Nanoparticle-Based in Vivo Investigation on Blood−Brain Barrier

Jun 29, 2004 - and Leu-Wei Lo*,§. Department of Applied Chemistry, National Chi-Nan University, Puli 545, Taiwan, Department of Education and Researc...
0 downloads 0 Views 275KB Size
Download PDF
Anal. Chem. 2004, 76, 4465-4471

Nanoparticle-Based in Vivo Investigation on Blood-Brain Barrier Permeability Following Ischemia and Reperfusion Chung-Shi Yang,†,‡ Chia-Hua Chang,§ Pi-Ju Tsai,| Wen-Yin Chen,| Fan-Gang Tseng,⊥ and Leu-Wei Lo*,§

Department of Applied Chemistry, National Chi-Nan University, Puli 545, Taiwan, Department of Education and Research, Taichung Veteran General Hospital, Taichung 400, Taiwan, Department of Engineering and System Science, National Tsing Hua University, Hsinchu 300, Taiwan, and Division of Medical Engineering Research and Center for Nanomedicine Research, National Health Research Institutes, Taipei 114, Taiwan

The blood-brain barrier (BBB) represents a significant impediment to a large variety of central nervous systemactive agents. In the current study, we applied fluorescent polystyrene nanospheres (20 nm) to study the BBB permeability following cerebral ischemia and reperfusion. A microdialysis probe was implanted in the cerebral cortex of an anesthetized rat injected with fluorescent polystyrene nanospheres. The circulating nanospheres extravasating to the brain extracellular fluids were collected by the probe. Fluorescence intensity in the microdialysates throughout the course of cerebral ischemia/ reperfusion was measured. Cerebral ischemia and reperfusion induced transient accumulations of extracellular nanospheres in the brain. The accumulation of nanospheres may result from their extravasation from the blood vessels. The concurrent cerebral oxygen levels monitored using oxygen-dependent quenching of phosphorescence decreased following ischemia and returned to their original levels after reperfusion. In conclusion, we demonstrated that high temporal resolution measurements of BBB permeability in vivo can be obtained using fluorescence polystyrene nanospheres and that these data correlate with changes of cerebral oxygen concentration. This present investigation indicates that nanoparticles have potential clinical applications involving drug delivery and determination of therapeutic efficacy and on-site diagnosis. Vascular permeability is a major determinant of local delivery of therapeutic drugs to target organs or tissues, for example, the delivery of chemotherapeutics to tumors. It is known that “pores” ranging from 2 to 20 nm in diameter mediate hydrophilic permeability in capillaries outside the brain.1 Brain features a * To whom the correspondence should be addressed. Phone: +886-226534401 ext 7185. Fax +886-2-26524141. E-mail:[email protected]. † National Chi-Nan University. ‡ Center for Nanomedicine Research, National Health Research Institutes. § Division of Medical Engineering Research, National Health Research Institutes. | Taichung Veteran General Hospital. ⊥ National Tsing Hua University. (1) Firth, J. A. J. Anat. 2002, 200, 541-548. 10.1021/ac035491v CCC: $27.50 Published on Web 06/29/2004

© 2004 American Chemical Society

blood-brain barrier (BBB) system, which is formed by the endothelium of the brain’s vessels, the basal membrane, and neuroglial cells. Although it prevents penetration of a large variety of central nervous system (CNS)-active agents, the BBB also protects against the transport of harmful substances to the brain. The BBB permeability can be increased following various CNS injuries including traumatic brain injury and stroke. This increase, which may result from the upregulation of pro-inflammatory cytokines such as tumor necrosis factor R or hypoxia-induced vascular endothelial growth factor (VEGF), is a central hallmark of inflammation and is the basis of edema in many acute and chronic disease states, including ischemia and reperfusion injury.2,3 The co-registration of positron emission tomography (PET) and magnetic resonance imaging (MRI) can be combined to determine changes in BBB transport characteristics.4 These techniques, though noninvasive, are expensive and too slow to detect changes in BBB permeability in response to dynamic variations in physiological conditions. High temporal resolution measurements for determination of BBB transport can be achieved using the indicator diffusion method, i.e., the carotid artery single injection technique.5 However, it is an indirect method and, unlike PET or MRI studies, cannot be used to establish the regional distribution of the BBB permeability. BBB permeability can also be determined using various in vitro models.6,7 However, such results may not correspond to in vivo findings. Another technique, intracerebral microdialysis, can provide continuous monitoring of levels of compounds within discrete brain regions of a single animal and has been widely used for pharmacological and physiological studies to assay endogenous neurotransmitters or exogenous compounds.8,9 (2) Mark, K. S.; Miller, D. W. Life Sci. 1999, 64, 1941-1953. (3) Schoch, H. J.; Fischer, S.; Marti, H. H. Brain 2002, 125, 2549-2557. (4) Lin, K. P.; Huang, S. C.; Baxter, L.; Phelps, M. E. IEEE Trans. Nucl. Sci. 1994, 41, 2850-2855. (5) Hasselbalch, S.; Knudsen, G.. M.; Holm, S. J. Cereb. Blood Flow Metab. 1996, 16, 659-666. (6) Dehouck, M.-P.; Cecchello, R.; Green, A. R.; Renftel, M.; Lundquist, S. Brain Res. 2002, 955, 229-235. (7) Lee, S. W.; Kim, W. J.; Choi, Y. K.; Song, H. S.; Son, M. J.; Gelman, I. H.; Kim, Y. J. Nat. Med. 2003, 9(7), 900-906. (8) Ungerstedt, U. In Measurement of Neurotransmitter Release In Vivo; Marsden, C. A., Ed.; Wiley and Sons: New York, 1984; pp 210-245. (9) De Lange, E. C. M.; Zurcher, C.; Danhof, M. Brain Res. 1995, 702, 261265.

Analytical Chemistry, Vol. 76, No. 15, August 1, 2004 4465

Figure 1. Positioning of the microdialysis probe. The probe was implanted in specific sites of the brains of anesthetized rats to in situ monitor the extravasation of administered fluorescent nanospheres into the brain across the blood-brain barrier. The nanospheres enter the extracellular interstitial fluids, with or without physiological stimuli such as ischemia and reperfusion, and are collected by microdialysis locus in quo for quantitative analysis.

Nanoparticles have been used to more efficiently deliver drugs, diagnostics, vaccines, and anticancer drugs, especially owing to their tendency to accumulate in inflamed areas of the body.10,11 Fluorescent polystyrene beads coupled to biologically active molecules such as proteins or nucleic acids can be used for selective cellular imaging.12,13 Though fluorescent nanoparticles have been used in various in vitro biomedical investigations, their use in vivo has not been commonly reported probably because in vivo monitoring is difficult. In this report, we used a novel approach involving in vivo microdialysis in combination with nanoparticles to determine the BBB permeability to nanoparticles (20 nm) under ischemic and reperfusion conditions. Specific sites in the brains of fluorescent nanoparticle-injected rats were monitored for accumulation of these particles with the aid of an implanted microdialysis probe. The nanospheres extravasated from blood vessels into the extracellular interstitial fluids with or without physiological stimuli (such as ischemia and reperfusion) were collected by microdialysis locus in quo for quantitative analysis (Figure 1). Since cerebral oxygenation controls vascular permeability by multiple mechanisms,3,14-17 we incorporated the method of oxygendependent quenching of phosphorescence into our nanoparticlebased in vivo microdialysis system to quantitatively determine the cerebral oxygen level. Oxygen-dependent quenching of phosphorescence can provide rapid and quantitative measurements of (10) Couvreur, P.; Kante, B.; Grislain, L.; Roland, M.; Speiser, P. J. Pharm. Sci. 1982, 71, 790-792. (11) Majeti, N. V.; Kumar, R. J. Pharm. Pharm. Sci. 2000, 3, 234-258. (12) Smith, C. L. J. Neurosci. 1994, 14 (1), 384-398. (13) Di, A.; Krupa, B.; Bindokas, V. P.; Chen, Y.; Brown, M. E.; Palfrey, H. C.; Naren, A. P.; Kirk, K. L.; Nelson, D. J. Nat. Cell Biol. 2002, 4(4), 279285. (14) Kreuter, J. In Encyclopedia of Pharmaceutical Technology; Swarbrick, J., Boylan, J. C., Eds.; Marcel Dekker: New York, 1994; Vol. 10, pp 165-190. (15) Kee, H. J.; Koh, J. T.; Kim, M.-Y.; Ahn, K. Y.; Kim, J. K.; Bae, C. S.; Park, S. S.; Kim, K. K. J. Cereb. Blood Flow Metab. 2002, 22, 1054-1067. (16) Marti, H. J.; Bernaudin, M.; Bellail, A.; Schoch, H.; Euler, M.; Petit, E.; Risau, W. Am. J. Pathol. 2000, 156, 965-976. (17) Kreuter, J. Adv. Drug Delivery Rev. 2001, 47, 65-81.

4466 Analytical Chemistry, Vol. 76, No. 15, August 1, 2004

tissue oxygen levels (i.e., two-dimensional imaging of oxygen distribution throughout the tissue).18,19 The recent development of this method now allows noninvasive assessment of tissue oxygenation in vivo using a multifrequency system to perform time-resolved phosphorescence analysis.20 The principle of phosphorescence quenching has been described in detail elsewhere.21-23 To real-time monitor cerebral oxygenation after middle cerebral artery occlusion (MCAO) and reperfusion, and after accounting for the weak collected phosphorescent signal and the high scattering properties of tissue, the phosphorescence lifetime measurements in the frequency domain are found to be preferable to those in the conventional time domain.24-26 In the present study, we developed an analytical method that combined the in vivo microdialysis sampling technique and the phosphorescence lifetime measurement to simultaneously monitor changes of extracellular nanosphere concentrations and oxygen levels in the brain. This method was applied to study correlations between oxygen concentration and blood-brain barrier permeability to extravasated nanospheres, following cerebral ischemia and reperfusion. EXPERIMENTAL SECTION Animal Model of Ischemia and Reperfusion. Male Sprague-Dawley rats (260-350 g) were used. The animals were anesthetized with urethane (1.2 g/kg, ip), and body temperature was maintained at 37 °C with a heating pad. Polyethylene catheters were inserted into the femoral artery for the administration of fluorescent nanospheres or G2 phosphorescent dye. The rat’s head was mounted on a stereotaxic apparatus (Davis Kopf Instruments, Tujunga, CA) with the nose bar positioned 3.3 mm below the interaural line. Following a midline incision, the skull was exposed and one burr hole was drilled on the skull for inserting a dialysis probe. The microdialysis probe was implanted into the cortex (-0.5 mm anterior and 5.5 mm lateral to the bregma and 4.0 mm from the brain surface). The bifurcated light guide for cerebral oxygen measurement was placed with its tip 1 mm from the surface of dura in the same cortex area where the microdialysis probe implanted. Cerebral ischemia was induced by the ligation of the bilateral common carotid arteries and unilateral middle cerebral artery. Intracranial cerebral blood flow was measured with a laser Doppler system. In Vivo Microdialysis and Nanospheres. The microdialysis system was perfused with 50 mM phosphate buffer solution (pH 7.4) at a flow rate of 2 µL/min. Microdialysis probes, made of polyethylene sulfonate with a molecular weight cutoff at 100 000, were purchased from CMA (Carnegie Medicine Association, Solna, Sweden). The microdialysates were allowed to flow through a polyimide-coated fused-silica capillary (360-µm o.d., 250-µm i.d.). (18) Wilson, D. F.; Cerniglia, G. J. Cancer Res. 1992, 52, 3988-3993. (19) Vinogradov, S. A.; Lo, L.-W.; Jenkins, W. T.; Evans, S. M.; Koch, C.; Wilson, D. F. Biophys. J. 1996, 70, 1609-1617. (20) Vinogradov, S. A.; Fernandez-Seara, M. A.; Dugan, B. W.; Wilson, D. F. Comp. Biochem. Physiol. A 2002, 132, 147-152. (21) Vanderkooi, J. M.; Maniara, G.; Green, T. J.; Wilson, D. F. J. Biol. Chem. 1987, 262, 5476-5482. (22) Gewehr, P. M.; Delpy, D. T. Med. Biol. Eng. Comput. 1993, 31, 2-21. (23) Lo, L.-W.; Koch, C. J.; Wilson, D. F. Anal. Biochem. 1996, 236, 153-160. (24) Pawlowski, M.; Wilson, D. F. Adv. Exp. Med. Biol. 1992, 316, 179-185. (25) Alcala, J. R.; Yu, C.; Yeh, G. J. Rev. Sci. Instrum. 1993, 64, 1554-1560. (26) Lakowicz, J. R. Principles of the Fluorescence Spectroscopy, 2nd ed.; Kluwer Academic and Plenum: New York, 1999.

A 2-cm segment of the capillary was stripped of the polyimide coating to serve as the detection window and placed inside a fluorescence detector (Argos 250 FL detector, Flux Instruments, Basel, Switzerland). Carboxylated polystyrene nanospheres, modified with fluorescence dye (F8781) were purchased from Molecular Probes, Inc. (Eugene, OR). Stock solutions (particle/mL) were diluted with the 50 mM phosphate buffer solution. The spectra of absorption and emission (absorption 365 nm, emission 412 nm) were then determined (Cary Eclipse Spectrophotometer, Varian Inc., Palo Alto, CA). The nanosphere solutions were administered intravenously (as a bolus) into the animals. The fluorescence intensity of nanoparticles in interstitial fluid sampled with microdialysis was monitored with a laser-induced fluorescence detector (Picometrics S. A., Ramonville, France). The excitation laser at 320 nm was used to prevent the Raman effect of water from contaminating the fluorescence emission, which was measured at 418 nm. In Vivo Cerebral Oxygen Monitoring and Phosphorescence Dye. The bifurcated light guide for measuring the phosphorescence was positioned 1 mm above the dura to one side of the implanted microdialysis probe without any physical interference. The light from the optical fiber passed through a 795 ( 40 nm interference filter (Omega Optical, Brattleboro VT) and a light guide adapter with ball lens was machined to filter and conduct the input phosphorescence signal onto the Hamamatsu C546001 APD module. The phosphorescence lifetime was resolved in the frequency domain (PMOD 5000, Oxygen Enterprises, Philadelphia, PA) and used to calculate the oxygen pressure. The near-infrared, oxygen-sensitive phosphorescence dye named G2 (Oxygen Enterprises) was used to monitor real-time changes in cerebral oxygenation. The spectral absorption and phosphorescence maximums of the dye were 636 and 805 nm, respectively (Cary Eclipse spectrophotometer). For phosphorescence measurements, 1.0 mL/kg G2 (3.0 mg/mL) was injected intravenously into the rat. Atomic Force Microscopy (AFM) for Nanosphere Validation. The atomic force microscope was used to validate the fluorescence signals detected by microdialysis were indeed attributed to the fluorescent nanospheres. A piece of Pyrex glass was soaked in acetone and sonicated for 5 min to clean the surface. The glass was then washed with double-distilled water and dried with an air blower. Microdialysis samples were sonicated for 10 min, and an aliquot of 2.5 µL of each sample was pipetd onto the Pyrex glass. The samples were stored in a cleanroom. AFM images were taken after the samples had been completely dried. All AFM images were acquired with JPK Instruments NanoWizard AFM. Contact-mode silicon cantilevers (CSC38, MicroMasch) had force constants of 0.01-0.08 N/m and tip apex radii of ∼10 nm. Images were aquired at 512 × 512 pixels with scan rate of 0.7-1.0 Hz. RESULTS AND DISCUSSION To monitor concentrations of interstitial nanospheres presumably extravasated from the blood-brain barrier and to correlate the changes simultaneously with the cerebral oxygen profile under ischemia and reperfusion, we developed an analytical method as illustrated in Figure 2. The system was a combination of microdialysis for the in situ sampling of fluorescent nanospheres and

Figure 2. Schematic diagram of the instrumentation. The combination of in vivo microdialysis with a fluorescence detection module and a phase-lock phosphorometer enable simultaneous measurement of extravasated nanosphere concentration in microdialysates and changes of intravascular oxygen concentration during ischemia and reperfusion.

phosphorescence lifetime measurement for the monitoring of cerebral oxygenation. To achieve both in situ measurements simultaneously, the optical characteristics of both chromophores have to be distinct, i.e., have minimal spectral crossover. The fluorescence nanosphere has absorption at 365 nm and emission at 412 nm (Figure 3A). In contrast, the G2 compound used for phosphorescence lifetime measurement characterizes its Q-band absorption at 636 nm and emission at 805 nm; both are in the near-infrared region of spectrum (Figure 3B). This combination was used to minimize concurrent “contamination” of optical measurements for both chromophores. The sampling capability of our microdialysis system for 20nm fluorescence nanospheres was validated in vitro at various nanosphere concentrations. Three concentrations of nanosphere solutions were tested (1.30 × 1013, 6.50 × 1012, and 3.25 × 1012 nanospheres/mL). When the microdialysis probe was placed in the solution and perfused at a flow rate of 2 µL/min with nanospheres, the nanosphere concentration in the perfusates gradually increased as a result of nanospheres crossing the membrane of the microdialysis probe. Approximately 20 min after the perfusion, the nanosphere concentrations reached a maximum, which reflected attainment of concentration equilibria between the inside and outside of the microdialysis probe. The fluorescence intensities corresponding to the three nanosphere concentrations were 259 ( 43.7, 131 ( 11.5, and 49 ( 2.7 mV, respectively (Table 1). These results from four separate measurements demonstrated that the microdialysis probe (made of polyethylenesulfonate) was effective in sampling the 20-nm nanospheres, and the on-line fluorescence detection module was effective in analyzing nanosphere concentration. As the microdialysis is a well-known technique competent for sampling of small molecules such as extracellular neurotransmitters, collecting of the relatively large particles such as 20-nm nanospheres, especially at higher concentration, through the probe membrane remains as a challenging task. It may reflect in our measurements of fluorescence intensity versus nanosphere concentration; the relative standard deviation increased with increase of the concentration of nanosphere. Nevertheless, our results suggested that the perfusion technique Analytical Chemistry, Vol. 76, No. 15, August 1, 2004

4467

Figure 4. In vivo measurement of BBB permeability under normoxia. After stabilization of the implanted microdialysis probe for 130 min, G2 solution (0.5 mL × 3.0 mg/mL) was injected iv at 161 min (arrow A) and the nanosphere solution (2.6 × 1014 nanospheres/mL) was injected (iv, bolus) at 195 min (arrow B) through the femoral vein to achieve the final concentration of ∼1.3 × 1013 nanospheres/mL in the blood. Under normoxia, nanosphere concentration in the microdialysates was maintained at baseline level, indicating that the intravenously injected nanospheres remained in the vasculature.

Figure 3. Optical characteristics of the fluorescent nanospheres and phosphorescent dye used to measure BBB permeability and cerebral oxygen levels. The fluorescent nanospheres have absorption and emission maximums at 365 and 412 nm, respectively (A), whereas G2 has absorption and emission maximums in the nearinfrared at 636 and 805 nm, respectively (B). This combination with minimal spectral crossover ensured both measurements could be made independently. Table 1. In Vitro Nanosphere Concentration-Dependent Microdialysis Measurementsa nanosphere concentration (particles/mL)

fluorescence intensity (mV)

1.30 × 1013 6.50 × 1012 3.25 × 1012

259 ( 43.7 131 ( 11.5 49 ( 2.7

a Values given are averages of repeated measurements from multiple experiments.

could be used to sample extracellular nanospheres in situ in the brain. Accordingly, a microdialysis probe was implanted into the brain of an anesthetized rat. After a 130-min microdialysis probe stabilization period, G2 solution (0.5 mL × 3.0 mg/mL) was injected iv at 161 min and the nanosphere solution (2.6 × 1014 nanospheres/mL) was injected iv at 195 min through the femoral vein to achieve the final concentration of ∼1.3 × 1013 nanospheres/ 4468 Analytical Chemistry, Vol. 76, No. 15, August 1, 2004

mL in the blood. Under normoxia, the nanosphere concentration in the microdialysates remained at baseline levels, indicating that the intravenously injected nanospheres remained in the vasculature (Figure 4). Under ischemia and reperfusion, a series of control experiments were performed to validate the feasibility of simultaneous in vivo measurements of BBB permeability and cerebral oxygen profile. Without the administration of G2 and fluorescent nanospheres, ischemia and reperfusion incurred no change of fluorescence intensity from microdialysates, suggesting no endogenous fluorescence substance was induced following such insults (Figure 5A). While administering G2 solution only, the fluorescence intensity maintained at its baseline throughout the ischemia and reperfusion episodes (Figure 5B). This indicated that the G2 compound has minimal interference on nanosphere fluorescence detection under the current optical setup (refer to Experimental Section). With the administration of fluorescent nanospheres alone, the fluorescence intensity measured in the microdialysates represented the nanospheres that passed the BBB and entered the implanted probe via the extracellular space in the brain. As shown in Figure 5C, the onset of ischemia induced by occlusion of the MCAO was followed by an increase in fluorescence intensity, which returned to the baseline after ∼10 min. Upon reperfusion, a second increase in fluorescence intensity was observed; however, it was more pronounced than the response to ischemia. Subsequently, the effect of ischemia and reperfusion on BBB permeability was investigated by simultaneously measuring extracellular nanosphere concentration and cerebral oxygen concentration during ischemic and reperfusion episodes. The G2 and nanosphere solutions were applied as with those for a normaxia experiment; the response of fluorescence intensity to ischemia and reperfusion was similar to that when administering nanosphere solution only (Figure 6). The pattern in changes of fluorescence intensity to ischemia and reperfusion was consistent

Figure 6. Effect of ischemia and reperfusion on BBB permeability to nanospheres, defined by extracellular nanosphere fluorescence intensity (s) and the cerebral oxygen concentration (- - -). The onset of MCAO was followed by an increase in fluorescence intensity, which returned to baseline after ∼10 min. Upon reperfusion, a second increase in fluorescence intensity was observed. The intravascular oxygen concentration (mainly in the middle cerebral artery) dropped rapidly from 60 mmHg at normaxia to ∼10 mmHg and then gradually returned to 20 mmHg during the 1-h ischemic episode. A rapid reperfusion-induced recovery of intravascular oxygen concentration to 60 mmHg in ∼10 min is shown. This is the representative from six independent animal experiments, all of which demonstrated similar profile of responses.

Figure 5. Control experiments for in vivo measurements of BBB permeability and cerebral oxygen profile following ischemia and reperfusion. (A) The microdialysis was positioned to the animal without preadministering of G2 and fluorescent nanosphere, the ischemia was induced by occlusion of the middle cerebral artery (MACO) at 60 min (arrow A), and the reperfusion was given at 90 min (arrow B) after the stabilization period, respectively. No change of fluorescence intensity from microdialysates was observed, suggesting no endogenous fluorescence substance was induced following such insults. (B) Under the condition of preinjection of G2 solution only, the fluorescence intensity maintained at its baseline throughout the ischemia (arrow A) and reperfusion (arrow B) episodes. As shown in (C), the onset of ischemia induced by MCAO (arrow A) was followed by an increase in fluorescence intensity, which returned to the baseline after ∼10 min. Upon reperfusion (arrow B), a second increase in fluorescence intensity was observed; however, it was more pronounced than the response to ischemia.

and independent of the sequence of G2 and nanosphere solutions introduced to the animal (data not shown). It was observed that the increase of fluorescence intensity consequent to ischemia and reperfusion was prominent but, in some cases, smaller in amplitude than that measured in an animal injected with nanospheres only. Thus, the possibility was not completely excluded that the Soret band absorption of G2 at 442 nm may quench the fluorescence of the nanosphere at 412 nm with a mechanism of either reabsorption or fluorescence resonance energy transfer. At any rate, it exerted no significant interference in our acquiring of a fluorescence profile in response to ischemia and reperfusion. The intravascular oxygen concentration was measured simultaneously, using oxygen-dependent quenching of phosphorescence, as mentioned in the introduction and Experimental Section. The intravascular oxygen concentration (mainly in the middle cerebral artery) dropped rapidly from 60 mmHg at normoxia to ∼10 mmHg at hypoxia and then gradually returned to 20 mmHg during the 1-h ischemic episode. Onset of reperfusion was marked by a rapid recovery of intravascular oxygen concentration to 60 mmHg in ∼10 min (Figure 6). Cerebral blood flow, before and after ischemia and reperfusion, was monitored by a laser blood flow monitor. The onset of ischemia caused a rapid drop in cerebral blood flow, and the blood flow remained at that low level during the 1-h ischemic period. The blood flow level returned to its preischemic level immediately after the onset of reperfusion (data not shown). The response of extracellular nanosphere fluorescence intensity to the ischemia and reperfusion episodes consistently lagged ∼5 min behind that of cerebral oxygen levels. The lag might be attributed to both the instrumental and physiological factors. Unlike the phosphorescence emission measurement of cerebral oxygen, which was immediate and on-site, the fluorescence measurement of the microdialysates required time for the extracellular nanospheres to perfuse into the implanted microdialysis Analytical Chemistry, Vol. 76, No. 15, August 1, 2004

4469

Figure 7. AFM images of the nanosphere in microdialysates prepared from both in vitro and in vivo experiments. The AFM images of in vitro microdialysates from two nanosphere solutions with concentrations of 2.6 × 1010 (A) and 2.6 × 1012 nanospheres/mL (B) showed the concentrationdependent increase of nanospheres that appeared in the scanning area. The AFM images of in vivo microdialysates collected from extracellular fluids of rat brain, before (C) and after (D) cerebral ischemia, illustrated a significant increase of nanospheres after the ischemic insult. The surface information of each AFM image was scanned with a contact-mode silicon cantilever and denoted with a color palette displayed aside each image.

probe and for the fluorescence detection module located ex vivo to respond. The time between nanosphere sampling and detection may be one of the reasons for the observed 5-min lag. Another reason could be that the onset of BBB permeability could be delayed by the time needed to activate the mediators of BBB permeability. Since the increase of intracranial pressure was reported as one of the most frequent neurological complications following acute ischemic stroke,27 it might contribute, at least in part, to the much faster sampling dynamics of nanosphere by microdialysis in vivo than that in vitro. To provide the physical evidence of the presence of nanospheres in the microdialysate to demonstrate the detected fluorescence signal was contributed from the nanosphere, the AFM was applied to acquire the nanosphere images of microdialysate samples prepared from both in vitro and in vivo experiments. The AFM images of two in vitro nanosphere microdialysates are shown (Figure 7A and B), which were perfused at a flow rate of 2 µL/min from nanosphere solutions with concentrations of 2.6 × 1010 and 2.6 × 1012 nanospheres/mL, respectively. The number of nanosphere appeared in the AFM imaging scanning area increased as the concentration of nanosphere (27) Weimar, C.; Roth, M. P.; Zillessen, G.; Glahn, J.; Wimmer, M. L. J.; Busse, O.; Haberl, R. L.; Diener, H.-C. Eur. Neurol. 2002, 48, 133-140.

4470 Analytical Chemistry, Vol. 76, No. 15, August 1, 2004

solution increased. The in vivo microdialysates collected from extracellular fluids of rat brain before and after cerebral ischemia were prepared for AFM imaging. Before ischemia, the nanosphere was rarely found in the microdialysate as illustrated with AFM (Figure 7C). On the contrary, after ischemia, AFM imaging indicated the increase of nanosphere concentration in the microdialysate (Figure 7D), suggesting the increase of BBB permeability for 20-nm nanospheres following ischemic insult. The surface information of each AFM image was scanned with contactmode silicon cantilever and displayed with a color palette denoting the surface height in the range of 0 to 30 nm. Microdialysis perfusion provides a minimally invasive method to sample extracellular fluids in anesthetized and awake animals. Although implanted probes are commonly used to sample small molecular weight extracellular neurotransmitters and metabolites, as well as polypeptides such as insulin (5500) and tumor necrosis factor R (12 500), they have not been used to sample and collect in vivo extracellular nanoparticles. In this report, we implanted a microdialysis probe with a dialysis membrane that had a higher molecular weight cutoff (MWCO) to sample the extracellular nanospheres extravasated from blood vessels following cerebral ischemia and reperfusion in an anesthetized rat. We believe that this is a minimally invasive way to assess in vivo BBB permeability

following experimental brain injury. BBB permeability has been characterized on the basis of the concentration (fluorescence intensity) of circulating nanospheres that entered the implanted microdialysis probe via the BBB. Several mechanisms of ischemia- and reperfusion-induced extravasation have been documented.17 Suppression of cerebral blood flow by MCAO may increase retention of the nanospheres in the brain’s capillaries, augment passive diffusion driven by local concentration gradients, and enhance the transport of nanospheres across the endothelial cell layer and thereby the delivery of nanospheres to the microdialysis probe. However, the passive diffusion alone would probably be insufficient to overcome the efflux generated by pumps located in the luminal side of the endothelial cell membranes, such as P-glycoprotein. The loss of blood flow alone has also been found not to acutely affect BBB function owing to the protective cascade of mechanisms involving cytokines and nitric oxide.28 In contrast, the increase of nanosphere extravasation from BBB might be due to cerebral ischemiainduced osmotic stress that leads to the opening of the tight junctions between endothelial cells.17 There is evidence that receptor-mediated endocytosis and transcytosis can provide a route for permeation of specific peptides and proteins (such as transferrin and insulin) through BBB. However, whether the bound peptide/protein is itself transcytosed remains an issue.29 All these mechanisms might be involved simultaneously or consecutively in the transport of nanospheres through BBB under ischemia and reperfusion. There are a number of reports indicating that hypoxia-induced VEGF expression plays a critical role in the augmentation of vascular leakage following tissue hypoxia, ischemia, or both.3,15 VEGF is a potent vascular permeability factor that increases microvascular permeability to blood plasma proteins within minutes after exposure, and administration of recombinant human VEGF165 after focal cerebral ischemia may exacerbate BBB permeability.30 Additionally, Marti et al. demonstrated that VEGF expression was strongly upregulated in cells within the ischemic border zone between 6 and 24 h after MCAO.16 However, we observed an immediate increase of extracellular nanosphere concentration following MCAO. With such a prompt response, (28) Krizanac-Bengez, L.; Kapural, M.; Parkinson, F.; Cucullo, L.; Hossian, M.; Mayberg, M. R.; Janigro, D. Brain Res. 2003, 977, 239-246. (29) Abbott, N. J. Cell. Mol. Neurobiol. 2000, 20(2), 131-147. (30) Zhang, Z. G.; Zhang, L.; Jiang, Q.; Zhang, R.; Davies, K.; Powers, C.; van Bruggen, N.; Chopp, M. J. Clin. Invest. 2000, 106, 829-838.

the increase of BBB permeability was unlikely attributed to the VEGF expression. CONCLUSIONS A microdialysis perfusion technique has been used to sample extracellular nanospheres in the brains of anesthetized rats. The effect of cerebral ischemia and reperfusion on the nanosphere extravasation was also examined. The intravascular concentration of nanosphere locus in quo was directly correlated to the measured fluorescence intensity of nanosphere-containing microdialysate. Since the blood flow dramatically decreased during the period of ischemic occlusion, it reduced the amount of local intravascular nanospheres entering the occluded site. It should be noted that, by normalizing the observed factor in the increase of fluorescence intensity with decreased blood flow, the augmentation of ischemiainduced BBB permeability, following MCAO, would be much more significant than what we demonstrated. However, this underestimation would not occur in the intravascular PO2 measurement since the lifetime of phosphorescence, in lieu of intensity, had been applied for analysis. We have demonstrated a novel method to monitor vascular permeability, especially during ischemia and reperfusion injury. Using a combination of fluorescence nanospheres, in vivo microdialysis, and oxygen-dependent quenching of phosphorescence, we were able to make high temporal resolution measurements of BBB permeability and correlate these with changes in intravascular oxygen concentration. ACKNOWLEDGMENT This research work was supported equally by NHRI Intramural Research Grant ME-092-PP-05 from the National Health Research Institutes of Taiwan and by Grant NSC 91-2213-E-260-013 from the National Science Council of Taiwan. We are truly indebted to Messrs. Yuan Tai Tseng and Chun-Jun Lin for their professional endeavors on the operation of atomic force microscopy and the interpretation of experimental data.

Received for review December 17, 2003. Accepted May 25, 2004. AC035491V

Analytical Chemistry, Vol. 76, No. 15, August 1, 2004

4471