*3.2. Cisplatin Treatment Significantly Decreases FBPase and G6Pase Enzyme Activity in Both WT and NHERF1 KO Mice*

Previous studies investigated the effect of cisplatin on gluconeogenesis [2,36,37]; however, the effect of NHERF1 protein loss on gluconeogenesis has not been investigated. FBPase is a critical regulatory enzyme in gluconeogenesis that catalyzes the hydrolysis of fructose-1,6-bisphosphate to fructose-6-phosphate and inorganic phosphate [38]. In order to determine if NHERF1 loss affected gluconeogenic enzyme activity alone or with cisplatin treatment, male 2–4-month-old mice were treated with vehicle or cisplatin to induce AKI and then sacrificed after 72 h. Kidney cortex tissue from these mice were used for the FBPase enzyme kinetic assay as described in the methods section. There were no significant differences between vehicle [(WT: 41.5 nmole/mg protein/min ± 6.5) (KO: 38.7 nmole/mg protein/min ± 4.1)] or cisplatin [(WT: 20.6 nmole/mg protein/min ± 0.4) (KO: 19.7 nmole/mg protein/min ± 1.2)] treated WT and NHERF KO kidneys (Figure 1A). However, cisplatin did decrease FBPase activity in both WT and NHERF1 KO kidneys (*p* = 0.0001) (Figure 1A).

*Activity in Mice* 

NHERF1 KO mice compared to vehicle saline controls.

*3.3. NHERF1 Deficiency or Cisplatin Treatment Does Not Significantly Affect LDH or MDH Enzyme* 

Lactate dehydrogenase (LDH) catalyzes the interconversion of lactate and pyruvate, concomitantly with the interconversion of NADH and NAD+ [40]. When oxygen is absent, LDH converts pyruvate, the final product of glycolysis, to lactate [40]. Thus, LDH activity was measured in vehicle and cisplatin mouse kidneys. NHERF1 loss (*p* = 0.65) or cisplatin treatment (*p* = 0.71) did not significantly affect lactate dehydrogenase activity in these mouse kidneys [(WT vehicle: 0.06 nmole/mg protein/min ± 0.02), (KO vehicle: 0.1 nmole/mg protein/min ± 0.05), (WT cisplatin: 0.1

**Figure 1.** Effect of cisplatin treatment on fructose-1,6-bisphosphatase and glucose-6-phosphase enzyme activity in wild-type (WT) and Na/H exchange regulatory factor 1 (NHERF1) knockout (KO) mouse kidneys. Two to 4-month-old male C57BL/6J WT and NHERF1 KO mice were given cisplatin (20 mg/kg dose intraperitoneally (IP)) or vehicle (saline) and sacrificed after 72 h as described in the Methods section. (**A**) Fructose-1,6-bisphosphatase (FBPase) enzyme activity was determined from the kidney cortex tissue of these mice. Data are means ± SEM (WT vehicle *n* = 3), (KO vehicle *n* = 4), (WT cisplatin *n* = 3), and (KO cisplatin *n* = 5). \*\*\* *p* = 0.001 cisplatin-treated WT and NHERF1 KO mice compared to vehicle saline controls. (**B**) Glucose-6-phosphatase (G6Pase) enzyme activity was determined from the kidney cortex tissue of these mice. Data are means ± SEM (WT vehicle *n* = 3), (KO vehicle *n* = 4), (WT cisplatin *n* = 3), and (KO cisplatin *n* = 5). \*\*\* *p* < 0.001 cisplatin-treated WT and **Figure 1.** Effect of cisplatin treatment on fructose-1,6-bisphosphatase and glucose-6-phosphase enzyme activity in wild-type (WT) and Na/H exchange regulatory factor 1 (NHERF1) knockout (KO) mouse kidneys. Two to 4-month-old male C57BL/6J WT and NHERF1 KO mice were given cisplatin (20 mg/kg dose intraperitoneally (IP)) or vehicle (saline) and sacrificed after 72 h as described in the Methods section. (**A**) Fructose-1,6-bisphosphatase (FBPase) enzyme activity was determined from the kidney cortex tissue of these mice. Data are means ± SEM (WT vehicle *n* = 3), (KO vehicle *n* = 4), (WT cisplatin *n* = 3), and (KO cisplatin *n* = 5). \*\*\* *p* = 0.001 cisplatin-treated WT and NHERF1 KO mice compared to vehicle saline controls. (**B**) Glucose-6-phosphatase (G6Pase) enzyme activity was determined from the kidney cortex tissue of these mice. Data are means ± SEM (WT vehicle *n* = 3), (KO vehicle *n* = 4), (WT cisplatin *n* = 3), and (KO cisplatin *n* = 5). \*\*\* *p* < 0.001 cisplatin-treated WT and NHERF1 KO mice compared to vehicle saline controls.

Malate dehydrogenase (MDH) is an enzyme involved in many metabolic pathways including the citric acid cycle. MDH reversibly catalyzes the oxidation of malate to oxaloacetate with the reduction of NAD+ to NADH [41]. In this study, we measured MDH activity as the conversion of oxaloacetate to malate and oxidation of NADH to NAD+. The effect of NHERF1 loss and/or cisplatin treatment on MDH activity was analyzed. In a similar way to LDH, NHERF1 loss or cisplatin G6Pase is the final step in gluconeogenesis, where it hydrolyzes glucose-6-phosphate to free glucose and a phosphate group [39]. Similarly to FBPase, G6Pase activity was comparable in WT and NHERF1 KO kidneys regardless of treatment [(WT vehicle: 92.3 nmole/mg protein/min ± 5.0), (KO vehicle: 103.0 nmole/mg protein/min ± 11.3), (WT cisplatin: 38.0 nmole/mg protein/min ± 5.1), and (KO cisplatin: 26.4 nmole/mg protein/min ± 3.8)]. Cisplatin led to a significant decrease in enzyme activity in both genotypes (*p* < 0.0001) (Figure 1B).

#### treatment did not significantly affect MDH activity in these mouse kidneys [(WT vehicle: 0.9 nmole/mg protein/min ± 0.06), (KO vehicle: 0.7 nmole/mg protein/min ± 0.02), (WT cisplatin: 0.8 nmole/mg protein/min ± 0.02), and (KO cisplatin: 0.81 nmole/mg protein/min ± 0.06)] (Figure 2B). *3.3. NHERF1 Deficiency or Cisplatin Treatment Does Not Significantly A*ff*ect LDH or MDH Enzyme Activity in Mice*

Lactate dehydrogenase (LDH) catalyzes the interconversion of lactate and pyruvate, concomitantly with the interconversion of NADH and NAD<sup>+</sup> [40]. When oxygen is absent, LDH converts pyruvate, the final product of glycolysis, to lactate [40]. Thus, LDH activity was measured in vehicle and cisplatin mouse kidneys. NHERF1 loss (*p* = 0.65) or cisplatin treatment (*p* = 0.71) did not significantly affect lactate dehydrogenase activity in these mouse kidneys [(WT vehicle: 0.06 nmole/mg protein/min ± 0.02), (KO vehicle: 0.1 nmole/mg protein/min ± 0.05), (WT cisplatin: 0.1 nmole/mg protein/min ± 0.8), and (KO cisplatin: 0.02 nmole/mg protein/min ± 0.006)] (Figure 2A).

**Figure 2.** Lactate dehydrogenase and malate dehydrogenase enzyme activity in WT and NHERF1 KO mouse kidneys. Two to 4-month-old male C57BL/6J WT and NHERF1 KO mice were given cisplatin (20 mg/kg dose IP) or vehicle (saline) and sacrificed after 72 h as described in the Methods section. (**A**) Lactate dehydrogenase (LDH) enzyme activity was determined from kidney cortex tissue of these mice. Data are mean ± SEM (WT vehicle *n* = 3), (KO vehicle *n* = 4), (WT cisplatin *n* = 3), and (NHERF1 KO cisplatin *n* = 5). No significant differences were recorded. (**B**) Malate dehydrogenase (MDH) enzyme activity was determined from the kidney cortex tissue of these mice. Data are mean ± SEM (WT vehicle *n* = 3), (KO vehicle *n* = 4), (WT cisplatin *n* = 3), (NHERF1 KO cisplatin *n* = 5). No significant differences were reported. **Figure 2.** Lactate dehydrogenase and malate dehydrogenase enzyme activity in WT and NHERF1 KO mouse kidneys. Two to 4-month-old male C57BL/6J WT and NHERF1 KO mice were given cisplatin (20 mg/kg dose IP) or vehicle (saline) and sacrificed after 72 h as described in the Methods section. (**A**) Lactate dehydrogenase (LDH) enzyme activity was determined from kidney cortex tissue of these mice. Data are mean ± SEM (WT vehicle *n* = 3), (KO vehicle *n* = 4), (WT cisplatin *n* = 3), and (NHERF1 KO cisplatin *n* = 5). No significant differences were recorded. (**B**) Malate dehydrogenase (MDH) enzyme activity was determined from the kidney cortex tissue of these mice. Data are mean ± SEM (WT vehicle *n* = 3), (KO vehicle *n* = 4), (WT cisplatin *n* = 3), (NHERF1 KO cisplatin *n* = 5). No significant differences were reported.

*3.4. NHERF1 Deficiency Upregulates ME and G6PD Activity*  Malic enzyme (ME) catalyzes the conversion of malic acid to pyruvate and produces NADPH [42]. ME serves as an additional source of NADPH for lipogenesis. In order to understand the effect that NHERF1 loss and/or cisplatin treatment may have on ME activity, kidney cortex tissue from vehicle or cisplatin-treated WT and NHERF1 KO were evaluated. Interestingly, there was a significant genotype effect on ME activity resulting in an increase in activity in NHERF1 KO kidneys (*P* = 0.0065) (Figure 3A). Additionally, a significant interaction was also noted between WT and Malate dehydrogenase (MDH) is an enzyme involved in many metabolic pathways including the citric acid cycle. MDH reversibly catalyzes the oxidation of malate to oxaloacetate with the reduction of NAD<sup>+</sup> to NADH [41]. In this study, we measured MDH activity as the conversion of oxaloacetate to malate and oxidation of NADH to NAD+. The effect of NHERF1 loss and/or cisplatin treatment on MDH activity was analyzed. In a similar way to LDH, NHERF1 loss or cisplatin treatment did not significantly affect MDH activity in these mouse kidneys [(WT vehicle: 0.9 nmole/mg protein/min ± 0.06), (KO vehicle: 0.7 nmole/mg protein/min ± 0.02), (WT cisplatin: 0.8 nmole/mg protein/min ± 0.02), and (KO cisplatin: 0.81 nmole/mg protein/min ± 0.06)] (Figure 2B).

#### NHERF1 KO kidneys after cisplatin treatment (*p* = 0.0005) [(WT vehicle: 0.07 nmole/mg protein/min ± 0.012), (KO vehicle: 0.21 nmole/mg protein/min ± 0.01), (WT cisplatin: 0.15 nmole/mg protein/min *3.4. NHERF1 Deficiency Upregulates ME and G6PD Activity*

± 0.012), and (KO cisplatin: 0.13 nmole/mg protein/min ± 0.02)] (Figure 3A). Cisplatin-induced AKI is known to decrease intermediates of the pentose phosphate pathway [2,7] in mice. Glucose-6-phosphate dehydrogenase (G6PD) is a cytosolic enzyme that participates in the pentose phosphate pathway, resulting in NADPH production [43]. This is accomplished when G6PD reduces NADP+ to NADPH while oxidizing glucose-6-phosphate [43]. G6PD enzyme activity was analyzed in vehicle and cisplatin-treated WT and NHERF1 KO kidney cortex to elucidate if NHERF1 loss and/or cisplatin treatment affected the pentose phosphate pathway. Similarly, to ME, there was a significant genotype effect on G6PD activity, resulting in an increase in activity in NHERF1 KO kidneys (*p* = 0.0033) (Figure 3B). Additionally, a significant interaction was also noted Malic enzyme (ME) catalyzes the conversion of malic acid to pyruvate and produces NADPH [42]. ME serves as an additional source of NADPH for lipogenesis. In order to understand the effect that NHERF1 loss and/or cisplatin treatment may have on ME activity, kidney cortex tissue from vehicle or cisplatin-treated WT and NHERF1 KO were evaluated. Interestingly, there was a significant genotype effect on ME activity resulting in an increase in activity in NHERF1 KO kidneys (*P* = 0.0065) (Figure 3A). Additionally, a significant interaction was also noted between WT and NHERF1 KO kidneys after cisplatin treatment (*p* = 0.0005) [(WT vehicle: 0.07 nmole/mg protein/min ± 0.012), (KO vehicle: 0.21 nmole/mg protein/min ± 0.01), (WT cisplatin: 0.15 nmole/mg protein/min ± 0.012), and (KO cisplatin: 0.13 nmole/mg protein/min ± 0.02)] (Figure 3A).

between WT and NHERF1 KO kidneys after cisplatin treatment (*p* = 0.00029) [(WT vehicle: 0.13 nmole/mg protein/min ± 0.02), (KO vehicle: 0.3 nmole/mg protein/min ± 0.03), (WT cisplatin: 0.3 nmole/mg protein/min ± 0.007), and (KO cisplatin: 0.3 nmole/mg protein/min ± 0.02)] (Figure 3B).

ATP provides energy to drive many cellular processes and is consumed during many metabolic

*3.5. NHERF1 Deficiency Does Not Affect ATP Abundance in Mouse Kidneys* 

*Area* 

nmoles/mg tissue ± 0.5) kidneys (*p* = 0.67) (Figure 4).

[2] the citric acid cycle or oxidative phosphorylation, and [3] beta-oxidation [44]. In order to determine if NHERF1 KO kidneys had differences in ATP content, kidneys were snap-frozen and processed while cold for LC-MS as described in the Methods section. LC-MS analysis revealed there were no significant differences in ATP amount in WT (3.4 nmoles/mg tissue ± 0.5) and NHERF1 KO (3.1

*3.6. NHERF1 Deficiency Does Not Affect Kidney Proximal Tubule Mitochondria Morphology, Number, or* 

The mitochondrial structure is essential for proper function; thus, EM images of WT and NHERF1 KO proximal tubule mitochondria were utilized to evaluate their morphology. These images were of 2–4-month-old male C57BL/6J WT and NHERF1 KO mice whose kidneys were

**Figure 3.** Effect of NHERF1 loss and cisplatin treatment on malic enzyme and glucose-6-phosphate dehydrogenase enzyme activity in WT and NHERF1 KO mouse kidneys. Two to 4-month-old male C57BL/6J WT and NHERF1 KO mice were given cisplatin (20 mg/kg dose IP) or vehicle (saline) and sacrificed after 72 h as described in the Methods section. (**A**) Malic enzyme (ME) activity was determined from the kidney cortex tissue of these mice. Data are mean ± SEM (WT vehicle *n* = 3), (KO vehicle *n* = 4), (WT cisplatin *n* = 3), and (KO cisplatin *n* = 5). \*\* *p* = 0.0065. Vehicle-treated NHERF1 KO mice compared to WT vehicle controls; \*\*\* *p* = 0.0005 interaction of cisplatin-treated NHERF1 KO mice to cisplatin-treated WT mice. (**B**) G6PD enzyme activity was determined from the kidney cortex tissue of these mice. Data are mean ± SEM (WT vehicle *n* = 3), (KO vehicle *n* = 4), (WT cisplatin *n* = 3), and (KO cisplatin *n* = 5). \*\* *p* = 0.0033 vehicle-treated NHERF1 KO mice compared to WT vehicle controls; \*\* *p* = 0.00029 interaction of cisplatin-treated NHERF1 KO mice to cisplatin-treated WT mice. **Figure 3.** Effect of NHERF1 loss and cisplatin treatment on malic enzyme and glucose-6-phosphate dehydrogenase enzyme activity in WT and NHERF1 KO mouse kidneys. Two to 4-month-old male C57BL/6J WT and NHERF1 KO mice were given cisplatin (20 mg/kg dose IP) or vehicle (saline) and sacrificed after 72 h as described in the Methods section. (**A**) Malic enzyme (ME) activity was determined from the kidney cortex tissue of these mice. Data are mean ± SEM (WT vehicle *n* = 3), (KO vehicle *n* = 4), (WT cisplatin *n* = 3), and (KO cisplatin *n* = 5). \*\* *p* = 0.0065. Vehicle-treated NHERF1 KO mice compared to WT vehicle controls; \*\*\* *p* = 0.0005 interaction of cisplatin-treated NHERF1 KO mice to cisplatin-treated WT mice. (**B**) G6PD enzyme activity was determined from the kidney cortex tissue of these mice. Data are mean ± SEM (WT vehicle *n* = 3), (KO vehicle *n* = 4), (WT cisplatin *n* = 3), and (KO cisplatin *n* = 5). \*\* *p* = 0.0033 vehicle-treated NHERF1 KO mice compared to WT vehicle controls; \*\* *p* = 0.00029 interaction of cisplatin-treated NHERF1 KO mice to cisplatin-treated WT mice.

Cisplatin-induced AKI is known to decrease intermediates of the pentose phosphate pathway [2,7] in mice. Glucose-6-phosphate dehydrogenase (G6PD) is a cytosolic enzyme that participates in the pentose phosphate pathway, resulting in NADPH production [43]. This is accomplished when G6PD reduces NADP<sup>+</sup> to NADPH while oxidizing glucose-6-phosphate [43]. G6PD enzyme activity was analyzed in vehicle and cisplatin-treated WT and NHERF1 KO kidney cortex to elucidate if NHERF1 loss and/or cisplatin treatment affected the pentose phosphate pathway. Similarly, to ME, there was a significant genotype effect on G6PD activity, resulting in an increase in activity in NHERF1 KO kidneys (*p* = 0.0033) (Figure 3B). Additionally, a significant interaction was also noted between WT and NHERF1 KO kidneys after cisplatin treatment (*p* = 0.00029) [(WT vehicle: 0.13 nmole/mg protein/min ± 0.02), (KO vehicle: 0.3 nmole/mg protein/min ± 0.03), (WT cisplatin: 0.3 nmole/mg protein/min ± 0.007), and (KO cisplatin: 0.3 nmole/mg protein/min ± 0.02)] (Figure 3B).

### *3.5. NHERF1 Deficiency Does Not A*ff*ect ATP Abundance in Mouse Kidneys*

ATP provides energy to drive many cellular processes and is consumed during many metabolic processes [44]. In eukaryotes, ATP is produced by three different metabolic pathways: [1] glycolysis, [2] the citric acid cycle or oxidative phosphorylation, and [3] beta-oxidation [44]. In order to determine if NHERF1 KO kidneys had differences in ATP content, kidneys were snap-frozen and processed while cold for LC-MS as described in the Methods section. LC-MS analysis revealed there were no significant differences in ATP amount in WT (3.4 nmoles/mg tissue ± 0.5) and NHERF1 KO (3.1 nmoles/mg tissue ± 0.5) kidneys (*p* = 0.67) (Figure 4).

**Figure 4.** ATP content of WT and NHERF1 KO mouse kidneys.LC-MS was utilized to evaluate the amount of ATP in these tissues as described in the Methods section. Data are means ± SEM (WT *n* = 5) and (KO *n* = **Figure 4.** ATP content of WT and NHERF1 KO mouse kidneys.LC-MS was utilized to evaluate the amount of ATP in these tissues as described in the Methods section. Data are means ± SEM (WT *n* = 5) and (KO *n* = 5). No significant differences were reported in these kidneys.

#### 5). No significant differences were reported in these kidneys. *3.6. NHERF1 Deficiency Does Not A*ff*ect Kidney Proximal Tubule Mitochondria Morphology, Number, or Area*

These were taken and evaluated by a renal pathologist for signs of injury, oxidative stress, and changes in cristae. There were no apparent changes in mitochondria morphology between WT and NHERF1 KO proximal tubules (Figure 5, panels A and B). The only injury reported was early ischemic changes most likely due to harvesting of the kidneys (Figure 5). Some endosomal swelling The mitochondrial structure is essential for proper function; thus, EM images of WT and NHERF1 KO proximal tubule mitochondria were utilized to evaluate their morphology. These images were of 2–4-month-old male C57BL/6J WT and NHERF1 KO mice whose kidneys were perfused with 3% glutaraldehyde prior to EM analysis.

was noted but occurred across both genotypes. Additionally, the density and distribution of mitochondria within the tubules were alike, and no apparent signs of oxidative stress were found in either genotype (Figure 5). Changes in mitochondrial number and a decrease in size have been associated with a decline in mitochondrial function [45]. Therefore, one goal of this study was to determine if mitochondrial number and/or size changed in NHERF1 KO proximal tubules when compared to WT. Images from electron microscopy (EM) of WT and NHERF1 KO kidney proximal tubules were utilized in order to These were taken and evaluated by a renal pathologist for signs of injury, oxidative stress, and changes in cristae. There were no apparent changes in mitochondria morphology between WT and NHERF1 KO proximal tubules (Figure 5, panels A and B). The only injury reported was early ischemic changes most likely due to harvesting of the kidneys (Figure 5). Some endosomal swelling was noted but occurred across both genotypes. Additionally, the density and distribution of mitochondria within the tubules were alike, and no apparent signs of oxidative stress were found in either genotype (Figure 5).

count the number of mitochondria and to calculate the average area via Image J. There was not a significant difference between the average number of mitochondria between WT and NHERF1 KO tubules (WT average number: 128.8) and (NHERF1 KO average number: 115) (*p* = 0.6) (Figure 6A). In addition, there was not a significant difference between the average area of mitochondria in WT and NHERF1 KO tubules (WT average area: 580,540.9 μm2) and (NHERF1 KO average area: 678,465.4 μm2) (*p* = 0.75) (Figure 6B). *3.7. WT and NHERF1 KO Mouse Kidney Mitochondria Have Similar Oxidative Capacities*  The mitochondria's capacity to reduce oxygen is a critical aspect in the process of mitochondrial electron transport and ATP synthesis. Therefore, measuring mitochondrial oxygen consumption can Changes in mitochondrial number and a decrease in size have been associated with a decline in mitochondrial function [45]. Therefore, one goal of this study was to determine if mitochondrial number and/or size changed in NHERF1 KO proximal tubules when compared to WT. Images from electron microscopy (EM) of WT and NHERF1 KO kidney proximal tubules were utilized in order to count the number of mitochondria and to calculate the average area via Image J. There was not a significant difference between the average number of mitochondria between WT and NHERF1 KO tubules (WT average number: 128.8) and (NHERF1 KO average number: 115) (*p* = 0.6) (Figure 6A). In addition, there was not a significant difference between the average area of mitochondria in WT and NHERF1 KO tubules (WT average area: 580,540.9 µm<sup>2</sup> ) and (NHERF1 KO average area: 678,465.4 µm<sup>2</sup> ) (*p* = 0.75) (Figure 6B).

provide a valuable method to assess mitochondrial function. One purpose of this work was to assess mitochondrial function by oxidative capacity in WT and NHERF1 KO kidneys using the Seahorse XF24 analyzer. In panel A of Figure 7, the oxygen consumption rate (OCR) of WT and NHERF1 KO kidney mitochondria are shown over time. Both WT and NHERF1 KO mitochondria exhibit a similar

phosphorylation will result in a change in the RCR when comparing isolated mitochondria [46].

Succinate/Rotenone plus ADP for the production of ATP, both genotypes exhibit a maximal increase in OCR. Moreover, both genotypes undergo a decrease in OCR after adding oligomycin, an inhibitor of complex V (formation of ATP from ADP via O2 consumption). Lastly, antimycin A shuts down all respiration, where the OCR is close to the basal OCR. The difference between the basal OCR and OCR

Changes in state 3 (conversion of ATP from ADP and consumption of O2) and state 4 (nonphosphorylating or resting respiration) respiration are commonly used to evaluate mitochondria oxidative capacity. Panel B of Figure 7 shows both state 3 [(WT: 60 pmoles/min/μg protein ± 15) and (NHERF1 KO: 44 pmoles/min/μg protein ± 6)] and state 4 [(WT: 37 pmoles/min/μg protein ± 15) and (NHERF1 KO: 28 pmoles/min/μg protein ± 6)] of WT and NHERF1 KO kidney mitochondria, where state 3 (*p* = 0.2) and 4 (*p* = 0.1) was determined to not be significantly different between the groups.

The respiratory control ratio (RCR) is the best general measure of mitochondrial function in isolated mitochondria. RCR is measured by taking state 3/state 4 respiration and sums up the main function of mitochondria: the ability to respond to ADP from a resting state by making ATP at high rates. The RCR has no absolute value that is diagnostic of mitochondrial dysfunction [46]. Thus, RCR values are substrate and tissue-dependent, making the RCR advantageous when measuring

after antimycin A is the non-mitochondrial respiration.

**Figure 5.** Electron microscopy of mitochondria in WT and NHERF1 KO proximal tubules. **Figure 5.** Electron microscopy of mitochondria in WT and NHERF1 KO proximal tubules. *Antioxidants* **2020**, *9*, x 13 of 20

Representative photomicrographs show a 4x field of WT and NHERF1 KO proximal tubule mitochondria. Panel A represents the proximal tubule mitochondria of WT mice, while panel B

**Figure 6.** Evaluation of mitochondrial number and area of WT and NHERF1 KO proximal tubules. (**A**) Number of mitochondria were counted in random 4× visual fields with the highest density of mitochondria. Data are means ± SEM (WT *n* = 6) and (NHERF1 KO *n* = 5). The mitochondria number of NHERF1 KO proximal tubules was insignificant when compared to WT. (**B**) Mitochondria area was calculated using electron microscopy (EM) images and Image J. Data are means ± SEM (WT *n* = 6) and (NHERF1 KO *n* = 5). Mitochondria area of NHERF1 KO proximal tubules were insignificant **Figure 6.** Evaluation of mitochondrial number and area of WT and NHERF1 KO proximal tubules. (**A**) Number of mitochondria were counted in random 4× visual fields with the highest density of mitochondria. Data are means ± SEM (WT *n* = 6) and (NHERF1 KO *n* = 5). The mitochondria number of NHERF1 KO proximal tubules was insignificant when compared to WT. (**B**) Mitochondria area was calculated using electron microscopy (EM) images and Image J. Data are means ± SEM (WT *n* = 6) and (NHERF1 KO *n* = 5). Mitochondria area of NHERF1 KO proximal tubules were insignificant when compared to WT.

#### when compared to WT. *3.7. WT and NHERF1 KO Mouse Kidney Mitochondria Have Similar Oxidative Capacities*

The mitochondria's capacity to reduce oxygen is a critical aspect in the process of mitochondrial electron transport and ATP synthesis. Therefore, measuring mitochondrial oxygen consumption can provide a valuable method to assess mitochondrial function. One purpose of this work was to assess mitochondrial function by oxidative capacity in WT and NHERF1 KO kidneys using the Seahorse XF24 analyzer. In panel A of Figure 7, the oxygen consumption rate (OCR) of WT and NHERF1 KO kidney mitochondria are shown over time. Both WT and NHERF1 KO mitochondria exhibit a similar trend and response to added substrates and inhibitors. When adding the substrate Succinate/Rotenone plus ADP for the production of ATP, both genotypes exhibit a maximal increase in OCR. Moreover, both genotypes undergo a decrease in OCR after adding oligomycin, an inhibitor of complex V (formation of ATP from ADP via O<sup>2</sup> consumption). Lastly, antimycin A shuts down all respiration, where the OCR is close to the basal OCR. The difference between the basal OCR and OCR after antimycin A is the non-mitochondrial respiration. *Antioxidants* **2020**, *9*, x 14 of 20

**Figure 7.** Mitochondrial function in isolated mitochondria of WT and NHERF1 KO kidneys. Mitochondria from two to 4-month-old male WT and NHERF1 KO mice were isolated and analyzed via the Seahorse XF24 for oxidative capacity as described in the Methods section. (**A**) Oxygen consumption rate (OCR) was recorded after the addition of both substrates (succinate/rotenone/ADP) and inhibitors (oligomycin and antimycin A) [(WT *n* = 6) and (NHERF1 KO *n* = 6)]. (**B**) State 3 and state 4 were calculated using the recorded OCRs of WT and NHERF1 KO mitochondria. Data are mean ± SD [(WT *n* = 6) and (NHERF1 KO *n* = 6)]. State 3 and state 4 respiration were considered insignificant between WT and NHERF1 KO mouse kidney mitochondria. (**C**) Respiratory control ratio (RCR) (state 3/state 4) was calculated between WT and NHERF1 KO kidney mitochondria. Data are represented as state 3/state 4 ratio [(WT *n* = 6) and (NHERF1 *n* = 6)]. RCR was insignificant between WT and NHERF1 KO mouse kidney mitochondria. Accordingly, the RCR was calculated between WT (1.63) and NHERF1 (1.61) KO kidney mitochondria and was also found to not be significantly different (Figure 7C). **Figure 7.** Mitochondrial function in isolated mitochondria of WT and NHERF1 KO kidneys. Mitochondria from two to 4-month-old male WT and NHERF1 KO mice were isolated and analyzed via the Seahorse XF24 for oxidative capacity as described in the Methods section. (**A**) Oxygen consumption rate (OCR) was recorded after the addition of both substrates (succinate/rotenone/ADP) and inhibitors (oligomycin and antimycin A) [(WT *n* = 6) and (NHERF1 KO *n* = 6)]. (**B**) State 3 and state 4 were calculated using the recorded OCRs of WT and NHERF1 KO mitochondria. Data are mean ± SD [(WT *n* = 6) and (NHERF1 KO *n* = 6)]. State 3 and state 4 respiration were considered insignificant between WT and NHERF1 KO mouse kidney mitochondria. (**C**) Respiratory control ratio (RCR) (state 3/state 4) was calculated between WT and NHERF1 KO kidney mitochondria. Data are represented as state 3/state 4 ratio [(WT *n* = 6) and (NHERF1 *n* = 6)]. RCR was insignificant between WT and NHERF1 KO mouse kidney mitochondria.

Changes in state 3 (conversion of ATP from ADP and consumption of O2) and state 4 (non-phosphorylating or resting respiration) respiration are commonly used to evaluate mitochondria oxidative capacity. Panel B of Figure 7 shows both state 3 [(WT: 60 pmoles/min/µg protein ± 15) and (NHERF1 KO: 44 pmoles/min/µg protein ± 6)] and state 4 [(WT: 37 pmoles/min/µg protein ± 15) and (NHERF1 KO: 28 pmoles/min/µg protein ± 6)] of WT and NHERF1 KO kidney mitochondria, where state 3 (*p* = 0.2) and 4 (*p* = 0.1) was determined to not be significantly different between the groups.

The respiratory control ratio (RCR) is the best general measure of mitochondrial function in isolated mitochondria. RCR is measured by taking state 3/state 4 respiration and sums up the main function of mitochondria: the ability to respond to ADP from a resting state by making ATP at high rates. The RCR has no absolute value that is diagnostic of mitochondrial dysfunction [46]. Thus, RCR values are substrate and tissue-dependent, making the RCR advantageous when measuring mitochondrial function in isolated mitochondria [46]. A change in almost any aspect of oxidative phosphorylation will result in a change in the RCR when comparing isolated mitochondria [46].

Representative photomicrographs show a 4x field of WT and NHERF1 KO proximal tubule mitochondria. Panel A represents the proximal tubule mitochondria of WT mice, while panel B represents NHERF1 KO proximal tubule mitochondria (WT *n* = 6) and (NHERF1 KO *n* = 5). The scale bars were set at 2 µm.

Accordingly, the RCR was calculated between WT (1.63) and NHERF1 (1.61) KO kidney mitochondria and was also found to not be significantly different (Figure 7C).

#### **4. Discussion**

This work aimed to examine two aspects of proximal tubule cell function: metabolic enzymatic pathways and mitochondrial structure and function. We proposed that changes in kidney metabolism and/or mitochondrial structure or function could predispose NHERF1 KO mice to cisplatin nephrotoxicity. This is the first study to find changes in the kidney pentose phosphate pathway enzymes with NHERF1 loss and a novel proposed mechanism of susceptibility to cisplatin-induced AKI. Recent studies have found that cisplatin alters renal cell metabolism, contributing to injury and the secondary result of chronic kidney disease (CKD) development [2–6]. Cisplatin treatment results in the depletion of amino acids in the kidney [2–5], reduces fatty acid oxidation while concomitantly accumulating fatty acids in the kidney [2,5,6], and decreases renal glycolytic enzymes and intermediates of the pentose phosphate pathway [2,36]. In addition to affecting metabolic pathways cisplatin, nephrotoxicity has been established in inducing apoptotic and necrotic cell death. The mechanisms involved in cisplatin-induced nephrotoxic cell death remain unclear. However, there is increasing evidence that ROS and mitochondrial function have an important role in cisplatin's mechanism of injury. These observations combined with the increased susceptibility to cisplatin-induced AKI suggested the hypothesis that NHERF1 KO mice have metabolic alterations and/or mitochondrial dysfunction that predispose them to cisplatin nephrotoxicity.

NHERF1 KO mice did not exhibit changes in gluconeogenic or glycolytic enzyme activity. Indeed, cisplatin treatment resulted in a parallel decrease in FBPase and G6Pase activity in NHERF1 KO and WT mice. Additionally, there were no significant changes with LDH and MDH activity between non-treated and treated WT and NHERF1 KO mice. These results are in agreement with previous studies [2,7]. Interestingly, NHERF1 KO mouse kidneys exhibit increased activity of ME and G6PD under baseline conditions when compared to WT mouse kidneys. The significance of this shift in metabolism toward a greater utilization of the pentose phosphate pathway is not entirely clear. However, these findings suggest a potential compensatory mechanism for increased NADPH production as protection against oxidative stress. ME and G6PD activity provide necessary NADPH, a key cofactor in redox control and reductive biosynthesis. ME plays a role in the production of pyruvate and serves as an additional source of NADPH for lipogenesis. Additionally, there is recent evidence for direct cross-talk between ME and the pentose phosphate pathway [47], where G6PD is a rate-limiting enzyme. A study using a cell culture model of diabetes observed that the increased activity of G6PD restored redox balance

in endothelial cells exposed to high glucose levels [48], where high glucose levels had previously decreased G6PD and increased levels of oxidative stress. A similar observation has been made in studies of liver cirrhosis in rats subject to oxidative stress, where an increase in ME and G6PD gene expression and activity [49] are also seen, presumably providing protection against the stress through an increased production of NADPH [49]. Multiple studies have noted the importance of cellular redox balance in both the development of and in protection from renal injury [50,51]. Additionally, one other investigator found that NHERF1 is a previously unidentified regulator of Nox1 (NADPH oxidase) and promotes Nox1 activity [52]. Taken together, these observations suggest that the kidneys of NHERF1 KO mice experience a greater degree of oxidative stress that is masked by increased NADPH production through the pentose phosphate pathway. If NHERF1 KO mouse kidneys are more reliant on the pentose phosphate pathway for maintenance of the cellular redox state, the decrease in the activity of the enzymes of the pentose phosphate pathway resulting from cisplatin toxicity could potentially result in more severe injury.

While proteomic data demonstrated changes in mitochondrial proteins in NHERF1 KO mice (Figure S1) (Table S1), we found no alterations in mitochondrial structure or function. The kidney ATP levels in NHERF1 KO mice are equivalent to WT. EM analysis revealed that NHERF1 KO mice have similar proximal tubule mitochondrial morphologies, size, distribution and number when compared to WT. This result is not surprising, as our proteomic analysis did not identify differences in any of the proteins associated with mitochondrial fission or fusion such as dynamin related protein 1 (Drp1), mitofusin-1 (Mfn1), or mitofusin-2 (Mfn2). Mitochondria from NHERF1 KO and WT kidneys were found to have similar oxidative capacities as demonstrated by the measurement of OCR and RCR (WT: 1.63 and KO: 1.61) (Figure 7C). However, these findings do not exclude completely a role for altered mitochondrial function as a contributor to enhanced susceptibility to cisplatin-induced AKI. Although isolated mitochondria from NHERF1 KO kidneys function normally, they may not do so in intact tissue. NHERF1 KO mice undergo phosphate wasting [28] due to the faulty trafficking of Npt2a to the apical membrane [21], which may result in intracellular phosphate deficiency. The loss of intracellular phosphate may create an intracellular environment where mitochondria cannot function properly. NHERF1 KO mouse kidneys may also sustain losses of other important nutrients. In addition to higher urine phosphate excretion, NHERF1 KO mice also demonstrate hypercalciuria and hyperuricosuria [26]. A full evaluation of the alterations in proximal tubule cell transport in the NHERF1 KO kidneys has not been examined. The absence of NHERF1 may also result in impaired signaling processes [53], alterations in intracellular or mitochondrial protein phosphorylation, loss of mitochondrial interaction with other organelles, or aberrant mitochondrial protein localization, as suggested by the proteomic data demonstrating changes in mitochondrial proteins associated with the renal cortical BBM (Figure S1) (Table S1). Thus, further studies to compare mitochondrial function of WT and NHERF1 KO in intact kidney tissue are warranted.

#### **5. Conclusions**

In conclusion, this study provides insight into metabolic and mitochondrial changes of NHERF1 KO mice and provides new avenues to explore regarding NHERF1 loss and susceptibility to cisplatin nephrotoxicity. We did not find changes in enzyme activity in gluconeogenesis or the citric acid cycle, ATP content, and mitochondrial morphology and function. However, we discovered that enzymes of the pentose phosphate pathway were found to be increased in NHERF1 KO mice and suggest these animals are expressing some differences in metabolic pathways and may be compensating for an underlying stress. The basis for these changes in the activity of this metabolic pathway and its significance for the increased susceptibility of NHERF1 KO mice to cisplatin nephrotoxicity remain unknown. These results provide another area to be explored in the future pertaining to NADPH levels in NHERF1 KO mouse kidneys. Further investigation into the bioenergetics of NHERF1 KO mouse kidneys may elucidate more insight into susceptibility to cisplatin injury and increase our understanding of the underlying mechanism of susceptibility to cisplatin-induced AKI. In the future, this information may provide novel therapeutic targets and/or biomarkers to use clinically for the prevention of cisplatin nephrotoxicity.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2076-3921/9/9/862/s1, Figure S1: Proteomic analysis of BBM from WT and NHERF1 KO mice, Table S1: Differential expression of BBM proteins from WT and NHERF1 KO mice.

**Author Contributions:** Conceptualization, A.B.-S. and E.L.; methodology, A.B.-S., E.L., M.T.B., S.J.K., B.G.H., K.K., J.H., A.S. and M.M.; Software, B.G.H., A.S., S.N.R., S.S., and M.M.; validation, A.B.-S., M.M., M.B., B.C., and E.L.; formal analysis, M.B., M.M., A.B.-S., S.N.R., and S.S.; investigation, A.B.-S., E.L., L.S., M.T.B., K.B.G., and K.K.; resources, B.G.H., L.S., M.M., M.B., M.T.B., and B.C.; data curation, A.B.-S., M.M., M.B., M.T.B., K.B.G. and J.H.; writing—original draft preparation, A.B.-S. and E.L.; writing—review and editing, M.M., B.G.H, S.J.K., K.K., B.C., M.T.B., and K.B.G.; visualization, A.B.-S., M.M., K.B.G., J.H., and E.L.; supervision, E.L., M.M., and M.T.B.; project administration, E.L.; funding acquisition, E.L. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work was supported in part by Merit Review Award Number 1BX000610 from the United States Department of Veterans Affairs BLR&D Service. The views in the manuscript are those of the author and do not reflect the views of the Department of Veterans Affairs. This work was also supported by NIH P20GM113226 UofL Hepatobiology and Toxicology CoBRE, NIH P30ES P30ES030283 UofL Center for Integrative Environmental Health Sciences.

**Acknowledgments:** We are grateful to Susan Isaacs for expert technical support and David Hoetker for performing the LC-MS analysis in this manuscript. We would like to thank the University of Louisville Division of Nephrology for providing financial support.

**Conflicts of Interest:** Dr. Lederer is a consultant for WebMD and HQSI.
