Viscotears

Induction of anaesthesia was achieved with a combination of fentanyl-fluanisone (Hypnorm, Vetapharm Ltd), midazolam (Hypnovel, Roche Ltd) and sterile water in the ratio 1:1:2 (1 ml/kg i.p). Surgical anaesthetic plane was maintained using isoflurane (0.25–0.8%) in 100% oxygen. Body temperature was monitored and maintained throughout surgery via a rectal thermometer and a homeothermic blanket respectively (Harvard Apparatus). Eyes were protected using Viscotears (Novartis). A scalpel was used to shave the head prior to the mouse being positioned in a stereotaxic frame (Kopf Instruments). Iodine was applied to the scalp and the scalp was removed. Using a dental drill, the bone covering the right somatosensory cortex was thinned to translucency to create the thinned optical window (~ 3 mm²). Cyanoacrylate glue was thinly applied across the window to strengthen the window and reduce optical specularities. Dental cement (Superbond C & B; Sun Medical) was applied to the bone on the contralateral side of the cranial window and a well was built up around the window to allow for a metal head plate to be attached for chronic imaging. Following surgery, mice were housed individually and given at least one week to recover before any imaging commenced.

Mice were anaesthetized with isoflurane and placed on a heating pad. A custom-built stereotaxic head holder allowed positioning and fixture of the mouse eye toward the objective (Fig. 1c). For imaging, an upright laser scanning confocal microscope (TCS SP5 II, Leica) equipped with a long-distance water-dipping objective (HCX IRAPO L 25x/0.95W, Leica) was used. Viscotears (Novartis, Basel, Switzerland) were used as the immersion liquid between the eye and the objective. At the end of the imaging sessions, the imaged eyes were treated with Oculentum simplex eye ointment to lubricate the cornea and help cure any possible abrasions or inflammation resulting from the eye manipulation. We detailed the microscopy imaging settings used for the different fluorescent probes and reporter proteins in Supplementary Table 2. Following these specifications, the grafts can be imaged safely, over multiple imaging sessions, with no photodamage to the cells.

Animals were anaesthetized by intraperitoneal injection of xylazine (10 mg/kg body weight) and ketamine (80 mg/kg body weight). Pupils were dilated with 1% tropicamide and 2.5% phenylephrine hydrochloride eye drops (Bausch & Lomb, Rochester, NY, USA) and a 6 mm coverslip positioned on gel lubricant (Viscotears, Novartis, Basel, Switzerland) was applied to the cornea. 1.5 µl of virus at 2.2 × 10¹² virus particles/ml was injected into the vitreous using a Hamilton syringe with a 10 mm 34-gauge needle (65N, Hamilton AG, Bonaduz, Switzerland) using a surgical microscope (M620 F20, Leica, Wetzlar, Germany). All animals received bilateral intravitreal injections of the same virus. Anaesthesia was reversed by intraperitoneal injection of atipamezole (10 mg/kg body weight). During recovery, 0.5% proxymetacaine hydrochloride and 0.5% chloramphenicol (Bausch & Lomb) was applied to injected eyes. All animals began testing 1 month after virus injection.

In vivo imaging of foam and/or islets in the anterior chamber of the eye of the transplanted animals, was performed as previously reported [27 (link)]. Briefly, mice were anesthetized with 2% isoflurane (Baxter) air mixture and placed on a heating pad, and the head was restrained with a head holder. The eyelid was carefully pulled back and the eye gently supported. For imaging, an upright laser scanning confocal microscope based on a Leica TCS-SP5 II (Leica Microsystems) was used together with long-distance water-dipping objectives (Leica HXC-APO 10x/0.3, 20x/0.5, and 40x/0.8 NA). Viscotears (Novartis) was used as an immersion liquid between the eye and the objective. Imaging of islet morphology was done by laser illumination at 633 nm and reflected light was collected between 630 and 637 nm. Scanning speed and laser intensities were adjusted to avoid cellular damage to the mouse eye, matrix or the islet graft. Pictures of vascularization were obtained by acquiring a set of z-stack frames through the whole islet with both laser at 633 nm and using a digital microscope color camera (DFC295, Leica Microsystems) to collect real-time images of vasculature. Post-processing, analysis, visualization and measurement of in vivo transplanted islets were performed with Volocity (PerkinElmer) software, to enable evaluation of size and vasculature of the islets in vivo.