Research ArticleBioengineering

Biocompatible near-infrared quantum dots delivered to the skin by microneedle patches record vaccination

See allHide authors and affiliations

Science Translational Medicine  18 Dec 2019:
Vol. 11, Issue 523, eaay7162
DOI: 10.1126/scitranslmed.aay7162
  • Fig. 1 Platform schematic and fluorescent probe characterization.

    (A) Fluorescent microparticles are distributed through an array of dissolvable microneedles in a distinct spatial pattern. (B) Microneedles are then applied to the skin for 2 to 5 min, resulting in dissolution of the microneedle matrix and retention of fluorescent microparticles. (C) A NIR LED and adapted smartphone are used to image patterns of fluorescent microparticles retained within the skin. By selectively embedding microparticles within microneedles used to deliver a vaccine, the resulting pattern of fluorescence detected in the skin can be used as an on-patient record of an individual’s vaccination history. (D) Rapid photobleaching of organic dyes covered with pigmented human skin under simulated solar light. (E) Emission profiles of QDs (solid lines) show a blue shift with increased shelling time. Dashed line depicts absorption by the 5-hour shelling sample. Arrows indicate the relevant y axis for absorption (left) and emission (right). a.u., arbitrary units. (F) PL QY as a function of shelling time under different excitation wavelengths. (G) Relative photoluminescence intensity comparison of QDs with the commercial inorganic dyes IRDC2 and IRDC3 (blue bars) and corresponding emission peaks (empty purple circles and square). (H) Photostability of QDs covered with pigmented human skin under simulated solar light. (I) Transmission electron microscopy (TEM) and high-resolution TEM (inset) showing the size and crystal structure of ZnS:Al-coated CuInSe2 QDs. Scale bars, 20 and 5 nm, respectively. n = 3 for all graphs containing error bars. Error bars indicate SD.

  • Fig. 2 Encapsulation and characterization of QDs in PMMA microspheres.

    Light microscopy images of PMMA microspheres loaded with (A) 0% (w/w) QDs, (B) 37.5% (w/w) QDs, and (C) 60% (w/w) QDs. (D) SEM image of PMMA microparticles containing S10C5H QDs. (E) Histogram of volumetric particle distribution smoothed using an 11-frame moving window smoothing function for improved clarity, n = 104 particles analyzed. (F) Photoluminescence profiles of S10C5H QDs before and after PMMA encapsulation showing a minimal shift in fluorescence emission wavelength. (G) Relative photoluminescence intensities (blue bars) using an 850-nm long-pass filter (n = 2) and emission peaks (empty purple circles). (H) Maintenance of photoluminescence intensity in PBS at 37°C over the course of months (n = 3), with representative NIR images inlaid at their respective time points; *P < 0.05 (one-way ANOVA with Tukey’s multiple comparisons). Dashed line indicates the camera saturation point. Error bars indicate SD.

  • Fig. 3 Microneedle modeling, fabrication, and evaluation.

    Optical images of microneedles and finite element analysis data of (A and B) a conical needle 1500 μm in height and 300 μm at its base; (C and D) a microneedle 300 μm at its base with a 750-μm cone atop a 750-μm cylinder; and (E and F) a microneedle 300 μm at its base with a 250-μm cone atop a 1250-μm cylinder. (G) SEM image of a dissolvable microneedle array based on the geometry shown in (C). (H) Ex vivo penetration force per needle based on microneedle geometry; n = 3; ***P < 0.001 and ****P < 0.0001 (one-way ANOVA with Tukey’s multiple comparisons). (I) Spacing-independent penetration force requirements in pig skin ex vivo (Student’s t test). In (B), (D), and (F), 0 and 100 represent the worst and best values, respectively, for each parameter for α between 0 and 1.

  • Fig. 4 Smartphone modifications and NIR marking detection in skin.

    (A) Photograph of disassembled LED used for NIR illumination at 780 nm combined with an 800-nm short-pass filter and aspheric condenser. (B) Photograph of disassembled NIR imaging smartphone consisting of a Google Nexus 5X smartphone with the internal short-pass filter removed and replaced with two external 850-nm long-pass filters set in a 3D-printed phone case. Images of a 16-needle microneedle patch containing PMMA-encapsulated QDs were collected with the adapted smartphone under ambient indoor lighting (C) without the 850-nm long-pass filters and (D) with the pair of 850-nm long-pass filters under LED illumination from the same distance. Inset shows an image at a higher exposure. (E) Optical and (F) SEM images of fluorescent microparticle-loaded microneedles before skin application. (G) Optical and (H) SEM images of microneedles after administration to explanted pig skin. Adapted smartphone images of pig skin before microneedle application (I) without and (J) with 850-nm long-pass filters. Adapted smartphone images of pig skin after application (K) without and (L) with 850-nm long-pass filters. Adapted smartphone images of pigmented human skin before microneedle application (M) without and (N) with the 850-nm long-pass filters. Smartphone images of human skin after application (O) without and (P) with the 850-nm long-pass filters. Note: Scale bars in NIR-filtered images are approximate with (J), (L), (N), and (P) taken at about the same distance. Components in (A) and (B) cropped for clarity.

  • Fig. 5 In vivo imaging of NIR patterns in rodent skin.

    Administration site after the delivery of a 4 × 4 microneedle patch containing (A) unencapsulated QDs or (B) PMMA-encapsulated QDs. (C) Short-term study of signal intensity after microneedle application of unencapsulated or PMMA-encapsulated QDs to rat skin, n = 4. Images of (D) circle, (E) cross, and (F) rectangle patterns imaged 24 weeks after administration of PMMA-encapsulated QDs to rats. Log-scale color maps of the same (G) circle, (H) cross, and (I) rectangle patterns shown in (D) and (F). (J) Number of markings detected 24 weeks after administration, n = 5. (K) Quantification of signal-to-noise ratio for the circle pattern showing no changes between 12, 24, and 36 weeks, n = 15 (one-way ANOVA with Tukey’s multiple comparisons). (L) Graph showing the average probability of the machine learning algorithm (all patterns correctly detected), n = 5. Grayscale images extracted from red channel of the adapted smartphone-generated Red Green Blue (RGB) image.

  • Fig. 6 Longitudinal imaging of NIR markings in rodent skin.

    Cropped, but otherwise raw, smartphone images collected from a fixed distance showing the intradermal NIR signal from PMMA-encapsulated QDs delivered via microneedle patches on rats 0, 12, and 24 weeks after administration. The text at the bottom of each image indicates the image collection settings ISO density and shutter speed (SS) in seconds.

  • Fig. 7 Biological response to PMMA-encapsulated QDs.

    (A) In vitro cytotoxicity of commercially available PbS QDs compared to unencapsulated and PMMA-encapsulated S10C5H QDs over 24 hours in a mouse macrophage cell line (Raw 264.7). n = 3, *P < 0.05, ***P < 0.001, and ****P < 0.0001 (two-way ANOVA with Tukey’s multiple comparisons). Representative histological samples collected from rats receiving microneedle-delivered PMMA particles containing S10C5H QDs (B and C) 1 day, (D and E) 2 weeks, and (F and G) 4 weeks after administration stained with hematoxylin and eosin or Masson’s trichrome, respectively. Arrows indicate the location of microparticles. (H) Total anti-poliovirus type 2 immunoglobulin G antibody titers and (I) neutralizing poliovirus type 2 antibody titers showing no differences after three doses of IPV2 delivered via subcutaneous injections or microneedles with or without PMMA-encapsulated QDs, n = 5. Dashed line indicates the threshold above which humans are considered immune.

Supplementary Materials

  • stm.sciencemag.org/cgi/content/full/11/523/eaay7162/DC1

    Materials and Methods

    Fig. S1. Optical properties of organic dyes.

    Fig. S2. Evolution of fluorescence emission properties with shelling time.

    Fig. S3. Fluorescence lifetime characterization of the S10C QD series.

    Fig. S4. Composition and physical properties of S10C5H QDs.

    Fig. S5. pH stability of PMMA-encapsulated QDs.

    Fig. S6. Finite element analysis of mechanical forces on microneedles.

    Fig. S7. Optimization of microneedle geometry using finite element analysis.

    Fig. S8. Machine learning training and validation.

    Table S1. Spectral characterization of custom QD formulations.

    Table S2. Multiexponential fitting parameters for photoluminescence decay curves.

    Movie S1. Intradermal administration and imaging of encapsulated QDs.

    Data file S1. Individual subject-level data.

    References (5462)

  • The PDF file includes:

    • Materials and Methods
    • Fig. S1. Optical properties of organic dyes.
    • Fig. S2. Evolution of fluorescence emission properties with shelling time.
    • Fig. S3. Fluorescence lifetime characterization of the S10C QD series.
    • Fig. S4. Composition and physical properties of S10C5H QDs.
    • Fig. S5. pH stability of PMMA-encapsulated QDs.
    • Fig. S6. Finite element analysis of mechanical forces on microneedles.
    • Fig. S7. Optimization of microneedle geometry using finite element analysis.
    • Fig. S8. Machine learning training and validation.
    • Table S1. Spectral characterization of custom QD formulations.
    • Table S2. Multiexponential fitting parameters for photoluminescence decay curves.
    • Legend for movie S1
    • References (5462)

    [Download PDF]

    Other Supplementary Material for this manuscript includes the following:

    • Movie S1 (.mp4 format). Intradermal administration and imaging of encapsulated QDs.
    • Data file S1 (Microsoft Excel format). Individual subject-level data.

Stay Connected to Science Translational Medicine

Navigate This Article