Mitochondrial HSC70-1 Regulates Polar Auxin Transport through ROS Homeostasis in Arabidopsis Roots

Arabidopsis mitochondrial-localized heat shock protein 70-1 (mtHSC70-1) modulates vegetative growth by assisting mitochondrial complex IV assembly and maintaining reactive oxygen species (ROS) homeostasis. In addition, mtHSC70-1 affects embryo development, and this effect is mediated by auxin. However, whether mtHSC70-1 regulates vegetative growth through auxin and knowledge of the link between ROS homeostasis and auxin distribution remain unclear. Here, we found that mtHSC70-1 knockout seedlings (mthsc70-1a) displayed shortened roots, decreased fresh root weight and lateral root number, increased root width and abnormal root morphology. The introduction of the mtHSC70-1 gene into mthsc70-1a restored the growth and development of roots to the level of the wild type. However, sugar and auxin supplementation could not help the mutant roots restore to normal. Moreover, mthsc70-1a seedlings showed a decrease in meristem length and activity, auxin transport carrier (PINs and AUX1) and auxin abundances in root tips. The application of exogenous reducing agents upregulated the levels of PINs in the mutant roots. The introduction of antioxidant enzyme genes (MSD1 or CAT1) into the mthsc70-1a mutant rescued the PIN and local auxin abundances and root growth and development. Taken together, our data suggest that mtHSC70-1 regulates polar auxin transport through ROS homeostasis in Arabidopsis roots.


Introduction
Reactive oxygen species (ROS), including superoxide anions (O 2 − ) and hydrogen peroxide (H 2 O 2 ), have received increasing attention as novel signal molecules that are involved in growth, differentiation, cell death and response to environmental stresses [1][2][3]. At present, it is clearly realized that there must be coordinated functioning of the signaling networks that govern ROS responses [4]. Under a physiological state, ROS levels are in dynamic balance, and ROS homeostasis is achieved through its production and scavenging [3,5,6]. Plants possess many ROS-scavenging enzymes, such as ascorbate peroxidases (APXs), catalases (CATs), peroxidases (POXs), superoxide dismutases (SODs) and some small molecular antioxidants, such as ascorbate (ASC, vitamin C), tocopherol (vitamin E), beta-carotene and glutathione (GSH) [6]. However, persistently high ROS levels that exceed the antioxidant capacities of cells are toxic and can cause oxidative damage [5,7]. It is well established that supraphysiological concentrations of ROS react non-specifically with proteins, lipids, nucleic acids and carbohydrates and generate other reactive species with potentially toxic consequences [8,9]. Moreover, high levels of ROS inhibit cell elongation, likely by stiffening cell walls [10]. Another possibility is that

Plant Materials and Growth Conditions
Arabidopsis thaliana (ecotype Columbia-0) seeds were sterilized with 75% ethanol and sown in 0.5 × Murashige-Skoog (MS) medium [50] including 1.0% (w/v) sucrose and 0.8% (w/v) agar. The sterilized seeds were kept at 4 • C for 3 d and then cultured in a growth chamber at 22 • C with a 16 h/8 h photoperiod and a light intensity of approximately 100 µmol m −2 s −1 . After two weeks of growth on MS medium, the seedlings were transplanted into the soil for cultivation in a greenhouse under the same conditions. The cultivated plants were watered once a week.
Genotypes of the transgenic lines used in this study are indicated in Table S2.

Methods of EdU Staining, FM4-64 Staining and GUS Staining
Ethynyl-29-deoxyuridine (EdU) staining was conducted as described by Xiong et al. [51] using an EdU detection cocktail (Life Technologies, Carlsbad, CA, USA). In brief, the roots of 4 days old seedlings grown at 22 • C were incubated in 1 µM EdU for 30 min and then fixed in phosphate-buffered saline (PBS, pH 7.2) including 4 % (w/v) formaldehyde and 0.1% Triton X-100 for 30 min, which was followed by 3 washes with PBS. The fixer was incubated in an EdU detection cocktail for 30 min in the dark and then washed 3 times with PBS.
FM4-64 staining was conducted as described by Gao et al. [52]. In brief, the roots of 4 days old seedlings were incubated in a staining solution containing 5 µM of the plasma membrane marker FM4-64 (Life Technologies, Carlsbad, CA, USA) for 15 min in the dark, which was followed by washing with PBS (pH 7.2).

Observation or Quantification of the Fluorescence Intensity or Distribution in the Stained or Transgenic Roots
The stained or transgenic samples were observed using a laser scanning confocal microscopy (LSCM) (FV3000, Olympus, Tokyo, Japan) with excitation/emission wavelengths of 561/572 nm for FM4-64, 488/510 nm for GFP and EdU and 510/527 nm for YFP, and photographed in their largest z planes. The same parameter settings were used for all genotypic samples in the same experiment. For GUS activity, the stained samples were observed using a stereomicroscope (SZX2-ILLT, Olympus, Tokyo, Japan). Fluorescence in the area of the root shown in the figure was visualized with ImageJ 1.42q software (National Institutes of Health, Bethesda, MD, USA).

Real-Time Quantitative RT-PCR
Ten days old A. thaliana seedlings grown at 22 • C and TriPure Reagent (Aidlab Biotechnologies Co., Ltd., Beijing, China) were used for the isolation of total RNA. Real-time quantitative RT-PCR (Q-PCR) was conducted following the method of Zhang et al. [53]. Primer Express software (Applied Biosystems, Carlsbad, CA, USA) was used to design Q-PCR primers (Table S1). PCR was conducted using an ABI Prism 7000 Sequence Detection System (Applied Biosystems, Carlsbad, CA, USA). The expression levels of genes tested in the WT or control plants were set to 1. ACTIN1 gene was used as the internal control.
2.6. Statistical Analysis STATISTICA 6.0 software (StatSoft, Inc., Tulsa, OK, USA) was used for statistical analyses. Difference significance was tested at p < 0.05 by ANOVA with Tukey's HSD test or Student's t-test.

Role of Glucose in Improving the Growth and Development of mtHSC70-1 Mutant Roots
A. thaliana mtHSC70-1 mutants exhibit higher total respiration, AOX respiratory pathway and ATP level, lower COX-dependent respiratory pathway and shorter roots compared to wild-type plants [47]. Glucose is the original substrate of the glycolysis pathway related to mitochondrial respiration and energy metabolism, so it can promote respiration and energy production. To confirm that the defects of mtHSC70-1 mutant roots are not due to energy deficiency, we needed to know whether sugar supplementation may rescue the growth and development of the mutant roots. To clarify this topic, we compared the phenotypes of roots of WT, mtHSC70-1 mutant (mthsc70-1a, Figure 1A) and two complementary mthsc70-1a lines (R2 and R3) grown on 0.5 × MS medium supplemented with different concentrations of glucose. In MS medium containing 1% glucose, the primary root of mthsc70-1a was 78.5% shorter than that of wild-type seedlings, with a 46% reduction in fresh weight and fewer lateral roots ( Figure 1B-E). Moreover, 59.6% of mthsc70-1a roots showed abnormal morphology, such as longer root hairs near the root tips, increased root width (>140 µm), shorter meristem length and larger and disordered cell arrangement in the elongation zone ( Figure 1F; Figure S1). Two complementary mthsc70-1a lines showed phenotypes similar to those of wild-type roots ( Figure 1B-F), indicating the roles of mtHSC70-1 in regulating root growth and development. Compared with 1% glucose, 3% and 5% glucose reduced primary root length, fresh root weight and lateral root number to some extent in all analyzed genotypic seedlings; however, they did not recover the root length, fresh weight and lateral root number of mthsc70-1a to WT level ( Figure 1B-E). In contrast, 3% glucose reduced the percentage of roots with abnormal morphology (24.5%); 5% glucose completely inhibited the emergence of roots with abnormal morphology ( Figure 1F). These results indicate that sugar supplementation could improve the morphogenesis of the mutant roots but had no positive effect on the root growth of the mutants. mary root of mthsc70-1a was 78.5% shorter than that of wild-type seedlings, with a 46% reduction in fresh weight and fewer lateral roots ( Figure 1B-E). Moreover, 59.6% of mthsc70-1a roots showed abnormal morphology, such as longer root hairs near the root tips, increased root width (>140 μm), shorter meristem length and larger and disordered cell arrangement in the elongation zone ( Figure 1F; Figure S1). Two complementary mthsc70-1a lines showed phenotypes similar to those of wild-type roots ( Figure 1B-F), indicating the roles of mtHSC70-1 in regulating root growth and development. Compared with 1% glucose, 3% and 5% glucose reduced primary root length, fresh root weight and lateral root number to some extent in all analyzed genotypic seedlings; however, they did not recover the root length, fresh weight and lateral root number of mthsc70-1a to WT level ( Figure 1B-E). In contrast, 3% glucose reduced the percentage of roots with abnormal morphology (24.5%); 5% glucose completely inhibited the emergence of roots with abnormal morphology ( Figure 1F). These results indicate that sugar supplementation could improve the morphogenesis of the mutant roots but had no positive effect on the root growth of the mutants. (C-E) Comparison of root length (C), fresh weight (D) and lateral root number (E) among WT, mthsc70-1a, R2 and R3 seedlings. The asterisks indicate significant differences from WT of the same treatment (t-test, *, p < 0.05; **, p < 0.01). (F) The roots of 10 days old seedlings grown in 0.5 × MS medium supplemented with 1%, 3% or 5% glucose were observed using a laser scanning confocal microscopy (LSCM). Representative images are shown. Scale bar = 40 µm. The roots of the mtHSC70-1 mutants were divided into two types according to their width (≤140 µm or >140 µm) in the transition position of the meristem zone and elongation zone. The percentages below the images represent the ratio of two types of roots. n, the total number of roots analyzed. WT, wild type; mthsc70-1a/b, two mtHSC70-1 T-DNA insertion mutant lines; R2 and R3, two complementary mthsc70-1a lines (mthsc70-1a/mtHSC70-1:mtHSC70-1).

Knockout of the mtHSC70-1 Gene Interferes with Polar Auxin Transport in Roots
Auxin is a major regulator of root system architecture and is known to regulate meristematic activities [11,12]. To clarify whether mtHSC70-1 affects root architecture by the auxin pathway, we compared the differences in meristem zone lengths and mitosis activities among WT, mthsc70-1a and R2 seedlings. Using FM4-64 (a plasma membrane marker dye) staining to highlight the cell boundaries, we found a pronounced decrease in the length of the meristem zone in the mthsc70-1a roots compared to the WT and R2 roots (Figure 2A,B; Figure S1). Using the thymidine analogue EdU for in situ detection of cell cycle S-phase entry, we demonstrated that mtHSC70-1 knockout greatly reduced root meristem activity ( Figure 2C). These results suggest lower cell division rates in the meristems of mthsc70-1a roots and imply that the auxin abundances in mthsc70-1a root tips might be lower than those of WT and R2 root tips. To confirm this, we crossed the mthsc70-1a plants with Col-0 plants harboring DR5:GFP or DR5:GUS (the synthetic auxin reporter) [24] and obtained homozygous mthsc70-1a/DR5:GFP plants or mthsc70-1a/DR5:GUS plants. The intensity of GFP fluorescence or GUS coloration was used to indicate the local auxin abundances in roots. Five days old Col/DR5:GUS seedlings exhibited a strong blue coloration in the quiescent center cells, columella cells and stele cells of the root tips ( Figure 3A,B). However, in 5 days old mthsc70-1a/DR5:GUS seedlings, the blue coloration was restricted to the root/hypocotyl transition zone, and no blue coloration was observed in the root tips ( Figure 3C-E). Five days old Col/DR5:GFP seedlings showed bright GFP fluorescence in the quiescent center cells, columella cells and stele cells ( Figure 3F). However, the GFP fluorescence was markedly weaker in the quiescent center cells and columella cells of mthsc70-1a/DR5:GFP roots ( Figure 3G,H). The results indicate that mtHSC70-1 knockout leads to decreased auxin abundance in root tips.     We wanted to know whether the defects of mtHSC70-1 mutant roots are due to the reduction of auxin synthesis or the blockage of PAT. To clarify this topic, we needed to investigate the effects of auxin supplementation or PAT inhibitor treatment on root growth and auxin distribution. Firstly, we compared the root length of WT and mthsc70-1a grown on 0.5 × MS medium, supplemented with different concentrations of 1-naphthlacetic acid (NAA), a synthetic auxin. The results showed that the treatment with exogenous NAA did not rescue the growth of the mutant roots ( Figure S2), implying that the root growth defects of mthsc70-1a may not be caused by the reduction of auxin synthesis. Subsequently, the effects of auxin transport inhibitor N-1-naphthylphthalamic acid (NPA) on root growth and auxin distribution were investigated. The WT, mthsc70-1a, R2 and R3 seeds were germinated and cultured on 0.5 × MS medium supplemented with or without 4 µM NPA. After 10 d of treatment, the root length and lateral root number of WT, R2 and R3 seedlings were significantly decreased compared with those of untreated seedlings, mimicking those of mthsc70-1a seedlings ( Figure 4A-C), implying that the defects of mthsc70-1a in root elongation and lateral root emergence may be due to the blockage of PAT. However, treatment with NPA had no significant effect on fresh root weight in all detected genotypic seedlings ( Figure 4D), implying a weaker role of PAT in substance accumulation. Further, the effect of NPA on auxin response was investigated using transgenic plants expressing DR5:GFP. After 5 d of NPA treatment, the distribution of GFP fluorescence in root tips was significantly different from those of untreated seedlings for Col/DR5:GFP, mthsc70-1a/DR5:GFP and R2/DR5:GFP plants. After NPA treatment, GFP fluorescence was accumulated in the quiescent center zone and reduced markedly in columella cells and stele cells ( Figure 4E), indicating that the auxin output from the quiescent center is affected. In addition, we noticed that NPA treatment only caused the change of GFP fluorescence distribution in root tips, and did not obviously reduce the fluorescence intensity. However, the knockout of mtHSC70-1 not only changed the GFP fluorescence distribution in root tips, but also significantly reduced the fluorescence intensity. The above results suggest that mtHSC70-1 knockout is likely to interfere with PAT in the roots.

Role of Exogenous Antioxidants in Restoring Polar Auxin Transport in mtHSC70-1 Mutant Roots
Mitochondria is a major site for redox reactions. To understand the contribution of the cellular redox state on the defected PAT in the mtHSC70-1 mutant, we investigated the effects of exogenous redox reagents on the PIN and auxin levels in WT and mtHSC70-1 mutants. H 2 O 2 and ASC/GSH were used to increase or decrease the ROS levels in cells. After treatment with 1 mM H 2 O 2 , the WT seedlings displayed a strong downregulation in the expressions of the PIN1, PIN2, PIN3 and PIN7 genes ( Figure 6A); however, the mtHSC70-1 mutant seedlings did not show an obvious difference in PIN expression levels from the untreated seedlings ( Figure S3A). After treatment with 4 mM ASC or 300 nM GSH, the mtHSC70-1 mutant seedlings showed a marked upregulation in the expressions of the PIN1, PIN2, PIN3 and PIN7 genes ( Figure 6B); however, the WT seedlings did not show marked difference in PIN expression levels from the untreated seedlings ( Figure S3B). Data obtained from the root tips of Col (or mthsc70-1a) plants harboring PIN:PIN-GFP confirmed the above results. After treatment with 1 mM H 2 O 2 , the Col/PIN1-, 2-, 3or 7-GFP seedlings exhibited a marked decrease in PIN protein abundance in the root tips compared to the untreated seedlings ( Figure 7A,B); however, mthsc70-1a/PIN1-, 2-, 3or 7-GFP seedlings did not show an obvious difference in PIN protein abundance in the root tips from the untreated seedlings ( Figure S4A,B). After treatment with 4 mM ASC or 300 nM GSH, the mthsc70-1a/PIN1-, 2-, 3or 7-GFP seedlings showed an increase in PIN protein abundance in the root tips compared to the untreated seedlings ( Figure 7C,D); however, the Col/PIN1-, 2-, 3or 7-GFP seedlings did not show marked difference in PIN protein abundance in the root tips from the untreated seedlings ( Figure S4C,D). The above results indicated that an imbalance towards a redox state might inhibit the expression of PIN genes and decrease the levels of PIN proteins in seedlings. Then, H 2 O 2 or GSH were applied to Col or mthsc70-1a seedlings harboring DR5:GFP to clarify their effects on auxin response. After the Col/DR5:GFP seedlings were treated with 1 mM H 2 O 2 , the GFP fluorescence intensities in the root tips were not an obvious difference from those before treatment ( Figure S5). After the mthsc70-1a/DR5:GFP seedlings were treated with 300 nM GSH, the GFP fluorescence intensities in the root tips were also not obvious difference from those before treatment ( Figure S5), which indicated that exogenous redox reagents had no evident effect on local auxin abundances in root tips.   22 °C for 5 d showed pronounced decreases in fluorescence intensity compared to the corresponding Col/PIN-GFP (or Col/AUX1-YFP) plants ( Figure 5). These results suggest that mtHSC70-1 plays a positive role in regulating the protein levels of PIN1, PIN2, PIN3, PIN7 or AUX1 in root tips and that mtHSC70-1 probably affects root development by regulating PAT.

Discussion
Mitochondria are the major sites of energy production and link redox metabol ATP synthesis through the respiratory electron transport chain, which rely on the tion of glycolysis product. Glucose plays an important role in energy metabolism

Discussion
Mitochondria are the major sites of energy production and link redox metabolism to ATP synthesis through the respiratory electron transport chain, which rely on the oxidation of glycolysis product. Glucose plays an important role in energy metabolism as a substrate of glycolysis reaction. Data presented here indicate that mtHSC70-1 mutation leads to root growth and development defects, and glucose supplementation restores some phenotypes of the mtHSC70-1 mutant, such as root width, root cell arrangement and root hairs, but has no effect on root length, fresh weight and lateral root number, suggesting the role of glucose in organ architecture. In previous studies, we unexpectedly found an increase in the O 2 consumption rate and ATP level in the mtHSC70-1 mutant [47]. Analysis of respiratory pathways in the mutant indicates an inhibition of the cytochrome c oxidase (COX) pathway and the activation of the alternative respiratory (AOX) pathway [47]. The higher O 2 consumption rate caused by the increased AOX pathway in the mtHSC70-1 mutant can compensate for the lower efficiency of ATP generation when electrons are channels through AOX [47]. Thus, the root growth defects in the mtHSC70-1 mutant may not be explained by the lack of energy, which might be the reason for the observed weak restoring effect of glucose on root growth phenotypes.
Mitochondria are also major sites of ROS production. In the process of electron transfer in the respiratory chain, a small number of electrons do not complete the whole transfer chain but leak and directly transfer to O 2 , generating ROS [54]. Complex I, Complex II and complex III are generally regarded as the main and direct sources of ROS [54][55][56][57][58][59]. It is reported that complex IV-deficient mutants can create electron leakage at the site of overreduced complex III [60]. Yeast cells lacking Rcf1 (a structural subunit of yeast COX) and Arabidopsis mitochondria-localized CYTOCHROME C OXIDASE DEFICIENT1 (COD1) mutants display decreased COX/complex IV activity and increased ROS production [60,61]. Our previous study indicates that mtHSC70-1 mutants exhibit reduced COX/complex IV activity and the accumulation of mitochondrial ROS, which might be responsible for shortened stems and roots and decreased rosette sizes [47]. Data presented here indicate that the introduction of ROS scavenging enzymes into mtHSC70-1 mutant rescues the growth and development of the roots, suggesting an inhibition role of high levels of ROS in plant growth and development.
It is reported that mtHSC70-1 regulates embryo development by auxin pathway in A. thaliana [49], and mtHSC70-1 knockout inhibits the growth of A. thaliana seedlings [47]. Whether mtHSC70-1 regulates vegetative growth through auxin is unclear. Data presented here demonstrate that mtHSC70-1 facilitates root growth and development through regulating PAT. Li et al. [49] reported that the loss of mtHSC70-1 function induces mitochondrial retrograde signaling, which results in abnormal auxin gradients and defective embryo development. However, the mitochondrial retrograde regulatory factor is still unknown. Our previous work demonstrated that the loss of mtHSC70-1 function leads to the accumulation of mitochondrial ROS [47]. Data presented here demonstrate that the removal of excess mitochondrial ROS in the mutant could rescue defective PAT. How mitochondrial ROS affect auxin transport and distribution in the cytosol and whether they play a role as the main mitochondrial retrograde regulator remain to be studied. It is reported that isolated mitochondria can release H 2 O 2 from the matrix to the extra-mitochondrial space [8]. However, the rate at which this occurs physiologically in intact cells remains unclear.
It is reported that the expression of CAT1 and MSD1 is rapidly induced after treatment with H 2 O 2 , and catalase and SOD activity increase obviously after treatments with oxidative stressors in Arabidopsis [62][63][64].Ślesak et al. [65] reported that three PAD downregulated transgenic lines display an increased H 2 O 2 level and higher activities of MSD1 and CAT in the leaves and shoots compared to the wild-type plants. Previously, we found that the expression levels of MSD1 and CAT1 genes in the mthsc70-1a seedlings are obviously high than those in WT, and ROS levels are also markedly high in the mutant [47], indicating that endogenous MSD1 and CAT1 are induced by the ROS, and endogenous MSD1 and CAT1 are not enough to eliminate the excess ROS in the mutant. Further, treatments with antioxidants ASC and GSH do not restore the growth of mthsc70-1a and mthsc70-1b roots [47]. Data presented here indicate that the application of exogenous antioxidants has no obvious effect on the local auxin abundances in root tips. This may be because the applications of exogenous antioxidants lack spatiotemporal specificity. Then, we investigated the effects of introducing ROS scavenging enzymes into the mthsc70-1a mutant on the PIN and auxin abundances and root growth. There are three catalase genes (CAT1, CAT2 and CAT3) and one manganese superoxide dismutase gene (MnSOD or MSD1) in the Arabidopsis genome [66,67]. Subcellular localization studies have suggested that MSD1 is localized in mitochondria [68,69]. Bioinformatics analysis shows that CAT1 is mainly localized in the peroxisome, mitochondria and cytosol (http://bar.utoronto.ca/cell_efp/ cgi-bin/cell_efp.cgi?primaryGene=At1g20630) (accessed on 3 December 2017). To ensure the effective removal of excess ROS produced by mtHSC70-1 deficiency, we introduced the mitochondria-targeted MSD1 or CAT1 gene, which are driven by the mtHSC70-1 promoter, into the mthsc70-1a mutant. The introduced MSD1 or CAT1 gene is expressed to some extent in several representative transgenic lines, and total MSD1 or CAT1 levels in the transgenic lines recover to a level similar to that in WT [47], which implies that ROS levels in these transgenic lines are decreased by the introduced MSD1 or CAT1. Data presented here show that the introduction of MSD1 or CAT1 into mtHSC70-1 mutants rescued the defects in PIN expression, auxin distribution and root growth and development. These results indicate that spatially and temporally distributions of ROS are essential for PAT and provide direct evidence for the role of ROS homeostasis in maintaining PAT.

Conclusions
Our previous results [47,48] and the data presented here support a working model in which mtHSC70-1 acts to assist respiratory chain complex IV assembly/activity and regulate ROS homeostasis, which maintains normal PAT and regulates plant growth and development ( Figure 11).
Antioxidants 2022, 11,2035 18 of 21 because the applications of exogenous antioxidants lack spatiotemporal specificity. Then, we investigated the effects of introducing ROS scavenging enzymes into the mthsc70-1a mutant on the PIN and auxin abundances and root growth. There are three catalase genes (CAT1, CAT2 and CAT3)and one manganese superoxide dismutase gene (MnSOD or MSD1) in the Arabidopsis genome [66,67]. Subcellular localization studies have suggested that MSD1 is localized in mitochondria [68,69]. Bioinformatics analysis shows that CAT1 is mainly localized in the peroxisome, mitochondria and cytosol (http://bar.utoronto.ca/cell_efp/cgi-bin/cell_efp.cgi?primaryGene=At1g20630) (accessed on 3 December 2017). To ensure the effective removal of excess ROS produced by mtHSC70-1 deficiency, we introduced the mitochondria-targeted MSD1 or CAT1 gene, which are driven by the mtHSC70-1 promoter, into the mthsc70-1a mutant. The introduced MSD1 or CAT1 gene is expressed to some extent in several representative transgenic lines, and total MSD1 or CAT1 levels in the transgenic lines recover to a level similar to that in WT [47], which implies that ROS levels in these transgenic lines are decreased by the introduced MSD1 or CAT1. Data presented here show that the introduction of MSD1 or CAT1 into mtHSC70-1 mutants rescued the defects in PIN expression, auxin distribution and root growth and development. These results indicate that spatially and temporally distributions of ROS are essential for PAT and provide direct evidence for the role of ROS homeostasis in maintaining PAT.

Conclusions
Our previous results [47,48] and the data presented here support a working model in which mtHSC70-1 acts to assist respiratory chain complex IV assembly/activity and regulate ROS homeostasis, which maintains normal PAT and regulates plant growth and development ( Figure 11). Figure 11. A working model for the function of mtHSC70-1 in the regulation of Arabidopsis growth and development. mtHSC70-1, as a molecular chaperone, assists in complex IV/cytochrome c oxidase (COX) assembly. The loss of mtHSC70-1 function inhibits the assembly/activity of COX, which results in the malfunction of the cytochrome respiratory pathway, which leads to the excessive production of ROS at the site of over-reduced complex III, and the activation of the alternative respiratory pathway. The high levels of ROS in mtHSC70-1 mutants interfere with polar auxin transport, subsequently impacting the growth and development of plants. Inhibited and increased pathways are indicated with black lines and red lines, respectively. The bars show blockage effects. Black dashed line indicates an unproven route. Complex IV and electron transfer disruption is shown with red crosses. CI, CIII and CIV, respiratory chain complexes I, III and IV; AOX, alternative oxidase; UQ, ubiquinone; Cyt c, cytochrome c; SOD, superoxide dismutase; CAT, catalases. Figure 11. A working model for the function of mtHSC70-1 in the regulation of Arabidopsis growth and development. mtHSC70-1, as a molecular chaperone, assists in complex IV/cytochrome c oxidase (COX) assembly. The loss of mtHSC70-1 function inhibits the assembly/activity of COX, which results in the malfunction of the cytochrome respiratory pathway, which leads to the excessive production of ROS at the site of over-reduced complex III, and the activation of the alternative respiratory pathway. The high levels of ROS in mtHSC70-1 mutants interfere with polar auxin transport, subsequently impacting the growth and development of plants. Inhibited and increased pathways are indicated with black lines and red lines, respectively. The bars show blockage effects. Black dashed line indicates an unproven route. Complex IV and electron transfer disruption is shown with red crosses. CI, CIII and CIV, respiratory chain complexes I, III and IV; AOX, alternative oxidase; UQ, ubiquinone; Cyt c, cytochrome c; SOD, superoxide dismutase; CAT, catalases.

Supplementary Materials:
The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/antiox11102035/s1. Figure S1. Comparison of root tip structure among wild type, mthsc70-1a and a complementary mthsc70-1a line. Figure S2. The effects of NAA treatment on primary root elongation in WT and mtHSC70-1 mutant seedlings. Figure S3. Effects of exogenous redox reagents on the expression of PIN genes in mtHSC70-1 mutant and WT seedlings by Q-PCR. Figure S4. Effects of exogenous redox reagents on PIN protein levels in mthsc70-1a/PIN-GFP and Col/PIN-GFP root tips. Figure S5. Effects of exogenous redox reagents on the auxin response. Table S1. List of primers used in this study. Table S2. Genotypes of the transgenic lines used in this study.