Visualizing and Quantifying Pharmaceutical Compounds within Skin using Coherent Raman Scattering Imaging

This methodology is significant in that it allows for reproducible and accurate method to quantify topically-applied drug products using coherent Raman imaging. Other methodologies provide bulky cutaneous pharmacokinetic information, while this methodology can provide both bulk and microscale cutaneous pharmacokinetic information such as the route of permeation of a drug. The visualization and quantification and permeation pathways will allow the development of drug products with a specific pathway in mind to reach the target site.

Tissue preparation will prove to be the most challenging part of this methodology. The tissue should be sufficiently thin so that one may be able to read through the skin. To begin the preparation of a nude mouse ear skin tissue, acquire euthanized nude mice and remove the ears of the mouse using forceps and microsurgical scissors.

Then place one ear in a large Petri dish of size 60 millimeters by 15 millimeters. Next, rinse the mouse ear with PBS twice and pat the ear dry each time with the task wiper. If used within 24 hours, place the ear in PBS in a small, 35 millimeters by 10 millimeters Petri dish at two to eight degrees Celsius.

To use after 24 hours of harvesting, place the ear in a Petri dish without PBS, then cover the dish with parafilm to store in a freezer at minus 20 degrees Celsius. While working under the biological hood, place the human skin tissue in a large Petri dish with the stratum corneum side facing down so that the subcutaneous fat is accessible. Then use forceps and microsurgical scissors to remove the subcutaneous fat.

Once subcutaneous fat can no longer be removed with the scissors, use a 10 blade disposable scalpel at a 45-degree angle to the skin to remove the remaining subcutaneous fat while holding the skin still with forceps. When the fat is removed, section the human skin into one centimeter by one centimeter pieces. For lipid imaging, place the anterior portion of the mouse ear facing toward the glass bottom of a 35-millimeter, number zero dish.

For the human skin, place the skin with the stratum corneum faced down to allow drug quantification from the superficial to the deeper layers. Once the tissue has been centered on the glass bottom, use a cotton tip applicator to ensure that the skin is flat and has complete contact with the cover slip surface of the glass bottom dish. To prevent any movement while imaging, place a washer on top of the skin and ensure that the tissue is visible through the center hole of the washer for Stimulated Raman Scattering or SRS transmission detection.

Pipette the pre-determined dose of the formulation onto the skin, then use a gloved finger to rub the formulation clockwise for 30 seconds. After the allotted time for the formulation to permeate has elapsed, remove the excess formulation before placing the skin with the formulation side facing toward the glass bottom dish. Next, proceed to the experimental setup by turning on the Microscope Control or MC software.

Looking through the eyepiece, adjust the axial focus with the adjustment knob to ensure that the tissue is within focus. When done, unblock both the pump and Stokes beams and confirm that ALG1 and ALG2 channels are enabled. Ensure the SRS photodiode is in position above the condenser, then click Focus x2 on the MC software to visualize the skin in the CARS and SRS channels.

In the Multi-Area Time Lapse or MATL module, go to the View tab and click on the Registered Point List to have the imaging list appear. Once the stratum corneum is identified, scroll through the axial focus or Z-focus to identify specific tissue stratifications. Specific depths can be added within the skin such as stratum corneum, sebaceous glands, adipocytes, or subcutaneous fat, as determined during live imaging with lipid-tuned contrast.

However, if entire depth stacks are to be taken, XY positions can be added, and the relative Z-position can be ignored. In the case of imaging specific depth for each skin stratification, select the Register Point to add it to the MATL queue. For full depth stacks, click Register Point for Each XY Position with Depth Selection in the Acquisition Parameters window.

Next, set the number of repeats to one in the MATL module and click Ready. When the Play arrow changes from gray to black, hit Play to begin imaging the preliminary lipid stack. Once the imaging cycle has been completed, block both the pump and Stokes beams and change the wavelength on the laser graphical user interface to the desired wavelength based upon the targeted wave number or Raman vibration.

Adjust the manual time delay stage while ensuring that the same powers are used to image the lipid and active pharmaceutical ingredient or API which are established a priori. Once the total number of cycle repeats has been chosen, unblock the pump beam and check if the power matches the desired power using the photodiode before pressing Play to begin automated imaging of the set points. Next, perform data analysis for each skin stratification by importing a sebaceous gland lipid image into Fiji and checking the box labeled Split channels to split the file into the CARS and SRS channels.

To open the Region of Interest or ROI Manager, click on Analyze, Tools, and ROI Manager. Using the SRS channel such that C is equal to one, de-marquee the sebaceous gland in the image and add the markings to the ROI Manager by clicking Add T in the ROI Manager. Repeat the process for each sebaceous gland within the image.

To mask out the lipid-rich regions, select each ROI and then click More, OR Combine, and Add tabs. To mask out the lipid-poor regions, use the Rectangle Tool in the Fiji menu and draw a square around the entire image. Add the region to the ROI Manager.

Click on the newly-added square ROI in addition to the ROI that selects all lipid-rich regions in the ROI Manager. Under More, select XOR to generate a mask of the lipid-poor regions and add it to the ROI Manager. Concatenate the images in numerical order by going to Image, Stacks, Tools, and concatenate tabs in order.

Alternatively, import the images by loading one into Fiji and then selecting an option called Group files with similar names on the setup page. While having the concatenated image active, go to the ROI Manager to select the lipid-rich regions and click the More tab before selecting Multi Measure, then wait for the Results window to appear. From the Results window, export the data to a spreadsheet and add a column titled Region to add lipid-rich regions to each row of data.

Add lipid poor to regions that were outside the lipids. To visualize the data, save and import the spreadsheet into R package, ggplot2. Then plot the data as a function of the image number versus mean intensity to estimate the concentration-time data.

Run non-compartmental analysis or NCA in RStudio on the intensity time data imported from the spreadsheet with the call for one layer. When done, plot and visually compare Jmax and AUCflux-all metrics. Additionally, carry out statistical comparisons across experimental conditions.

The representative analysis depicts stratum corneum, sebaceous glands, adipocytes, subcutaneous fat from the nude mouse ear skin, and stratum corneum, papillary dermis, and a sebaceous gland from the human skin. In nude mouse ear skin, the limited tissue movement of sebaceous glands was observed with the same tissue depth of the glands at the time of formulation application and 120 minutes after application. The substantial tissue movement in the skin showed barely-visible sebaceous glands after 120 minutes of drug application, demonstrating the inefficiency of the original protocol.

In several studies, the intensity versus time profiles showed high initial concentrations, followed by decreasing concentrations, indicating no further absorption of the drug into the tissue. On the other hand, some studies indicated a rise in the intensities which later declined over the experimental duration, demonstrating that the flux into the tissue had not reached a maximum prior to imaging. NCA analysis of concentration-time profiles of two formulations for the same API indicated that the ethoxy-ethoxyethanol provided extended API permeation compared to that of the gel formulation regardless of the skin layer.

The total exposure analysis between the same formulations displayed similar observations in the deeper layers of the skin. The preparation of human skin and proper skin placement on the glass bottom dish after drug application are critical to the experiment's success. The clear solutions here demonstrated the suitability of this methodology.

Further research will involve marketed formulations such as creams to comprehend how the formulation impacts API penetration and permeation.