Re: plant protease 식물 단백질 분해제... 탐구!!

[문형철]

REVIEW article

Front. Plant Sci. , 18 April 2023

Sec. Plant Abiotic Stress

Volume 14 - 2023 | https://doi.org/10.3389/fpls.2023.1165845

This article is part of the Research TopicCrop Resistance Mechanisms to Alleviate Climate Change-Related StressView all 15 articles

The roles of plant proteases and protease inhibitors in drought response: a review

Sellwane Jeanette Moloi

Rudo Ngara*

• Department of Plant Sciences, University of the Free Sate, Qwaqwa Campus, Phuthaditjhaba, South Africa

Upon exposure to drought, plants undergo complex signal transduction events with concomitant changes in the expression‎ of genes, proteins and metabolites. For example, proteomics studies continue to identify multitudes of drought-responsive proteins with diverse roles in drought adaptation. Among these are protein degradation processes that activate enzymes and signalling peptides, recycle nitrogen sources, and maintain protein turnover and homeostasis under stressful environments. Here, we review the differential expression‎ and functional activities of plant protease and protease inhibitor proteins under drought stress, mainly focusing on comparative studies involving genotypes of contrasting drought phenotypes. We further explore studies of transgenic plants either overexpressing or repressing proteases or their inhibitors under drought conditions and discuss the potential roles of these transgenes in drought response. Overall, the review highlights the integral role of protein degradation during plant survival under water deficits, irrespective of the genotypes’ level of drought resilience. However, drought-sensitive genotypes exhibit higher proteolytic activities, while drought-tolerant genotypes tend to protect proteins from degradation by expressing more protease inhibitors. In addition, transgenic plant biology studies implicate proteases and protease inhibitors in various other physiological functions under drought stress. These include the regulation of stomatal closure, maintenance of relative water content, phytohormonal signalling systems including abscisic acid (ABA) signalling, and the induction of ABA-related stress genes, all of which are essential for maintaining cellular homeostasis under water deficits. Therefore, more validation studies are required to explore the various functions of proteases and their inhibitors under water limitation and their contributions towards drought adaptation.

가뭄에 노출되면 식물은 유전자, 단백질, 대사 산물의 발현 변화와 함께 복잡한 신호 전달 과정을 겪습니다. 예를 들어, 단백질체학 연구는 가뭄 적응에 다양한 역할을 하는 가뭄 반응성 단백질을 계속해서 확인하고 있습니다. 그 중에는 효소와 신호 펩타이드를 활성화하고, 질소원을 재활용하며, 스트레스 환경에서 단백질 회전율과 항상성을 유지하는 단백질 분해 과정이 있습니다. 여기에서는 가뭄 스트레스 하에서 식물 프로테아제와 프로테아제 억제 단백질의 발현 차이와 기능적 활동을 검토합니다. 주로 대조되는 가뭄 표현형의 유전자형을 포함하는 비교 연구에 초점을 맞춥니다. 또한 가뭄 조건에서 프로테아제 또는 그 억제제를 과발현하거나 억제하는 형질전환 식물에 대한 연구를 탐구하고, 이러한 형질전환 유전자가 가뭄 반응에서 잠재적으로 어떤 역할을 하는지 논의합니다. 전반적으로, 이 검토는 식물 생존에 있어 단백질 분해의 필수적인 역할을 강조하고 있으며, 이는 가뭄에 대한 내성 유전자의 수준과 무관합니다. 그러나 가뭄에 민감한 유전자형은 더 높은 단백질 분해 활성을 나타내는 반면, 가뭄에 내성이 있는 유전자형은 더 많은 프로테아제 억제제를 발현하여 단백질을 분해로부터 보호하는 경향이 있습니다. 또한, 유전자 변형 식물 생물학 연구에 따르면, 가뭄 스트레스 하에서 프로테아제와 프로테아제 억제제가 다양한 다른 생리 기능에 관여하는 것으로 밝혀졌습니다. 여기에는 기공 폐쇄 조절, 상대적 수분 함량 유지, 아브시신산(ABA) 신호를 포함한 식물호르몬 신호 전달 시스템, 그리고 ABA 관련 스트레스 유전자 유도가 포함되는데, 이 모든 것들은 물 부족 상황에서 세포의 항상성을 유지하는 데 필수적입니다. 따라서 물 부족 상황에서 프로테아제와 그 억제제의 다양한 기능과 가뭄 적응에 대한 기여도를 조사하기 위해서는 더 많은 검증 연구가 필요합니다.

1 Introduction

Plants require an optimum supply of light, water, temperature, and mineral nutrients for normal growth and development (Taiz and Zeiger, 2012). In nature, however, plants are often exposed to diverse abiotic stresses, including drought, high salinity, extreme temperatures, chemical toxicity and nutrient deficiency, which negatively affect their survival (Wang et al., 2003; Mahajan and Tuteja, 2005). In the case of crops, these stress factors may reduce yield, causing negative impacts on food supply chains (Mittler, 2006). Among the environmental stresses, drought is the major limiting factor of crop production worldwide (Farooq et al., 2009; da Silva et al., 2013), and its effects are further exacerbated by climate change and global warming. As climate models continue to predict the occurrence of frequent and severe drought episodes, more famines are likely to be experienced in drought-prone areas (IPCC, 2007; Nelson et al., 2009).

Plants are exposed to water deficit stress when rainfall declines during the growing season and when the rate of transpiration exceeds that of water absorption by roots due to hot and dry conditions (Turner and Begg, 1981; Bray, 1997). Consequently, osmotic stress develops, causing adverse effects on plant physiology, metabolism, and growth patterns (Taiz and Zeiger, 2012). For example, water deficit disrupts cell structure and function, including membrane integrity, photosynthesis, respiration, and growth processes (Anjum et al., 2011; Kumar et al., 2018; Kapoor et al., 2020). The extent of such effects, however, depends on the duration and severity of the water limitation, the plant species, genotype and/or developmental stage (Mullet and Whitsitt, 1996; Bray, 1997; Anjum et al., 2011), and whether water scarcity occurs in combination with other biotic and/or abiotic stresses (Rizhsky et al., 2004; Suzuki et al., 2014; Zandalinas et al., 2018). Nonetheless, plants have developed various mechanisms to cope with the prevailing environmental stresses to maintain cell structure, function and growth (Xiong and Zhu, 2002; Farooq et al., 2009; Shao et al., 2009; Osakabe et al., 2014).

One of the earliest responses of plants to dehydration stress is stomatal closure which is induced by the phytohormone abscisic acid (ABA) (Zhang et al., 2006; Lim et al., 2015; Kuromori et al., 2018). ABA is synthesised in roots and leaves in response to reduced water content in drying soil (Davies and Zhang, 1991; Taiz and Zeiger, 2012). As stomata close, transpiration water loss is reduced, and water is conserved. However, stomatal closure also restricts the absorption of CO2 required for photosynthesis and the uptake of nutrients for plant growth (Basu et al., 2016). The reduction in CO2 absorption and assimilation leads to an electron-rich environment in cells that is conducive for increased accumulation of reactive oxygen species (ROS) (Salehi-Lisar and Bakhshayeshan-Agdam, 2016), causing oxidative stress (Gill and Tuteja, 2010; Sharma et al., 2012), a secondary effect of many environmental stresses, including drought (Levitt, 1980a; Levitt, 1980b). If ROS molecules are not maintained at relatively low levels and/or effectively detoxified, they may damage lipids, nucleic acids and proteins, causing catastrophic disruptions to cell structure and metabolism (Wang et al., 2003).

To mitigate the adverse effects of both osmotic and oxidative stresses on cells, plants accumulate or activate a variety of stress-related genes, proteins, and metabolites through changes in cellular metabolism (Xiong and Zhu, 2002; Shao et al., 2009). Some of these molecular changes are mediated by ABA via the ABA-dependent pathway of stress response, while others form part of the ABA-independent pathway (Shinozaki and Yamaguchi-Shinozaki, 1997; Shinozaki and Yamaguchi-Shinozaki, 2007; Yoshida et al., 2014), and have various signalling, gene regulatory and protective functions (Shinozaki and Yamaguchi-Shinozaki, 1997; Shinozaki and Yamaguchi-Shinozaki, 2007). Ultimately, cellular homeostasis is maintained to promote plant survival under stressful conditions (Xiong and Zhu, 2002; Vinocur and Altman, 2005).

Apart from ABA, other phytohormones such as jasmonic acid, salicylic acid, ethylene, auxins, gibberellins, cytokinins, brassinosteroids, and small molecular peptides also mediate plant responses to drought through complex interactions of their signalling pathways (Clarke and Durley, 1981; Hale and Orcutt, 1987; Ullah et al., 2018; Jogawat et al., 2021; Salvi et al., 2021; Iqbal et al., 2022). The crosstalk between phytohormones may have positive or negative effects on the interacting hormones and their ultimate effect in alleviating drought stress (Ullah et al., 2018; Jogawat et al., 2021). Furthermore, the biosynthesis, catabolism, and transport of phytohormones, and their conversion between bioactive, inactive, and storage forms also influence how plants reprogram growth and developmental processes to survive drought stress (Clarke and Durley, 1981; Hale and Orcutt, 1987). For example, the levels of ABA, auxins, salicylic acid, jasmonic acid, brassinosteroids, and ethylene may increase in response to water deficits to facilitate stomata regulation of transpiration, osmotic adjustment, ROS scavenging, and increased root growth.

Conversely, the contents of gibberellins and cytokinins tend to decline under drought, and these hormones have opposite effects on stomata conductance and shoot and root meristem activity compared to ABA (Clarke and Durley, 1981; Hale and Orcutt, 1987; Ullah et al., 2018; Salvi et al., 2021; Iqbal et al., 2022). Furthermore, a decline in gibberellins content in plants under drought results in the accumulation of growth-repressor proteins such as DELLA and the subsequent development of growth-retarded plant phenotypes that are more tolerant to drought (Achard and Genschik, 2009; Salvi et al., 2021). Likewise, drought-induced reduction in cytokinins is associated with shoot growth inhibition and enhanced root growth facilitating water absorption from drying soils (Salvi et al., 2021). Collectively, phytohormone interactions in plants under drought stress modulate plant responses through complex morpho-physiological and molecular mechanisms, including transcriptional regulation of drought stress-related genes to maximise survival.

Research on plant responses to drought has increased exponentially in the last two decades, as evidenced by the growing number of related publications available on the PubMed database https://pubmed.ncbi.nlm.nih.gov/. In addition, genomes of over 788 different plant species (Sun et al., 2022), including those of the model plant Arabidopsis (Arabidopsis thaliana) (The Arabidopsis Genome Initiative, 2000), and major crops (Bolger et al., 2014) have been sequenced. Several reviews have discussed the applications of genome sequences (Edwards and Batley, 2010; Bolger et al., 2014) and genome editing tools (Arora and Narula, 2017; Bao et al., 2019; Zhang et al., 2020) for crop improvement, including drought resilience. Furthermore, genome sequence data are invaluable reference tools for high-throughput “omics” studies involving many aspects of plant biology. Consequently, innumerable transcriptomics, proteomics, and metabolomics studies on plant responses to drought have been published and reviewed (Cramer et al., 2011; Ngara and Ndimba, 2014; Shanker et al., 2014; Barkla, 2016; Ngara et al., 2021; Singh et al., 2022; Thanmalagan et al., 2022). In addition, the generalised functional catalogue of plant genes outlined by Bevan et al. (1998) has become an invaluable tool for assigning putative roles to drought-responsive proteins identified in proteomics studies (Nouri and Komatsu, 2010; Aranjuelo et al., 2011; Zhao et al., 2011; Mohammadi et al., 2012a; Wang et al., 2014; Tamburino et al., 2017; Goche et al., 2020).

Undoubtedly, the drought models employed in such studies are quite diverse, and so are the results generated (Osmolovskaya et al., 2018). Nonetheless, “omics” studies continue to broaden our understanding of how plants reprogram cellular metabolism to maximise survival under unfavourable environmental conditions (Kido et al., 2016; Ngara et al., 2021; Baldoni, 2022; Rakkammal et al., 2022; Singh et al., 2022). Furthermore, comparative proteomics studies of plant genotypes with contrasting drought phenotypes provide new insights into drought response mechanisms (Barkla, 2016). For example, common plant responses to water deprivation between drought-tolerant and sensitive cultivars include the down-regulation of metabolism-related proteins, possibly as an energy-saving mechanism, and the up-regulation of defence-related proteins for protective functions (Ford et al., 2011; Jedmowski et al., 2014; Faghani et al., 2015; Cheng et al., 2016; Wu et al., 2016; Zeng et al., 2019; Goche et al., 2020; Moosavi et al., 2020). Conversely, increased alternative splicing events (Fracasso et al., 2016) and higher constitutive expression‎ of secondary metabolism, redox homeostasis, and translation-related genes (Fracasso et al., 2016; Azzouz-Olden et al., 2020) possibly contribute towards the drought-superior traits of some varieties. These tolerant varieties also exhibit greater drought-induced accumulation of proteins involved in signal transduction, osmolyte biosynthesis, transcription, translation, and several protective roles such as antioxidant enzymes, dehydrins, late embryogenesis abundant proteins, chaperons, and regulators of proteolysis (Wang et al., 2015; Cheng et al., 2016; Chmielewska et al., 2016; Goche et al., 2020). Overall, such genes and proteins are pivotal during drought adaptation (Ingram and Bartels, 1996; Ramanjulu and Bartels, 2002) and could serve as potential biomarkers for crop improvement strategies (Barkla, 2016).

Here, we review the differential expression‎ and functional activities of plant protease and protease inhibitor proteins under drought stress, mainly focusing on comparative studies of cereal crops involving genotypes of contrasting drought phenotypes. We further explore studies of transgenic plants that either overexpress or repress proteases or their inhibitors under drought conditions and discuss the potential roles of these transgenes in drought response.

1 서론

식물은 정상적인 성장과 발달을 위해 최적의 빛, 물, 온도, 미네랄 영양분을 필요로 합니다(Taiz and Zeiger, 2012). 그러나 자연에서 식물은 가뭄, 높은 염도, 극한의 온도, 화학적 독성, 영양소 결핍 등 생존에 부정적인 영향을 미치는 다양한 비생물적 스트레스에 노출되는 경우가 많습니다(Wang et al., 2003; Mahajan and Tuteja, 2005). 작물의 경우, 이러한 스트레스 요인은 수확량을 감소시켜 식량 공급망에 부정적인 영향을 미칠 수 있습니다(Mittler, 2006). 환경 스트레스 중 가뭄은 전 세계적으로 작물 생산의 주요 제한 요인입니다(Farooq et al., 2009; da Silva et al., 2013). 그리고 그 영향은 기후 변화와 지구 온난화로 인해 더욱 악화됩니다. 기후 모델이 빈번하고 심각한 가뭄의 발생을 계속 예측함에 따라, 가뭄이 잦은 지역에서 기근이 더 많이 발생할 가능성이 높습니다(IPCC, 2007; Nelson et al., 2009).

식물은 성장기에 강우량이 감소하고 고온 건조한 조건으로 인해 뿌리의 수분 흡수 속도보다 증산 속도가 더 빠르면 물 부족 스트레스에 노출됩니다(Turner and Begg, 1981; Bray, 1997). 결과적으로 삼투압 스트레스가 발생하여 식물 생리학, 신진대사, 성장 패턴에 악영향을 미칩니다(Taiz and Zeiger, 2012). 예를 들어, 물 부족은 세포막의 완전성, 광합성, 호흡, 성장 과정 등 세포 구조와 기능을 방해합니다(Anjum et al., 2011; Kumar et al., 2018; Kapoor et al., 2020). 그러나 이러한 영향의 정도는 물 부족의 지속 기간과 심각성, 식물 종, 유전자형 및/또는 발달 단계(Mullet and Whitsitt, 1996; Bray, 1997; Anjum et al., 2011), 그리고 물 부족이 다른 생물적 및/또는 비생물적 스트레스와 함께 발생하는지 여부(Rizhsky et al., 2004; Suzuki et al., 2014; Zandalinas et al., 2018). 그럼에도 불구하고, 식물들은 세포 구조, 기능, 성장을 유지하기 위해 환경 스트레스에 대처하기 위한 다양한 메커니즘을 개발해 왔습니다(Xiong and Zhu, 2002; Farooq et al., 2009; Shao et al., 2009; Osakabe et al., 2014).

식물이 탈수 스트레스에 대한 가장 초기 반응 중 하나는 식물호르몬인 압시산(ABA)에 의해 유발되는 기공 폐쇄입니다(Zhang et al., 2006; Lim et al., 2015; Kuromori et al., 2018). ABA는 건조한 토양에서 수분 함량이 감소함에 따라 뿌리와 잎에서 합성됩니다(Davies and Zhang, 1991; Taiz and Zeiger, 2012). 기공이 닫히면 증발 수분 손실이 감소하고 수분이 보존됩니다. 그러나 기공이 닫히면 광합성에 필요한 CO2의 흡수와 식물 성장을 위한 영양분의 흡수도 제한됩니다(Basu et al., 2016). CO2 흡수 및 동화 작용의 감소는 세포 내 전자 풍부 환경을 유도하여 활성 산소(ROS)의 축적을 증가시킵니다(Salehi-Lisar and Bakhshayeshan-Agdam, 2016). 이로 인해 산화 스트레스(Gill and Tuteja, 2010; Sharma et al., 2012)가 발생하며, 이는 가뭄을 포함한 많은 환경 스트레스의 2차 효과입니다(Levitt, 1980a; Levitt, 1980b). ROS 분자가 상대적으로 낮은 수준으로 유지되지 않거나 효과적으로 해독되지 않으면 지질, 핵산, 단백질에 손상을 입혀 세포 구조와 신진대사에 치명적인 혼란을 초래할 수 있습니다(Wang et al., 2003).

세포에 대한 삼투압과 산화 스트레스의 부정적인 영향을 완화하기 위해, 식물은 세포 대사의 변화를 통해 다양한 스트레스 관련 유전자, 단백질, 대사 산물을 축적하거나 활성화합니다(Xiong and Zhu, 2002; Shao et al., 2009). 이러한 분자 변화 중 일부는 스트레스 반응의 ABA 의존 경로를 통해 ABA에 의해 매개되는 반면, 다른 일부는 ABA 독립 경로의 일부를 형성합니다(Shinozaki and Yamaguchi-Shinozaki, 1997; Shinozaki and Yamaguchi-Shinozaki, 2007; Yoshida et al., 2014), 다양한 신호 전달을 가지고 있으며, 유전자 조절 및 보호 기능(Shinozaki and Yamaguchi-Shinozaki, 1997; Shinozaki and Yamaguchi-Shinozaki, 2007). 궁극적으로 스트레스가 많은 조건에서 식물 생존을 촉진하기 위해 세포 항상성이 유지됩니다(Xiong and Zhu, 2002; Vinocur and Altman, 2005).

ABA 외에도, 자스몬산, 살리실산, 에틸렌, 옥신, 지베렐린, 사이토키닌, 브라시노스테로이드, 그리고 작은 분자 펩타이드와 같은 다른 식물호르몬도 신호 전달 경로의 복잡한 상호작용을 통해 가뭄에 대한 식물 반응을 매개합니다(Clarke and Durley, 1981; Hale and Orcutt, 1987; Ullah et al., 2018년; Jogawat 외, 2021년; Salvi 외, 2021년; Iqbal 외, 2022년). 식물호르몬 간의 상호작용은 상호작용하는 호르몬과 가뭄 스트레스를 완화하는 궁극적인 효과에 긍정적 또는 부정적 영향을 미칠 수 있습니다(Ullah 외, 2018년; Jogawat 외, 2021년). 또한, 식물 호르몬의 생합성, 이화 작용, 수송, 그리고 생체 활성, 비활성, 저장 형태 사이의 전환도 식물이 가뭄 스트레스를 견디기 위해 성장과 발달 과정을 재프로그램하는 방식에 영향을 미칩니다(Clarke and Durley, 1981; Hale and Orcutt, 1987). 예를 들어, ABA, 옥신, 살리실산, 자소산, 브라시노스테로이드, 에틸렌의 수준은 수분 부족에 반응하여 기공 조절을 촉진하여 증산, 삼투압 조절, 활성산소 소거, 뿌리 성장 증가를 촉진할 수 있습니다.

반대로, 지베렐린과 사이토키닌의 함량은 가뭄이 지속될 때 감소하는 경향이 있으며, 이 호르몬들은 ABA와 비교했을 때 기공 전도도와 싹과 뿌리 분열조직 활동에 반대되는 영향을 미칩니다(Clarke and Durley, 1981; Hale and Orcutt, 1987; Ullah et al., 2018; Salvi et al., 2021; Iqbal et al., 2022년). 또한, 가뭄에 노출된 식물의 지베렐린 함량이 감소하면 DELLA와 같은 성장 억제 단백질이 축적되고, 그 결과 가뭄에 더 잘 견디는 성장 지연 식물 표현형이 발달합니다(Achard and Genschik, 2009; Salvi et al., 2021). 마찬가지로, 가뭄으로 인한 사이토키닌의 감소는 싹의 성장 억제 및 건조한 토양에서 수분 흡수를 촉진하는 뿌리의 성장 증가와 관련이 있습니다(Salvi et al., 2021). 종합적으로, 가뭄 스트레스에 노출된 식물의 식물호르몬 상호작용은 생존을 극대화하기 위해 가뭄 스트레스 관련 유전자의 전사 조절을 포함한 복잡한 형태-생리학적 및 분자적 메커니즘을 통해 식물 반응을 조절합니다.

가뭄에 대한 식물의 반응에 대한 연구는 지난 20년 동안 기하급수적으로 증가했으며, PubMed 데이터베이스 https://pubmed.ncbi.nlm.nih.gov/에 관련 출판물이 증가하고 있다는 사실만 보더라도 이를 확인할 수 있습니다. 또한, 모델 식물인 애기장대(Arabidopsis thaliana) (The Arabidopsis Genome Initiative, 2000)와 주요 작물(Bolger et al., 2014)을 포함한 788종 이상의 식물 종(Sun et al., 2022)의 게놈이 염기서열 분석되었습니다. 여러 리뷰에서 가뭄 저항성을 포함한 작물 개량을 위한 게놈 서열(Edwards and Batley, 2010; Bolger et al., 2014)과 게놈 편집 도구(Arora and Narula, 2017; Bao et al., 2019; Zhang et al., 2020)의 적용에 대해 논의했습니다. 또한, 게놈 서열 데이터는 식물 생물학의 여러 측면을 포함하는 고처리 “옴니버스” 연구에 매우 중요한 참고 자료입니다. 따라서, 가뭄에 대한 식물의 반응에 관한 수많은 전사체학, 단백질체학, 대사체학 연구가 발표되고 검토되었습니다(Cramer et al., 2011; Ngara and Ndimba, 2014; Shanker et al., 2014; Barkla, 2016년; Ngara 외, 2021년; Singh 외, 2022년; Thanmalagan 외, 2022년). 또한, Bevan 외. (1998년)이 개괄한 식물 유전자의 일반화된 기능 카탈로그는 단백질체학 연구에서 확인된 가뭄 반응 단백질에 추정 역할을 할당하는 데 매우 유용한 도구가 되었습니다(Nouri와 Komatsu, 2010년; Aranjuelo 외, 2011년; Zhao 외, 2011년; Mohammadi 외, 2012년 a호; Wang 외, 2014년; Tamburino 외, 2017년; Goche 외, 2020년).

이러한 연구에 사용된 가뭄 모델은 매우 다양하며, 그로 인해 생성되는 결과도 다양합니다(Osmolovskaya et al., 2018). 그럼에도 불구하고, “옴니버스” 연구는 불리한 환경 조건에서 생존을 극대화하기 위해 식물이 세포 대사를 어떻게 재프로그램 하는지에 대한 이해를 넓혀가고 있습니다(Kido et al., 2016; Ngara et al., 2021; Baldoni, 2022; Rakkammal et al., 2022; Singh et al., 2022). 또한, 가뭄에 대한 표현형이 대조적인 식물 유전자형을 비교하는 단백질체학 연구는 가뭄에 대한 반응 메커니즘에 대한 새로운 통찰을 제공합니다(Barkla, 2016). 예를 들어, 가뭄에 강한 품종과 민감한 품종 사이의 물 부족에 대한 일반적인 식물 반응에는 에너지 절약 메커니즘으로 추정되는 대사 관련 단백질의 하향 조절과 보호 기능을 위한 방어 관련 단백질의 상향 조절이 포함됩니다(Ford et al., 2011; Jedmowski et al., 2014년; Faghani 외, 2015년; Cheng 외, 2016년; Wu 외, 2016년; Zeng 외, 2019년; Goche 외, 2020년; Moosavi 외, 2020년). 반대로, 증가된 대체 스플라이싱 이벤트(Fracasso et al., 2016)와 2차 대사, 산화 환원 항상성, 번역 관련 유전자의 더 높은 구성적 발현(Fracasso et al., 2016; Azzouz-Olden et al., 2020)은 일부 품종의 가뭄에 대한 우수한 특성에 기여할 수 있습니다. 이러한 내성 품종은 또한 신호 전달에 관여하는 단백질의 가뭄에 의한 축적을 더 많이 나타내며, osmolyte 생합성, 전사, 번역, 그리고 항산화 효소, 탈수소효소, 후기 배발생 풍부 단백질, 샤페론, 단백질 분해 조절기(Wang et al., 2015; Cheng et al., 2016; Chmielewska et al., 2016; Goche et al., 2020)와 같은 여러 가지 보호 역할. 전체적으로, 이러한 유전자와 단백질은 가뭄 적응 과정에서 중추적인 역할을 하며(Ingram and Bartels, 1996; Ramanjulu and Bartels, 2002), 작물 개량 전략의 잠재적 바이오마커로 사용될 수 있습니다(Barkla, 2016).

여기에서는 가뭄 스트레스 하에서 식물 프로테아제와 프로테아제 억제 단백질의 차별적 발현과 기능적 활동을 검토합니다. 주로 가뭄 표현형이 대조적인 곡물 작물의 유전자형에 대한 비교 연구에 초점을 맞추고 있습니다. 또한 가뭄 조건에서 프로테아제 또는 그 억제제를 과발현하거나 억제하는 형질전환 식물에 대한 연구를 탐구하고, 이러한 형질전환 유전자가 가뭄 반응에 미치는 잠재적 역할에 대해 논의합니다.

2 Plant proteases and protease inhibitors

Plant proteases are proteolytic enzymes that hydrolyse peptide bonds in proteins and are found in various plant tissues and organs (Palma et al., 2002; Schaller, 2004; van der Hoorn, 2008; Sharma and Gayen, 2021). Their activities are tightly regulated through the transcriptional control of protease transcripts, post-translational modifications of their proenzymes, actions of endogenous protease inhibitors, and/or compartmentalization into organelles and cellular compartments to avoid random acts of protein degradation (Brzin and Kidrič, 1996; Vierstra, 1996; Diaz-Mendoza et al., 2016). For instance, plant proteases are located in the cytosol, chloroplasts, vacuoles, nuclei, endoplasmic reticulum, proteasome, mitochondria, and cell walls (Vierstra, 1996; Kidric et al., 2014; Diaz-Mendoza et al., 2016), and also secreted into the extracellular matrix (Ngara et al., 2018; Ngcala et al., 2020; Godson and van der Hoorn, 2021). Each of these cellular compartments may possess specialised proteolytic pathways. For example, in the cytosol, protein degradation is mainly carried out by the highly selective ubiquitin-proteasome system (UPS), which consists of ubiquitin, the proteasome and associated components (Hopkins and Huner, 2009; Xu and Xue, 2019). Other proteolytic machinery found in plants include the caseinolytic protease (Clp) system in plastids and mitochondria and the vacuolar processing enzymes in vacuoles (Kato and Sakamoto, 2010; Vaseva et al., 2012; Nishimura and van Wijk, 2015; van Wijk, 2015; Ali and Baek, 2020).

Proteases are structurally and functionally diverse and are classified based on their catalytic activity, such as aspartic, cysteine, serine and threonine peptidases (Callis, 1995; Schaller, 2004). Alternatively, proteolytic enzymes are grouped into endo- and exo-peptidases depending on the site of cleavage on the peptide chain (Palma et al., 2002; Schaller, 2004). Examples of endopeptidases include serine, cysteine, aspartic, threonine and metalloendopeptidases, while exopeptidases include aminopeptidases, dipeptidases, carboxypeptidase, didpeptidyl-peptidases, omega peptidases and peptidyl-dipeptidases (Beers et al., 2000; Palma et al., 2002; Vaseva et al., 2012; Kidric et al., 2014). Likewise, protease inhibitors are diverse small molecules of either protein or non-protein nature (Polya, 2003); and are located in various plant organs and cellular compartments (Mosolov and Valueva, 2011; Kidric et al., 2014). They are differentiated based on their reaction mechanism or the type of proteases they inhibit (Mosolov and Valueva, 2011; Kidric et al., 2014). Examples of endogenous plant protease inhibitors include phytocystatins, serpins, Kunitz protease inhibitors, Bowman-Birk inhibitors, a-amylase inhibitors, bifunctional trypsin inhibitors, metallocarboxypeptidase inhibitors, mustard trypsin inhibitors, and potato-type inhibitors (Mosolov and Valueva, 2011; Vaseva et al., 2012; Kidric et al., 2014; Hellinger and Gruber, 2019). Additional information and online resource tools for plant proteases and their endogenous inhibitors can be found in the listed databases (Table 1).

2 식물 프로테아제와 프로테아제 억제제

식물 프로테아제는 단백질의 펩티드 결합을 가수분해하는 단백질 분해 효소이며, 다양한 식물 조직과 기관에서 발견됩니다(Palma et al., 2002; Schaller, 2004; van der Hoorn, 2008; Sharma and Gayen, 2021). 그들의 활동은 프로테아제 전사체의 전사 조절, 프로엔자임의 번역 후 변형, 내인성 프로테아제 억제제의 작용, 그리고/또는 단백질 분해의 무작위적 작용을 피하기 위한 세포 기관과 세포 구획으로의 분할을 통해 엄격하게 규제됩니다(Brzin and Kidrič, 1996; Vierstra, 1996; Diaz-Mendoza et al., 2016). 예를 들어, 식물 프로테아제는 세포질, 엽록체, 액포, 핵, 소포체, 프로테아좀, 미토콘드리아, 세포벽에 위치해 있으며(Vierstra, 1996; Kidric et al., 2014; Diaz-Mendoza et al., 2016), 또한 세포 외 기질로 분비됩니다(Ngara et al., 2018; Ngcala et al., 2020; Godson and van der Hoorn, 2021). 이러한 세포 소기관들 각각은 특화된 단백질 분해 경로를 가지고 있을 수 있습니다. 예를 들어, 세포질에서 단백질 분해는 주로 유비퀴틴, 프로테아좀, 그리고 관련 구성 요소들로 구성된 매우 선택적인 유비퀴틴-프로테아좀 시스템(UPS)에 의해 수행됩니다(Hopkins and Huner, 2009; Xu and Xue, 2019). 식물에서 발견되는 다른 단백질 분해 기구로는 플라스티드와 미토콘드리아의 카제인 분해성 프로테아제(Clp) 시스템과 액포의 액포 처리 효소(Kato and Sakamoto, 2010; Vaseva et al., 2012; Nishimura and van Wijk, 2015; van Wijk, 2015; Ali and Baek, 2020)가 있습니다.

프로테아제는 구조적, 기능적으로 다양하며, 촉매 활성에 따라 아스파르트, 시스테인, 세린, 트레오닌 펩티다아제 등으로 분류됩니다(Callis, 1995; Schaller, 2004). 또는, 단백질 분해효소는 펩티드 사슬의 절단 부위에 따라 엔도펩티다제와 엑소펩티다제로 분류됩니다(Palma et al., 2002; Schaller, 2004). 엔도펩티다제의 예로는 세린, 시스테인, 아스파르트, 트레오닌, 메탈로엔도펩티다제가 있으며, 엑소펩티다제의 예로는 아미노펩티다제가 있습니다. 디펩티다제, 카르복시펩티다제, 디디펩티딜펩티다제, 오메가펩티다제, 펩티딜디펩티다제(Beers et al., 2000; Palma et al., 2002; Vaseva et al., 2012; Kidric et al., 2014). 마찬가지로, 프로테아제 억제제는 단백질 또는 비단백질 성질의 다양한 소분자(Polya, 2003)이며, 다양한 식물 기관과 세포 구획(Mosolov and Valueva, 2011; Kidric et al., 2014)에 위치합니다. 이들은 반응 메커니즘이나 억제하는 프로테아제의 유형에 따라 구분됩니다(Mosolov and Valueva, 2011; Kidric et al., 2014). 내인성 식물 프로테아제 억제제의 예로는 피토시스타틴, 세르핀, 쿠니츠 프로테아제 억제제, 보우만-버크 억제제, α-아밀라제 억제제, 이중 기능성 트립신 억제제, 메탈로카르복시펩티다제 억제제, 겨자 트립신 억제제, 감자형 억제제(Mosolov and Valueva, 2011; Vaseva et al., 2012; Kidric et al., 2014; Hellinger and Gruber, 2019). 식물성 프로테아제와 그 내인성 억제제에 대한 추가 정보와 온라인 자료 도구는 나열된 데이터베이스에서 찾을 수 있습니다(표 1).

TABLE 1

Table 1 Online databases and resource tools for plant proteases and protease inhibitors.

2.1 Putative protein substrates and functions of plant proteases

The mechanism of action of proteases, their inhibitors, and different proteolytic pathways in plants have been extensively reviewed elsewhere (Vierstra, 1996; Vaseva et al., 2012; Kidric et al., 2014) and can be accessed for in-depth reading. Other reviews have discussed proteolytic machinery in plastids, mitochondria and peroxisomes, as well as their roles in maintaining protein homeostasis, also called proteostasis in these organelles (Kato and Sakamoto, 2010; Nishimura and van Wijk, 2015; van Wijk, 2015; Nishimura et al., 2016; Nishimura et al., 2017; Sun et al., 2021). The extensive review by van Wijk (2015) also lists substrates of plastidal, mitochondrial and peroxisomal proteases and is recommended for further reading, together with other references therein. Readers are also encouraged to access the comprehensive review chapter by Kato and Sakamoto (2010) on major chloroplast proteases, their substrates and functions. Nishimura et al. (2016) also provide a comprehensive pictorial depiction of the different types of metallo, serine, and aspartic proteases in plastids of land plants, their suborganellar location, and apply case studies to illustrate how the major chloroplastic proteases maintain organellar proteostasis. These earlier reviews highlight the different types of proteolytic machinery in plant cells, organelles and compartments, as well as the integral role of proteolysis during protein processing, maturation and quantity and quality control.

A common notion from these reviews is the extensive diversity of plant proteases, their multiple protein substrates, subcellular locations, and specialised functions. These proteolytic systems also work in a coordinated manner to maintain protein homeostasis (Kato and Sakamoto, 2010). However, the identities of specific protein substrates and mechanisms of action for many of these proteases are only partially known (Kato and Sakamoto, 2010; van Wijk, 2015). Nevertheless, enormous progress has been made in profiling the physiological substrates of various plant proteases. For example, several studies have used in vitro recombinant proteins, mutant systems, substrate-trapping assays, and N-terminal degradome techniques coupled with quantitative proteomics and mass spectrometry technologies for substrate identification (Moberg et al., 2003; Tsiatsiani et al., 2013; Sako et al., 2014; Carrie et al., 2015; Galiullina et al., 2015; Opalinska et al., 2017; Moreno et al., 2018; Welsch et al., 2018; Rowland et al., 2022).

In a recent study, Rowland et al. (2022) used several mutant types, the terminal amine stable isotopic labelling of substrates (TAILS), and label-free proteomic methods to investigate the coordination between Clp and presequence protease (PreP) proteases in maintaining chloroplast proteostasis. The results showed synergistic interactions between the two protease systems, their effects on embryo lethality, and changes in N-terminal protein processing activities. Furthermore, the loss-of-function of the Clp systems resulted in the over-accumulation of chloroplast stromal chaperons such as heat shock protein 90 (HSP90), chaperon protein CLPB3, chloroplastic (CLPB3) and chaperonins (CPN60/10/20), which possibly re-fold and stabilise abnormally folded and aggregated proteins. However, none of these three chaperons was affected in the prep1prep2 mutant plants relative to the wild-type (Rowland et al., 2022) further highlighting the different effects of proteases in plants. Using several clp mutants, co-immunoprecipitation, and proteomic analyses, Welsch et al. (2018) experimentally showed that phytoene synthase, an enzyme in carotenoid biosynthesis, is a substrate for the Clp protease system. Therefore, the authors argued that proteolysis is critical in ensuring quantity control of phytoene synthase for adequate carotenoid biosynthesis (Welsch et al., 2018).

We acknowledge that a complete listing of known substrates of various plant proteases is beyond the scope of the current review. However, below is a brief account of how plants maintain protein homeostasis using diverse proteolytic machinery and their substrates. In many cases, they operate in tandem to complete target protein degradation (Kato and Sakamoto, 2010; van Wijk, 2015). A summary list of protein substrates of selected plant proteolytic machinery is also presented in Table 2. For example, once imported into their target organelles, nucleus-encoded chloroplast and mitochondrial proteins are further processed through the proteolytic cleavage of chloroplast transit peptides (cTP) and mitochondrial targeting peptides (mTP) before maturation. These activities are carried out by the stromal processing peptidase (SPP) (Richter and Lamppa, 1999; Richter et al., 2005) and mitochondrial processing peptidase (MPP) (Ghifari et al., 2019) metalloproteases in chloroplasts and mitochondria, respectively or other yet to be identified peptidases (Rowland et al., 2022). In peroxisomes, the cleavable N-terminal peroxisomal targeting signal (PTS2) of nucleus-encoded proteins is cleaved off by the degradation of the periplasmic protein (Deg)15 proteases, while PTS1 in non-cleavable (Schuhmann et al., 2008). Deg proteases are ATP-independent serine endopeptidases with essential roles in protein quality control in nearly all organisms, including plants (Schuhmann et al., 2008; Schuhmann et al., 2012).

TABLE 2

Table 2 Examples of protein substrates for selected proteases and peptidases during plant growth and development.

Schuhmann et al. (2008) experimentally verified that PTS2-containing presequences of glyoxysomal malate dehydrogenase, 3-keto-acyl-CoA thiolase, and a long-chain acyl-CoA synthetase 6 are substrates of Deg15. The cleaved cTP, mTP and PTS2 peptides are subsequently degraded by PreP (Moberg et al., 2003; Kmiec and Glaser, 2012) or organellar oligopeptidase (OPP) (Ghifari et al., 2019) to prevent the accumulation of potentially toxic non-functional peptides and to facilitate amino acid recycling (van Wijk, 2015; Rowland et al., 2022). Other in vitro studies using recombinant proteins have shown that various Deg family proteins may be involved in maintaining chloroplast proteostasis by degrading thylakoid lumen proteins such as plastocyanin and the oxygen-evolving complex (OEC) protein 33 of the photosystem II (PSII), and heat stressed or photo-damaged D1 reaction centre protein (Kato and Sakamoto, 2010; van Wijk, 2015). Chloroplasts and mitochondria also contain other specialised endopeptidases, such as the type-I signal peptidase family (SPase I) and the plastidal type-I signal peptidase (Plsp1), that process luminal and intermembrane precursor proteins by cleaving off signal peptides for full maturation and activation (Tuteja, 2005; Kato and Sakamoto, 2010). For example, Plsp1 is involved in the maturation of the translocon at the outer envelope membrane of chloroplasts, 75kDA (Toc75), a member of the protein translocation complex at the outer envelope membrane of plastids (Baldwin et al., 2005). Other chloroplast and mitochondrion-encoded proteins are further post-translationally modified by the proteolytic removal of an N-terminal methionine using methionine aminopeptidases (MAPs) to achieve normal protein function and leaf development (Ghifari et al., 2019).

Due to their wide structural diversity, substrate specificities, selectivity, and/or subcellular locations (Vierstra, 1996; Kato and Sakamoto, 2010; Vaseva et al., 2012; Kidric et al., 2014; van Wijk, 2015), plant proteases are involved in numerous functions during normal plant growth and development and in response to biotic and abiotic stresses (Figure 1). For instance, proteolytic activities are essential during nitrogen recycling and remobilisation, leaf senescence and programmed cell death, controlled degradation of damaged or abnormal proteins, activation and maturation of proteins and peptide hormones, seed germination, seedling growth, cellular housekeeping, regulating protein turnover, and in response to biotic and abiotic stresses (Brzin and Kidrič, 1996; Vierstra, 1996; Schaller, 2004; van der Hoorn, 2008; Hopkins and Huner, 2009; Kato and Sakamoto, 2010; Kidric et al., 2014; van Wijk, 2015; Diaz-Mendoza et al., 2016). On the other hand, protease inhibitors regulate the function of proteases by inhibiting their catalytic activity and may participate in plant defence and protective roles in response to biotic and abiotic stresses (Brzin and Kidrič, 1996; Mosolov and Valueva, 2005; Mosolov and Valueva, 2011).

FIGURE 1

Figure 1 General functions of plant proteases in plant growth and development. (Brzin and Kidrič, 1996; Vierstra, 1996; Schaller, 2004; van der Hoorn, 2008; Hopkins and Huner, 2009; Kato and Sakamoto, 2010; Kidric et al., 2014; van Wijk, 2015; Diaz-Mendoza et al., 2016; Sharma and Gayen, 2021).

2.2 Plant protease and protease inhibitor proteins under drought stress2.2.1 Comparative proteomics studies

The involvement of plant proteases and protease inhibitors in drought response has been reported in various plant species using transcriptomics, proteomics and/or enzyme activity assays. Earlier reviews have also discussed the roles of plant proteases and their inhibitors under normal growth and in response to various biotic and abiotic stresses (Brzin and Kidrič, 1996; Mosolov and Valueva, 2011; Kidric et al., 2014; Godson and van der Hoorn, 2021), including senescence (Diaz-Mendoza et al., 2016). Xu and Xue (2019) reviewed the components of the UPS and their regulation in response to biotic and abiotic stresses. Others have outlined the potential function of plant proteases in signal transduction systems under environmental stresses (Sharma and Gayen, 2021), including drought (D’Ippólito et al., 2021). Due to the extensive collection of “omics” studies available in the public domain, we opted to focus this review on the drought-induced expression‎al changes of protease and protease inhibitor proteins as depicted in comparative proteomic studies of cereal crops (Table 3). However, the general literature on proteomics approaches (Monteoliva and Albar, 2004; Wu et al., 2006; Abdallah et al., 2012; Zargar et al., 2016) and plant proteome analyses under drought stress and applications in crop improvement (Mohammadi et al., 2012b; Barkla, 2016; Wang et al., 2016; Ghatak et al., 2017) has been extensively reviewed elsewhere and will not be discussed in the current review.

TABLE 3

Table 3 Studies showing differential accumulation and functional activities of protease and protease inhibitor proteins in plants under drought stress.

Wu et al. (2016) used label-free and tandem mass tag multiplexing methods to analyse leaf proteome changes in two rice (Oryza sativa) varieties subjected to water deficits and re-watering. Amongst the observed unique proteome changes was the increased accumulation of a ClpD1 protease in the drought-tolerant IAC1131 rice variety under severe drought conditions and its decreased accumulation upon re-watering. ClpD1 belongs to the caseinolytic protease (Clps) family of proteins (Kato and Sakamoto, 2010; Roberts et al., 2012; Vaseva et al., 2012; Nishimura and van Wijk, 2015) and is encoded by the drought-inducible gene OsClpD1 in rice (Wu et al., 2016). Clp proteases are involved in degrading damaged proteins in the plastids (Kato and Sakamoto, 2010; Ali and Baek, 2020). Wu et al. (2016) suggested that the drought-induced accumulation of ClpD1 in the drought-tolerant variety increased the genotype’s chances of coping with extreme drought conditions through the regulation of protein quality.

In a comparative gel-based proteomic study, an array of proteolysis-related proteins were up-regulated in leaves of wheat (Triticum aestivum) plants subjected to drought stress for 18, 24 and 48 hours (Cheng et al., 2016). For example, proteasome subunit alpha type-2, 20S proteasome beta 7 subunit, an aspartic proteinase precursor protein and a triticain alpha subunit were up-regulated in Xihan No. 2, a drought-tolerant wheat variety. Likewise, increased abundances of proteasome subunit alpha type-7-A, proteasome subunit alpha type-1 and an ATP-dependent Clp protease proteolytic subunit were observed in the drought-sensitive variety following water deprivation. The authors suggested that proteolysis is an essential mechanism for maintaining protein quality under stress, irrespective of a cultivar’s degree of drought resilience. Furthermore, when coupled with adequate protein refolding by chaperons, proteolytic activity helps plants to maintain protein homeostasis under dehydration stress (Cheng et al., 2016). In another study, a proteasome alpha-type protein accumulated in leaves of a drought-sensitive wheat variety but not the drought-tolerant one upon exposure to water limitation (Faghani et al., 2015). Overall, the observations highlight the importance of selective protein degradation by the ubiquitin-proteasome system in plants under water deficit stress.

In another comparative gel-based proteomic study of sorghum (Sorghum bicolor) leaves, an aspartate protease and a mitochondrial processing peptidase were up-regulated in the drought-susceptible variety in response to drought stress (Jedmowski et al., 2014). Aspartic proteases are widely distributed within the plant kingdom and are involved in protein degradation during normal plant development, nitrogen recycling, programmed cell death, and stress response (Mutlu and Gal, 1999; Simoes and Faro, 2004). As discussed earlier, mitochondrial processing peptidases are involved in the removal of N-terminal signal peptides also called the presequence, from nuclear-encoded mitochondrial proteins during or after their import into the mitochondrion (Glaser et al., 1998; Kwasniak et al., 2012; Ghifari et al., 2019). As such, the observed drought-increased accumulation of aspartic protease could indicate high levels of degraded proteins in the susceptible variety, while the mitochondrial processing peptidase could be responsible for presequence removal (Jedmowski et al., 2014), and thus maturation of various newly synthesised mitochondrial precursors (Glaser et al., 1998; Gakh et al., 2002; van Wijk, 2015).

In one of our studies, Goche et al. (2020) conducted a comparative root proteomic analysis of two contrasting sorghum varieties subjected to limited watering. Among the observed differences in protein expression‎ patterns were the up-regulation of proteolytic enzymes in ICSB338, the drought-sensitive variety, while the more drought-tolerant SA1441 accumulated proteins involved in transcription, protein synthesis, and the inhibition of protease activities (Goche, 2018; Goche et al., 2020). For example, dipeptidyl peptidase, serine-type, cysteine-type, and aspartic-type endopeptidases were up-regulated in the drought-susceptible sorghum variety upon exposure to drought stress. On the other hand, the drought-tolerant variety predominantly up-regulated protease inhibitors such as the proteinase inhibitor, potato inhibitor I (Goche, 2018; Goche et al., 2020). Although the significance of such contrasting proteome expression‎ patterns warrants further functional validation studies, it is evident that the drought-resilient SA1441 reprograms gene expression‎, possibly to facilitate the accumulation of diverse stress proteins and protect them from proteolytic degradation. It is also plausible that the drought-sensitive sorghum variety requires extensive protein degradation to remove damaged proteins in many cellular locations and to recycle amino acids for targeted protein synthesis (Hopkins and Huner, 2009).

Ford et al. (2011) also conducted an extensive comparative leaf proteomic analysis of one drought-sensitive Kukri and two drought-tolerant RAC875 and Excalibur wheat cultivars in response to cyclic drought treatment. Leaf samples were subsequently taken during the initial period of water deficit, two wilting time-points, and following re-watering for relative water content (RWC) estimations and proteome analysis. As observed from the RWC results, this experimental system provided a platform for investigating complex physiological changes of wheat under repetitive cycles of water deficit and recovery, similar to those experienced under natural field conditions. Comparative analyses of the drought-responsive proteins were conducted at the four sampling time points using isobaric tags for relative and absolute quantification (iTRAQ) and tandem mass spectrometry. Overall, the results revealed complex proteome responses of wheat to the cyclic drought treatment, genotypic differences between cultivars irrespective of their levels of drought resistance, and some common trends in metabolic changes under conditions of water deprivation (Ford et al., 2011). For example, proteins related to photosynthesis and the Calvin cycle were down-regulated across the three cultivars, while ROS-scavenging enzymes and dehydrins were up-regulated.

The study also revealed increased accumulation of type II metacaspase and leucine aminopeptidase proteins in drought-tolerant cultivars, especially towards the later stages of water deprivation. However, after re-watering, the type II metacaspase protein levels reverted to control levels in both drought-tolerant cultivars but increased in the drought-susceptible cultivar (Ford et al., 2011). Conversely, the accumulation pattern of the leucine aminopeptidase protein was not consistent between the tolerant cultivars during the water deficit but increased upon re-watering in both cultivars. Generally, the study highlights the complexity of plant proteome responses during drought stress and recovery, the diversity of proteases and their selective functions during drought stress, and the differential responses of proteases in drought-stressed wheat plants with different drought tolerance levels. For example, since metacaspases play a role in programmed cell death (Tsiatsiani et al., 2011), their early expression‎ in drought-tolerant cultivars may implicate them in protein degradation processes that rapidly sacrifice some cells to ensure plant survival under stress (Ford et al., 2011). In addition, leucine aminopeptidases activate proteins and regulate protein turnover in plants (Walling and Gu, 1996; Matsui et al., 2006). As such, their increased accumulation upon re-watering in the drought-tolerant cultivars may indicate the importance of protein degradation, amino acid recycling and activation processes as cells reset their protein component during recovery from stress.

In another comparative proteomic study, proteases were differentially expressed in barley (Hordeum vulgare) plants subjected to drought stress (Ashoub et al., 2013). The study used gel-based proteomic methods to analyse the leaf proteome changes between accessions #15141 and #15163, which are tolerant and sensitive to drought, respectively. For example, two ATP-dependent Clp proteases were up-regulated in both varieties in response to drought stress. In contrast, a third ATP-dependent Clp protease and two leucine aminopeptidases were up-regulated only in the drought-susceptible accession, while a zinc metalloprotease had increased accumulation only in the tolerant accession. The authors also reported that the drought-susceptible variety showed an increased accumulation of proteases in drought response compared to the drought-tolerant variety, which accumulated more proteins involved in protective mechanisms against drought stress (Ashoub et al., 2013). Overall, the studies mentioned above emphasize the diversity of proteolytic enzymes, their selective specificities, and pivotal roles in various protein degradation processes during drought response in plants.

Although the current review aimed at surveying the drought-induced expression‎ changes of protease and protease inhibitor proteins as depicted in comparative proteomic studies, we also scanned through a few similar studies on comparative transcriptomics to establish trends in transcript levels of these two groups of proteins. Indeed, the correlation between mRNA and proteins of various biological systems, including the human liver (Anderson and Seilhame, 1997), yeast (Gygi et al., 1999), and plants (Carpentier et al., 2008; Ponnala et al., 2014) has been debated for years, with poor correlation trends being attributed to various factors including the differential turnover rates of mRNA and proteins and posttranslational modifications of proteins (Abbott, 1999; Salekdeh et al., 2002; Vogel and Marcotte, 2012). Nevertheless, different “omics” technologies and the data thereof should be regarded as complementary, as each analytical technique may over and/or under-represent particular groups of biological molecules depending on their inherent characteristics (Carpentier et al., 2008). However, we found recent comparative transcriptomics studies that reported the drought-induced differential gene expression‎ patterns of proteases and/or their inhibitors between plant genotypes with contrasting levels of drought tolerance (Zenda et al., 2019; Abdel-Ghany et al., 2020; Bhogireddy et al., 2020; Shivhare et al., 2020).

For example, Abdel-Ghany et al. (2020) conducted an extensive comparative transcriptome analysis between two drought-resistant and two drought-sensitive sorghum varieties in response to 20% polyethylene glycol (PEG)-induced osmotic stress. The results revealed differential expression‎ of protein degradation-related transcripts between seedlings of the two groups of genotypes in response to 1 and 6 hours of osmotic stress treatment. For example, a cysteine protease gene and a senescence-associated gene were up-regulated in all four sorghum genotypes in response to the 1 hour and 6 hours of stress, respectively, highlighting the critical role of proteolysis and senescence during drought response in plants. In addition, following 1 hour of PEG treatment, a cysteine protease inhibitor and three cysteine protease genes were among the top 20 up-regulated genes in both drought-resistant sorghum varieties. However, at 6 hours, four seed storage protein-related protease inhibitors are down-regulated in the same drought-resistant varieties. Surprisingly, no protease or protease inhibitor genes were responsive to the PEG treatments at either time point in the drought-susceptible sorghum varieties. The authors suggested that these proteases and inhibitor genes possibly functioned in early-stress responses in the drought-resistant varieties (Abdel-Ghany et al., 2020).

Bhogireddy et al. (2020) also conducted a comparative leaf transcriptome analysis of two peanut (Arachis hypogaea L.) genotypes subjected to drought stress. The study reported a higher constitutive expression‎ of several drought-responsive genes including protease inhibitors and senescence-related proteases in the drought-tolerant genotype than the sensitive one. Following water deprivation, both genotypes had increased expression‎ of several proteolysis-related genes, but the change was generally much greater in the drought-susceptible variety. The authors suggested that high constitutive expression‎ of drought-stress genes contributes towards the increased drought resilience of the drought-tolerant peanut genotype (Bhogireddy et al., 2020), and these findings are consistent with those reported in other drought transcriptomics studies of sorghum (reviewed in Ngara et al., 2021).

Protein degradation-related genes were also differentially expressed in maize (Zea mays) plants subjected to limited watering (Zenda et al., 2019). For example, several UPS-related genes were up-regulated in leaves of the drought-tolerant maize genotype following stress. Furthermore, a comparative analysis of transcripts between the drought-stressed samples of the susceptible genotype versus the tolerant genotype revealed an up-regulation of several ubiquitin-related and the really interesting new gene (RING) zinc-finger protein related genes. The authors suggested that protein ubiquitination and proteolytic processes are critical in facilitating protein turnover and homeostasis in plants under drought stress (Zenda et al., 2019). While the above account describes the trends in a few studies, future systematic reviews with meta-analyses of various studies must eval‎uate and synthesise the trends in constitutive and drought-induced expression‎ of these proteolysis-related genes and proteins between multiple plants and genotypes.

2.2.2 Functional validation studies

Indeed, proteomics technologies continue to broaden our understanding of plant molecular changes under drought. Such alterations affect many functional classes of plant genes (Bevan et al., 1998), including protein degradation. However, expression‎ proteomics data still needs to be validated to ascertain the biological functions of the identified stress-responsive proteins (Rabilloud and Lescuyer, 2014; Handler et al., 2018). Undoubtedly, such proteomic to biological inferences are laden with hurdles that have been critically reviewed by Rabilloud and Lescuyer (2014). Nevertheless, different system biology approaches (Kitano, 2002), such as western blotting, quantitative polymerase chain reaction (qPCR), and enzyme activity assays, are often used to validate quantitative proteomics data (Rabilloud and Lescuyer, 2014; Handler et al., 2018). Other functional validation approaches include transgenic plant systems that either overexpress or repress specific genes under altered environmental conditions (Oh et al., 2005; Kondou et al., 2010; Bolle et al., 2011; Rhee and Mutwil, 2014; Watanabe and Hoefgen, 2019). Therefore, here we discuss a few studies that have utilised protease activity assays in cereals and legumes (Table 4), and transgenic overexpression‎ (Table 5) and knockout or knockdown mutant (Table 6) plant systems to elucidate the involvement and functions of proteases and/or protease inhibitors in plants subjected to drought stress. The reviewed transgenic studies are mainly on the model plant Arabidopsis. The study of plant gene function using gain-of-function or loss-of-function mutants, such as overexpression‎ or knockout/down, has been reviewed (Kondou et al., 2010; Bolle et al., 2011). However, the pros and cons of the above-mentioned experimental validation systems are outside the scope of the current review and, thus, will not be discussed. Readers are directed to excellent reviews where such issues are critically discussed (Bajaj et al., 1999; Sharma et al., 2002; Bhatnagar-Mathur et al., 2008; Kondou et al., 2010; Bolle et al., 2011; Rabilloud and Lescuyer, 2014; Rhee and Mutwil, 2014; Khan et al., 2016; Handler et al., 2018).

TABLE 4

Table 4 Studies showing differential functional activities of proteases in plants under drought stress.

TABLE 5

Table 5 List of transgenic overexpression‎ studies illustrating the roles of protease and protease inhibitors in plants under drought stress.

TABLE 6

Table 6 List of transgenic knockout/down studies illustrating the roles of protease and protease inhibitors in plants under drought stress.

2.2.2.1 Enzyme activity assays

Using a wide range of substrates and protease inhibitors, Hieng et al. (2004) eval‎uated the activities of different proteolytic enzymes in leaf extracts of common bean (Phaseolus vulgaris) cultivars subjected to water limitation. The cultivars used, Tiber, Zorin and Starozagorski exhibited various degrees of sensitivity to drought stress, with Tiber being more drought-tolerant. The results showed a general decrease in leaf protein content in all three cultivars, possibly due to increased proteolytic activity. However, the drought-tolerant cultivar exhibited the least protein content decrease and the least amount of proteolytic activity under severe drought stress. Activity assays for a 65 kDa serine protease with a pH optimum of 8.5 showed the greatest increase in the drought-susceptible varieties Starozagorski and Zorin, and a decrease in Tiber, the tolerant cultivar under severe stress. In contrast, the activities of two other serine proteases increased in all three cultivars, possibly illustrating the existence of a large group of proteases with different roles under drought stress in common beans. The study also reported increased activity of an aminopeptidase in the most drought-sensitive Starozagorski and a decrease in the drought-tolerant cultivar. Overall, the findings of the study support the notion of the presence of a wide range of proteolytic enzymes with different substrate specificities, subcellular locations and specific inhibitors in plants. The authors also suggested that regulation of serine proteases in the tolerant variety could guard against premature drought-included leaf senescence in common bean (Hieng et al., 2004).

In another study, Simova-Stoilova et al. (2010) investigated the activities of acidic proteases and aminopeptidases in winter wheat genotypes under drought and recovery at biochemical and transcriptional levels. The cultivars used were Katya (drought-resistant), Pobeda (cold-resistant) and Sadovo (disease-resistant), with Pobeda and Sadovo being more drought-sensitive than Katya. The results showed significant variations in leaf total protein content of control plants of all three cultivars. Furthermore, upon exposure to water limitation, the protein content of the drought-susceptible Pobeda and Sadovo varieties decreased by almost 50% relative to the respective controls but was restored after re-watering. On the other hand, minimal protein loss was observed in the drought-resistant cultivar under drought stress. The results also showed increased endopeptidase activity for all three cultivars under severe drought stress, with both susceptible varieties exhibiting the highest activity. The high basal level of azocaseinolytic activity observed in the drought-resistant variety, but the least drought-induced effect is noteworthy. As suggested in sorghum transcriptomics studies under water limitation (Fracasso et al., 2016; Azzouz-Olden et al., 2020) high constitutive expression‎ of genes in drought-tolerant genotypes could contribute towards superior traits to dehydration stress. Overall, the authors suggested that low levels of proteolytic activity observed in the drought-resistant Katya cultivar could be regarded as a marker for drought tolerance (Simova-Stoilova et al., 2010).

Likewise, Cruz de Carvalho et al. (2001) investigated the leaf endoprotease activity of common bean and cowpea (Vigna unguiculata) cultivars in response to drought stress. The cowpea cultivars EPACE-1 and TI83D were reportedly more drought-tolerant than the common bean cultivars Carioca and IPA, with an overall tolerance trend of EPACE-1 > TI83D > IPA > Carioca. The results showed differences in endoproteolytic activity across all four cultivars, being higher in the most drought-susceptible common bean cultivars than in cowpea. For example, under mild-drought conditions, the total endoproteolytic activity increased by 235%, 119%, 95% and 58% for cultivar Carioca, IPA, TI83D and EPACE-1, respectively. The study also investigated the involvement of cysteine, aspartic, serine, and metalloproteases in the drought response of the most drought-sensitive Carioca cultivar using class-specific protease inhibitors. The results suggested limited involvement of metalloproteases in the common bean cultivar under mild stress, but some enzyme activity of cysteine and serine proteases. Furthermore, pepstatin A inhibited about 25% of the total proteolytic activity under mild-drought conditions, thus indicating the presence of drought-responsive aspartic proteases in the bean leaf extract. Subsequent analysis of aspartic protease activity in all four cultivars using a specific peptide substrate revealed that water deficit induced aspartic protease activity in cowpea and bean leaf tissues. However, this enzyme activity increased with increasing levels of sensitivity to drought. The authors suggested that this increase in aspartic protease enzyme activity in the drought-sensitive bean cultivars could be a drought-adaptive response for remobilizing nitrogen to other plant parts under stressful environments (Cruz de Carvalho et al., 2001).

2.2.2.2 Transgenic plant biology2.2.2.2.1 Using overexpression‎ mutant lines

Drought studies using transgenic plants overexpressing protease and protease inhibitors genes (Table 5) are unravelling the roles of protein degradation under conditions of water deprivation (Huang et al., 2007; Zhang et al., 2008; Chen et al., 2010; Yao et al., 2012; Tiwari et al., 2015; Tan et al., 2017; Roux et al., 2019; Malefo et al., 2020; Sebastián et al., 2020). For example, Yao et al. (2012) conducted an extensive molecular study of an Arabidopsis gene ASPG1 (aspartic protease in guard cell 1) and its biological functions in drought response. The study used wild-type and ASPG1-overexpressing (ASPG1-OE) Arabidopsis plants to investigate tissue-specific gene expression‎ patterns and the roles of ASPG1 in ABA-signalling systems associated with drought. After exposure to a 14-day drought stress treatment followed by re-watering, ASPG1-OE plants exhibited a greater recovery from wilting symptoms and higher survival rates. Gene expression‎ analysis revealed that ASPG1 is drought-inducible in both wild-type and ASPG1-OE plants and is expressed in young seedlings, leaves, stems, flowers, and siliques but not in roots. The gene is also preferentially localized in guard cells where it participates in the ABA-signalling processes, including ROS accumulation triggering stomatal closure. The increased hydrogen peroxide contents are further detoxified by high antioxidant activities of superoxide dismutase and catalase, thus alleviating oxidative stress effects. In addition, increased ABA sensitivity was observed in ASPG1-OE plants under drought stress which could account for increased stomatal closure and a significant reduction in transpiration water loss. The authors suggested that ASPG1 participates in drought response via the ABA-dependent pathway (Yao et al., 2012).

Sebastián et al. (2020) also investigated the participation of an aspartic protease gene (APA1) in drought response using transgenic Arabidopsis plants overexpressing APA1 (OE-APA1), wild-type and apa1 insertional lines under well-watered and mild-water deficit conditions. The study revealed that OE-APA1 lines tolerated drought stress better than the wild-type and apa1 lines. In addition, drought stress did not affect the overall plant phenotype, total chlorophyll content, and principal root length of the OE-APA1 lines relative to the well-watered plants. However, APA1 overexpressing plants exhibited reduced stomatal pore aperture and water loss under mild-drought stress compared to the wild-type and apa1 lines, yet exogenous ABA did not intensify stomatal closure. APA1 was shown to be ABA-responsive, exhibiting a 4-fold increase in expression‎ in well-watered wild-type plants supplemented with exogenous ABA. Furthermore, under well-watered conditions, OE-APA1 plants up-regulated the expression‎ of ABA biosynthetic and signalling genes relative to the wild-type, while the same genes were down-regulation in apa1 insertional lines. As such, the authors suggested that APA1 participates in drought response via the regulation of the ABA-signalling pathway and its location in leaves, vascular tissues, epidermal, and guard cells implicate it in stomatal closure as a water-saving mechanism under conditions of water scarcity (Sebastián et al., 2020).

Likewise, Chen et al. (2010) also investigated the role of sweet potato (Ipomoea batatas) papain-like cysteine protease (SPCP2) in transgenic Arabidopsis plants. Qualitative phenotypic analyses showed that transgenic Arabidopsis plants overexpressing SPCP2 contained a higher number of incompletely developed siliques, and exhibited earlier flowering, reduced fresh weight per seed, and lower germination rates compared to the wild-type controls. It is plausible that the SPCP2 protein degraded silique storage proteins resulting in incomplete development. Furthermore, SPCP2 gene expression‎ was induced during natural leaf senescence, thus suggesting a gene function in senescence. SPCP2 transgenic Arabidopsis plants also showed high tolerance to salt and drought stresses compared to the wild-type, further implicating the gene in osmotic-stress response (Chen et al., 2010).

Bowman-Birk inhibitors (BBI) are compound inhibitors that inhibit both trypsin and chymotrypsin (Habib and Fazili, 2007; Vaseva et al., 2012). Malefo et al. (2020) generated transgenic A. thaliana plants overexpressing a BBI gene from maize and studied its role in drought tolerance. Drought stress was imposed on 4-week-old plants by withholding water for 9 days, and comparisons were made between the wild-type and transgenic plants under well-watered and drought-stress conditions. The results showed that the transgenic plants exhibited less wilting symptoms, greater survival rate, higher RWC, and higher fresh leaf biomass than the wild-type under drought conditions. Furthermore, lipid peroxidation was reduced, while glutathione-S-transferase activity was enhanced in transgenic lines compared to the wild-type exposed to water limitation. The authors suggested that the improved performance of the BBI-overexpressing transgenic plants under drought could be attributed partly to the reduction in drought-induced oxidative stress (Malefo et al., 2020).

In another study, an Oryza sativa chymotrypsin inhibitor-like 1 (OCPI1) gene was transformed into rice plants and characterized at the reproductive stage under drought-stressed field conditions (Huang et al., 2007). The results showed that the positive transgenic plants inhibited endogenous chymotrypsin activity resulting in higher total protein content of leaves and panicles compared to the wild-type under drought conditions. Furthermore, the OCPI1-overexpressing plants had higher grain yield due to higher seed setting rates than the wild-type under similar drought conditions, further supporting the potential usefulness of OCPI1 in crop improvement strategies (Huang et al., 2007). Likewise, the over-expression‎ of an OCPI2 gene in Arabidopsis resulted in enhanced tolerance to salt and PEG- and mannitol-induced osmotic stresses in transgenic plants relative to the wild-type (Tiwari et al., 2015). Transgenic Arabidopsis OCPI2 plants also exhibited greater vegetative and reproductive potential, higher biomass, seed yield, RWC, membrane stability and proline content. The authors suggested that these enhanced characteristics could have contributed towards the better performance of transgenic plants under salt and osmotic stresses (Tiwari et al., 2015).

Apart from serine protease inhibitors discussed above, plants also contain cystatins which inhibit cysteine proteases of papain and legumain families (Grudkowska and Zagdańska, 2004; Martínez et al., 2012). Some phytocystatins are drought-inducible as evidenced by increased gene expression‎ of a triticale cystatin TrcC-8 in triticate leaf and root tissue under dehydration (Chojnacka et al., 2015). Western blotting analysis also showed increased drought-induced accumulation of the protein but its decline upon re-watering in the same tissues. In addition, the recombinant TrcC-8 protein exhibited inhibitory effects on wheat and triticale leaf cysteine proteases under water deficit stress (Chojnacka et al., 2015). Therefore, cystatins may function as regulatory elements of cysteine protease activity under drought stress (Diop et al., 2004; Zhang et al., 2008).

Tan et al. (2017) studied the biological roles of cystatins under drought stress by overexpressing a Malus prunifolia cystatin gene (MpCYS4) in transgenic Arabidopsis and apple (Malus domestica). The study showed that the MpCYS4 protein is localized in the nucleus, plasma membrane and cytoplasm, consistent with its known cellular locations (Kidric et al., 2014; Diaz-Mendoza et al., 2016). A range of drought assays showed that transgenic Arabidopsis plants overexpressing MpCYS4 had improved drought tolerance as exhibited by the plants’ decreased water loss, increased survival rate, increased RWC and greater stomatal closure under drought, relative to the wild-type. Physiological assessments of the transgenic apple plants overexpressing MpCYS4 also exhibited higher RWC, chlorophyll content and stomatal closure, and reduced electrolyte leakage under water deficits compared to the wild-type. Generally, the MpCYS4 gene resulted in higher transcriptional expression‎ of known ABA- and stress-responsive genes in both transgenic Arabidopsis and apple plants, further highlighting the involvement of MpCYS4 in ABA-signalling processes and thus drought tolerance (Tan et al., 2017).

2.2.2.2.2 Using knockout or knockdown mutant lines

Plant gene function can also be studied using knockout or knockdown mutants with total or partial loss of gene function (Bolle et al., 2011). Several studies have utilized such mutants of proteases or protease inhibitors to study their role during plant growth and drought response (Islam et al., 2017; Gomez-sanchez et al., 2019; Mishra et al., 2021; Wang et al., 2022). A summary of these studies is also given in Table 6. For example, Wang et al. (2022) investigated the role of the maize (Z. mays) senescence-associated gene, ZmSAG39 encoding a cysteine protease in maize plants subjected to either darkness or drought stress. The genotypes used were wild-type, transgenic ZmSAG39 overexpression‎ (OE) lines and ZmSAG39 knockout lines. The results indicated that ZmSAG39 gene expression‎ was induced by darkness and drought stress in the leaves of wild-type plants. The study also investigated seed germination rates in wild-type, ZmSAG39 knockout and ZmSAG39-OE plants under 8% and 12% polyethylene glycol (PEG)-6000 treatment. The results showed that the knockout lines had higher germination rates when compared to the wild and ZmSAG39-OE. Furthermore, the ZmSAG39-OE lines exhibited severe leaf senescence under complete darkness for 5 days, with lower chlorophyll content but higher membrane ion leakage rate and malondialdehyde (MDA) content than wild-type and knockout lines.

Similarly, following a 14-day drought-stress treatment imposed by withholding water, the ZmSAG39-OE lines showed greater levels of leaf wilting and senescence than the wild-type. In contrast, the ZmSAG39 knockout lines had an upright and greener phenotype. In addition, the ZmSAG39-OE lines exhibited reduced survival rates and chlorophyll content, higher membrane leakage levels, MDA, ROS, and oxidative stress damage compared to wild-type and knockout lines. Interestingly, the activities of a range of antioxidant enzymes were much greater in the ZmSAG39 knockout lines than in the wild-type and ZmSAG39-OE lines, thus, indicating the enhanced antioxidant capacity of knockout mutants. The expression‎ levels of various stress-responsive genes were also analyzed between the wild-type and transgenic plants. The results showed expression‎ levels of stress-related genes (lipid-transfer protein, caleosin-related family protein, and salt-overly sensitive 1) and chlorophyll synthesis-related genes (chlorophyll a oxygenase) were higher in the drought-stressed knockout lines compared to the wild-type.

On the other hand, senescence-related genes (microRNA 3, cysteine protease, and senescence-associated gene 12) and a chlorophyll degradation-related gene (non-yellow colouring 1) were higher in the drought-stressed ZmSAG39-OE lines compared to the wild-type. These results indicate the negative regulatory effect of ZmSAG39 on the analysed stress-related and chlorophyll-synthesis-related genes under drought stress. However, under the same conditions of water deficits, ZmSAG39 exhibited a positive regulatory effect on genes associated with senescence and chlorophyll degradation. The authors concluded that ZmSAG39 is associated with leaf senescence, and its increased expression‎ under darkness and drought enhances the sensitivity of maize plants to stress. As such, the knocking out of ZmSAG3 from maize enhanced the resistance of maize plants to darkness and drought tolerance and delayed leaf senescence (Wang et al., 2022).

Gomez-sanchez et al. (2019) also investigated the effect of individually knocking down drought-induced cysteine protease genes, HvPap1 and HvPap19, in leaves of barley (Hordeum vulgare) under drought stress. Seven-day-old plants of wild-type, knockdown HvPap1 (KDPap1), and knockdown HvPap19 (KDPap19) were subjected to drought stress by withholding water for 14 days. The wild-type plants were more susceptible to water deprivation, as evidenced by reduced leaf turgor, while both knockdown genotypes remained upright. Furthermore, KDPap1 plants exhibited delayed leaf senescence during their normal growth cycle under well-watered conditions compared to the wild-type. KDPap1 also had a thicker upper epidermis cuticle under normal and drought conditions compared to the wild-type and KDPap19. However, the drought-induced reduction in chlorophyll and carotenoid contents was not significantly different between the three genotypes. Nevertheless, drought-stressed KDPap1 and KDPap19 plants exhibited higher total protein content than wild-type plants, possibly indicating reduced proteolysis in the knockdown mutant lines.

The study also analysed the levels of selected phytohormones in the three genotypes under well-watered and drought-stress conditions (Gomez-sanchez et al., 2019). The results showed differences in the responses of the phytohormones to water deprivation. For example, ABA levels increased upon exposure to drought in wild-type, KDPap1 and KDPap19 lines, possibly underscoring the critical role of this hormone during drought response. However, ABA levels were much higher in both knockdown lines compared to the wild-type. Conversely, 12-oxo-phytodienoic acid (OPDA), a precursor for jasmonic acid (Taiz and Zeiger, 2012), decreased in all three genotypes following water deprivation. This decline could be attributed to the observed increase in jasmonic acid and its bioactive form, jasmonic-isoleucine, in drought-stressed wild-type and KDPap1 plants. However, independent lines of KDPap19 gave inconsistent trends in jasmonic acid and jasmonic-isoleucine levels. On the other hand, salicylic acid remained unchanged under drought stress conditions in wild-type and transgenic plants. Although increases in ABA and jasmonic acid levels are known to regulate stomatal closure and water loss under drought stress conditions (Salvi et al., 2021), the observed stomatal behaviour in the knockdown plants was unusual. The authors, however, concluded that HvPap1 and HvPap19 are drought-inducible genes and may cause differential effects on barley plants subjected to drought (Gomez-sanchez et al., 2019).

Islam et al. (2017) investigated the functions of Kunitz proteinase inhibitors (KPIs) in white clover (Trifolium repens L.) plants subjected to water deficits using several knockdown lines of different Tr-KPIs. The study utilised two distinct water withholding regimes, a Non-PreStress treatment with direct exposure of plants to water deficits and a PreStress treatment in which plants were initially exposed to drought stress and rehydration, followed by a final drought stress treatment. Both Tr-KPI1 and Tr-KPI5 genes were drought-inducible in three untransformed white clover genotypes with varying degrees of drought tolerance. However, significantly higher gene expression‎ levels were observed in the PreStress treatment, and the genes exhibited different induction levels in leaf and root tissues. The study also showed that proline accumulation and gene expression‎ of 9-cis-epoxycarotenoid dioxygenase (NCED1), a key enzyme in the ABA biosynthesis pathway (Taiz and Zeiger, 2012) are drought-inducible in Non-PreStress treated plants but not PreStress treatment. The authors suggested that PreStress water deprivation treatment primes plants for future exposure to water limitation, resulting in reduced proline accumulation and no significant changes in NCED gene expression‎.

To further study the role of Tr-KPI genes in drought response, four knockdown mutant lines were generated, 35S::tr-kpi1, 35S::tr-kpi2, 35S::tr-kpi4 and 35S::tr-kpi5; however, for the sake of brevity in this review, the knockdown lines will be referred to as kpi1, kpi2, kpi4 and kpi5, respectively. Leaf proline levels were much higher in well-watered plants of kpi1 and kpi5 lines and even greater during PreStress treatment than the wild-type plants. However, kpi2 and kpi4 lines exhibited lower proline levels under well-watered conditions compared to the wild-type and were not used further in the study. These results possibly highlight the additive effects of high endogenous proline levels in kpi1 and kpi5 lines in response to PreStress treatment. Both kpi1 and kpi5 plants also exhibited increased drought-induced transcript abundance of ethylene biosynthesis genes, 1-aminocyclopropane-1-carboxylic acid (ACC) synthase 1 (ACS1) and ACC oxidase 1 (ACO1) compared to the wild-type plants. The authors suggested that knockdown lines may experience some level of constitutive stress under well-watered conditions; hence the increased proline accumulation possibly acts as a ROS scavenger. Furthermore, Tr-KPIs may have different tissue-specific expression‎ patterns and target various proteases, resulting in a functionally diverse group of active proteinase inhibitors during plant growth and drought response (Islam et al., 2017).

In an extensive study, Mishra et al. (2021) investigated the effect of an A. thaliana filamentous temperature-sensitive H (FtsH) pseudo-protease (AtFtsHi3) on the growth and drought tolerance of Arabidopsis plants using knockdown mutants. Pseudo-proteases are catalytically inactive proteolytic enzymes with cell regulatory functions (Reynolds and Fischer, 2015). Gene expression‎ analysis of AtFtsHi3 indicated that the transcript was expressed in seedlings, young flowers, roots, leaves, siliques, and stems during the growth of wild-type Arabidopsis plants. The generated knockdown mutant, ftshi3-I (kd), exhibited a 99% reduction in the transcript expression‎ of the target gene AtFtsHi3 relative to the wild-type. In addition, the mutant seedlings and plants had a pale-green phenotype, were much smaller in rosette size, and had reduced root growth, with lower numbers of lateral roots than the wild-type. The study also generated complementation lines by expressing the FtsHi3 gene in the ftshi3-I (kd) background and designated them ftshi3-I (Comp). Further analysis of chloroplasts ultrastructure between the wild-type, ftshi3-I (Comp), and ftshi3-I (kd) showed that the knockdown mutant lines had distorted chloroplasts with thinner membranes, fewer and distorted thylakoid membranes, and fewer starch granules relative to the other wild-type and Comp lines.

Severe drought stress was also imposed on 4-week-old Arabidopsis plants of wild-type, ftshi3-I (Comp), and ftshi3-I (kd) lines by withholding watering for up to 20 days and observing the drought-related phenotypes. The results showed that ftshi3-I (kd) plants were more tolerant to 12-days of drought treatment, while the wild-type and ftshi3-I (Comp) plants were more drought sensitive with wilting and chlorotic phenotypes. Furthermore, 80% of the ftshi3-I(kd) plants recovered after 20 days of drought stress, followed by 14 days of re-watering, while only 20% of the wild-type and ftshi3-I (Comp) recovered under the same conditions. While all three genotypes had relatively similar leaf ABA levels under control conditions, upon exposure to water limitation, the wild-type and ftshi3-I (Comp) plants had higher ABA content in leaves than the knockdown line. The authors suggested that high endogenous levels of ABA in cotyledons and roots under well-watered conditions could better prime the ftshi3-I(kd) plants for drought than the wild-type. Furthermore, the ftshi3-I(kd) plants had reduced stomatal density but larger stomatal size than the other two genotypes. Overall, this study showed that ftshi3-I(kd) plants had improved drought tolerance compared to wild-type and ftshi3-I (Comp) plants (Mishra et al., 2021).

2.3 Overall functions: more than just acts of protein degradation and its regulation

In summary, plants encounter drought-induced osmotic and oxidative stresses, which disrupt cell structure and function (Levitt, 1980b; Ingram and Bartels, 1996; Bray, 1997). In turn, drought signalling events alter the expression‎ patterns of various genes and proteins with diverse functions in drought adaptation, such as proteases and protease inhibitors (Figure 2A). Consequently, damaged, misfolded, or aggregated proteins are either removed through protein degradation activities of proteases (Vierstra, 1996; Vaseva et al., 2012) or stabilized by chaperons and osmoprotectants (Ingram and Bartels, 1996; Bray, 1997) to restore cellular homeostasis. Such proteolytic activities are essential for preventing the accumulation of potentially toxic non-functional proteins and peptides, recycling nitrogen sources and providing free amino acid pools for renewed protein synthesis well-suited for stressful conditions (Vierstra, 1996; van Wijk, 2015; Rowland et al., 2022). The accumulated drought-responsive protease inhibitor proteins primarily regulate the catalytic activities of proteases (Mosolov and Valueva, 2011). Nonetheless, the roles of plant proteases and their inhibitors under drought stress are more than just acts of protein degradation and regulation (Figure 2).

FIGURE 2

Figure 2 Regulation and functions of plant proteases and protease inhibitors extrapolated from various drought-stress studies. (A) A simplified schematic diagram showing that drought stress signalling in plants results in changes in gene and protein expression‎ patterns such as those of proteases and protease inhibitors. (B) Differential regulation of plant proteases and protease inhibitors in drought-susceptible and tolerant plant varieties under drought stress. (C) Physiological effects of transgenic overexpression‎ and repression of proteases and protease inhibitors in plants under drought stress. Red arrows indicate up-regulation or increase, while black arrows indicate down-regulation or decrease. ABA, abscisic acid; Clp, caseinolytic protease; MDA, malondialdehyde; ROS, reactive oxygen species; RWC, relative water content; OE-plants, overexpression‎-plants. (Cruz de Carvalho et al., 2001; Hieng et al., 2004; Huang et al., 2007; Chen et al., 2010; Simova-Stoilova et al., 2010; Ford et al., 2011; Yao et al., 2012; Ashoub et al., 2013; Jedmowski et al., 2014; Faghani et al., 2015; Tiwari et al., 2015; Wang et al., 2015; Cheng et al., 2016; Chmielewska et al., 2016; Wu et al., 2016; Islam et al., 2017; Tan et al., 2017; Gomez-sanchez et al., 2019; Goche et al., 2020; Malefo et al., 2020; Sebastián et al., 2020; Mishra et al., 2021; Wang et al., 2022).

Results from comparative proteomics and enzyme activity assays (Figure 2B) and transgenic biology studies (Figure 2C) provide experimental evidence for the diverse roles of plant proteases and their inhibitors under drought stress (Tables 3-6). For example, while most plants undergo protein damage due to osmotic and/or oxidative stresses, the drought-sensitive genotypes suffer the most, as evidenced by their increased endoproteolytic activities and reduced protein content under drought stress (Cruz de Carvalho et al., 2001; Hieng et al., 2004; Simova-Stoilova et al., 2010). On the contrary, drought-tolerant genotypes protect their proteins from drought-induced damage by increasing the efficiency of their protective machinery, such as antioxidant capacity and chaperon activities, and inhibiting proteolysis (Cheng et al., 2016; Goche et al., 2020). Furthermore, the reviewed transgenic plant biology studies (Tables 5, 6) add new insights into other roles of proteases and protease inhibitors, including maintaining high RWC and water conservation through enhanced stomatal control (Yao et al., 2012; Tiwari et al., 2015; Tan et al., 2017; Malefo et al., 2020; Sebastián et al., 2020). In addition, proteases and their inhibitors participate in ABA-signalling events and the up-regulation of ABA-responsive genes essential for drought adaptation (Yao et al., 2012; Tan et al., 2017; Sebastián et al., 2020). As a result, transgenic plants overexpressing either protease or protease inhibitor genes exhibit enhanced performance, plant growth and yield, survival rate, and resilience to water deficits (Huang et al., 2007; Yao et al., 2012; Tiwari et al., 2015; Tan et al., 2017; Malefo et al., 2020).

Furthermore, gene function studies using loss-of-function technologies such as knockout/down mutants also provide additional experimental evidence on the broader roles of plant proteases and their inhibitors in drought response (Tables 5, 6; Figure 2C). Nevertheless, the experimental challenges related to the unintended effects of gene manipulation technologies on the overall plant phenotype are real. For example, Gomez-sanchez et al. (2019) suggested that the observed unexpected increase in cuticle thickness of KDPap1, an HvPap1 knockdown mutant line, could have been due to pleiotropic effects associated with the knocked-down cysteine protease gene. In addition, knockdown mutants exhibit some residual mRNA expression‎ or protein accumulation of the knocked-down gene to varying extents (Islam et al., 2017; Gomez-sanchez et al., 2019; Mishra et al., 2021), hence the knockdown character (Gomez-sanchez et al., 2019). For example, while Mishra et al. (2021) reported a 99% reduction in the transcript expression‎ of the target AtFtsHi3 gene in the generated knockdown mutant, ftshi3-I (kd), relative to the wild-type, other studies reported much lower reduction rates of the target genes (Islam et al., 2017).

Furthermore, although the knockdown procedure would have specifically intended to reduce the expression‎ level of a target gene, in some cases, untargeted genes are also affected (Islam et al., 2017). For example, Islam et al. (2017) produced different knockdown mutants of specific Tr-KPIs in their study. However, the RNA interference procedure also affected the expression‎ of other untargeted members of the Tr-KPI gene family. Consequently, eval‎uating resultant plant mutant phenotypes after exposure to drought stress becomes challenging. Therefore, despite the steady progress in gene function investigations using transgenic mutant plants, more studies with multiple transgenic lines are required to ascertain the bona fide physiological effects and roles of target protease and protease inhibitor genes.

As discussed earlier, phytohormones may interact during plant growth and development, as well as in response to drought stress, ultimately contributing towards the overall plant phenotype (Clarke and Durley, 1981; Hale and Orcutt, 1987; Ullah et al., 2018; Jogawat et al., 2021; Salvi et al., 2021; Iqbal et al., 2022). In some cases, the proteolytic machinery regulates phytohormone activity and signalling (Ullah et al., 2018; Salvi et al., 2021; Iqbal et al., 2022). This is evidenced by the results of the reviewed transgenic studies (Tables 5, 6) involving genes of aspartic proteases (Yao et al., 2012; Sebastián et al., 2020), a cystatin (Tan et al., 2017), senescence-associated gene (Wang et al., 2022), cathepsin-like proteases (Gomez-sanchez et al., 2019), Kunitz proteinase inhibitors (Islam et al., 2017) and a pseudo-protease FtsHi3 (Mishra et al., 2021).

For example, gibberellins regulate gene expression‎ through DELLA proteins, which are repressors of plant growth and development (Achard and Genschik, 2009). Under normal growth conditions, gibberellins facilitate optimal plant growth and development by mediating the 26S proteasome degradation of DELLA proteins. Conversely, when exposed to drought conditions, gibberellins levels decrease, and DELLA proteins accumulate, resulting in plant growth inhibition and the development of dwarf plant phenotypes (Achard and Genschik, 2009; Taiz and Zeiger, 2012; Colebrook et al., 2014; Salvi et al., 2021). Similarly, the 26S proteasome degradation pathway is involved in the jasmonic acid signalling pathway (Taiz and Zeiger, 2012). During drought stress, the biosynthesis and accumulation of jasmonic acid increases and jasmonate ZIM-domain (JAZ) proteins are degraded by the 26S proteasome, thus allowing transcription of downstream stress-related genes. JAZ proteins are repressors of DNA-binding transcriptional factors of the jasmonate hormonal signalling system (Taiz and Zeiger, 2012; Wager and Browse, 2012). Therefore, their presence under normal growth conditions regulates the transcriptional control of jasmonic acid-related gene expression‎ (Wager and Browse, 2012). Thus, the 26S proteasome machinery regulates phytohormone signalling systems in plants.

Other studies have reported that exogenous application of ABA increases chymotrypsin inhibitory activity in barley leaves, while jasmonic acid treatment enhances trypsin inhibitor activity (Casaretto et al., 2004). These observations highlight the role of phytohormones in regulating proteolytic activities and vice versa during normal plant growth and drought response. Indeed, the current review of comparative proteomics, enzyme activity assays and transgenic studies expands our understanding of the broader functions of proteases and protease inhibitors in drought response. Some of the inferred physiological roles of protein degradation activities and their regulatory processes discussed above are summarised in Figure 3.

FIGURE 3

Figure 3 Physiological functions of selected proteases and protease inhibitors in drought-stressed plants inferred from transgenic studies. ABA, Abscisic acid; RWC, relative water content. (Chen et al., 2010; Yao et al., 2012; Tan et al., 2017; Malefo et al., 2020; Sebastián et al., 2020; Mishra et al., 2021).

3 Concluding remarks and perspectives

Protein degradation and its regulatory processes in plants are crucial for maintaining cellular homeostasis under water deficits; otherwise, damaged proteins would accumulate and disrupt cell structure and metabolism. However, the drought-induced differential expression‎ of proteases and protease inhibitor proteins in contrasting genotypes are also unravelling the complex networking events during drought responses in plants. For example, increased proteolysis and reduced total protein in drought-sensitive genotypes could indicate massive protein damage events and less robust protective machinery against drought and its secondary effects, which result in sub-optimal metabolic and growth processes. In contrast, the more drought-tolerant genotypes exhibit reduced levels of proteolysis due to fewer proteases and more protease inhibitors. Together with enhanced protective machinery and improved targeted stress-induced gene transcriptional and translational activities, these factors collectively contribute towards greater performance and survival rates under water-limitation conditions.

Nevertheless, the types and functions of proteases and protease inhibitors are quite diverse, and these proteins could be involved in many more physiological roles, such as stomatal control and ABA, jasmonic acid, and ethylene signalling. Furthermore, with this wide structural and functional diversity, their differential expression‎ patterns between drought-tolerant and susceptible genotypes possibly make proteases and protease inhibitors important groups of proteins with potential use in crop improvement strategies. For instance, could the reduced level of protein degradation in drought-tolerant genotypes be used as a marker for drought tolerance and vice versa? Nevertheless, more functional validation studies are required to unravel additional roles of protein degradation events in plants under drought stress as we continue to find ways of generating crops with enhanced resilience to hot and drier climates. Furthermore, meta-analyses of protease and protease inhibitor gene and/or protein expression‎ datasets between contrasting plant species and genotypes are required to unravel the expansive role of protein degradation in plant drought response.

Author contributions

Conceptualisation, RN. Writing-manuscript preparation, SJM, RN. Writing-review and editing, RN. Funding acquisition, supervision, RN. Both authors contributed to the article and approved the submitted version.

Funding

This work is based on the research supported by the National Research Foundation of South Africa Grant Numbers: 113966 and 129884 awarded to RN. The Frontiers Fee Support and the University of the Free State (UFS) library Open Access Publications Fund all contributed towards the article processing cost.

Acknowledgments

We thank the National Research Foundation of South Africa for the research grant awards to RN and postgraduate student scholarship for SJM. We also thank the Frontiers Fee Support and the UFS Library Open Access Publications Fund for their contributions towards the article processing cost.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be eval‎uated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

Abbott, A. (1999). A post-genomic challenge: learning to read patterns of protein synthesis. Nature 402, 715–720. doi: 10.1038/45350

PubMed Abstract | CrossRef Full Text | Google Scholar

Abdallah, C., Dumas-gaudot, E., Renaut, J., Sergeant, K. (2012). Gel-based and gel-free quantitative proteomics approaches at a glance. Int. J. Plant Genomics 2012, 1–17. doi: 10.1155/2012/494572

CrossRef Full Text | Google Scholar

Abdel-Ghany, S. E., Ullah, F., Ben-Hur, A., Reddy, A. S. N. (2020). Transcriptome analysis of drought-resistant and drought-sensitive sorghum (Sorghum bicolor) genotypes in response to PEG-induced drought stress. Int. J. Mol. Sci. 21, 772. doi: 10.3390/ijms21030772

PubMed Abstract | CrossRef Full Text | Google Scholar

Achard, P., Genschik, P. (2009). Releasing the brakes of plant growth: how GAs shutdown DELLA proteins. J. Exp. Bot. 60, 1085–1092. doi: 10.1093/jxb/ern301

PubMed Abstract | CrossRef Full Text | Google Scholar

Ali, M. S., Baek, K. H. (2020). Protective roles of cytosolic and plastidal proteasomes on abiotic stress and pathogen invasion. Plants 9, 832. doi: 10.3390/plants9070832

PubMed Abstract | CrossRef Full Text | Google Scholar

Anderson, L., Seilhame, J. (1997). A comparison of selected mRNA and protein abundances in human liver. Electrophoresis 18, 533–537. doi: 10.1002/elps.1150180333

PubMed Abstract | CrossRef Full Text | Google Scholar

Anjum, S. A., Xie, X.-Y., Wang, L.-C., Saleem, M. F., Man, C., Lei, W. (2011). Morphological, physiological and biochemical responses of plants to drought stress. Afr. J. Agric. Res. 6, 2026–2032. doi: 10.5897/AJAR10.027

CrossRef Full Text | Google Scholar

Aranjuelo, I., Molero, G., Erice, G., Avice, J. C., Nogués, S. (2011). Plant physiology and proteomics reveals the leaf response to drought in alfalfa (Medicago sativa L.). J. Exp. Bot. 62, 111–123. doi: 10.1093/JXB/ERQ249

PubMed Abstract | CrossRef Full Text | Google Scholar

Arora, L., Narula, A. (2017). Gene editing and crop improvement using CRISPR-cas9 system. Front. Plant Sci. 8. 1932. doi: 10.3389/fpls.2017.01932

PubMed Abstract | CrossRef Full Text | Google Scholar

Ashoub, A., Beckhaus, T., Berberich, T., Karas, M., Bruggemann, W. (2013). Comparative analysis of barley leaf proteome as affected by drought stress. Planta 237, 771–781. doi: 10.1007/s00425-012-1798-4

PubMed Abstract | CrossRef Full Text | Google Scholar

Azzouz-Olden, F., Hunt, A. G., Dinkins, R. (2020). Transcriptome analysis of drought-tolerant sorghum genotype SC56 in response to water stress reveals an oxidative stress defense strategy. Mol. Biol. Rep. 47, 3291–3303. doi: 10.1007/s11033-020-05396-5

PubMed Abstract | CrossRef Full Text | Google Scholar

Bajaj, S., Targolli, J., Liu, L.-F., David Ho, T.-H., Wu, R. (1999). Transgenic approaches to increase dehydration-stress tolerance in plants. Mol. Breed. 5, 493–503. doi: 10.1023/A:1009660413133

CrossRef Full Text | Google Scholar

Baldoni, E. (2022). Improving drought tolerance: Can comparative transcriptomics support strategic rice breeding? Plant Stress 3, 100058. doi: 10.1016/J.STRESS.2022.100058

CrossRef Full Text | Google Scholar