Fabrication of electrode-embedded fibers and IMOD
The fabrication of electrode-embedded fiber was based on reported procedures44. In brief, a layer of polycarbonate (PC) film with moderate thickness was rolled onto a teflon rod and consolidated in a vacuum oven to prepare a macroscopic preform that resembled the final fiber geometry. Two slots were machined 180 degree apart for electrode embedding, followed by a layer of polyvinylidene difluoride (PVDF) film till the proper diameter was reached. The preform was consolidated to enable a stable fiber drawing process. After the second consolidation, the teflon rod was removed from the preform and two copper wires were inserted into the two machined slots separately prior to the thermal drawing process. During the fiber thermal drawing process, copper wire was pulled from the coil which was mounted on a customized stage above the drawing furnace. As a result of a convergence fiber drawing, the outcoming fiber has a structure of one hollow channel in the center and two copper wires opposite of each other outside the hollow channel (see Fig. 1a). To assemble the fiber with the integrated circuit (IC) chip for impedance measurement, copper wires in the fiber were manually exposed by a scalpel and electrically connected to external copper wires with conductive silver paint (SPI Supplies, West Chester, PA). These external copper wires were soldered to the headpins of the IC chip (Sullins Connector Solutions, San Marcos, CA) and epoxy was applied to protect the connection and avoid short circuits.
Optical fiber coating and drug loading
To prepare a polymer solution of hydrophobic molecules (e.g., rhodamine B) for fiber coating, a mixture of poloxamer (1000 mg, pluronic F127, Sigma-Aldrich), PLGA (330 mg, poly(d,l-lactide-co-glycolide), molecular weight: 30–60k, Sigma-Aldrich) in tetrahydrofuran (THF, 5 mL) was agitated vigorously in a 8 mL reaction vial until PLGA solution was homogenous and clear. Saturated rhodamine B/acetone solution was added into the PLGA/THF solution until solution was homogeneous and pink. Similarly, verteporfin (Cayman Chemical, Ann Arbor, MI)/PLGA solution was prepared by adding 2.5 mg/mL of verteporfin/THF solution and 5 mL PLGA solution into an 8 mL reaction vial.
To coat the fiber layer-by-layer with the polymer solution, a polycarbonate optical fiber was dipped into PLA/THF (100 mg/mL) solution and removed slowly so as to create a smooth coating. Next, it was placed into an air-tight chamber connected to vacuum. The coated fiber was dried under vacuum for 30 min before it was dipped into the rhodamine B / PLGA solution. The fiber was dried under vacuum again for 60 min. This was repeated until there were 11 layers of alternating polymer coatings (additional 1 PLA layer outside of rhodamine B/ PLGA). Verteporfin-coated fiber was prepared similarly. To extend the release of verteporfin, three additional layers of PLA coatings were coated onto the verteporfin fibers.
Animal studies
All animal work was conducted in accordance to the National Institutes of Health Guide for the Care and Use of Laboratory Animals under protocols approved by Virginia Tech Institutional Animal Care and Use Committee and Institutional Biosafety Committee (animal protocol numbers: 15-220, 18-120, 19-002, 21-029).
Mice and cell lines
BALB/c and C57BL/6 female mice (age 4–6 weeks from Jackson Laboratories or Envigo), nu/nu female mice (age 4 to 6 weeks from Charles River), FVB/N-Tg(MMTV-PyVT)634Mul/J male (age 4–6 weeks from Jackson Laboratories; JAX stock #002374) and FVB/NJ female mice (age 4–6 weeks from Jackson Laboratories; JAX stock #001800) were used and maintained according to approved animal protocol. MMTV-PyMT female mice with spontaneously developed breast cancer (age 6–7 weeks) were crosses of FVB/N-Tg(MMTV-PyVT)634Mul/J male and FVB/NJ female mice, and genotyped before 3 weeks of age (Transnetyx). All mice were group-housed (5 mice per cage) and maintained under a regular light-dark cycle altered every 12 h with free access to water and food under pathogen-free conditions in a barrier facility at Virginia Tech. Murine breast cancer E0771 cell line was purchased from CH3 Biosystem (Amherst, NY, catalog # 940001); murine breast cancer 4T1 (catalog # CRL-2539), murine colon cancer CT26 (catalog # CRL-2638), and murine melanoma B16F10 (catalog # CRL-6475) cell lines were originally purchased from American Type Culture Collection (ATCC, Manassas, VA). All cells were maintained at 37 °C with 5% CO2 in culture medium according to instructions, supplemented with 10% heat-inactivated FBS and penicillin/streptomycin (all from Life Technology, Grand Island, NY). All cells were tested to be free of mycoplasma.
Subcutaneous tumor inoculation
For subcutaneous (s.c.) inoculation of single B16F10, 4T1, E0771, and CT26 tumors, 5 × 105 cells in 50 µL of sterile PBS (1×) were subcutaneously injected into the back of neck of C57BL/6 or BALB/c female mice after hair was removed. Treatments were started when tumors reached ~50 mm3. The body weight and tumor size were measured every 1–2 days after the treatment started. Tumor length and width were measured with a digital caliper, and the tumor volume was calculated using the following equation: tumor volume = length × width × width/289. Mice were euthanized when their tumor volumes reached a predetermined end point (1000 mm3) or when their body weights dropped over 10%. For tumor rechallenge studies, mice that overcame tumors were inoculated with the same amount of tumor cells (5 × 105 cells) in the right flank and monitored for another 2 weeks. For evaluation of anti-metastatic effect, mice that overcame tumors were intravenously injected with 106 E0771 tumor cells via tail-vein injection and monitored for another 4 weeks.
IMOD implantation and treatments
Mice receiving treatments through IMOD were placed under anesthesia, with s.c. injection of carprofen (5 mg/kg) as analgesia prior to implantation. Implantation of IMOD was started by making a small slit on the edge of tumor using sterile surgical scissors. The optical fiber end of IMOD was gently inserted into the tumor and the device was positioned under skin with refillable tubing remaining above skin level (see Fig. 1). Skin was pulled taut and adhered around device using 3 M Vetbond Tissue Adhesive. An ethanol solution (70%) was applied appropriately to maintain cleanliness of surgical wound. Intratumoral treatments of ICB antibodies via IMOD were administered in 100 µL sterile PBS through refillable channel on IMOD. In total, 100 µg of anti-CTLA-4 (clone 9H10, BioXCell, Lebanon, NH; catalog # BE0146, diluted to 1 mg/mL) and anti-PD-1 (clone RMP1-14, BioXCell, catalog # BE0146, diluted to 1 mg/mL) were administered each per dose. Note that once the tumor had shrunk from large tumor size of over 100 mm3 to small size of below 75–100 mm3, the device was carefully removed from mice and the treatment was stopped.
For photodynamic therapy, verteporfin loaded fiber was trimmed to 1 cm per dosage and implanted into tumor. At 4 h post-implantation, the mouse was anesthetized with isoflurane. The tumor area was irradiated with near infrared light (Kessil H150-red LED light source of 34 W, 20 s, 600–700 nm) while the rest of the mouse body was covered with aluminum foil. For groups treated with IMOD/PDT× 2, 3 days after the first treatment, mice were anesthetized with isoflurane and the tumor site was irradiated with near infrared light for 20 s in order to activate the released verteporfin from the device. For the group receiving i.p. or i.t. injected verteporfin (5 μg, equivalent to 0.25 mg/kg), light irradiation was also applied onto tumor site for 20 s at 4 h post-administration.
To measure the light intensity of IMOD, a two-centimeter long IMOD was coupled to a silica fiber patchcord (Thorlabs) which was air-coupled to the NIR light source (Kessil H150-red LED light source of 34 W, 600-700 nm). The light output was measured by a power meter (Thorlabs) with a photodetector (Thorlabs) attached, and the average power recorded was 62.4 mW/cm2 leading to a light dose of 1.25 J/cm2 that was delivered to the tissue in 20 s. To measure the light intensity for externally applied photodynamic therapy, the same light source (positioned away from the animal to avoid the heat’s impact) and measuring power meter were used. For a typical exposed tissue area for light irradiation with diameter ~ 0.5 cm, the average power was 63.82 mW/cm2, similar to that of IMOD.
In vivo tumor impedance measurement
For tumor impedance measurements, tumors were inoculated and allowed to form tumor mass of approximately 50-100 mm3, as described above. IMOD was implanted and impedance reading was recorded by connecting Gamry Interface 1000™ Potentiostat to IC chip on IMOD. Mice were placed under anesthesia while all impedance readings were taken. Treatments were given after impedance readings were measured for better consistency. For both measuring and dosing of drugs via IMOD, Pt / Cu were used to replace Cu / Cu electrodes due to its stability for impedance measuring over weeks. Normalized value was calculated based on the starting value of the measurement (set as 1 for both size and impedance readings at 10 kHz).
Whole-body fluorescence imaging of mice
Whole body fluorescence imaging of mice was performed with LI-COR Biosciences Odyssey Infrared Imaging System with emission wavelength at 700 nm (for verteporfin) and 800 nm (for Cy7-BSA). The nu/nu mice were first subcutaneously inoculated with 5 × 104 4T1 cells in 50 µL PBS (1×) into the back of neck. Similar to the descriptions above, IMOD with verteporfin loaded optical fiber was implanted once tumor length was determined to be approximately 8 mm. The nu/nu mice were fed with alfalfa-free food for at least one week prior to the study in order to minimize the gastrointestinal background autofluorescence90. For imaging studies, 100 µg/mL Cy7-BSA in 30 µL sterile PBS (1×) was injected through the refillable tubing on IMOD into the 4T1 tumor after PDT irradiation. Fluorescence images of nu/nu mice were taken at designated time points. The analysis of the results was carried out using Origin software (Northampton, MA) by fitting the normalized fluorescence intensity (I) – time (t) curve into an exponential decay model according to Eq. (1).
$$I=acdot exp(-t/{T}_{1})+b$$
(1)
where a, b, T1 are all constants fitted by Origin software. The starting point for the curve fitting was the peak value of the normalized fluorescence intensity, usually at 4 or 12 h. The obtained T1 value was used to calculate the decay half-life τ based on Eq. (2).
$$tau =ln(2)cdot {T}_{1}$$
(2)
Note that the fluorescence intensity at t = 0 was normalized as 1.
Flow cytometry analysis of tumor lymphocytes
Tumors, lymph nodes and spleens were resected from mice, weighted and gently ground to generate single cell suspensions through a 70-mm cell strainer. Red blood cells were lysed with RBC Lysis Buffer (Biolegend, San Diego, CA, catalog # 420302). Cells were counted and resuspended in Cell Staining Buffer (Biolegend) and used for flow cytometry staining. Non-specific immunofluorescent staining was prevented by incubating cells with TruStain fcX™ (anti-mouse CD16/32) (clone 93, catalog # 101302, dilution 1:50, Biolegend) antibody in 100 µl volume for ~10 min on ice. Cell-surface staining with antibody was performed according to manufacturer’s instructions (Biolegend) with a dilution ratio of 1/100 (also see reporting summary of this paper). Fluorescent antibodies (all from Biolegend) used included CD45 (clone 30-F11, catalog # 103108, dilution 1:100), CD3 (clone 17A2, catalog # 100220, dilution 1:100), CTLA-4 (CD152, clone UC10-4B9, catalog # 106312, dilution 1:100), CD11c (clone N418, catalog # 117348, dilution 1:100), CD49b (clone HMα2, catalog # 103518, dilution 1:100), PD-1 (CD279, clone RMP1-30, catalog # 109116, dilution 1:100), F4/80 (clone BM8, catalog # 123131, dilution 1:100), CD19 (clone 6D5, catalog # 115510, dilution 1:100), CD4 (clone GK1.5, catalog # 100460, dilution 1:100), Ly-6G (clone 1A8, catalog # 127608, dilution 1:100), CD11b (clone M1/70, catalog # 101262, dilution 1:100), CD25 (clone PC61, catalog # 102012, dilution 1:100), CD8a (clone 53-6.7, catalog # 100722, dilution 1:100), CD44 (clone IM7, catalog # 103008, dilution 1:100), CD62L (clone MEL-14, catalog # 104412, dilution 1:100), Ly-6C (clone HK1.4, catalog # 128016, dilution 1:100), CD38 (clone 90, catalog # 102722, dilution 1:100), Tim-3 (CD366, clone RMT3-23, catalog # 119704, dilution 1:100), LAG-3 (CD223, clone C9B7W, catalog # 125221, dilution 1:100), Slamf6 (Ly-108, clone 330-AJ, catalog # 134610, dilution 1:100), IFN-γ (clone XMG1.2, catalog # 505806, dilution 1:100), TNF-α (clone MP6-XT22, catalog # 506308, dilution 1:100), IL-2 (clone JES6-5H4, catalog # 503808, dilution 1:100), Granzyme B (clone QA16A02, catalog # 372214, dilution 1:100), ki-67 (clone 16A8, catalog # 652411, dilution 1:100), and FoxP3 (clone MF-14, catalog # 126404, dilution 1:100). Fixable live/dead cell discrimination was performed using Zombie Aqua™ Fixable Viability Kit according to manufacturer’s protocol (Biolegend, catalog # 423102). For intracellular staining, cells were fixed and permeabilized with the True-Nuclear™ Transcription Factor Staining kit (Biolegend, catalog # 424401) following manufacturer’s instruction before being stained with antibodies. All flow cytometric data collection was performed using BD FACSARIA™ flow cytometer (BD Biosciences, San Jose, CA) and analyzed using FlowJo software (Ashland, OR). Gating strategies are provided in Supplementary Figs. 9 and 15.
CD8+ T cell isolation, activation and cytokine analysis
Tumor CD8+ TILs were isolated using MojoSort TM mouse CD8 cell isolation kit (Biolegend, catalog # 480035) following manufacturer’s instruction. Isolated CD8+ TILs were numerated and plated at a density of approximately 1 × 106 cells per well in a 96-well plate. The suspension was incubated with cell activation cocktail (containing brefeldin A, phorbol 12-myristate 13-acetate, and ionomycin; Biolegend, catalog # 423304) for 6 h at 37 °C with 5% CO2 following manufacturer’s instruction, followed by surface and intracellular flow cytometry staining to detect IFN-γ, TNF-α, other cytokines and cell markers.
Immunohistological staining
For immunohistological staining of tumor tissues, 5 mm formalin fixed, paraffin embedded tumor tissue sections were deparaffinized with xylene, rehydrated, and heated in sodium citrate H.I.E.R. (Heat Induced Epitope Retrieval, pH=6, Biologend, catalog # 420902) for 30 min at 95 °C in a slide glass rack to perform antigen retrieval. Slides were blocked in 3% BSA for 30 min at room temperature, and stained with primary antibody rabbit anti-CD8 antibody (Abcam; catalog # ab203035, dilution 1:200, room temperature, 60 min), then Ultra Streptavidin HRP Kit (Biolegend catalog #929501) containing diaminobenzidine (DAB) chromogen was applied following the manufacture’s protocol for visualization, in which brown staining represented presence of the targeted molecule. They were then counterstained with hematoxylin (Vector Laboratories, catalog # H-3401) for 5 min and rinsed with running tap water.
Statistical analysis
Statistical analysis was performed using Origin Software (Northampton, MA). Most comparisons between groups were assessed using Mann–Whitney U-test (two-sided). Kaplan–Meier survival curves were compared using log-rank test. Significance was represented as follows: *P < 0.05, **P < 0.01, ***P < 0.001, and not significant (n.s.). The n values and specific statistical methods are indicated in figure legends.
