|
Fig. 1. The Cre-dependent cartilage labeling system. (A, B) This system uses an inducer transgenic line (pDPCol2.3rtTA-TRECre-HS4) that expresses Cre recombinase in cartilages to act on loxP sites in a Cre-reporter transgene in a second line (pCLFR-SceI)�resulting in DsRed2 replacing CFP in chondrocytes, thereby establishing DsRed2 expression in cartilage under control of the constitutive CMV pro- moter. The inducer line was designed for doxycycline (Dox) activation of Cre expression through the rtTA/TRE (TetOn) system. However, as shown in the next panel, col2a1 regulatory elements drive both rtTA and Cre expression, bypassing the tetracycline/doxycycline response element (TRE) promoter and Dox control resulting in loss of inducible control. (C, D, E) Inducer and Cre-reporter transgenic lines were crossed and 2-week-old tadpoles not treated with Dox are shown in bright field (C), blue (D), and red (E) fluorescent images. Double transgenics have both the inducer, revealed by GFP (green fluorescent protein) expression in the eyes driven by the lens-specific gamma crystallin (CRY) promoter, and the reporter, revealed by ubiquitous CFP (cyan fluorescent protein) driven by the CMV (cytomega- lovirus) promoter. The four expected types of progeny were identified: (1) non-transgenic, (2) pCLFR-SceI single transgenics (CFP in the bodies), (3) pDPcol2.3rtTA-TRECre-HS4 single transgenics (GFP in the eyes, seen as a blue-green color in the blue images), and (4) double transgenics (GFP in the eyes and CFP in the body). Note that in the absence of Dox, DsRed2 fluorescence is visible in larval car- tilages in double transgenic animals (E, right corner � see text for explanation).
|
|
Fig. 2. The Cre-dependent system labels early forming larval cartilages with DsRed2 under the short-lived cytomegalovirus (CMV) pro- moter in double transgenic tadpoles. Images show dorsal and ventral views of DsRed2-labeled cartilage at the indicated Nieuwkoop- Faber stage (NF) stages. Note highest intensity of signal in the remodeling palatoquadrate (PQ) and cranial trabeculae (TR; A�C), and eventual loss of signal in all dorsal cartilages (D). Also note Meckel�s cartilage (MC) and ceratohyal (CH) persist in modified form through NF66 (E�H). No expression was detected in either the nasal capsule cartilages or otic capsules (see text for explanations.) CB, cerato- branchial; HY, hyale; PA, planum antorbital; PQA, ascending process of the palatoquadrate; PQO, otic process of the palatoquadrate.
|
|
Fig. 3. The rtTA-dependent cartilage labeling system. (A, B) This system labels cartilage using the same transgenic inducer line as the Cre-dependent system and a reporter line with Dox-inducible green fluorescent protein (GFP) expression under control of the tetracy- cline/doxycycline response element (TRE) promoter. The longevity of inducible GFP protein in newly formed cartilage is used to label and track chondrocytes. The reporter construct allows Dox-inducible expression of GFP wherever rtTA from the inducer line is expressed (i.e., chondrocytes). (C�E) Inducer and GFP reporter transgenic lines were crossed and 2-week-old tadpoles were treated with or without 50 lg/mL Dox for 24 h. Bright field (C) and green (D) and red (E) fluorescent images are shown. The four expected types of progeny were identified by examining fluorescent protein expression in the eyes (GFP for the inducer and DsRed2 for the reporter). Note that car- tilage-specific GFP expression is evident only in double transgenic tadpoles treated with Dox.
|
|
Fig. 4. Green fluorescent protein (GFP)mRNA and protein stability in transgenic animals from the rtTA-dependent system. (A) Double transgenic tadpoles were treated with 50 lg/mL Dox for 3 days at NF57. Green fluorescence images of the hind limbs were taken on Days 0�3, 5 and 7. GFP protein became strongly visible starting on Day 2 and remained intense through Day 7 � 4 days after Dox removal. (B) GFP mRNA was measured using quantitative real-time polymerase chain reaction (PCR) from hind limbs corresponding to the images in the previous panel, n = 3 pairs of limbs per day. Minimal GFP mRNA was detectable in transgenic animals in the absence of Dox, and GFP mRNA expression reached a peak at 2�3 days of Dox treatment. By 2�4 days after Dox removal, GFP mRNA quickly returned to pre-Dox levels. Note that the persistence of inducible GFP protein expression, after GFP mRNA has degraded, renders this system useful for labeling and tracking tadpole cartilages through metamorphosis.
|
|
Fig. 5. Tracing dorsal cranial cartilages labeled with green fluo- rescent protein (GFP) in the rtTA-dependent system. Double transgenic tadpoles were treated with Dox at the indicated stages and then imaged at the indicated stages. (A) Dox-induc- tion at NF55 labels larval cartilage elements. (B) By NF63, only the otic capsules (OTC) and tectum posterius (TP) carry strong GFP protein expression. Faint expression is also found in the tec- tum anterius (TA). (C) Dox-induction at NF57 still labels most lar- val cartilages, though at lower intensity compared to Dox- induction at NF55. In addition, GFP labeling becomes evident in precursors of the adult alary cartilages (AC). (D) Alary, tecti anteri- us and posterius, and otic capsule labeling is retained. Faint expression also surrounds the braincase, corresponding to the area covered by the frontoparietal bone (FP). (E) Dox-induction at NF59 fails to label larval cartilages, except weak expression in the otic capsule, which persists to NF63. The nasal capsule carti- lages are strongly labeled and (F) remain so through NF63. Addi- tional abbreviations: NS,nasal septum; OBC, oblique cartilage; PA, planum antorbitale; PQ, palatoquadrate; Sc, scapula; SP, suprarostral plate; VC, vertebral column
|
|
Fig. 6. Tracing ventral cranial cartilages labeled with GFP using the rtTA-dependent system. (A) Dox-induction at NF55 labels lar- val cartilage elements, while (B) only the hyale (HY) is detectable by NF63. (C) Dox-induction at NF57 still labels all larval cartilage, though at lower intensity compared to Dox-induction at NF55. (D) GFP labeling in the hyale is not detectable at NF63 following later induction, but is apparent in the adult alary cartilages (AC; see Figure 5). (E) Dox-induction at NF59 is barely detectable GFP-labeling in distal (DMC) and medial (MMC) ends Meckels�s cartilage. (F) By NF63, distal (DMC) and medial (MMC) ends Meckels�s cartilage are still labeled but in their remodeled posi- tion. Portions of the nasal capsule from the dorsal side show though at NF63. CB, ceratobranchial; CH, ceratohyal; MC, Meckel�s cartilage.
|
|
Fig. 7. Tracing cranial cartilages through T3-induced metamorphosis using the rtTA-dependent system. One-week-old double trans- genic tadpoles were treated with 50 lg/mL Dox for 2 days then treated with 10 nmol/L T3 for 7 days in the absence of Dox. Brightfield and green fluorescence dorsal and ventral images were taken on Days 0, 3, 4, 5, and 7 of T3 treatment. All larval cartilages were well- labeled by the beginning of the T3 treatment. Note that the otic capsule (OTC) is the only labeled cartilage remaining in dorsal view, while the remodeling hyale (HY), elongating Meckel�s cartilage (MC), and degenerating ceratobranchials (CB) are visible in the ventral view.
|
|
Fig. 8. Tracing cranial cartilages after initiation of T3 treatment in early tadpoles using the rtTA-dependent system. One-week-old double transgenic tadpoles were treated with 10 nM T3 for 3 days then treated with 10 nmol/L T3 and 50 lg/mL Dox for the next 4 days. Sib- ling double transgenic tadpoles treated with Dox and not T3 served as controls. No food was given during T3 treatment in treated or control tadpoles. Brightfield and green fluorescence dorsal and ventral images were taken on Day 7 of T3 treatment. Compared to Fig. 7, GFP labeling intensity in the non-T3-treated controls was much reduced, likely due to starvation and lack of new cartilage forma- tion. In the T3-treated tadpoles, newly formed cartilage is marked, representing nasal capsule cartilages (NC), elongating Meckel�s carti- lage (MC), and remodeling hyale (HY). Punctate expression also occurs on the largely resorbed palatoquadrate (PQ) and ceratobranchials (CB).
|
|
Fig. S1. Enhancer activity of transgenic Col2.3 promoter. (A) Tadpoles single transgenic for the inducer construct (Fig. 1A) were imaged with brightfield and green fluorescence. Exposure of the fluorescence image was longer than used in the other figures to reveal low level green fluorescent protein (GFP) expression specifically in cartilage. Such cartilage-specific GFP expression can be explained by enhancer activity of col2a1 regulatory elements inducing the CRY-linked GFP at a distance beyond both the rtTA and Cre coding regions. Thus, presumably the Col2.3 promoter also activates Cre, explaining the cartilage-specific expression of DsRed2 in Figure 1B. (B) Cre-dependent system tadpoles (pDPCol2.3rtTA-TRECre-HS4 � pCLFR-SceI) at NF46 treated with or without 50 μg/mL Dox for 2 days were imaged with brightfield and red fluorescence. Note that the intensity and extent of cartilage labeling with DsRed2 is independent of Dox. Because col2a1 regulatory elements can activate GFP and thus also Cre, which is closer to the Col2.3 promoter than GFP, and because Dox has no effect on DsRed2 labeling, we conclude that the undesirable DsRed2 expression in cartilage occurs via cartilage-specific Cre expression driven by the col2a1 regulatory elements.
|
|
Fig. S2. Tracing tibiofibular cartilages labeled with green fluorescent protein (GFP) using the rtTA-dependent system. (A) Lateral view. Double transgenic tadpoles were imaged immediately after treatment for 3 days with 50 μg/mL Dox at the indicated stages. Note that at this earliest labeling, only the distal ends of the cartilaginous elements have GFP. Arrows indicate early GFP expression detectable at the base of the forming digits, approximately corresponding to the future location of the metatarsals. (B) Double transgenic tadpoles were treated with 50 μg/mL Dox for 2�3 days at the indicated stages, and the fused tibiofibula (TF) was imaged at NF63. Note that treatment with Dox at later stages resulted in GFP expression increasingly restricted to epiphyseal ends. Fe, femur Is, ishium; Me, metatarsals; Ph, phalanges; tf, tibiale fibulare.
|
|
Fig. S3. Tracing pectoral girdle cartilages labeled with GFP using the rtTA-dependent system. (A, C, E) Lateral perspective. (B, D) Ventral perspective. (A) Dox-induction at NF59 labels all pectoral girdle cartilages, being evenly distributed across the distal and proximal suprascapular cartilage (DSC and PSC, in relation to the shoulder joint) and concentrated at the medial region of the procoracoid (PR) and epicoracoid (EC). (B, C) Green fluorescent protein (GFP) labeling was even more concentrated in the medial portions of the procoracoid and epicoracoid, and the proximal suprascapular cartilage by NF63. (D) Similar to the tibiofibular cartilage, later GFP induction (NF63) preferentially labels the ends of these cartilages, and there is only midline expression in the procoracoid and epicoracoid. (E) Distal suprascapular GFP expression remains punctate following Dox-induction during either NF59 or NF63.
|
