2.1.

Chemicals and Reagents

Sodium fluorescein and $\beta$-glucuronidase were purchased from Sigma Aldrich (St. Louis, MO). Ilium xylazil and ketamine hydrochloride were obtained from Bayer Australia (Pymble NSW, Australia) and Parnell Laboratories, Australia, respectively. Reagents used for western blot analysis were purchased from Bio-rad (Protein std precision plus BIO-Rad, IMMuno-blot membrane PVDF, $10\times$ TGS Buffer, $10\times$ TG Buffer, 30% Acrylamide/Bis Solution, $29:1$, TEMED and Ammonium Persulfate); Thermo Fisher (Supersensitive West Femto Maximum Sensitivity, Pierce BCA Protein Assay Kit, Dental Developer replenisher, Dental fixer replenisher and T-mat g film); and Whatman (Whatman filter paper). MRP2 monoclonal [M2 III-6] antibody (catalogue number AB3373) was purchased from Sapphire Bioscience (Sydney NSW, Australia); OATP2 (catalogue number AB3572P) was purchased from Merck Millipore (Kilsyth VIC, Australia); and rabbit antibody MRP3 from Sigma Aldrich. The reagents used in real-time polymerase chain reaction (RT-PCR) were obtained from Invitrogen to Life Technologies Australia (Mulgrave, Victoria, Australia) (Sybr master mix) and Sigma Aldrich [quantitative PCR (qPCR) primers].

2.2.

Animals

Male Wistar rats, weighing $320\pm 20\mathrm{g}$, purchased from the University of Queensland Biological Resources, were used in all experiments. All studies were approved by the Animal Ethics Committee at the University of Queensland. Rats were housed at the University of Queensland in Pharmacy Australia Centre of Excellence, where the temperature was maintained at $20\pm 1°\mathrm{C}$ and humidity at 60 to 75%, with artificial light for 12 h (7 a.m. to 7 p.m.) daily. All animals had unlimited access to food and water.

2.3.

Surgical Procedures and I/R Model

Rats were anaesthetized initially by an intraperitoneal injection of ketamine hydrochloride ($80\mathrm{mg}/\mathrm{kg}$) and xylazine ($10\mathrm{mg}/\mathrm{kg}$). Anesthesia was maintained throughout the experiment by administering ketamine ($2.2\mathrm{mg}/100\mathrm{g}$) and xylazil ($0.25\mathrm{mg}/100\mathrm{g}$). Body temperature was controlled by placing rats on a heating pad set to 37°C. Ischemia was induced locally in approximately 70% of the liver, by clamping the portal vein, hepatic artery, and bile duct supplying the median and left lobes using a microvascular clamp. After 60 min, the clamp was removed to allow reperfusion of the liver for 4 h and rats were allowed to recover during this time under close monitoring. Sham rats (i.e., untreated controls) underwent the same procedure without clamping the vessels. After 3 h of reperfusion, rats were again anaesthetized. The jugular vein and carotid artery were cleared of surrounding tissue and cannulated using polyethylene (PE)-tubing for administration of sodium fluorescein (intravenous bolus $10\mathrm{mg}/\mathrm{kg}$ at 4 h of reperfusion) and blood collection. A midline laparotomy was performed to expose the liver and to cannulate the bile duct, which was cleared of surrounding fat tissue and cannulated using PE-tubing. The left lobe was slightly raised above the intraperitoneal cavity to allow imaging of fluorescein distribution. The liver and intraperitoneal cavity were kept moist by continuous administration of 0.9% saline throughout the experiment.

2.4.

Tissue Collection

Blood (0.2 mL) was collected during the surgery from the inferior vena cava using a 30 gauge needle and plasma concentration of alanine transaminase (ALT) was measured using a Hitachi 747 analyzer (Hitachi Ltd., Tokyo, Japan). Blood for fluorescein determination was collected at 5, 10, 30, 60, 120, and 180 min from the carotid artery. Bile was collected in preweighted tubes every 10 min for 1 h and every hour after that for 3 h. At the end of the experiment, the left and median lobes were excised and portions were snap frozen in liquid nitrogen and stored at $-70°\mathrm{C}$ for later analysis with western blot and RT-PCR.

2.5.

Multiphoton Intravital Imaging

MPT images were recorded using a DermaInspect system (JenLab GmbH, Jena, Germany) equipped with an ultrashort (85 fs pulse width) pulsed mode-locked 80-MHz Titanium:Sapphire MaiTai laser (Spectra Physics, 25 Mountain View, USA). The excitation wavelength was set to 740 nm for liver autofluorescence and 920 nm for fluorescein fluorescence, with an emission signal range of 350 to 650 nm established through the use of a BG39 bandpass filter. Images were recorded with $10\times$ (low magnification) or $40\times$ (high magnification) objectives. The laser power was set to 15 or 20 mW and the acquisition time for obtaining the images was 13.4 or 7.4 s per frame for high- and low-magnification imaging, respectively. Each image was $179\times 179\mu \text{m}$ wide at a resolution of $512\times 512\mathrm{pixels}$. The left lobe of the liver was placed on a metal plate, attached to an adjustable stand that could be elevated or lowered as required. The plate was slightly raised above the intraperitoneal cavity to reduce movement in the image from respiration and to minimize pressure on the organs underneath. Images were recorded before injection of fluorescein and then continuously throughout the experiment up to 180 min.

FLIM data were collected with a time-correlated single-photon counting (TCSPC) SPC-830 detector (Becker & Hickl GmbH, Berlin, Germany) integrated into the MPT system. The TCSPC module assembles a photon distribution of the scan area across the $x$ and $y$ coordinates taken from four photon counters, with only three used for this study. The module determines the time of arrival for each photon detected within the fluorescence decay. Fluorescence emission was spectrally resolved between three linearly arranged photon counters through the use of three dichroic filters in the beam path, spectrally dividing the emission light into three channels for each photon counter: 350 to 450 nm {reduced nicotinamide adenine dinucleotide phosphate [NAD(P)H]}, 450 to 515 nm [NAD(P)H, flavin adenine dinucleotide (FAD)], and 515 to 620 nm (FAD, fluorescein).^10,31 Each FLIM image was collected at an exposure of 13.4 s and acquisition image size of $214\times 214\mu \text{m}$.

2.6.

Image Processing

All image processing was done in ImageJ 1.44p (National Institutes of Health, USA). To trace the distribution of fluorescein in blood and hepatocytes, these areas were selected in the images recorded by low magnification. For analysis of fluorescein distribution in bile, this area was selected in images recorded by high magnification. The mean fluorescence intensity in these areas was calculated and converted into concentrations using a standard curve prepared in homogenized liver. The profile of fluorescein concentration (in sinusoid or hepatocyte) versus time was fitted by an empirical triexponential model.

Model fitting was performed using software SCIENTIST® (Micromath Scientist, Salt Lake City, UT).

2.7.

FLIM Data Analysis

FLIM data analysis was performed using SPCImage 3.9.7 software (Becker & Hickl GmbH). FLIM data is composed of an array of pixels containing several time channels distributed across the fluorescence decay curve. The decay curve represents a sum of multiple exponentials, or components, as each pixel contains an overlay of fluorescence from several endogenous fluorophores at various conformations. For the data we obtained, a double-exponential decay model function was used.

The fitted decay curve establishes two lifetimes: the short (${\tau }_1$) and long (${\tau }_2$) fluorescence decay lifetimes (ps) and with corresponding relative amplitude coefficients ${\alpha }_1$ and ${\alpha }_2$ (%), respectively.

The instrument response function (IRF) was calibrated using a sucrose crystal standard first analysed by FLIM (Ajax Finechem Pty Ltd, Sydney, NSW, Australia) was used. Each individual FLIM image is convoluted with $\mathrm{IRF}(t)$ in the model function, $f(t)$, to calculate the function $F(t)$. The function below corresponds to the shape of the signal if the measured optical signal was identical to the model function $f(t)$.

Using the $F(t)$ function, the measured data points are obtained and analyzed to determine the average lifetime and relative contributions of each endogenous molecule.

2.8.

High-Performance Liquid Chromatography

The concentration of fluorescein and fluorescein monoglucuronide in each collected bile sample was determined by high-performance liquid chromatography (HPLC, Shimadzu, Kyoto, Japan) using 1 mL/min flow rate with a C18 column [Agilent HC-C18 (2) $4.6\times 150\mathrm{mm}$, 5 μm] that included a security guard column (Phenomex C18 $4\times 3\mathrm{mm}$). Fluorescence excitation and emission wavelengths were 488 and 515 nm, respectively, with a retention time of 5 min. The mobile phase consisted of 50 mM ${\mathrm{NaH}}_2{\mathrm{PO}}_4$ and methanol ($2:3$).

Analysis of fluorescein involved diluting 5 μl of bile to 500 μl with 0.2 M acetate buffer. For analysis of fluorescein monoglucuronide (FG), a 5-μl bile sample was diluted in 50 μl $\beta$-glucuronidase (10 U/100 μl). All samples were then incubated for 1 h at 37°C in a water bath. Fluorescein samples were further diluted in deionized water ($1:10$), whereas FG samples were first diluted in an acetate buffer ($1:20$) and then in water ($1:10$). Samples (10 μl) were injected into the HPLC system. Bile was collected in preweighted tubes, enabling calculation of the excretion rates for fluorescein and fluorescein monoglucuronide.

2.9.

RT-PCR

RNA was extracted from 100 to 150 mg liver tissue using TRIzol Reagent as instructed by the manufacturer. DNA contaminations were removed by incubation with RNase-free DNAse-1. First strand cDNA was synthesized from 1 μg total RNA using SuperScript III reverse transcriptase (Invitrogen) and oligo dT as instructed by the manufacturer.

cDNA was diluted ($1:20$) in water, and 2 μl of the sample was added to each well in duplicates. A mixture containing 5 μl platinum SYBR® Green qPCR, 0.2 μl of forward and 0.2 μl of reverse primer and 2.6 μl water, was added to each well, after which the plate was centrifuged to mix the solutions. RT-PCR was read using a Roche LightCycler480.

2.10.

Western Blot

Liver slices were homogenized in protein lysis buffer solution ($1:20$, weight:volume), after which triton ($10\mu \text{L}/\mathrm{ml}$ buffer) was added and stored on ice. The solution was centrifuged at 13,000 rpm at 4°C for 20 min and the supernatant was collected. The supernatant, equivalent to 30 or 50 μg of protein, was mixed with loading buffer and heated for 5 min at 100°C. Acrylamide resolving gel (10%) and 4% stacking gel, both with sodium dodecyl sulfate, were prepared with tris-glycine as a running buffer. Electrophoresis was run at 150 V for 45 min, after which the gel was transferred to membranes at 100 V for 1 h at 4°C. The membranes were blocked in 10% skim milk in blocking buffer [tris-buffered saline with 0.1% Tween 20 (TBS-T)] overnight at 4°C while shaking. Membranes were incubated with primary antibody ($1:500$ for OATP2 and MRP2 and $1:1000$ for MRP3) in blocking buffer for 1.5 h, after which it was washed in TBS-T. Membranes were incubated with the secondary antibody ($1:100\mathrm{K}$ for OATP2, $1:300\mathrm{K}$ for MRP3, and $1:20\mathrm{K}$ for MRP2) for 1.5 h, followed by washing in TBS-T first and then TBS. The membrane was developed onto film using a luminescence kit (SuperSignal® West Femto) with a reaction time of 5 min. $\beta$-actin was used as a loading control ($1:1000$ for primary and $1:200\mathrm{K}$ for secondary antibody). Protein bands were quantified using ImageJ and adjusted with density of $\beta$-actin.

2.11.

Statistical Analysis

The data are expressed as $\mathrm{mean}\pm \mathrm{standard}\mathrm{error}\mathrm{of}\mathrm{the}\text{mean}$ ($n=3$ to 7). A student’s $t$-test was used for statistical comparison of serum ALT levels, fluorescence lifetime data, and pharmacokinetic parameters between sham and I/R injury groups. To analyze statistical differences in mRNA or protein levels between sham and I/R injury groups at 4 and 24 h, a one-way analysis of variance was used with a post hoc Newman-Keuls test. Statistical analysis was done using GraphPad Prism v5.03 (GraphPad Software Inc., La Jolla, California) with a $p$ value less than 0.05 considered statistically significant.

3.1.

I/R-Induced Elevation in Serum ALT Levels

Figure 2 demonstrates extensive liver damage as indicated by significantly elevated ($p<0.05$; $n=4$) serum ALT levels in rats subjected to I/R injury ($1200\pm 260\mathrm{U}/\mathrm{L}$) compared to sham ($54\pm 7\mathrm{U}/\mathrm{L}$).

Fig. 2

Alanine transaminase levels in plasma after I/R injury. Values are presented as $\mathrm{mean}\pm \mathrm{standard}\mathrm{error}\mathrm{of}\mathrm{the}\mathrm{mean}$ (SEM) ($n=4$ for sham, $n=6$ for I/R injury), * $p<0.05$ significantly different from sham.

3.2.

Intravital FLIM Imaging of the Rat Liver after Sham or I/R Injury

Figure 3(a) shows representative intravital FLIM images (pseudocolored according to average weighted lifetime ${\tau }_m$), across three spectral channels (i.e., 350 to 450, 450 to 515, and 515 to 620 nm), of hepatocytes from within the rat liver following sham or I/R injury. The data demonstrate that in the second spectral channel [450 to 515 nm; NAD(P)H/FAD], ${\tau }_m$ appears to be smaller in the rat liver after I/R injury (blue) relative to sham (yellow-green). This observation is also seen in Fig. 3(b), indicated by the left-shift of the ${\tau }_m$ histogram derived from the FLIM image. The average ${\tau }_m$ [450 to 515 nm; NAD(P)H/FAD] of hepatocytes from I/R-treated rats ($1575.3\pm 41.9\mathrm{ps}$) was significantly lower ($p<0.001$; $n=6$) than the sham control ($1225.4\pm 16.3\mathrm{ps}$) [Fig. 3(c)]. Further examination revealed that the fast decay lifetime component (${\tau }_1$) within hepatocytes was significantly smaller ($p<0.01$) in I/R-treated rats ($1054.5\pm 31.4\mathrm{ps}$) compared to the sham control ($932.4\pm 17.4\mathrm{ps}$) [Fig. 3(d)]. The tendency for the long decay lifetime component (${\tau }_2$) in I/R-treated rats to be smaller than the sham control was not statistically significant. In addition, while the ${\tau }_m$ of I/R-treated rats, in the 350 to 450 nm [NAD(P)H] spectral channel appeared smaller than the sham control [Fig. 3(b)], this difference was not statistically significant.

Fig. 3

I/R injury associated changes in the average weighted fluorescence lifetime (${\tau }_m$) within the liver in vivo. (a) Representative ${\tau }_m$ intravital images of sham and I/R injury in the liver. FLIM data were measured within three spectral ranges [pseudocolor range (blue-to-red) indicated in parenthesis]: 350 to 450 (750 to 1500 ps), 450 to 515 (1000 to 1750 ps), and 515 to 620 nm (500 to 1750 ps) to capture NAD(P)H, NAD(P)H/FAD, and FAD/fluorescein, respectively. Each image is $214\times 214\times 1{\mu \text{m}}^3$. (b) ${\tau }_m$ lifetime histograms of sham and I/R injury liver NAD(P)H, NAD(P)H/FAD, and FAD/fluorescein, respectively. The histograms represent the average pixel intensity for each ${\tau }_m$ lifetime. (c) The corresponding average ${\tau }_m$ lifetimes from the three spectral channels are charted for the sham and I/R injury groups. (d) The average short (${\tau }_1$) and long (${\tau }_2$) decay lifetimes obtained from the NAD(P)H/FAD (450 to 515 nm) channel for the sham and I/R injury groups. Values are presented as the mean±SEM ($n=6$), $**p<0.01$ or $**p<0.001$ significantly different from sham.

Similar to Fig. 3, Fig. 4 shows representative redox FLIM images of hepatocytes from I/R and sham control rats. The images were pseudocolored according to the ratio of amplitude coefficients ${\alpha }_1$ and ${\alpha }_2$. Figure 4(a) demonstrates that in the 350 to 450 nm [NAD(P)H] spectral channel, the redox ratio (${\alpha }_1/{\alpha }_2$) appears higher in hepatocytes after I/R injury (red) relative to sham (blue-green). This tendency is reflected in the ${\alpha }_1/{\alpha }_2$ ratio histograms in Fig. 4(b). However, the average ${\alpha }_1/{\alpha }_2$ ratio in I/R-treated rats ($2.663\pm 0.05$) is not significantly higher ($p=0.187$) than the sham control ($2.501\pm 0.12$) [Fig. 4(c)].

Fig. 4

I/R injury associated changes in the redox ratio (${\alpha }_1/{\alpha }_2$) within the liver in vivo. (a) Representative redox intravital images of sham and I/R injury in the liver. FLIM data were measured within three spectral ranges [pseudocolor ${\alpha }_1/{\alpha }_2$ range: 0 to 5 (blue-to-red)]: 350 to 450, 450 to 515, and 515 to 620 nm to capture NAD(P)H, NAD(P)H/FAD, and FAD/fluorescein, respectively. Each image is $214\times 214\times 1{\mu \text{m}}^3$. (b) Redox ratio histograms of sham and I/R injury liver NAD(P)H, NAD(P)H/FAD, and FAD/fluorescein, respectively. The histograms represent the average pixel intensity for each redox ratio value. (c) The corresponding average redox ratios from the three spectral channels are charted for the sham and I/R injury groups. Values are presented as the mean±SEM ($n=6$).

3.3.

Fluorescein Uptake, Distribution, and Clearance from the Rat Liver after I/R Injury

Figure 5 shows representative fluorescence intravital images of the rat liver, from sham and I/R injury groups, over time after injection of fluorescein. Representative intravital images, taken at $\times 10$ and $\times 40$ magnifications [Fig. 5(a)], demonstrate elevated hepatocyte autofluorescence [two photon excitation (${2\mathrm{P}}_{\mathrm{Exc}}$): 740 nm) in the sham control rats compared to I/R injury group. Figure 5(b) and 5(c) shows representative images of fluorescein (${2\mathrm{P}}_{\mathrm{Exc}}$: 920 nm) uptake, distribution, and clearance within the liver of sham and I/R injury rats after injection at selected time points over a 180-min time-course. Fluorescein was detected within minutes after injection of fluorescein in sham and I/R injury rats. Fluorescein appears to be taken up and cleared from the liver faster in the sham group compared to the I/R injury group, which displayed delayed uptake and elevated retention within the liver over time [Fig. 5(b) and 5(c)].

Fig. 5

Representative intravital images of fluorescein uptake in the sham and I/R injury rat liver groups. (a) Fluorescence intensity images recorded in vivo at ${\lambda }_{\mathrm{Exc}/\mathrm{Em}}$: $740/350$ to 650 nm (endogenous autofluorescence) within sham and I/R injury (4 h of reperfusion) groups at low ($10\times$) and high ($40\times$) magnifications. (b) and (c) Fluorescence intensity images recorded in vivo at ${\lambda }_{\mathrm{Exc}/\mathrm{Em}}$: $920/350$ to 650 nm (fluorescein fluorescence alone) in sham and I/R injury groups at various time points following 4 h of reperfusion at low ($10\times$) and high ($40\times$) magnifications, respectively. Narrow arrows indicate blood vessels, filled white arrows show hepatocytes, and opened arrows show bile duct. Scale bars for images taken at $10\times$ and $40\times$ magnifications indicate a length of 200 and 40 μm, respectively.

Figure 6(a) to 6(c) show the changes in fluorescein concentration within the sinusoids, hepatocytes, and bile, respectively, of sham and I/R injury rats after fluorescein injections. It appears that the fluorescein clearance rate from the sinusoids was slower for I/R injury, as indicated by a shallower slope [Fig. 6(a)], which was statistically significant ($p<0.05$) at 10 to 25 min. Similarly, the excretion rate from the hepatocytes also appeared slower in the I/R group [Fig. 6(b)].

Fig. 6

Change in fluorescein concentration over time within the sinusoid (a), hepatocytes (b), and bile (c) of sham and I/R injury groups, measured by intravital imaging of fluorescein fluorescence within the rat liver and bile. A standard curve of fluorescein fluorescence intensity to concentration was established to measure unknown concentrations by intravital imaging. Data represent the mean concentration of fluorescein $\pm \mathrm{SEM}$ ($n=3$). Significant differences are indicated with an asterisk ($*p<0.05$).

Figure 7(a) and 7(b) shows the AUC value and elimination half-life, respectively, for fluorescein in the sinusoids and hepatocytes, calculated by fitting data from Fig. 6(a) and 6(b) to a triple exponential equation. The data show a significantly increased ($p<0.05$) AUC in the hepatocytes ($5100\pm 340\mu \text{g}\cdot \min /\mathrm{mL}$) for the I/R injury group compared to the sham group ($2800\pm 640\mu \text{g}\cdot \min /\mathrm{mL}$). The AUC in the sinusoids and the half-life in the hepatocytes tended to be higher for the I/R injury group ($2030\pm 510\mu \text{g}\cdot \min /\mathrm{mL}$ and $430\pm 61\min$, respectively) relative to the sham group ($1200\pm 390\mu \text{g}\cdot \min /\mathrm{mL}$ and $250\pm 55\min$, respectively), but these differences were not statistically significant.

Fig. 7

Pharmacokinetic parameters of fluorescein clearance within the hepatocytes of sham and I/R injury groups. Fluorescein accumulation (area under the curve) (a) and elimination half-life (b) were calculated from fluorescein fluorescence intensity changes within the hepatocytes. Data represent the mean±SEM ($n=3$). Significant differences are indicated with an asterisk ($*p<0.05$).

3.4.

Concentration of Fluorescein in Plasma and Excretion of Fluorescein and Metabolites in Bile

Figure 8(a) demonstrates that the plasma concentration of fluorescein, measured by HPLC, was similar in sham and I/R injury groups at 4 h of reperfusion. The biliary excretion rate of fluorescein [Fig. 8(b)] showed a slight tendency to be lower in the I/R injury group relative to the sham control group, though these differences were not statistically significant. In contrast, the biliary excretion rate of fluorescein monoglucuronide for sham and I/R injury groups was comparable over time [Fig. 8(c)].

Fig. 8

Plasma concentration (a), fluorescein (b), and fluorescein monoglucuronide (c) biliary excretion rate over time measured by HPLC in sham and I/R injury groups. Data represent the $\mathrm{mean}\pm \mathrm{SEM}$ ($n=5$ to 6).

3.5.

mRNA and Protein Expression of OATP2, MRP2, and MRP3

Figure 9 shows the mRNA and protein expression of OATP2, MRP2, and MRP3 within excised liver tissue from sham and I/R injury groups. mRNA expression of the transporter gene MRP2 tended to be higher in the I/R injury group compared to the sham group after 4 h of reperfusion, respectively, while OATP2 and MRP3 transporter expression appeared to decrease after 4 and 24 h of reperfusion, respectively [Fig. 9(a)]. However, none of these differences were statistically significant. Similar to mRNA expression, OATP2 protein levels for the I/R injury group appeared to be higher than the sham group after 4 h of reperfusion [Fig. 9(b) and 9(c)]. MRP2 protein levels were comparable between sham and I/R injury groups at both time points. After 4 h of reperfusion, MRP3 levels in the I/R injury group were significantly lower than the sham [Fig. 9(c)]. At 24 h, MRP3 protein levels were comparable to the sham group.

Fig. 9

mRNA (a) and protein expression (b) and (c) of the transporters (MRP2, OATP2, and MRP3) involved in the uptake and excretion of both fluorescein and fluorescein monoglucuronide within the rat liver of sham and I/R injury groups. (b) Western blots of MRP2, OATP2, MRP3, and $\beta$-actin expression from four independent rats in sham and I/R injury groups (4 and 24 h after reperfusion). (c) Densitometry quantification of protein expression of OATP2, MRP2, and MRP3 (normalized to $\beta$-actin) of sham and I/R injury groups (4 and 24 h of reperfusion). Data represent the $\mathrm{mean}\pm \mathrm{SEM}$ ($n=4$). Significant differences are indicated with an asterisk ($*p<0.05$).