Demonstration of impaired neurovascular coupling responses in TG2576 mouse model of Alzheimer’s disease using functional laser speckle contrast imaging

Abstract

Increasing evidence from epidemiological, clinical, and experimental studies indicates that cerebromicrovascular dysfunction and microcirculatory damage play critical roles in the pathogenesis of many types of dementia in the elderly, including both vascular cognitive impairment (VCI) and Alzheimer’s disease. Vascular contributions to cognitive impairment and dementia (VCID) include impairment of neurovascular coupling responses/functional hyperemia (“neurovascular uncoupling”). Due to the growing interest in understanding and pharmacologically targeting pathophysiological mechanisms of VCID, there is an increasing need for sensitive, easy-to-establish methods to assess neurovascular coupling responses. Laser speckle contrast imaging (LSCI) is a technique that allows rapid and minimally invasive visualization of changes in regional cerebromicrovascular blood perfusion. This type of imaging technique combines high resolution and speed to provide great spatiotemporal accuracy to measure moment-to-moment changes in cerebral blood flow induced by neuronal activation. Here, we provide detailed protocols for the successful measurement in neurovascular coupling responses in anesthetized mice equipped with a thinned-skull cranial window using LSCI. This method can be used to evaluate the effects of anti-aging or anti-AD treatments on cerebromicrovascular health.

Neurovascular uncoupling in aging and Alzheimer’s disease

It is well recognized that the brain consumes more energy than any other human organ. Over 20% of the body’s total energy requirements are spent to fuel the brain, which in turn only accounts for 2% of the total body mass. Moment-to-moment regulation of cerebral blood flow (CBF) is crucial since inadequate supply of glucose and oxygen to an active region of the brain would cause cell dysfunction or injury within a very short time frame. In healthy subject during times of increased neural activity, a homeostatic mechanism termed neurovascular coupling (functional hyperemia) matches the localized demand for glucose and oxygen with increased blood supply to ensure normal brain function. Neurovascular coupling is a feed-forward mechanism which requires the coordinated cellular interaction between neurons, astrocytes, pericytes, vascular endothelial and smooth muscle cells (Petzold and Murthy 2011; Stobart et al. 2013; Wells et al. 2015; Chen et al. 2014; Tarantini et al. 2016). A large body of evidence derived from both clinical and experimental studies demonstrate that aging significantly impairs neurovascular coupling responses, which likely contribute to cognitive decline in the elderly (Balbi et al. 2015; Fabiani et al. 2013; Sorond et al. 2013; Tong et al. 2012; Toth et al. 2014; Zaletel et al. 2005; Park et al. 2007). There is also growing evidence for microvascular pathophysiological alterations having a causal role in the development of cognitive decline associated with Alzheimer’s disease (AD) (Tarantini et al. 2016; Snyder et al. 2015).

An early role of vascular dysregulation in the progression of AD was underscored by recent studies of late onset AD using brain images and plasma biomarkers from the Alzheimer’s Disease Imaging Initiative (ADNI) (Iturria-Medina et al. 2016). Vascular dysregulation in AD includes deficiencies in cerebrovascular reactivity, CBF, and neurovascular coupling responses (Girouard and Iadecola 2006; Gorelick et al. 2011; Hock et al. 1997; Rombouts et al. 2000). Neurovascular coupling dysfunction of AD has been replicated in experimental studies showing that in mouse models of AD, neurovascular coupling is also significantly impaired (Rancillac et al. 2012; Shin et al. 2007; Royea et al. 2017), at least in part, due to enhanced oxidative stress (Nicolakakis et al. 2008; Park et al. 2008; Park et al. 2005) arising from mitochondrial dysfunction and inflammation (Lacoste et al. 2013; Ongali et al. 2014). Importantly, recent evidence suggests that pharmacological interventions that rescue functional hyperemia result in improved cognitive function in mice with AD pathologies (Tong et al. 2012; Nicolakakis et al. 2008). Due to the increased realization that understanding of the mechanisms underlying neurovascular dysfunction is critical for developing novel therapeutic interventions to prevent or treat AD, there is an increasing need in many laboratories to adapt methodologies to investigate neurovascular coupling responses in mouse models of aging and AD (Lacoste et al. 2013; Ongali et al. 2014; Papadopoulos et al. 2016; Hamel et al. 2016; Nicolakakis and Hamel 2011; Papadopoulos et al. 2014). In this paper, published as part of the “Methods for Geroscience” series in the “Translational Geroscience” initiative of the journal (Callisaya et al. 2017; Kane et al. 2017; Kim et al. 2017; Liu et al. 2017; Meschiari et al. 2017; Perrott et al. 2017; Shobin et al. 2017; Ashpole et al. 2017; Bennis et al. 2017; Deepa et al. 2017; Grimmig et al. 2017; Hancock et al. 2017; Konopka et al. 2017; Podlutsky et al. 2017; Sierra and Kohanski 2017; Tenk et al. 2017; Ungvari et al. 2017a; Ungvari et al. 2017b; Urfer et al. 2017a; Urfer et al. 2017b), we present an easy-to-adapt protocol for assessment of neurovascular coupling responses in mice in both geroscience and AD research. As in these studies, experimental animals usually undergo behavioral testing prior to terminal experimentation; we aimed to provide a protocol that is relatively fast allowing investigators to process larger cohorts of animals. In our experience, assessment of neurovascular coupling responses in 10–15 animals per week is realistic using this protocol.

Laser speckle contrast imaging (LSCI) for assessment of neurovascular coupling responses

The accurate measurement of changes in local CBF in response to neuronal activation is essential for the assessment of the efficacy of physiological neurovascular coupling or its age- or disease-related dysfunction in experimental models. The traditional, real-time monitoring of local CBF in the cerebral cortex relies on laser Doppler flowmetry (i.e., the measurement of velocity with the aid of the frequency shift caused by the Doppler effect), which is a valid approach with excellent temporal resolution, but provides no information as to the spatial variation of flow. Still, spatial resolution is desired when a small microvascular bed responding to the activity of a distinct neuron population needs to be identified and monitored (e.g., within the barrel cortex of the mouse) (Ayata et al. 2004), or when irregular flow patterns are to be visualized in experimental models of cerebral ischemia (Bere et al. 2014). Laser Doppler flowmetry can be applied in a scanning mode to obtain two-dimensional relative flow maps (Lauritzen and Fabricius 1995), with the limitation that the mechanical scan may not provide high enough resolution (Briers 2001; Tew et al. 2011).

As an alternative to Laser Doppler flowmetry, LSCI was first introduced for the mapping of microvascular perfusion in various tissues including the skin and the retina (Briers 2001; Ruth 1990; Tamaki et al. 1994), and was later adapted and found highly suitable to create flow maps of the superficial layers of cerebral cortex (Dunn et al. 2001). LSCI flow maps are computed using fluctuating intensity of the random interference effect known as speckle; still, LSCI and laser Doppler flowmetry both derive flow information on the basis of the same physical phenomenon and yield comparable results (Briers 2001; Tew et al. 2011). With regard to the cerebral cortex, comprehensive evaluation of LSCI against laser Doppler flowmetry has demonstrated that the two approaches deliver correlating flow data and are equally valid and powerful, with LSCI having the advantage of a good spatial resolution (Ayata et al. 2004). In particular, laser Doppler flowmetry and LSCI were found similarly suitable for the characterization of CBF changes in response to whisker stimulation, CO₂ challenge, or after middle cerebral artery occlusion in rodents (Ayata et al. 2004).

A distinct additional benefit of using LSCI is that it can be effectively combined with other imaging modalities, allowing the exact spatial and temporal correlation of optical signals. For instance, relative changes in cerebral blood volume and hemoglobin saturation can be achieved by recording intrinsic optical signals at specified wave lengths (i.e., green or red, respectively) simultaneous with CBF variations visualized by LSCI (Bere et al. 2014; Farkas et al. 2010). In addition, spectroscopic measurements using multiple wavelengths—rather than a single light source of a specific, narrow range—can yield quantitative data on hemoglobin saturation parallel with relative changes in CBF assessed by LSCI (Dunn et al. 2003). Finally, LSCI has been very successfully integrated into multi-modal imaging systems, which visualize membrane potential changes in the cerebral cortex (i.e., the intensity of the optical signal emitted by a voltage-sensitive fluorescent dye increases with decreasing transmembrane potential) (Farkas et al. 2010; Obrenovitch et al. 2009), or image variations of pH in the nervous tissue (i.e., fluorescence intensity of a pH-sensitive dye increases with deepening acidosis) (Menyhart et al. 2017). These approaches are highly pertinent and very powerful, because the exact spatial and temporal match of individual modalities offers the opportunity to draw specific conclusions about their coupling patterns (i.e., neuronal activation, metabolic status, and CBF).

Imaging apparatus

Many laboratories build their own setups for LSCI using a CCD camera with optics and custom-written image acquisition software. The protocol below was specifically optimized for experiments in geroscience and AD research for laboratories, whose primary expertise is not in cerebrovascular physiology, but who want to quickly adopt LSCI-based methods to evaluate cerebromicrovascular health and/or assess potential therapeutic interventions. The experiments shown in Fig. 1A were conducted using the commercially available high-resolution PeriCam PSI laser speckle imager (Perimed, Järfälla, Sweden) in 11-month old C57BL/6 and TG2576 mice overexpressing human APP. This device has high magnification optics, which resolves details of 20 μm/pixel in areas up to 20 × 27 mm with a fixed working distance of 10 cm. Individual data points of CBF changes in response to whisker stimulation are represented in Fig. 1B.

Fig. 1.

a Representative neurovascular coupling responses between C57BL/6 and TG2576 mouse overexpressing human APP. Representative images of blood perfusion maps (upper panel) obtained using laser speckle contrast imaging in age-matched wild-type control (left) and in the mice overexpressing human amyloid precursor protein (right). The differential perfusion maps in the middle and bottom panels show regional increases in cerebral blood flow (white arrows), specifically in contralateral somatosensory whisker barrel cortex during mechanical whisker stimulation. Anatomically, the whisker barrel cortex is located 1 mm rostral and 3 mm lateral from the bregma. Thinned skull preparations do not require the skull to be completely transparent (i.e., extremely thin) as laser speckle imaging can be performed through a partially thinned and smoothed skull. b Overexpression of human APP in TG2576 mice results in decreased neurovascular coupling responses. The figure represents individual data points of cerebral blood flow changes in response to whisker stimulation in C57BL/6 and TG2576 mice

Experimental procedures

1. Experiments using laboratory animals must be performed in accordance with institutional and federal guidelines. The procedures described here have been approved by the Animal Care and Use Committees of the participating institutions.

2. The surgeries described in the protocol are terminal. The methods are optimized for quick processing of larger cohorts of animals. This protocol can be completed within 3 h.

3. Mouse anesthesia: The following methods of anesthesia are appropriate for assessment of neurovascular coupling measurements in rodents: (1) isoflurane (Masamoto et al. 2007), (2) ketamine (85 mg/kg, i.m.) and xylazine (3 mg/kg, i.m.) (Tong et al. 2012), and (3) alpha chloralose (Norup Nielsen and Lauritzen 2001; Hillman et al. 2007). For isoflurane use, induce anesthesia with 4% isoflurane in oxygen mix in an induction chamber using a surgical isoflurane vaporizer (Harvard Apparatus). Monitor the surgical depth by observing absence of the toe pinch reflex. Mice sedated with alpha chloralose can be first induced with 4% isoflurane, and then given a 114-mg/kg i.p. injection of alpha chloralose (Low et al. 2016; White and Field 1987). Isoflurane dose should be decreased for approximately 15 min until the alpha chloralose takes effect. Alpha chloralose can be dissolved in an 80:20 mixture of 1× phosphate buffered saline: polyethylene glycol. At 60 min, alpha chloralose-sedated mice should be given an additional half-dose bolus. Note that alpha chloralose may have unwanted effects on heart rate and pCO₂, which may confound measurements, so careful monitoring of the animals is recommended (Low et al. 2016).

4. If desired, inject 1 mg/kg of dexamethasone (s.c., in the scruff of the neck) to reduce cerebral swelling and reduce airway secretions during surgery (Winship 2014).

5. Mice are endotracheally intubated and ventilated (MousVent G500; Kent Scientific Co., Torrington, CT). For endotracheal intubation, use the 20G plastic tube from intravenous catheter without the provided metal guide (Safelet Cath, Nipro Corp.). Connect the endotracheal tube to the mouse ventilator and monitor end-tidal CO₂ to keep blood gas values within the physiological range (Tarantini et al. 2015; Toth et al. 2015). Blood gases (pO₂ and pCO₂) and pH should be measured at the beginning and at the end of the experiment

6. Apply eye ointment (Liposic ophthalmic gel, Bausch and Lomb) onto the eyes to prevent desiccation.

7. Use a homeothermic temperature controller (Kent Scientific Co., Torrington, CT) to maintain rectal temperature at 37 °C (Toth et al. 2014).

8. Cannulate the right femoral artery to continuously monitor arterial blood pressure using a pressure transducer (Living Systems Instrumentations, Burlington, VT) (Toth et al. 2014). The femoral artery catheter can be used also for systemic drug administration or, alternatively, a venous catheter can be placed in the femoral vein.

9. Shave the skin overlying the desired imaging location.

10. Place the mouse in a stereotaxic frame (Leica Microsystems, Buffalo Grove, IL).

11. Inject 0.01 ml of the local anesthetic bupivicaine (5 mg/ml in saline, s.c.) at the incision path. Make a 1-cm longitudinal incision along the midline of the skull. Pull aside the skin to expose the skull and hold in place with bulldog serrefines. Remove the periosteum with fine forceps; clean the surface of the skull with sterile gauze and cotton tip applicators until dry and clear of blood.

12. Define the borders of the planned thinned-skull cranial window using a permanent marker.

13. Use a precision dental drill for thinning the skull over the region of interest until translucent. To avoid producing excess heat and potentially damaging the brain, continually move the drill bit around region of interest, using a stochastic pattern. Regularly flush the thinned area with cold HEPES-buffered ACSF to avoid heat-induced damage to the superficial layers of the brain (Fig. 2A). Use a scalpel for the final stages of thinning. The thickness of the skull is appropriate for LSCI when pial vessels are visible. HEPES-buffered ACSF: in 500 ml of distilled H₂O, NaCl 3.94 g, KCl 0.2 g, MgCl₂ × 6H₂O 0.102 g, CaCl₂ × 2H₂O 0.132 g, NaHEPES 0.651 g; adjust the pH to 7.4.

14. Once the skull is thinned, wipe dry the surface and apply a drop of cyanoacrylate (Fig. 2B). Once cured (after 5 min), administer a thin layer of nitrocellulose lacquer to the skull to allow for even light spread on the thinned bone surface (Fig. 2C). An alternative would be to apply a layer of low-melt agarose and cover it with a coverslip.

15. After 5 more minutes of curing time, place the mouse and frame under the laser speckle contrast imager (Perimed, Järfälla, Sweden) for imaging (Fig. 2D). The laser speckle contrast imager is placed 10 cm above the thinned skull.

16. The depth of the anesthesia should be monitored throughout the experiment (tail pinch). If isoflurane anesthesia is used at this time the isoflurane is lowered to 1% maintenance dose. Higher dose of isoflurane may result in loss of autoregulation. The arterial blood pressure should be monitored and be within the physiological range throughout the experiments (90–110 mmHg).

17. Acquire a stable baseline CBF measurement.

18. To achieve the highest CBF responses, the right whiskers/whisker pad can be stimulated either mechanically of electrically. For mechanical stimulation of the whiskers, a cotton swab is used to carefully and gently brush the mouse whiskers from side to side for 30 s at ~5 Hz while recording the changes in blood flow. Alternatively, the right whisker pad can also be stimulated by a bipolar stimulating electrode placed to the ramus infraorbitalis of the trigeminal nerve and into the masticatory muscles. The stimulation protocol used to investigate neurovascular coupling consists of ten stimulation presentation trials with an intertrial interval of 70 s, each delivering a 30-s train of electrical pulses (2 Hz, 0.2 mA, intensity, and 0.3-ms pulse width) to the mystacial pad after a 10-s prestimulation baseline period.

19. Capture differential perfusion maps of the brain surface. Changes in CBF should be assessed above the left barrel cortex in ~six trials, separated by 5 min intervals. Specific neurovascular coupling responses are manifested in a well-defined region in the contralateral barrel cortex (Fig. 2). To demonstrate specificity of the responses in Fig. 2, the simultaneous measurement of blood flow changes to unilateral whisker stimulation in both hemispheres is shown. It is recommended that the side of whisker stimulation be alternated once to check the contralateral responses.

20. Average changes in CBF and express the values as percent (%) increase from the baseline value (Kazama et al. 2004). It is recommended that the experimenter be blinded to the treatment of the animals.

21. At the end of the experiments, transcardially perfuse and decapitate the animal. The brains should be immediately removed and hemisected for subsequent biochemical and histological analyses (e.g., measurement of AD-specific brain biomarkers).

Fig. 2.

Illustration of the procedures for preparation of an acute thinned-skull closed cranial window for laser speckle contrast imaging. a Place the mouse into stereotaxic frame. Remove the hair and perform the midline skin incision and retract the skin to expose the skull surface. Thin the skull over the brain region of interest (over the whisker barrel cortex) on both sides using a precision dental drill. Use cold artificial CSF to prevent overheating. b Once the skull is thinned, wipe-dry the skull surface and apply a drop of cyanoacrylate evenly over the cranial window and allow it to dry for 5 min. c Once cured, cover the cyanoacrylate layer with nitrocellulose lacquer and allow to dry for 5 min. d Position the laser speckle contrast imager 10 cm above the cranial window