Illuminating Mitochondrial Function and Dysfunction using Multiphoton Technology

Studying differential cell function or dysfunction within the kidney is made difficult by cellular heterogeneity, alterations in regional blood flow, oxygen tension, and interstitial tonicity, resulting in extremely complex anatomic and physiologic arrangements. Creative investigators have developed techniques to reduce this heterogeneity but often with loss or alteration of three-dimensional cellular associations, anatomic and physiologic cell-to-cell interactions, and harsh isolation and experimental conditions. These limitations have made “absolutes” difficult to determine, especially in relation to pathologic processes such as cell type involvement in renal ischemia. In this issue of JASN, Hall et al.¹ use multiphoton microscopy of kidney slices to compare mitochondrial parameters and differential cellular responses to stress. This approach allows them to quantify several key parameters of mitochondrial function and dysfunction in proximal tubular (PT) cells and directly compare different PT segments with each other and with distal tubular (DT) cells of the thick ascending limb (TAL). Because mitochondrial alterations play a central role in normal cell function and in response to injury, the article adds to our previous knowledge about an important area.

To understand the significance and limitations of their contribution, one must first understand what is known about differences between PT and TAL cells. Abundant mitochondria in cortical and outer medullary renal epithelial cells are necessary to meet the ATP demands of sodium transport by high-capacity aerobic metabolism. They compose 33% of the volume of proximal convoluted tubular S1 cells, 39% of the cells in the S2 segment, and 22% of the volume of cells in the proximal straight S3 segment; in the medullary and cortical TAL cells, mitochondria account for 30 to 44% of cell volume.² Differential tubular cell metabolism is also notable for the absence of aerobic and anaerobic glycolysis in the proximal convoluted tubule (S1),^3,4 which removes an important mechanism for preserving cell ATP and viability during injury.⁵ In contrast, DT segments, including the TAL, have well-developed glycolytic pathways.^3,4,6,7 Glycolysis occurs in the S3 segment of the proximal tubule, albeit to a lesser extent than in distal tubules, and contributes to maintaining ATP there when mitochondrial function is impaired.^3,8

Mitochondrial ATP production depends on substrate-supported, electron transport–mediated proton extrusion from the mitochondrial matrix that generates a proton gradient across the inner mitochondrial membrane. In turn, this is used to drive phosphorylation of ADP to ATP by proton movement down the gradient back into the matrix through the inner membrane F₁F_O-ATPase. Electron transport–driven proton extrusion is also responsible for the net potential across the inner membrane (ΔΨ_m), and ΔΨ_m in combination with the pH gradient accounts for the net proton electrochemical gradient across the membrane, also called the proton motive force. ΔΨ_m is the larger of the two components of proton motive force under both normal and pathologic conditions⁹; therefore, it serves as a valuable index of the state of the entire system, as shown in recent studies documenting a major role for nonesterified fatty acids in the persistent mitochondrial dysfunction seen in re-oxygenated proximal tubules.^9,10 As discussed by Hall et al.,¹ when ΔΨ_m is low and ATP is available, the F₁F_O-ATPase reverses the flow of protons by hydrolyzing ATP to extrude protons and restore ΔΨ_m.

Fluorescence approaches, using both chemical probes and autofluorescence of components of the mitochondrial electron transport chain, provide powerful tools for dynamically following determinants of mitochondrial function and have been widely applied to studies of isolated mitochondria and whole cells. Parameters related to mitochondrial function including ΔΨ_m, superoxide production, cellular GSH levels, and the redox state of NADH and FAD²⁺ can be quantified in individual cells and ΔΨ_m within individual mitochondria.¹¹ Hall et al.¹ show that mitochondrial uptake of two different lipophilic cationic probes (tetramethyl rhodamine methyl ester and rhodamine 123), as a measure of ΔΨ_m, follows the pattern TAL>proximal convoluted tubule>proximal straight tubule. This pattern is not changed by inhibition of the multidrug resistance transporter with verapamil, suggesting it is due to differences in ΔΨ_m between tubular segments rather than in transport of the probe across the plasma membrane of individual cells; however, during chemical anoxia, ΔΨ_m is well maintained in TAL cells but not in PT cells. Inhibition of the F₁F_O-ATPase with oligomycin abolishes this difference, indicating that differences between PT and TAL cells are due to support of ΔΨ_m by F₁F_O-ATPase–mediated ATP hydrolysis. Proximal tubule cells had a lower ratio of ATPase to IF1, an endogenous low pH–activated inhibitor of the ATPase.¹² Less inhibition of the ATPase by IF1 can adaptively conserve residual ATP under conditions such as ischemia, in which mitochondrial aerobic metabolism is absent,¹² but can also be deleterious insofar as a low ΔΨ_m favors opening of the mitochondrial permeability transition pore, which leads to necrosis or apoptosis.¹³ The maintenance of ΔΨ_m in the TALs also indicates persistence of significant glycolytic ATP production in those cells despite omission of glucose from the medium during the period of chemical hypoxia.

Additional biochemical questions that could be addressed with this approach come to mind. Would keeping pH 7.4 during chemical hypoxia to limit activation of IF1 allow for better preservation of ΔΨ_m in the PTs? Are there differences between the S2 and S3 segments of the PT in that regard that reveal differences in glycolytic capacity? Would perfusion of glucose during chemical hypoxia help compensate for their lack of glycogen stores relative to the TAL cells⁷ and produce better preservation of ΔΨ_m? The observations on superoxide production and redox state could also be expanded in informative ways in future studies.

Fluorescence techniques are rapidly becoming the approach of choice for studying mitochondrial function and individual mitochondria within a cell can now be studied using membrane-permeant cations and fluorescence resonance energy transfer techniques.¹¹ In addition, calcein entry can be used to measure the mitochondrial permeability transition.¹³ This has important implications regarding understanding disease pathophysiology and therapy and can be used in vivo.^14,15 Multiphoton microscopy increases the depth of tissue penetration, reduces phototoxicity, and can follow three different fluorescence probes at once. This allows for direct correlation of multiple dynamic processes in the kidney.¹⁶ Recent advances in the use of infrared-multiphoton microscopy now offer doubling of imaging depth and further reduction of phototoxicity.¹⁷ Enhanced speed of image capture rates is now available through line-screening confocal microscopy, enhancing temporal resolution to real-time speeds.¹⁸

Certain limitations of the Hall et al. studies must be kept in mind. First, the studies use multiphoton techniques in tissue sections. Tissue sections allow for enhanced labeling with fluorescent probes and increased control over the environment and experimental conditions. They also allow for elucidating new information about complex pathways in fully differentiated cells in their normal anatomic tissue environment; however, tubules are not being perfused with glomerular filtrate, O₂ and substrate delivery occurs by diffusion, transport across cells is markedly decreased, and certain limitations of multiphoton microscopy must be considered. Previous qualitative intravital studies showed that PT cells had a greater Ψ_m than DT cells under physiologic conditions when using rhodamine 123 as the probe.¹⁵ After intravenous delivery, rhodamine 123 enters cells from the basolateral aspect, indicating adequate access of the dye to all cells (unpublished observations). The reason for this apparent difference in ΔΨ_m between tissue sections and intravital measurement is unclear. In addition, because mitochondrial diameter is approximately 1 μm and multiphoton optical sections are slightly less than 1 μm, ΔΨ_m in some mitochondria is underestimated; therefore, creation of three-dimensional volumes would help to minimize this inherent concern.

In summary, the new multiphoton data provided by Hall et al.¹ focus our understanding of mitochondrial function to individual cells with direct comparisons between adjacent PT and TAL cells. The data provide additional information as to why PT cells are more sensitive to ischemia. Application and extension to individual mitochondria within cells, quantifying their response to stress and toxins, and use of this knowledge to understand therapies further await future studies.^13,14 Indeed, the future is bright for multiphoton microscopy's advancing our mechanistic understanding to individual cells and organelles with enhanced spatial and temporal resolution and increased imaging depth.

DISCLOSURES

None.

J.M.W. is funded by National Institutes of Health grant DK034275, and B.A.M. is funded from National Institutes of Health grants P30-DK079312 and R01-DK069408, a VA Merit Review, and an INGEN (Indiana Genomics Initiative) grant from the Lilly Foundation to Indiana University School of Medicine.