SEARCH

SEARCH BY CITATION

Keywords:

  • dermatology;
  • hair shaft;
  • skin;
  • thick sample imaging

Abstract

  1. Top of page
  2. Abstract
  3. Background
  4. Experimental design
  5. Results and Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Please cite this paper as: Point scanning confocal microscopy facilitates 3D human hair follicle imaging in tissue sections. Experimental Dermatology 2010; 19: 691–694.

Abstract:  Efficiency is a key factor in determining whether a scientific method becomes widely accepted in practical applications. In dermatology, morphological characterisation of intact hair follicles by traditional methods can be rather inefficient. Samples are embedded, sliced, imaged and digitally reconstructed, which can be time-consuming. Confocal microscopy, on the other hand, is more efficient and readily applicable to study intact hair follicles. Modern confocal microscopes deliver and collect light very efficiently and thus allow high spatial resolution imaging of relatively thick samples. In this letter, we report that we successfully imaged entire intact human hair follicles using point scanning confocal microscopy. Light delivery and light-collection were further improved by preparing the samples in 2,2′-Thiodiethanol (TDE), thus reducing refractive index gradients. The relatively short total scan times and the high quality of the acquired 3D images make confocal microscopy a desirable method for studying intact hair follicles under normal and pathological conditions.


Background

  1. Top of page
  2. Abstract
  3. Background
  4. Experimental design
  5. Results and Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Currently used traditional methods to reveal the 3D structure of human hair follicles (HFs) within the scalp (1–3) are usually time-consuming and labour-intensive. They involve sample embedding, cutting and staining, followed by imaging of all the sections and complex digital imaging computations, altogether a very lengthy process. In this report, we show representative results from a study where we applied point scanning confocal microscopy (4,5) as a more efficient imaging technique for histological samples to characterise the spatial structure of human HFs. This approach may also provide a faster and more precise tool to reveal hair shaft (HS) disorders, or test agent-induced changes in the HS quality of organ-cultured human scalp HFs (6–8), based on changes in the 3D structure and fluorescence properties of HFs.

Experimental design

  1. Top of page
  2. Abstract
  3. Background
  4. Experimental design
  5. Results and Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Traditionally, the structural characterisation of HFs by histology includes creating vertically oriented sections of the skin such that the epidermis is located at one pole and the subcutis at the other pole. This technique is inherently error-prone mainly because of the frequent mismatch between the spatial orientation of the HF, which rises to the skin surface in an oblique angle, and the orientation of the cut section. This results in a misaligned and incomplete reconstruction of the 3D structure of the HF within the scalp. Each cut section is about 5–8 μm thick, requiring a large number of slices to assess the entire length and diameter of the HF and its shaft. Furthermore, each stained slice/section has to be imaged individually, and the images must be registered to correct for the inevitable sample distortions and misalignments. These technical complexities make the observation of a complete longitudinal section of the HF from the bulb to the epidermis often exceedingly difficult.

For the approach proposed here, we used microdissected human scalp HFs, as described before (9–11). Immediately after dissection, the native HFs underwent a mounting procedure using gradually higher concentrations of 2,2′-Thiodiethanol (TDE) diluted with phosphate-buffered saline (PBS) and water (12) or whole HFs were stained for the inhibitory β1 integrin antibody mAb13, as described before (1) and then mounted in TDE. The TDE mounting medium allows high spatial resolution imaging at greater depths than traditional mounting media, because of the substitution of water by TDE inside the tissue. Because the refractive index (RI) of concentrated TDE is very close to that of immersion oil, imaging a TDE-saturated tissue with an oil immersion objective through a glass coverslip will decrease the optical aberrations caused by RI differences (12).

The HFs were placed on glass slides in small droplets of 99% TDE, covered with 0.17-μm glass coverslips. Spacers in between the coverslip and the slide helped to avoid compressing the HF. The samples were imaged with Leica TCS SP5 confocal microscopes (Leica Microsystems, Germany) built on Leica DM6000 upright microscopes, using 20×/0.75 NA multi-immersion objectives with type F immersion oil. HFs were imaged in the xyz mode using depth compensation, which was necessary because of the significant loss of light experienced with thick samples.

Results and Discussion

  1. Top of page
  2. Abstract
  3. Background
  4. Experimental design
  5. Results and Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

The HS of all HFs studied here (n = 32) showed bright autofluorescence, presumably arising from melanin (13), when illuminated with a laser line between 470 and 580 nm (Fig. S1). We chose the 561-nm excitation line that produced the strongest autofluorescence signal; the same laser line was also utilised to generate differential interference contrast (DIC) images (14,15). The HS of a HF (up to 100 μm thick) could be imaged in one Z stack, using 1-μm Z step size. The longitudinal extent of the HF (i.e. shaft and bulb) could be imaged in three or four segments. One Z stack could be recorded in about 15 min; thus, the entire HF required only 45–60 min to be completely characterised at high spatial resolution.

The resulting Z stacks can be viewed either separately, or after appropriate ‘stitching’, a procedure that aligns image tiles. The longitudinal cross-section of the whole length of a HS is shown in Fig. 1b, where we utilised the Leica LAS AF 2.0 software. For this report, the 3D reconstructions were performed with Amira 5.2.0 (Visage Imaging GmbH, Berlin, Germany) (Fig. 1c). Figure 1a and b show that it was easy to identify all three layers of the HS. The medulla is the innermost layer of the HS, which appears as a dark zone in the 2D reconstruction. The hair fibres of the HS cortex comprise keratin filaments embedded in a sulphur-rich matrix. Here, they are coloured red enclosing the medulla and being surrounded by the bright orange cuticle. Further structural details could be revealed by transmission electron microscopy (16–20). Using traditional methods to visualise the HS, it is dubious if one would ever see all three layers of the HS in one image, emphasising the usefulness of confocal microscopy in HF studies.

image

Figure 1.  Confocal microscopy images of an intact hair follicle (HF) from the human scalp. (a) 2D cross-section of a HF, drawn perpendicular to the axis of the hair shaft (HS) and reconstructed from a 3D image stack; this panel clearly shows that the shape and diameter of the HS can easily and precisely be deducted from this image (HS in red, surrounding tissue in grey from DIC image stack, extracted from b – see dotted line). (b) 2D longitudinal section along the mid-plane of the 3D reconstruction of the HF. The HS is visualised via autofluorescence, induced by 561-nm excitation. All other layers of the HF are visualised by DIC imaging and appear in shades of grey. (c) 3D reconstruction of the HF shown in b. The HS appears in red (autofluorescence by 561-nm excitation), the surrounding tissue is imaged via DIC and appears in grey. Abbreviations: HS, hair shaft; IRS, inner root sheath; ORS, outer root sheath; CTS, connective tissue sheath; m, medulla of the HS; co, cortex of the HS; cu, cuticle of the HS. Sample orientation is indicated by the xyz axes in the bottom left corner; confocal images were acquired in the xy plane. Scale bars: 100 μm. Total volume of the HS region shown equals 1.7 × 10−3 mm3. Laser power of the 561-nm line was 0.9 mW at the sample’s focal plane.

Download figure to PowerPoint

Figure 2a shows a HF stained for the inhibitory β1 integrin antibody mAb13 using the traditional method (1). Using point scanning microscopy, one can easily visualise the same immunoreactivity pattern in a 2D image (Fig. 2b) as well as in a 3D reconstruction (Fig. 2c,d). A total of 22 antibody-stained HFs were imaged in this study.

image

Figure 2.  2D and 3D image reconstructions of human scalp hair follicles stained for mAb13. (a) Hair follicle stained for the β1 integrin inhibitory antibody mAb13 (positive immunoreactivity for mAb13 in red; nucleus counterstaining in blue). Picture taken with Keyence BZ-8100 Microscope (Keyence, Osaka, Japan). (b) 2D longitudinal section along the mid-plane of the 3D reconstruction of the hair follicle (autofluorescence induced by 561-nm excitation in red; immunoreactivity for mAb13 induced by 488-nm excitation in green). (c) 3D reconstruction of a control hair follicle where the primary antibody was omitted during the staining procedure. Autofluorescence of the hair shaft is shown in red at 561-nm excitation. (d) 3D reconstruction of the hair follicle shown in b. The HS appears in red (autofluorescence by 561-nm excitation), the surrounding outer root sheath and connective tissue sheath are imaged via 488-nm excitation and appear in green (artefact ‘a’ is autofluorescent in red). Abbreviations: CTS, connective tissue sheath; HS, hair shaft; ORS, outer root sheath; a, artefact. Sample orientation is indicated by the xyz axes in the top left corners. Scale bars: 100 μm.

Download figure to PowerPoint

In this brief report, we recommend a fast and non-destructive method that greatly facilitates the 3D assessment of intact human HFs. The method of point scanning confocal microscopy of TDE-infused HFs does not require the time and effort that comes with the classic method of producing dozens of serial sections, long staining procedures and acquiring large numbers of images that also need to be registered. Using confocal microscopy to image the entire HS only needs a single microdissected intact HF, a quarter of an hour time per segment, and results in a detailed view of the HS. For the analysis of HS disorders, changes in HS structure in health and disease as well as for preclinical and clinical hair growth trials this method could become a comfortable, instructive and widely used research tool.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Background
  4. Experimental design
  5. Results and Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

We thank Leica Microsystems UK Ltd for providing access to their Leica SP5X demo system for the duration of the scans of HF excitation spectra, an example of which is shown in Fig. S1. ZC is supported by the Wellcome Trust 079204/Z/06/Z. JK was supported by a Junior Grant from the Medical Faculty, University of Luebeck, Germany. TB is a recipient of the Janos Bolyai Scholarship of the Hungarian Academy of Sciences.

References

  1. Top of page
  2. Abstract
  3. Background
  4. Experimental design
  5. Results and Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Background
  4. Experimental design
  5. Results and Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Figure S1. Excitation spectrum of autofluorescence of the HS of an unstained/untreated hair follicle. The spectrum was recorded on a Leica SP5X confocal system equipped with a White Light Laser, which provided a continuous spectrum excitation laser light source. The scan was performed at 3-nm bandwidth with excitation between 470 and 670 nm, whilst emission was recorded from a flexible wavelength range that started at 5 nm above excitation and ended at 800 nm. The fluorescence values were calculated from a region of interest selected from the distal end of the HS, without including any background.

Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

FilenameFormatSizeDescription
EXD_1110_sm_fS1.tif30KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.