DEV030700 Supplementary Material

Files in this Data Supplement:

• Supplemental Figure S1 -

  Fig. S1. Hair follicles lacking partial Notch signaling gradually transform into epidermal cysts in adult mice. (A) Ventral images of adult compound mutant mice demonstrate progressive alopecia in the Msx2-Cre expression domain. As more alleles are lost, visible keratin cysts (white arrows) appear to replace hair follicles. All mice are photographed at P120, except the Msx2-Cre/+; Notch1^flox/flox; Notch2^flox/+; Notch3^−/− (N1N2hN3CKO) mouse, which is 60 days old. (B) Immunofluorescence and H&E staining reveal that Notch1-deficient adult hair follicles in Msx2-Cre/+; Notch1^flox/flox (N1CKO) revert to the epidermal differentiation program (asterisks) late in the life of the animal. Scale bars: 50 µm.

• Supplemental Figure S2 -

  Fig. S2. Rbpj-deficient bulge stem cells give rise to keratinocytes forming the anagen hair follicle in the absence of exogenous hair cycle induction. (A) Diagram of the timeline used to treat 50-day-old K15CrePGR/+; Rbpj^flox/flox; Rosa26R (K15-RBP-jCKO) and control animals with RU486 in order to delete Rbpj in adult hair follicle stem cells at the beginning of the second telogen phase. The hair follicles are examined 35 days later at the point when some have entered the anagen phase of a new hair cycle. The status of telogen hair is also documented 14 days after RU486 treatment (arrowhead). Note that no hair induction step is included. (B) X-Gal staining shows blue bulge cells in K15-RBP-jCKO and control telogen hair follicles 14 days after Cre activation. Arrow points to a blue cell in the isthmus. (C) After 35 days, hair follicles have entered anagen. X-Gal staining demonstrates that blue Rbpj-deleted cells in K15-RBP-jCKO hair follicles have contributed to all layers of the anagen hair follicle. Scale bars: 25 µm.

• Supplemental Figure S3 -

  Fig. S3. Evidence of bifurcation in the Notch signaling pathway downstream of γ-secretase. (A) The Rbpj-deficient hair follicles display a milder disruption than that seen with loss of Notch receptors (Msx2-Cre/+; Notch1^flox/flox; Notch2^flox/flox, or N1N2CKO) or γ-secretase (Msx2-Cre/+; PS1^flox/flox; PS2^−/−, or PSDCKO). This would be expected in a linear pathway if a few crucial permissive Notch targets were derepressed upon Rbpj loss, ameliorating the phenotype (Koelzer and Klein, 2006). Alternatively, the Notch pathway might have an Rbpj-independent yet γ-secretase-dependent arm that contributes to the severity of the hair phenotypes observed in the Notch-deficient hair follicle (Demehri et al., 2008). To distinguish between these possibilities, mice with γ-secretase and Rbpj deletion in hair follicles are examined (Msx2-Cre/+; PS1^flox/flox; PS2^−/−; Rbpj^flox/flox, or PSDRBP-jCKO) (Demehri et al., 2008). In this genetic combination, if derepressed Notch targets are important in hair follicle maintenance, they will suppress the PSDCKO phenotype, and the PSDRBP-jCKO hair will display a milder, RBP-jCKO-like phenotype (Koelzer and Klein, 2006). On the other hand, if Rbpj-independent targets of Notch contribute to hair follicle maintenance, PSDRBP-jCKO hair follicles will have the PSDCKO phenotype. A combination of both effects will generate an intermediate phenotype. PSDRBP-jCKO mice display a hair phenotype similar to PSDCKO, a result supporting the assertion that Rbpj-independent signals downstream of Notch activation contribute to hair follicle homeostasis (Demehri et al., 2008; Rangarajan et al., 2001; Talora et al., 2002). To confirm that these Rbpj-independent signals are regulated by Notch, mice lacking Notch1, Notch2 and Rbpj in hair follicles (Msx2-Cre/+; Notch1^flox/flox; Notch2^−/−; Rbpj^flox/flox, or N1N2RBP-jCKO) are examined. As expected, these mice have a hair phenotype that is similar to N1N2CKO, which is more severe than that of RBP-jCKO mice. (B) Immunohistochemical staining for Rbpj shows that Rbpj protein is absent in PSDRBP-jCKO skin at P9. Scale bars: 100 µm in A; 50 µm in B.

  References

  Koelzer, S. and Klein, T. (2006). Regulation of expression of Vg and establishment of the dorsoventral compartment boundary in the wing imaginal disc by Suppressor of Hairless. Dev. Biol. 289, 77-90.

  Talora, C., Sgroi, D. C., Crum, C. P. and Dotto, G. P. (2002). Specific down-modulation of Notch1 signaling in cervical cancer cells is required for sustained HPV-E6/E7 expression and late steps of malignant transformation. Genes Dev. 16, 2252-2263.

• Supplemental Figure S4 -

  Fig. S4. Mice lacking all Notch receptors in the bulge stem cells develop keratin cysts and intense inflammation 10 weeks after RU486-mediated gene deletion. CD45 staining of X-Gal-stained skin from animals described in Fig. 3D shows a marked infiltration of immune cells in the dermis of K15-N1N2N3CKO animals. Note that K15-N1CKO animals contain many fewer CD45⁺ leukocytes in their dermis at this point relative to K15-N1N2N3CKO mice. No counterstaining was performed. Scale bars: 25 µm.

• Supplemental Figure S5 -

  Fig. S5. Thirty-five days after RU486 treatment, K15-RBP-jCKO anagen hair follicles contain Rbpj-deficient blue cells above the bulge region (dashed line), which are not found in control animals. The hair follicles with a streak of blue cells extending into the epidermis were counted and their percentage over the total number of blue-cell-containing anagen hair follicles was calculated. Scale bars: 25 µm.

• Supplemental Figure S6 -

  Fig. S6. Following RU486 application, blue Rbpj-deficient bulge cells migrate upward toward the epidermis in hair follicles that have not yet entered anagen. X-Gal staining shows blue bulge cells in a K15-RBP-jCKO hair follicle, which is in telogen 35 days after Cre activation according to the induction protocol described in Fig. S3A. Arrows highlight blue cells in the isthmus. Scale bar: 25 µm.