Intravital imaging of real-time endogenous actin dysregulation in proximal and distal tubules at the onset of severe ischemia-reperfusion injury

Abstract

Severe renal ischemia-reperfusion injury (IRI) can lead to acute and chronic kidney dysfunction. Cytoskeletal modifications are among the main effects of this condition. The majority of studies that have contributed to the current understanding of IRI have relied on histological analyses using exogenous probes after the fact. Here we report the successful real-time visualization of actin cytoskeletal alterations in live proximal and distal tubules that arise at the onset of severe IRI. To achieve this, we induced fluorescent actin expression in these segments in rats with hydrodynamic gene delivery (HGD). Using intravital two-photon microscopy we then tracked and quantified endogenous actin dysregulation that occurred by subjecting these animals to 60 min of bilateral renal ischemia. Rapid (by 1-h post-reperfusion) and significant (up to 50%) declines in actin content were observed. The decline in fluorescence within proximal tubules was significantly greater than that observed in distal tubules. Actin-based fluorescence was not recovered during the measurement period extending 24 h post-reperfusion. Such injury decimated the renal architecture, in particular, actin brush borders, and hampered the reabsorptive and filtrative capacities of these tubular compartments. Thus, for the first time, we show that the combination of HGD and intravital microscopy can serve as an experimental tool to better understand how IRI modifies the cytoskeleton in vivo and provide an extension to current histopathological techniques.

Figure 1

Brightfield and confocal microscopic images highlight the presence of actin in the renal brush border ex vivo. Images were taken with 60× objectives (2× digital zoom) using brightfield (A) and confocal (B, C) microscopes. These images of the proximal tubule (PT) in cortical kidney sections outline the innate actin localization (identified by arrows) along brush borders using exogenous probes (H&E, A, and Texas-red phalloidin, B). Similarly, (C) highlights the intense presence of actin along the brush border in the PT of cortical kidney sections obtained from rats that expressed EGFP-actin fusion proteins using HGD. (B) was taken using only the red-pseudo-color channel, and (C) was taken using only the green-pseudo-color channel. Scale bars represent 10 µm.

Figure 2

HGD allowed the visualization of the actin-rich renal brush border in vivo. (A) (taken at 2× optical zoom) shows innate autofluorescent patterns that are used to routinely distinguish the proximal tubule (PT) from the distal tubule (DT) segments but were unable to outline brush border segments. In comparison, (B and C) (taken at 2× optical zoom), as well as (D and E) (taken at 1× optical zoom), highlight the presence of the actin-rich brush border in vivo (identified by arrows) in the proximal tubule. The region outlined in (D) (dashed-line) is presented as (C), to focus on the brush border as we did in (B). (A and B) were formed by merging the green- and red-pseudo-color channels, while (C and D) were formed by merging the blue-, green- and red-pseudo-color channels. We presented different combinations of the pseudo-channels shown in (D) to create (E and F), to better highlight EGFP-actin expression in the tubules. Specifically, (E) was created by merging the green- and blue-pseudo-colors, and (F) was created by merging the blue- and red-pseudo-colors. Overall, the presence of Hoechst 33342 and 150-kDa TRITC-dextran dyes in (C–F) delineated the tubular and supporting vasculature architectures. Scale bars represent 20 µm.

Figure 3

Intravital two-photon micrographs were taken with a ×60 objective from a live rat that received HGD. All images were formed by merging the green-and red-pseudo-color channels and shows the homogenous distribution of EGFP-actin expression in both proximal and distal tubules that did not appear to be affected by continuous imaging over a 1-h period. Scale bar represents 20 µm.

Figure 4

Extensive alterations to tubular structure that occurred after one hour of reperfusion. Intravital two-photon micrographs taken with a 60X objective show the effects of severe IRI in vivo. Animals that received sham injuries maintained intact tubular structure (A and C). (A) was obtained from an animal that did not receive HGD (group 1), while (C) (which displays the actin-rich brush border) was obtained from an animal that received HGD (group 3). In comparison, we also observed substantial damage to both proximal and distal tubules in (B) and (D), which were taken from animals that were subjected to severe IRI (animals in group 2 did not receive HDG, (B), and animals in group 4 received HGD, D). (C and D) can also be found in Fig. 2 (as F) and Fig. 6 (as F), respectively. Such injury dysregulated the actin cytoskeleton, and specifically, stripped the proximal tubule of its characteristic brush border that was visible in (C). Scale bars represent 20 µm.

Figure 5

Time-lapse images outline disruptions to normal renal filtrative and endocytic capacities that resulted from severe IRI. Intravital two-photon micrographs were taken with a 60× objective from a live rat in group 4, which received HGD and was subjected to IRI. After reinstating blood flow to the kidney, we observed a substantial injury 24 h after reperfusion. At that time point, we infused a mixture of 4-kDa FITC and 150-kDa TRITC dextrans, along with Hoechst 33342, via the jugular vein of the animal, to track renal dynamics. (A–I) illustrate the loss of EGFP-actin expression, reductions in the thicknesses of the vasculature (V), aggregated red blood cells (rouleaux) in the peritubular capillaries (dashed line in D), absence of endocytic uptake of low-molecular-weight FITC dextran molecules by the proximal tubules, and simultaneous entry of both FITC and TRITC dyes in the lumen. Moreover, these images illustrate the initial presence of the TRITC dye entering the peritubular vasculature (B) and then the entry of the FITC dye (C). After that, there was a reduced level of fluorescence within the vasculature observed in image I. Scale bar represents 20 µm. The time-lapse video for this event is presented in Supplemental Video 1.

Figure 6

Time-lapse images tracked alterations in actin-based fluorescence observed during the first 60 min of reperfusion. Intravital two-photon micrographs were taken with a ×60 objective from a live rat in group 4 across 60 min (this is the same imaging field that is previously presented in Fig. 3D). This animal received HGD and was subjected to ischemia-reperfusion injury (IRI). All images were formed by merging the green-and red-pseudo-color channels and shows the expression of EGFP actin in both proximal and distal tubules. This fluorescent protein expression allowed us to visualize the live and real-time changes in tubular structure and function that resulted from IRI (arrows identified changes in proximal tubules, and arrowheads identified changes in distal tubules). At the 10-min mark, EGFP-actin appeared more heterogeneously distributed and clumped in tubular segments. The dashed ovals in (B and E) track the outlined region and show how cells have sloughed off the proximal tubule segment and migrated into the lumen to generate ghost tubules (tubules mostly devoid of living cells) by the 50-min mark. It should be noted that there were minor shifts in the field during the 60-min imaging period that resulted from the vibration caused by respiration. Scale bar represents 20 µm. A time-lapse video showing portions of this event is presented in Supplemental Video 2.

Figure 7

In vivo changes in mean fluorescence intensities obtained from proximal and distal tubular segments. There were no considerable differences in fluorescence intensity recorded from proximal and distal tubular segments from animals that did not receive HGD (group 1), but there were larger variations in autofluorescence that resulted from ischemia-reperfusion injury (group 2) across the 60-min measurement period. In comparison, we observed substantial decreases in fluorescence intensities in proximal and distal tubular segments recorded from animals that received HGD (groups 3 and 4).