed Central ADS Google Scholar
Clough, S. J. & Bent, A. F. Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J. 16, 735–743 (1998).
CAS PubMed Google Scholar
Cheung, A. Y. & Wu, H.-M. Overexpression of an Arabidopsis formin stimulates supernumerary actin cable formation from pollen tube cell membrane. Plant Cell 16, 257–269 (2004).
CAS PubMed PubMed Central Google Scholar
de Graaf, B. H. J. et al. Rab11 GTPase-regulated membrane trafficking is crucial for tip-focused pollen tube growth in tobacco. Plant Cell 17, 2564–2579 (2005).
PubMed PubMed Central Google Scholar
Meng, J. G. et al. Integration of ovular signals and exocytosis of a Ca²⁺ channel by MLOs in pollen tube guidance. Nat. Plants 6, 143–154 (2020).
CAS PubMed Google Scholar
Bolte, S. et al. FM-dyes as experimental probes for dissecting vesicle trafficking in living plant cells. J. Microsc. 214, 159–173 (2004).
MathSciNet CAS PubMed Google Scholar
Hooper, C. M., Castleden, I. R., Tanz, S. K., Aryamanesh, N. & Millar, A. H. SUBA4: the interactive data analysis centre for Arabidopsis subcellular protein locations. Nucleic Acids Res. 45, D1064–D1074 (2017).
CAS PubMed Google Scholar
Jefferson, R. A., Kavanagh, T. A. & Bevan, M. W. GUS fusions: beta-glucuronidase as a sensitive and versatile gene fusion marker in higher plants. EMBO J. 6, 3901–3907 (1987).
CAS PubMed PubMed Central Google Scholar
Zhong, S. et al. Cysteine-rich peptides promote interspecific genetic isolation in Arabidopsis. Science 364, 851 (2019).
Google Scholar
Mecchia, M. A. et al. RALF4/19 peptides interact with LRX proteins to control pollen tube growth in Arabidopsis. Science 358, 1600–1603 (2017).
CAS PubMed ADS Google Scholar
Monshausen, G. B., Bibikova, T. N., Weisenseel, M. H. & Gilroy, S. Ca²⁺ regulates reactive oxygen species production and pH during mechanosensing in Arabidopsis roots. Plant Cell 21, 2341–2356 (2009).
CAS PubMed PubMed Central Google Scholar
Denninger, P. et al. Male–female communication triggers calcium signatures during fertilization in Arabidopsis. Nat. Commun. 5, 4645 (2014).
CAS PubMed ADS Google Scholar
Toyota, M. et al. Glutamate triggers long-distance, calcium-based plant defense signaling. Science 361, 1112–1115 (2018).
CAS PubMed ADS Google Scholar
Damineli, D. S. C., Portes, M. T. & Feijo, J. A. Oscillatory signatures underlie growth regimes in Arabidopsis pollen tubes: computational methods to estimate tip ___location, periodicity, and synchronization in growing cells. J. Exp. Bot. 68, 3267–3281 (2017).
CAS PubMed PubMed Central Google Scholar
Guo, J., He, J., Dehesh, K., Cui, X. & Yang, Z. CamelliA-based simultaneous imaging of Ca²⁺ dynamics in subcellular compartments. Plant Physiol. 188, 2253–2271 (2022).
CAS PubMed PubMed Central Google Scholar
Ruiz, M. C. M. et al. Circadian oscillations of cytosolic free calcium regulate the Arabidopsis circadian clock. Nat. Plants 4, 690–698 (2018).
PubMed Central Google Scholar
Basu, D. & Haswell, E. S. The mechanosensitive ion channel MSL10 potentiates responses to cell swelling in Arabidopsis seedlings. Curr. Biol. 30, 2716–2728 (2020).
CAS PubMed Google Scholar
Morgan, J. M. Osmoregulation and water-stress in higher-plants. Annu. Rev. Plant Physiol. 35, 299–319 (1984).
Google Scholar
Munns, R. & Tester, M. Mechanisms of salinity tolerance. Annu. Rev. Plant. Biol. 59, 651–681 (2008).
CAS PubMed Google Scholar
Maggio, A., Zhu, J.-K., Hasegawa, P. M. & Bressan, R. A. Osmogenetics: aristotle to Arabidopsis. Plant Cell 18, 1542–1557 (2006).
CAS PubMed PubMed Central Google Scholar
Wang, Z.-Y. & Tobin, E. M. Constitutive expression of the CIRCADIAN CLOCK ASSOCIATED 1 (CCA1) gene disrupts circadian rhythms and suppresses its own expression. Cell 93, 1207–1217 (1998).
CAS PubMed Google Scholar
Newton, A. C., Bootman, M. D. & Scott, J. D. Second messengers. Cold Spring Harbor Perspect. Biol. 8, a005926 (2016).
Google Scholar
Berridge, M. J., Bootman, M. D. & Roderick, H. L. Calcium signalling: dynamics, homeostasis and remodelling. Nat. Rev. Mol. Cell Biol. 4, 517–529 (2003).
CAS PubMed Google Scholar
Bailey-Serres, J., Parker, J. E., Ainsworth, E. A., Oldroyd, G. E. D. & Schroeder, J. I. Genetic strategies for improving crop yields. Nature 575, 109–118 (2019).
CAS PubMed PubMed Central ADS Google Scholar
Yu, X., Feng, B., He, P. & Shan, L. From chaos to harmony: responses and signaling upon microbial pattern recognition. Annu. Rev. Phytopathol. 55, 109–137 (2017).
CAS PubMed PubMed Central Google Scholar
Edel, K. H., Marchadier, E., Brownlee, C., Kudla, J. & Hetherington, A. M. The evolution of calcium-based signalling in plants. Curr. Biol. 27, R667–R679 (2017).
CAS PubMed Google Scholar
Oldroyd, G. E. D. Speak, friend, and enter: signalling systems that promote beneficial symbiotic associations in plants. Nat. Rev. Microbiol. 11, 252–263 (2013).
CAS PubMed Google Scholar
Helliwell, K. E. et al. Spatiotemporal patterns of intracellular Ca²⁺ signalling govern hypo-osmotic stress resilience in marine diatoms. New Phytol. 230, 155–170 (2021).
CAS PubMed Google Scholar

Download references

Acknowledgements

The authors thank M. R. Knight and S. Gilroy for aequorin- and YC3.6-expressing Arabidopsis seeds; S. Miller for MscS and MscL E. coli strains; D. Julius, W. Liedtke and J. Grandl for TRP clones; L. Cameron, Y. Gao, W. Yang and A. Costa for GCaMP and GFP imaging and discussions; T.-P. Sun, D. R. McClay, J. Grandl, S.-Y. Lee and P. N. Benfey for helpful discussions, sharing equipment and/or reading of the manuscript; members of our laboratories for their support; and J.-Z. Pei for administrative management. S.X., Z.J. and F.W. were supported in part by fellowships from China Scholarship Council; S.P., Y.L., Yan Wang and S.D. were supported in part by Hunan Shennong Academic Program (10154, 549912008701); Q.T., W.L., Q.S., X. Wang, K.C., M.C., Q.Z., Yiqi Wang, H.L. and B.Y. by National Youth Program. This work was supported by grants from National Key Research and Development Program of China (2023YFF1001200), Hunan Shennong Scholar Fund (20354, 540732100199) and Chinese NSF (31571461) to F.Y., Chinese NSF (U19A2028, 32072574) to X.Z. and F.L., DOE (DE-SC0014077) to T.V.-D. and Z.-M.P., and NSF (IOS-1457257) to Z.-M.P.

Author information

These authors contributed equally: Songyu Pei, Qi Tao, Wenke Li, Guoning Qi
These authors jointly supervised this work: Zhen-Ming Pei, Fang Yuan

Authors and Affiliations

College of Horticulture and Landscape, Hunan Agricultural University, Changsha, China
Songyu Pei, Yan Wang, Shiwen Dai, Zhonghua Liu, Xuexiao Zou, Feng Liu & Fang Yuan
Department of Biology, Duke University, Durham, NC, USA
Songyu Pei, Shijian Xu, Lynn Theprungsirikul, Jingyuan Zhang, Bryan Krichilsky, Jessica Pei, Karen Song, Zhonghao Jiang, Feihua Wu, Feng Liu, Zhen-Ming Pei & Fang Yuan
Fitzpatrick Institute for Photonics, Duke University, Durham, NC, USA
Songyu Pei, Bridget M. Crawford, Tuan Vo-Dinh & Zhen-Ming Pei
College of Life Sciences, Zhejiang University, Hangzhou, China
Songyu Pei, Qi Tao, Wenke Li, Xi Wang, Yuantao Liu & Qi Zhang
College of Life and Environmental Sciences, Hangzhou Normal University, Hangzhou, China
Songyu Pei, Qi Tao, Wenke Li, Guoning Qi, Borong Wang, Qiujing Shen, Xi Wang, Xiaomei Wu, Liang Liang, Yuantao Liu, Kena Chen, Yang Shen, Mengjia Cheng, Qi Zhang, Yiqi Wang, Hongli Liu, Benguang Yang & Fang Yuan
Fuqua School of Business, Duke University, Durham, NC, USA
Jessica Pei
Department of Physics, Duke University, Durham, NC, USA
Douglas M. Johnson & Gary B. Swift
Department of Biochemistry, Duke University, Durham, NC, USA
Huanghe Yang

Authors

Songyu Pei
View author publications
Search author on:PubMed Google Scholar
Qi Tao
View author publications
Search author on:PubMed Google Scholar
Wenke Li
View author publications
Search author on:PubMed Google Scholar
Guoning Qi
View author publications
Search author on:PubMed Google Scholar
Borong Wang
View author publications
Search author on:PubMed Google Scholar
Yan Wang
View author publications
Search author on:PubMed Google Scholar
Shiwen Dai
View author publications
Search author on:PubMed Google Scholar
Qiujing Shen
View author publications
Search author on:PubMed Google Scholar
Xi Wang
View author publications
Search author on:PubMed Google Scholar
Xiaomei Wu
View author publications
Search author on:PubMed Google Scholar
Shijian Xu
View author publications
Search author on:PubMed Google Scholar
Lynn Theprungsirikul
View author publications
Search author on:PubMed Google Scholar
Jingyuan Zhang
View author publications
Search author on:PubMed Google Scholar
Liang Liang
View author publications
Search author on:PubMed Google Scholar
Yuantao Liu
View author publications
Search author on:PubMed Google Scholar
Kena Chen
View author publications
Search author on:PubMed Google Scholar
Yang Shen
View author publications
Search author on:PubMed Google Scholar
Bridget M. Crawford
View author publications
Search author on:PubMed Google Scholar
Mengjia Cheng
View author publications
Search author on:PubMed Google Scholar
Qi Zhang
View author publications
Search author on:PubMed Google Scholar
Yiqi Wang
View author publications
Search author on:PubMed Google Scholar
Hongli Liu
View author publications
Search author on:PubMed Google Scholar
Benguang Yang
View author publications
Search author on:PubMed Google Scholar
Bryan Krichilsky
View author publications
Search author on:PubMed Google Scholar
Jessica Pei
View author publications
Search author on:PubMed Google Scholar
Karen Song
View author publications
Search author on:PubMed Google Scholar
Douglas M. Johnson
View author publications
Search author on:PubMed Google Scholar
Zhonghao Jiang
View author publications
Search author on:PubMed Google Scholar
Feihua Wu
View author publications
Search author on:PubMed Google Scholar
Gary B. Swift
View author publications
Search author on:PubMed Google Scholar
Huanghe Yang
View author publications
Search author on:PubMed Google Scholar
Zhonghua Liu
View author publications
Search author on:PubMed Google Scholar
Xuexiao Zou
View author publications
Search author on:PubMed Google Scholar
Tuan Vo-Dinh
View author publications
Search author on:PubMed Google Scholar
Feng Liu
View author publications
Search author on:PubMed Google Scholar
Zhen-Ming Pei
View author publications
Search author on:PubMed Google Scholar
Fang Yuan
View author publications
Search author on:PubMed Google Scholar

Contributions

F.Y., F.L. and Z.-M.P. conceived the project and supervised the work. F.Y. and Z.-M.P. designed the experiments on pollen, and F.Y., F.L. and Z.-M.P. designed the experiments on E. coli. S.P., Q.T., W.L., G.Q., Q.S., B.W. and F.Y. characterized pollen phenotypes. F.Y., F.L., S.X., L.T., Y.S., J.P. and J.Z. conducted E. coli growth analyses, verified T-DNA lines, and generated most double and triple mutants. S.P., Q.T., W.L., G.Q., B.M.C., B.K., L.T., S.X., Y.L., H.L., J.Z., K.S., L.L. and B.Y. analysed T-DNA lines and transgenic materials. S.P., Q.T., W.L., K.C., M.C., Yan Wang, S.D., Q.Z., Yiqi Wang, X. Wang, Z.J., F.W., X. Wu, D.M.J., G.B.S., Z.L., B.M.C., H.Y. and F.Y. carried out Ca²⁺ imaging and analysis and maintained facilities. F.Y., F.L., X.Z., T.V.-D. and Z.-M.P. provided funding and project management. F.Y., S.P. and Z.-M.P. wrote the paper with input from F.L. and J.P. All other authors edited and commented on the manuscript.

Corresponding authors

Correspondence to Feng Liu, Zhen-Ming Pei or Fang Yuan.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature thanks the anonymous reviewer(s) for their contribution to the peer review of this work.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data figures and tables

Extended Data Fig. 1 Schematic illustration of the perception of extracellular stimuli by cell-surface receptors, and the characteristics of analyzing hypo-osmosensors in plants.

a, The extracellular stimuli (first messengers) are perceived by cell-surface receptors/sensors, and converted to cytosolic signals (second messengers). It is essential for living organisms to sense environmental signals to program accordingly growth, development and survival^{1,4,5,21,23,34,46,89,90,91,92,93,94}. The environmental signals include physical, chemical, and biological stimuli, which are called the first messengers. The cytosolic signals include diffusible small molecules and ions, which are called the second messengers⁸⁹. These receptors are usually ion channels, receptor kinases (RTKs)/receptor-like kinases (RLKs), and G-protein-coupled receptors (GPCRs) in animals and plants, and their activation often leads to increases in [Ca²⁺]_i through either Ca²⁺ influx across the plasma membrane or Ca²⁺ release from internal stores^4,5,23,90. It is well known that Ca²⁺ ion serves as a versatile second messenger for various external stimuli in living organisms^{1,19,21,89,90}. b,c, Comparison between the assays for the activities of hyper-osmosensors and hypo-osmosensors in plants. With respect to the analysis of cytosolic signals, for example, [Ca²⁺]_i, the stable baseline at the resting state should be established first. Subsequently, the extracellular stimulus, such as hyper-osmotic stress (b) or hypo-osmotic shock (c), could be applied as indicated by the arrow, and the dynamic of cytosolic signals could be recorded continuously, which reflects the activity of the cell-surface receptor corresponding to the stimulus applied. These terms of “osmosensing”, “osmosensitive” and “osmosensor” have been widely used in microorganisms, plants, and animals^2,4,11,12,87. In general, as long as cells face the amount of available water less than “normal” in the extracellular spaces, which causes cells to lose water from the cytosol, this is called as hyper-osmotic stress/stimulation. On the other hand, if cells face the amount of water more than “normal”, which leads to water influx into the cell, this is called as hypo-osmotic shock/stimulation. Note that, the definition of hyper- or hypo-osmolarity is not about the absolute values to some extent, rather the relationship to the osmolarity in the cytosol, similar to the de-polarization and hyper-polarization in relative to the resting membrane potential. Note that, under the normal growth conditions without any osmotic stress, vegetative tissues (seedlings) are well hydrated, and thus by default, the vegetative tissues are at the stable baseline of the hydrated state (b). Hyper-osmotic stress could be applied directly to these vegetative tissues, and cytosolic signaling processes could be recorded accordingly^6,85,87. In contrast, also under the normal growth conditions, pollen grains and seeds are well dehydrated. By default, they are at the stable baseline of the dehydrated state, and hypo-osmotic shock could be applied directly to analyze hypo-osmosensitive activities (c)^{41,43,48,72,95}. Apparently, without the need of any pretreatment, vegetative tissues are good systems for the analysis of hyper-osmotic stress, while pollen grains and seeds are good systems for the analysis of hypo-osmotic stress. d, The assay for hypo-osmotic shock-induced signaling or physiological changes in vegetative tissues is much more challenging technically. The analyses of hypo-osmotic shock-induced responses in vegetative tissues have been well described in a number of studies^16,17,18,84. Essentially, these vegetative tissues have to be pre-treated with hyper-osmotic conditions for a relatively long period of time (hours or days) first to establish the stable baseline of the dehydrated state, then hypo-osmotic shock could be applied for the analysis of hypo-osmotic signaling processes (in minutes) or physiological changes (in hours and days). Therefore, it has been hard to establish Ca²⁺-imaging-based genetic screens for mutants defective in HOSCA, in contrast to mutants defective in hyper-osmotic stress-induced Ca²⁺ increases⁶. Evidently, if vegetative tissues are treated with drought stress by withdraw water supply for days to establish the dehydrated state, and then water is resupplied to allow rehydration¹⁴, the phenotypes of growth and survival rate could reflect the complex effects on both the dehydration and/or rehydration processes.

Extended Data Fig. 2 hypo-osmosensitive OSCA2.1 and OSCA1.3 are identified through functional expression screen in E. coli.

a, Images show E. coli cells expressing 15 OSCA genes after 16 h of growth on the low salt hypo-osmotic Lysogeny broth (LSHypo-LB) agar media (5 mM NaCl, 3 mM KCl; 70 mOsm) in the absence or presence of IPTG induction. Under the IPTG induction, 2 mM LaCl₃ (a Ca²⁺ channel blocker) or 200 mM sorbitol (for elevated osmolarity, 270 mOsm) were added to the media. The growth rates of similar experiments were analyzed and shown in Fig. 1a–d. *, E. coli cells expressing OSCAs with the reduced growth under LSHypo-LB agar media. b, Side-by-side assay of E. coli cells expressing OSCA2.1, OSCA1.3 and OSCA2.5 in the presence or absence of Ca²⁺ channel blocker Gd³⁺ (2 mM GdCl₃) in experiments similar to these in a. Gd³⁺ restored the E. coli growth, suggesting that similar to La³⁺, Gd³⁺ could block OSCA channel activities to alleviate their toxicity. c,d, Growth rates of E. coli cells expressing hypo-osmosensitive OSCA1.3 in the presence of LaCl₃ (c) or 200 mM sorbitol (d) from experiments similar to these in Fig. 1e. Note that, OSCA1.3 was less toxic to E. coli than OSCA2.1 when expressed under hypo-osmotic conditions, suggesting that OSCA2.1 might be more hypo-osmosensitive than OSCA1.3 under the imposed conditions. Quantifications of growth rates are shown (mean ± s.d.; n = 3, 4 or 5 independent experiments). e,f, Quantification of growth rates of E. coli cells expressing the toxic OSCA2.5 from experiments similar to these in Fig. 1e. The E. coli growth was rescued by LaCl₃, suggesting that the toxicity of OSCA2.5 was caused by its Ca²⁺-permeable channel activity (e); but was not rescued by the hyper-osmotic treatment (200 mM sorbitol; f), indicating that OSCA2.5 was activated by factors other than the hypo-osmolarity. Data are mean ± s.d. (n = 3 or 4 independent experiments). Note that, using this stringent growth screen, we could only identify these hypo-osmosensitive OSCAs, whose hypo-osmosensitive channel activities were expressed to a level toxic to E. coli, but could not rule out others having hypo-osmosensitivities, because their activities did not rise to the toxic threshold, possibly due to low protein expression levels.

Source Data

Extended Data Fig. 3 The hypo-osmosensitive OSCA2.1 and its close homologue OSCA2.2 are both required for pollen germination.

a,b, Genotyping and gene expression analysis of the osca2.1/2.2 double-knockout mutant. PCR reactions showed a flanking DNA fragment upstream (LP) and downstream of the insertion site (RP) in wild type (WT) but not in the osca2.1/2.2 double mutant, and a DNA fragment flanking the T-DNA border (BP) and the downstream of the insertion site (RP) in osca2.1/2.2 but not WT, which demonstrated that osca2.1/2.2 is a homozygous double T-DNA insertion line (a). The mRNAs of OSCA2.1 and OSCA2.2 were absent in the osca2.1/2.2 double mutant, indicating that osca2.1/2.2 is a knock-out/null mutant (b). Similar results were seen in > 3 times. c, Representative micrographs of pollen grains from wild type, the osca2.1 and osca2.2 single mutants, the osca2.1/2.2 double mutant, and the OSCA2.1 osca2.1/2.2 and OSCA2.2 osca2.1/2.2 complementation lines, which were placed on the standard germination media (535 mOsm) for 6 h. Similar experiments were carried out and shown in Fig. 2c. The complementation lines are these #9 or #10 lines that were described below in i. Note that, although OSCA2.1 and 2.2 were both hypo-osmosensitive channels and redundant in pollen germination, the reason that only OSCA2.1 was identified in the E. coli assay (Fig. 1) was that the ion-transporting activity of OSCA2.2 hypo-osmosensitive channels was not expressed functionally to the toxic threshold as compared to that of OSCA2.1, possibly due to low protein expression of OSCA2.2. Conversely, although OSCA1.3 hypo-osmosensitivity was identified in E. coli, the reason that the osca1.3/2.1 double mutant did not show a pollen germination defect as compared to osca2.1/2.2 (Fig. 2b) might be that the activity of OSCA1.3 was probably lower than that of OSCA2.2 in pollen, and knocking out OSCA1.3 together with OSCA2.1 could not lower the activity to a threshold sufficient to disrupt germination. Scale bars are 50 μm. Similar results were seen in > 3 times. d–e, Analysis of time course of pollen germination on wild-type stigmas via aniline blue staining. The flowers from wild-type plants were emasculated 24 h prior to pollination, and pollen grains from wild type and osca2.1/2.2 plants were dabbed onto pre-emasculated stigmas. After 1, 1.5, and 2 h of pollination, stigmas with pollen grains were stained with the aniline blue, fluorescence images were taken, and the numbers of pollen grains that stayed on the stigma were counted. Representative micrographs of pollen grains from wild type and osca2.1/2.2 taken at 1 h after pollination (d) as well as at 2 h (Fig. 2d) are shown. Scale bars are 50 μm. The numbers of pollen grains from these experiments were plotted as the function of time after pollination (e; mean ± s.e.; n_WT = 23, 13, 27 stigmas for each data point, n_osca2.1/2.2 = 9, 13, 39 stigmas for each data point; two-way ANOVA). With increases in the germination time, the numbers of germinated pollen grains increased in both wild type and osca2.1/2.2, but consistently less germinated pollen grains in osca2.1/2.2, suggesting that osca2.1/2.2 was defective in pollen germination under the imposed in vivo condition, i.e. at a given osmolarity at the stigma. Similar results were seen in > 5 times. f, Analysis of changes in GFP fluorescence during in vitro pollen germination. Pollen grains from wild-type plants constitutively expressing GFP, which was driven by a pollen specific promoter LAT52, were placed on the standard in vitro germination media, and GFP fluorescence images were taken every 30 s for 300 min. Representative images taken for the same three wild-type pollen grains before germination (top panel) at the time of 10 min and after one pollen grain germinated (bottom panel) at the time about 100 min are shown here. Similar results were seen in > 5 times. The three wild-type pollen grains were labeled as #1, #2 and #3. The #1 wild-type grain germinated and was indicated by an arrow. The #2 and #3 WT grains did not germinate and were labeled by “x”. After germination, the #1 grain displayed low fluorescence with a circular pattern, while the fluorescence of other two non-germinated #2 and #3 grains displayed strong fluorescence. This phenotype of changes in GFP fluorescence during germination was used to score germinated vs non-germinated pollen grains for the assay of in vivo pollen germination on stigma in g,h. Scale bars are 5 μm. g–h, Analysis of pollen germination on WT stigmas using pollen grains constitutively expressing GFP. As described in experiments in Fig. 2d,e, for in vivo pollen germination on stigma, the flowers from wild-type plants (without GFP expression) were emasculated 24 h prior to pollination, and pollen grains from wild-type and osca2.1/2.2 plants (with GFP expression as in f) were dabbed onto the surfaces of pre-emasculated stigmas. After 30 min of pollination, stigmas with pollen grains were analyzed, and GFP fluorescence images were collected (g). Scale bars are 20 μm. Based on images in f, germinated pollen grains displayed weaker GFP fluorescence, notably much weaker in the center of pollen grains, and were labeled with arrows. In contrast, the non-germinated grains displayed solid fluorescence patterns, which were labeled with x (see the insert). Germination rates were calculated, and data from more than 5 independent experiments were shown (h; mean ± s.d., n_WT = 10 stigmas, n_osca2.1/2.2 = 11 stigmas). i, Complementation of the pollen germination defects in osca2.1/2.2 by genomic DNA of OSCA2.1 or OSCA2.2 under the standard germination media (535 mOsm). The osca2.1/2.2 mutant was the double T-DNA insertion homozygous line, and genomic segments containing their promoter and the gene were introduced into osca2.1/2.2 (OSCA2.1 osca2.1/2.2 or OSCA2.2 osca2.1/2.2). More than 100 transforms were selected, and ten representative individual transgenic lines (out of > 50), which were sorted by their germination rates (#1 to #10), were showed to have higher germination rates than osca2.1/2.2 (mean ± s.d.; n = 3 independent experiments). Note that, based on their wild-type-like germination rates, the #9 and #10 lines were selected for further analysis, such as for pollen germination in c. These #9 and #10 lines were also used for the generation of GCaMP6-expression lines (OSCA2.1 osca2.1/2.2 GCaMP6 or OSCA2.2 osca2.1/2.2 GCaMP6). j,k, No difference in variability between wild-type and osca2.1/2.2 pollen grains. Micrographs showed the viability of wild-type and osca2.1/2.2 pollen grains after hydration (j). Scale bars are 50 μm. Pollen grains were stained with fluorescein diacetate and propidium iodide dyes for viable and unviable pollen grains, respectively, after incubation for 30 min on the 420 mOsm germination media. Note that, these pollen grains with the cytosol stained by fluorescein diacetate were pointed by arrows and considered unviable. Viability rates of WT and osca2.1/2.2 pollen grains from plants grown under low (35%), normal (60%), or high relative humidity (80%) in the growth chamber were quantified (k). Under these growth conditions, the viability of osca2.1/2.2 pollen grains was not affected also. Data are presented as mean ± s.d. (n = 3 or 4 independent experiments).

Source Data

Extended Data Fig. 4 Medium Ca²⁺ effects on pollen germination, and the subcellular localization of OSCA2.1/2.2 and their requirement for hypo-osmolarity-induced [Ca²⁺]_i increases (HOSCA) in pollen grains.

a,b, Representative micrographs of pollen grains from wild-type and osca2.1/2.2 plants (a). Pollen grains were placed on the hyper-osmotic solid germination media (680 mOsm) containing variable Ca²⁺ concentrations of 2 mM, 5 mM, or 10 mM CaCl₂ (black bars at the bottom) for 6 h, and micrographs were taken. Scale bars are 50 μm. Germination rates of wild-type and osca2.1/2.2 pollen grains were quantified (b). Data are presented as mean ± s.d. (n = 3 trails). Similar results were seen in > 10 times. c–e, hypo-osmolarity-induced [Ca²⁺]_i increases (HOSCA) in pollen grains were not affected in the osca2.1 and osca2.1 single mutants and the osca2.1/2.2 complementation lines (OSCA2.1 osca2.1/2.2 and OSCA2.2 osca2.1/2.2). Pollen grains expressing the Ca²⁺ indicator GCaMP6 were hydrated in the germination media containing 300 mM sorbitol for 1 h, and GCaMP6 fluorescence images were taken every 1 s after the treatment with the germination solution containing 0 mM sorbitol, similar to these experiments in Fig. 3a,b. Representative GCaMP6 images scaled by a pseudo-color bar before and after hypo-osmotic treatment were shown (c). Scale bars are 20 μm. HOSCA in wild-type, osca2.1, osca2.2, OSCA2.1 osca2.1/2.2 and OSCA2.2 osca2.1/2.2 pollen grains from experiments as in c were analyzed for kinetics, respectively (d), and for peak changes in HOSCA (e). Data were mean ± s.d. (n = 3 independent experiments). f, OSCA2.1-GUS and OSCA2.2-GUS expression levels in pollen grains from plants expression pOSCA2.1::OSCA2.1-GUS and pOSCA2.2::OSCA2.2-GUS, respectively, were placed on the common germination media at 0 min and 120 min, and the GUS assay was performed. Scale bars are 50 μm. g, Pollen grains over-expressing GFP alone (pLAT52::GFP) were placed on the standard germination media at 0 min and 120 min, respectively, and GFP fluorescence was analyzed using confocal microcopy. pLAT52::GFP was used as a control for pOSCA2.1::OSCA2.1-YFP and pOSCA2.2::OSCA2.2-YFP localization in pollen grains in Fig. 3f. Scale bars are 5 μm. h, The plasma-membrane localization of OSCA2.1/2.2 in pollen tubes. The fluorescence of pOSCA2.1::OSCA2.1-YFP, pOSCA2.2::OSCA2.2-YFP, or FM4-64 fluorescence was analyze using confocal microscopy in newly developed pollen tubes from experiments similar to these in Fig. 3f. Merged images showed FM4-64 red fluorescence was well associated with OSCA2.1-GFP or OSCA2.2-GFP fluorescence in both the cell surface and the cytosol, but not GFP-alone fluorescence, suggesting that OSCA2.1-GFP and OSCA2.2-GFP were localized to the plasma membrane, and underwent typical endocytosis, which was seen in pollen tube tips^43,71,72,73. Scale bars are 10 μm.

Source Data

Extended Data Fig. 5 OSCA2.1 and OSCA2.2 form hypo-osmolarity-gated Ca²⁺-permeable channels in HEK293 cells.

a, The cell-surface localization of OSCA2.1 and OSCA2.2 in HEK293 cells. HEK293 cells were transiently transfected with OSCA2.1–GFP or OSCA2.2–GFP plasmids with the GFP plasmids as a control, and GFP fluorescence was analyzed by confocal microscopy. The GFP fluorescence of OSCA2.1 and OSCA2.2 was seen in the vicinity of the plasma membrane, while the GFP alone was seen throughout the cell. Similar data were shown in the insets in Fig. 3i,j. Scale bars are 20 μm. b–e, [Ca²⁺]_i increases in response to hypo-osmolarity treatment in HEK293 cells expressing empty vector pcDNA3.2 or the positive control of TRPV4. The cells were incubated in standard bath solution (300 mOsm; − hypo-osmo) and then treated with the hypo-osmotic solution (140 mOsm; + hypo-osmo), and [Ca²⁺]_i increases were analyzed by Fura-2 emission ratios (F₃₄₀/F₃₈₀) and scaled with a pseudo-color bar (b). Scale bars are 50 μm. [Ca²⁺]_i dynamics in cells expressing empty vector pcDNA3.2 (c) or TRPV4 (d) from experiments as in a. Data are mean ± s.d. (n = 60 cells; r.u., relative unit). Arrows indicate the time of administration of hypo-osmolarity solution. Peak increases in [Ca²⁺]_i were calculated from experiments similar to these in c and d (e). Data are mean ± s.d. (n = 3 or 4 independent experiments). f–i, hypo-osmolarity-gated OSCA2.1 and OSCA2.2 channel activities were blocked by La³⁺. The cells expressing empty vector pcDNA3.2, OSCA2.1, or OSCA2.2 were incubated in standard bath solution (300 mOsm; − hypo-osmo) for 30 min and then treated with the hypo-osmotic solution containing 0.5 mM LaCl₃ (140 mOsm; + hypo-osmo), and [Ca²⁺]_i were analyzed by Fura-2 emission ratios (f). Scale bars are 50 μm. Dynamic analyses of [Ca²⁺]_i increases in cells expressing empty vector pcDNA3.2 (g), OSCA2.1 (h), or OSCA2.2 (i) from experiments as in f. Data are mean ± s.d. (n = 6 ROI).

Source Data

Extended Data Fig. 6 Desensitization of hypo-osmosensitive OSCA2.1 and OSCA2.2 Ca²⁺ channels expressed in HEK293 cells.

a, Representative plots show HOSCA desensitization of OSCA2.1, OSCA2.2, and TRPV4 when expressed in HEK293 cells with pcDNA3.2 as a control. HEK293 cells were treated with the isosmotic solution (300 mOsm; black) and hypo-osmotic solution (140 mOsm; gray) repetitively three times, and [Ca²⁺]_i were analyzed by Fura-2 emission ratios. Data are mean ± s.d. (n = 3 independent experiments). Similar results were seen in > 10 times. b, Quantification of HOSCA desensitization from experiments similar to a. OSCA2.1 and OSCA2.2 showed over 50% of desensitization, while TRPV4 showed only about 10–20% of desensitization. The cells expressing pcDNA3.2 almost did not show any desensitization. The properties of HOSCA kinetics and desensitization for OSCA2.1 and OSCA2.2 expressed in HEK293 cells were consistent with the HOSCA kinetics in planta as shown in Fig. 3a,b. The 1^st HOSIC was arbitrarily set to 1, and data are mean ± s.d. (n = 3 or 4 independent experiments). Similar results were seen in > 10 times.

Source Data

Extended Data Fig. 7 Ca²⁺ oscillations prior to germination in pollen grains placed on the standard media (535 mOsm) are impaired in the osca2.1/2.2 double mutant.

a,b, Representative cumulative Ca²⁺ images scaled by a pseudo-color bar at 8–33 min (left) and 150–175 min (right) are shown in pollen grains expressing GCaMP6 (a). Scale bars are 50 μm. Pollen grains from plants expressing GCaMP6 were placed on the common germination media (535 mOsm), and GCaMP6 fluorescence images were taken every 30 s for 300 min as shown in Fig. 4. Pollen GCaMP6 images for a total of 25 min at the indicated interval time points (8–33 min and 150–175 min) were pooled together, and scaled by a pseudo-color bar. The cumulative Ca²⁺ signals were analyzed using ImageJ and similar experiments were quantified (b; mean ± s.d.; n_WT = 91 grains, n_osca2.1/2.2 = 82 grains). Similar results were seen in > 10 times. c,d, Cumulative pollen Ca²⁺ images for 300 min from experiments as in a and b were shown in 3D surface pseudo-color images (c) or bar graphs (d), respectively (mean ± s.d.; n_WT = 91 grains, n_osca2.1/2.2 = 82 grains).

Source Data

Extended Data Fig. 8 Pollen Ca²⁺-oscillation phenotype prior to germination in the osca2.1/2.2 double mutant is rescued under the hypo-osmotic media (350 mOsm) and exacerbated under hyper-osmotic media (680 mOsm).

a,b, Cumulative GCaMP6-fluorescence images were pooled together at the intervals of 8–33 min (left) and 138–163 min (right) for a total of 25 min, respectively, after pollen grains were spread on the hypo-osmotic (350 mOsm) germination media from WT and osca2.1/2.2 plants (a), and quantified (b). Scale bars are 50 μm. GCaMP6-fluorescence images were collected every 30 s for 300 min, and are scaled by a pseudo-color bar. Data are mean ± s.d. (n_WT = 69 grains, n_osca2.1/2.2 = 72 grains). Under the hypo-osmotic media WT pollen grains displayed an accelerated hydration, and at the time of 8–33 min the [Ca²⁺]_i were already increased to the level of these at the time of 138–163 min (as a saturated status), as shown in a and b. However, osca2.1/2.2 pollen grains displayed a “normal” hydration as seen for the WT at the standard 535 mOsm media (Fig. 4a), and at the time of 8–33 min the [Ca²⁺]_i were lower, and then increased to the level at the time of 138–163 min. c,d, Cumulative pollen Ca²⁺ images for 300 min from the same experiments as in a and b were shown in 3D surface pseudo-color images (c) or bar graphs (d), respectively. Representative 3D surface plots of cumulative Ca²⁺ signals for 300 min after pollen grains were spread on the 350 mOsm germination media from WT and osca2.1/2.2 plants. Data are mean ± s.d. (n_WT = 69 grains, n_osca2.1/2.2 = 72 grains). Similar results were seen in > 10 times. e,f, Representative Ca²⁺ oscillations in a WT (e) or an osca2.1/2.2 (b) pollen grain expressing GCaMP6 on the hypo-osmotic germination media (350 mOsm). GCaMP6-fluorescence images were taken every 30 s for 300 min. The time-lapse acquisitions show the peaks, troughs and resting phases of Ca²⁺ oscillations. Similar to the patterns Ca²⁺ oscillations in WT pollen grains in Fig. 4a, the resting phases (RePh₁, RePh₂ and RePh₃) were separated with Ca²⁺ oscillations with small amplitude (CaOsc^S) and large amplitude (CaOsc^L) in both WT and osca2.1/2.2 pollen grains. P_L/P_S, peak amplitude of CaOsc^L/CaOsc^S; T_L/T_S, trough amplitude of CaOsc^L/CaOsc^S. Similar results were seen in > 10 times. g,h, Cumulative GCaMP6-fluorescence images were pooled together at the intervals of 8–33 min (left) and 165–190 min (right) for a total of 25 min, respectively, after pollen grains were spread on the hyper-osmotic germination media (680 mOsm) from WT and osca2.1/2.2 plants (g), and quantified (h). Scale bars are 50 μm. GCaMP6-fluorescence images were collected every 30 s for 300 min, and analyzed (mean ± s.d.; n_WT = 91 grains, n_osca2.1/2.2 = 100 grains). i,j, Cumulative pollen Ca²⁺ images for 300 min from the same experiments as in g and h were shown in 3D surface pseudo-color images (i) or bar graphs (j), respectively. Data are mean ± s.d. (n_WT = 91 grains, n_osca2.1/2.2 = 100 grains). k,l, Representative Ca²⁺ oscillations in a WT (k) or an osca2.1/2.2 (l) pollen grain expressing GCaMP6 on the hyper-osmotic germination media (680 mOsm). GCaMP6-fluorescence images were taken every 30 s for 300 min. Largely altered patterns Ca²⁺ oscillations especially for osca2.1/2.2 pollen grains feature extended the resting phase (RePh₁) for osca2.1/2.2 pollen grains and Ca²⁺ oscillations with small amplitude (CaOsc^S) for WT pollen grains. P_L/P_S, peak amplitude of CaOsc^L/CaOsc^S; T_L/T_S, trough amplitude of CaOsc^L/CaOsc^S. Similar results were seen in > 10 times.

Source Data

Extended Data Fig. 9 Restoration of pollen Ca²⁺-oscillation defects in osca2.1/2.2 germination under the hypo-osmotic media (420 mOsm).

a,b, Example traces of GCaMP6 fluorescence from 10 representative WT pollen grains placed on hypo-osmotic media (420 mOsm; a) showed Ca²⁺ spiking and germination initiation (arrows). The colors and their representations were the same as in Fig. 4 and Fig. 5a,b. Results from three independent experiments were shown. The amplitudes, durations, periods, and frequencies of these Ca²⁺ traces were quantified (b; mean ± s.e.m., n = 112 grains). These oscillations were superimposed to illustrate the properties for CaOsc^S and CaOsc^L, as shown in Fig. 5c,d. Similar results were seen in > 5 times. c,d, Example traces of GCaMP6 fluorescence from 10 representative osca2.1/2.2 pollen grains placed on hypo-osmotic media (420 mOsm; c) showed Ca²⁺ spiking and germination initiation, as seed in Fig. 5e,f. The Ca²⁺ spiking patterns under the hypo-osmotic condition were quantified (d; mean ± s.e.m., n = 118 grains). These oscillations were superimposed to illustrate the properties for CaOsc^S and CaOsc^L, as seen in Fig. 5g,h. Similar results were seen in > 5 times.

Source Data

Extended Data Fig. 10 Impaired osmolarity regulation of [Ca²⁺]_i oscillations during pollen germination in osca2.1/2.2.

a,b, Data on Ca²⁺ spiking patterns in pollen grains placed on hypo-osmotic (350 mOsm) or hyper-osmotic media (680 mOsm) were from the same experiments in Fig. 5a,b,e,f, and quantified for the periods of CaOsc^S (a) and CaOsc^L (b) similarly as in Fig. 5c,d,g,h (mean ± s.d.; n₃₅₀ = 114 grains for WT; n₃₅₀ = 235 grains for osca2.1/2.2; n₆₈₀ = 122 grains for WT; n₆₈₀ = 198 grains for osca2.1/2.2). The time between the peaks of two Ca²⁺ spicks was calculated as a period. The oscillation periods of CaOsc^S and CaOsc^L were increased under high osmolarity media in WT, suggesting that low water availability suppressed Ca²⁺ oscillations. Overall, the oscillation periods of CaOsc^S and CaOsc^L were longer in osca2.2/2.2 than those in WT, suggesting that osca2.2/2.2 had weaker Ca²⁺ oscillations. c,d, Data on Ca²⁺ spiking patterns in pollen grains placed on hypo-osmotic media (350 mOsm) or hyper-osmotic media (680 mOsm) were from the same experiments in Fig. 5a,b,e,f, and quantified for the oscillation duration times of CaOsc^S (c) and CaOsc^L (d) similarly as in Fig. 5c,d,g,h (mean ± s.d.; n₃₅₀ = 114 grains for WT; n₃₅₀ = 235 grains for osca2.1/2.2; n₆₈₀ = 122 grains for WT; n₆₈₀ = 198 grains for osca2.1/2.2). The duration time of an individual Ca²⁺ spick was calculated as the time between the troughs immediately before and after the peak. The duration times of CaOsc^S and CaOsc^L were increased under high osmolarity media in WT, also indicating that the water availability was positively associated with Ca²⁺ oscillations. Overall, the duration times of CaOsc^S and CaOsc^L were longer in osca2.2/2.2 than those in WT, except for CaOsc^L at high osmolarity media. Note that, the shorter duration times of CaOsc^L at 680 mOsm in osca2.2/2.2 were largely due to very few or no CaOsc^L occurring as seen in Fig. 5f, preventing accurate analysis. These results suggest that OSCA2.1/2.2 are responsible for sensing water, and that Ca²⁺ oscillations are the second messengers for water. e,f, Effects of media osmolarity on the period (e) and amplitude (f) of CaOsc^S from the same experiments as in Fig. 5i,j (mean ± s.d.; n = 5 independent experiments; two-way ANOVA, P < 0.001). With the decreases in media osmolarity, i.e. increases in water availability, the oscillation amplitudes were elevated and the oscillation periods were reduced, suggesting that the water availability was positively associated with Ca²⁺ oscillations. Overall, the osca2.2/2.2 was less sensitive to the increases in water availability, consistent with OSCA2.1/2.2 being hypo-osmosensors. g–j, Complementation of defective Ca²⁺ oscillations in osca2.1/2.2 under the standard germination media (535 mOsm) by expression of genomic DNA of OSCA2.1 or OSCA2.2. Data on Ca²⁺ spiking patterns in pollen grains placed on the 535 mOsm media were from the experiments similar to these in Fig. 4, and the periods and amplitudes of CaOsc^S (g, h) and CaOsc^L (i, j) were quantified similarly as in a–d (mean ± s.d.; n = 3 independent experiments). The periods and amplitudes of CaOsc^S and CaOsc^L in the osca2.1 and osca2.1 single mutants were similar to these in WT, consistent with their phenotypes in pollen germination (Fig. 2c) and HOSCA (Extended Data Fig. 4c–e). In addition, these periods and amplitudes of CaOsc^S and CaOsc^L in the complementation lines (OSCA2.1 osca2.1/2.2 and OSCA2.2 osca2.1/2.2) were comparable to these in WT, also suggesting that OSCA2.1 and OSCA2.2 had functional redundancy in regulating Ca²⁺ oscillations. Notably and consistently, the periods of CaOsc^S and CaOsc^L in osca2.1/2.2 were significantly longer than these of the rest of genotypes, and the amplitudes of CaOsc^S and CaOsc^L in osca2.1/2.2 were also significantly smaller than these of the rest of genotypes. Taken together, these WT-like phenotypes on Ca²⁺ oscillations, germination, and HOSCA in the osca2.1 and osca2.1 single mutants and complementation lines suggest that OSCA2.1 and OSCA2.2 were redundant and function together to control pollen grain perception of hypo-osmolarity conditions. k, Model of osmosensors and water potential relationships in plant cells and their environment. Soil water potential (Ψw_soil) drops from saturated state (0 MPa) to field capacity (−0.03 MPa), at which both air and water are in the macropores, to permanent wilting point (−1.5 MPa), at which soil water is held by solid particles too tightly to be taken up by plants^7,85,87. The air water potential is about −100 MPa, which provides the driving force for water transport through the soil−plant-atmosphere continuum. Notably, the extracellular spaces (called the apoplast) are connected largely with their environments (illustrated by a dashed line surrounding cell walls and arrows). The extracellular spaces in roots are directly connected to the soil, and stomatal pores allow the extracellular spaces in the leaf to connect to the air, forming the soil-plant-atmosphere continuum. Therefore, the water potential in the extracellular spaces (outside of the cell, Ψw_o) is not set to a point, in contrast to that in mammalian cells, and varies enormously from −0.03 to −4.5 MPa (equivalent to the change in solution osmolarity from 12 to 1820 mOsm). As the osmosensing OSCA1 family has expended greatly compared to three members in animals, and co-evolved with the plant transition from water to land³⁵, these OSCAs may be major osmosensors in plants. l, Model of osmosensors and water potential relationships in mammalian cells and their environment. The extracellular spaces are separated with their environment (illustrated by a solid line and stop arrows). The water potential of the extracellular fluid (ECF, Ψw_o) is set to a point, and the ECF osmolarity is tightly regulated around this set-point of ∼ 300 ± 30 mOsm (equivalent to the change in water potential from −0.66 to −0.82 MPa)¹².

Source Data

Supplementary information

Supplementary Information

This file contains Supplementary Figure 1: Gel source data; Supplementary Table 1: OSCA T-DNA insertion lines and primers used for genotyping; and Supplementary Table 2: PCR primers and vectors.

Reporting Summary

Supplementary Video 1

Hypo-osmolarity-induced Ca²⁺ increases (HOSCA) are disrupted in osca2.1/2.2 pollen grains. Pollen grains from wild type and osca2.1/2.2 plants expressing the Ca²⁺ indicator GCaMP6 were hydrated in the germination medium containing 300 mM sorbitol for 1 h, and GCaMP6 fluorescence images were taken every 1 sec for 300 sec after treated with the germination solution containing 0 mM sorbitol. The same GCaMP6 images scaled by a pseudocolour bar were shown in Fig. 3a and quantified in Fig. 3b,c.

Supplementary Video 2

Pollen Ca²⁺ oscillations prior to germination are impaired in osca2.1/2.2. Pollen grains from wild type and osca2.1/2.2 plants expressing the Ca²⁺ indicator GCaMP6 were placed on the standard germination medium (535 mOsm), and GFP fluorescence images were taken every 30 s for 300 min. The same GCaMP6 images scaled with a pseudocolour bar from one representative pollen grains are shown in Fig. 4a for wild-type and Fig. 4b for osca2.1/2.2.

Supplementary Video 3

The impaired Ca²⁺ oscillations prior to germination in osca2.1/2.2 pollen grains are rescued under hypo-osmolarity media. Pollen grains from wild type and osca2.1/2.2 plants expressing the Ca²⁺ indicator GCaMP6 were placed on the hypo-osmotic (350 mOsm) germination medium, and GFP fluorescence images were taken every 30 s for 300 min. Similar GCaMP6 images scaled with a pseudocolour bar from representative wild-type and osca2.1/2.2 pollen grains were shown in Fig. 5a and Fig. 5e, respectively.

Supplementary Video 4

Ca²⁺ oscillations prior to germination are attenuated under hyper-osmolarity media in both wild-type and osca2.1/2.2 pollen grains, but more in osca2.1/2.2. Pollen grains from wild type and osca2.1/2.2 plants expressing the Ca²⁺ indicator GCaMP6 were placed on the hyper-osmotic (680 mOsm) germination medium, and GFP fluorescence images were taken every 30 s for 300 min. Similar GCaMP6 images scaled with a pseudocolour bar from representative wild-type and osca2.1/2.2 pollen grains were shown in Fig. 5b and Fig. 5f, respectively.

Source data

Source Data Fig. 1

Source Data Fig. 2

Source Data Fig. 3

Source Data Fig. 4

Source Data Fig. 5

Source Data Extended Data Fig. 2

Source Data Extended Data Fig. 3

Source Data Extended Data Fig. 4

Source Data Extended Data Fig. 5

Source Data Extended Data Fig. 6

Source Data Extended Data Fig. 7

Source Data Extended Data Fig. 8

Source Data Extended Data Fig. 9

Source Data Extended Data Fig. 10

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Pei, S., Tao, Q., Li, W. et al. Osmosensor-mediated control of Ca²⁺ spiking in pollen germination. Nature 629, 1118–1125 (2024). https://doi.org/10.1038/s41586-024-07445-6

Download citation

Received: 08 August 2022
Accepted: 19 April 2024
Published: 22 May 2024
Issue Date: 30 May 2024
DOI: https://doi.org/10.1038/s41586-024-07445-6

This article is cited by

Tackling abiotic stress in plants: recent insights and trends
- Heng Zhang
- Zhaobo Lang
- Pengcheng Wang
Stress Biology (2025)

Subjects

Abstract

Similar content being viewed by others

Main

Screening for hypo-osmosensitive OSCAs

OSCAs are essential for pollen germination

Regulation of pollen germination via OSCAs

OSCAs are vital for hypo-osmotic Ca2+ spiking

OSCAs form hypo-osmosensing Ca2+ channels

OSCAs control pollen Ca2+ oscillations

OSCAs couple external osmolarity to Ca2+ spiking

Discussion

Methods

Screen based on functional expression of OSCA in E. coli

Plant materials and growth conditions

In vitro pollen germination assay

In vivo pollen germination assays

Imaging of [Ca2+]i in HEK293 cells

Analysis of OSCA–GFP/YFP expression

Histochemical GUS activity analysis

Pollen grain viability assay

Imaging of [Ca2+]i in pollen grains

Water potential

DNA constructs and transgenic lines

OSCA mRNA analysis

Statistics and reproducibility

Reporting summary

Data availability

References

Acknowledgements

Author information

Authors and Affiliations

Contributions

Corresponding authors

Ethics declarations

Competing interests

Peer review

Peer review information

Additional information

Extended data figures and tables

Supplementary information

Source data

Rights and permissions

About this article

Cite this article

Share this article

This article is cited by

Search

Quick links

OSCAs are vital for hypo-osmotic Ca²⁺ spiking

OSCAs form hypo-osmosensing Ca²⁺ channels

OSCAs control pollen Ca²⁺ oscillations

OSCAs couple external osmolarity to Ca²⁺ spiking

Imaging of [Ca²⁺]_i in HEK293 cells

Imaging of [Ca²⁺]_i in pollen grains