Site-Specific Somatic Mutagenesis in the Retinal Pigment Epithelium

Abstract

purpose. Generate site-specific somatic mutations selectively in the retinal pigment epithelium (RPE) in mice.

methods. A transgenic mouse line expressing the Cre recombinase under the control of the tyrosinase-related protein (TRP)-1 promoter was generated. The presence of Cre was determined by in situ hybridization, and Cre-mediated excision of DNA was analyzed by PCR and alkaline phosphatase (AP) histochemistry in reporter mice carrying a loxP-flanked (floxed) retinoid X receptor α (RXRa) gene and in Z/AP mice, respectively.

results. Cre was expressed in the RPE from embryonic day 10.5 to postnatal day 12, resulting in efficient floxed excision of DNA in the RPE from embryonic day 10.5 to adulthood in TRP1-Cre mice. Expressed Cre and excision of DNA were also detected in the ciliary margin of the retina and in some cells in the neural retina, but not in the embryonic periocular mesenchyme or in the choroid.

conclusions. The TRP1-Cre mouse line, which induces efficient Cre-mediated excision of DNA selectively in the RPE, provides a new, powerful tool to study gene functions in the RPE in vivo.

The retinal pigment epithelium (RPE) is a cell monolayer lying between the retina and choroid of the eye. In the adult, the RPE plays essential roles in maintenance of the blood-retinal barrier, which regulates movements of ions, water, and other molecules; transport and metabolism of retinoids required for vision; and phagocytosis of photoreceptor outer segments. ¹ In developing eyes, the RPE can be induced to transdifferentiate into retinal cells in several species, including mammals. ² Moreover, the pigmented ciliary epithelium, an extension of the RPE in the ciliary body, appears to contain retinal stem cells in adult mammalian eyes. ^{3 4} The molecular mechanisms underlying the functions and differentiation of the RPE cells remain poorly understood. 

Recent advances in mouse genetics have enabled tissue-specific gene targeting using the Cre recombinase. ^{5 6} Cre is a P1 bacteriophage protein that catalyzes recombination of DNA between two 34-bp recognition sites (loxP sites). Expression of Cre, driven by a tissue-specific promoter in animals carrying a loxP-flanked (floxed) DNA segment, allows selective deletion of the DNA segment in that tissue. The tyrosinase-related protein (TRP)-1 gene is expressed in melanin-synthesizing cells, and various promoter regions of this gene have been shown to drive transgene expression in the RPE (reviewed by Beermann ⁷ ). In the current study, we generated a mouse line that expresses Cre under the control of the TRP1 promoter that catalyzes efficient excision of floxed DNA segments selectively in the RPE. 

Methods

Transgenic Mice

The 4.4 kb BamHI-SalI DNA segment containing the mouse TRP1 promoter ⁸ was isolated, blunt-ended, and cloned into the blunt-ended SalI site of pGS-Cre, ⁹ resulting in pTRP1-Cre. Transgenic mice were established with the 6.2-kb NotI fragment of pTRP1-Cre (Fig. 1) as described. ¹⁰ Genotyping for Cre was as described. ¹¹ RXRα^L2/+ mice, ¹² Z/AP mice, ¹³ or Z/APΔ mice (in which the floxed lacZ gene of the Z/AP reporter gene is excised by Cre in the germ line) ¹⁴ have been described elsewhere. The study protocol adhered to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. 

PCR Analysis of Cre-Mediated Excision of DNA

To test for Cre activity, founder mice from each transgenic line were crossed with mice carrying floxed retinoid X receptor α alleles (RXRα^L2/L2). ¹² Genomic DNA was isolated from tails of postnatal day 7 (P7) bigenic offspring (Cre^(tg/0):RXRα^L2/+) and from various organs of 4-week-old mice, as described. ¹¹ The retina, RPE, and choroid from adult mice were separately collected from eyeballs that were treated with 3 mg/mL collagenase type-I (Worthington, Lakewood, NJ) and 1 mg/mL trypsin (Sigma, St. Louis, MO) in DMEM (Life Technologies, Cergy Pontoise, France) for 1 hour at 37°C. The RPE-free retinas were further microdissected to obtain peripheral (P; 0.5-mm wide) and central (C; 1-mm diameter) portions without optic nerve. Polymerase chain reaction (PCR) was performed at a thermal sequence of 94°C (30 seconds) to 58°C (30 seconds) for 40 (or as indicated) cycles using primers UN128 (5′-CTCAAGTGAGGTGGACATTA-3′), WH231 (5′-GAGCTATTGTGCCCTGGAAG-3′), and WH233 (5′-GGAAGCCCAAGATGACCCTC-3′) to detect wild-type (WT; 150 bp; UN128/WH231), floxed (L2; 200 bp; UN128/WH231), and recombined (L⁻; 100 bp; UN128/WH233) RXRα alleles. The PCR products were separated in 2.5% agarose gels and visualized by ethidium bromide staining. Photographs of the stained gels were optically scanned, and the density of each fragment was quantified with NIH Image software (version 1.56; provided in the public domain by the National Institutes of Health, Bethesda, MD, and available at http://rsb.info.nih.gov/nih-image/) to evaluate the extent of L2 to L⁻ conversion. 

In Situ Hybridization

Ten-micrometer-thick fresh-frozen sections of whole embryos (embryonic day [E]10.5, E11.5, and E14.5), heads (postnatal day [P]1), and eyes (P12 and adult) of TRP1-Cre mice were fixed in 4% paraformaldehyde in PBS for 10 minutes at room temperature, hybridized with a ³⁵S-labeled Cre antisense riboprobe, and processed as described. ¹⁴  

Histochemical Analysis of Cre-Mediated Excision of DNA

Cre transgenic mice were crossed with the Z/AP reporter mice. ¹³ Male albino offspring carrying both the Cre and Z/AP transgenes were crossed with CD1 females to obtain albino bigenic animals at E10.5 and E14.5, at P2 and P12, and in adults (2 months old). Embryos (E8.5 and E13.5) and adult eyes were also obtained from Z/APΔ mice in which the floxed LacZ gene is excised in the germ line. Whole embryos, pup heads, and adult eyes were fixed in 4% paraformaldehyde in PBS for 16 hours at 4°C and cryoprotected in 20% sucrose in PBS for 10 hours at 4°C. Ten-micrometer-thick frozen sections were subjected to alkaline phosphatase (AP) histochemical detection, as described. ¹³ Briefly, sections were heat inactivated for endogenous AP activity in PBS at 70°C for 1 hour, equilibrated with a reaction buffer (100 mM Tris-HCl [pH 9.5], 100 mM NaCl, 50 mM MgCl₂, 0.1% Tween 20, and 2 mM levamisole [Vector Laboratories, Burlingame, CA]) for 15 minutes at room temperature, and incubated with 0.4 mg/mL nitroblue tetrazolium chloride, 0.19 mg/mL 5-bromo-4-chloro-3-indolyl-phosphate, 100 mM Tris (pH 9.5), and 50 mM MgSO₄ (NBT/BCIP tablets; Roche Molecular Biochemicals, Mannheim, Germany) at room temperature for 10 to 20 minutes. 

Results

Injection of the TRP1-Cre transgene into fertilized eggs resulted in 12 independent transgenic mouse lines. Founder animals from each line were crossed with RXRα^L2/L2 mice, and floxed excision of DNA was monitored by PCR. In bigenic animals from seven lines, excised RXRα alleles (RXRα L⁻) were detected in the tail DNA at P7 (data not shown). One of the other five lines exhibiting no excision of DNA in the tail showed Cre-mediated floxed excision of DNA predominantly in the RPE at 4 weeks of age (Fig. 2A , and data not shown). In this line, referred to as TRP1-Cre, some conversion of the RXRα L2 allele to RXRα L⁻ allele also occurred in the neural retina (Fig. 2A) . Semiquantitative PCR indicated that the RXRα L2 allele was excised in more than 95% of the cells in the RPE and in less than 30% and 5% of the cells from the periphery and center of the neural retina, respectively (Fig. 2B) . As expected, no excision of DNA was observed in any of the tissues from control littermates that did not carry the TRP1-Cre transgene (data not shown). 

To further characterize the TRP1-Cre line, the distribution of Cre transcripts was analyzed by in situ hybridization (Fig. 3 ; Table 1 ). At E10.5, Cre transcripts were detected in the dorsal portion of the presumptive RPE (Fig. 3A) , and at E11.5, E14.5, P2, and P12, high levels of transcripts were detected throughout the RPE (Figs. 3B 3C 3D 3E) . Note that low levels of Cre transcripts were observed in other ocular tissues: the ciliary margin of the retina (CM, Fig. 3B ) and optic stalk (OS, Fig. 3G ) at E11.5; the optic nerve (ON, Fig. 3H ) as well as in subsets of cells in the ganglion cell layer (GCL) and outer portion of the neuroblastic layer (NBL) of the retina (Figs. 3C 3H) at E14.5; and the optic nerve (ON), iris (I), and ciliary body (CB) at P2 and P12 (Figs. 3I 3J , and data not shown). It is noteworthy that no Cre transcripts were detected in the embryonic periocular mesenchyme and its derivatives (i.e., the choroid and sclera), at any stage of eye development (Figs. 3A 3B 3C 3D 3E 3F 3G 3H) . At E10.5 and E14.5, Cre transcripts were detected at low levels in small subsets of cells in extraocular tissues: the mesencephalon (MES, Fig. 3K ), trigeminal nerve ganglion (TG, Fig. 3G ), and dorsal root ganglia (DRG, Fig. 3L ). However, no Cre transcripts were detected in any of the tissues from adult (2-month-old) TRP1-Cre mice, including the eye (Fig. 3F) , central nervous system, lung, pancreas, liver, spleen, skin, muscle, stomach, and intestine (data not shown). 

To analyze Cre-mediated recombination of DNA at the cellular level, TRP1-Cre mice were crossed with Z/AP reporter mice. These reporter mice were suitable for histologic detection of Cre-mediated excision of DNA as the Z/APΔ mice, in which the floxed LacZ in the Z/AP reporter gene is excised in the germ line, exhibited strong and ubiquitous AP activity on histologic sections from E8.5 embryos, E13.5 embryo heads, and adult eyes (Figs. 4A 4B , and data not shown). Bigenic TRP1-Cre:Z/AP mice displayed high levels of AP activity in most, if not all, cells of the RPE at E10.5 and E14.5, at P2 and P12, and in adulthood (Figs. 5A 5B 5C 5D 5E) . The AP activity in the neural retina was much weaker than that in the RPE and was distributed, from E14.5 to adulthood, as a gradient increasing from the optic nerve exit point toward the retinal periphery (Figs. 5B 5F , and data not shown). The optic stalk at E10.5 (OS, Fig. 5G ) and optic nerve at E14.5, P2, P12, and adulthood (ON, Figs. 5B 5H , and data not shown) were also positive for AP activity. At P2, P12, and adulthood, AP activity was observed in the iris (I) and ciliary body (CB, respectively, Figs. 5F 5I 5J , and data not shown). Outside the eye region, AP activity was detected in the trigeminal nerve ganglion (TG, Figs. 5A 5K ), dorsal root ganglia at E10.5 and E14.5 (DRG, Fig. 5M , and data not shown), as well as in a small subset of cells in the mesencephalon at E10.5 (MES, Fig. 5L ). In adult TRP1-Cre:Z/AP mice, no AP activity was detected in the central nervous system, lung, pancreas, liver, spleen, skin, and muscle (data not shown). The stomach and intestine, which displayed strong endogenous AP activity, could not be analyzed (data not shown). 

Discussion

In TRP1-Cre transgenic mice, floxed DNA segments were excised selectively in the RPE from midgestation to adulthood. If one takes into account the facts that floxed DNA remains excised in the cells even after they have ceased to express Cre and that the deletion is stably transmitted to the cell progeny, the sites of Cre’s expression and the excision of floxed DNA segments were well correlated in the various organs analyzed, with the notable exception of the neural retina (Table 1) . Indeed, the retinal cells in which floxed DNA segments are excised were distributed as a gradient, being more abundant in the periphery than in the center from E14.5 to adulthood. This gradient cannot be ascribed to heterogeneity in expression of AP from the excised Z/AP gene, because AP was uniformly expressed in the Z/APΔ retina (Fig. 5 and data not shown). In contrast, strong expression of Cre was detected at E10.5 in cells of the ciliary margin of the retina (Fig. 3) , which is reported to contain retinal progenitors in fish and amphibians. ^{15 16} The graded pattern in the retina could be accounted for, at least partly, by incorporation into the developing retina of the progeny of the cells in the ciliary margin in which Cre excises the floxed DNA at E10.5. In any event, our data provide support to the idea that ciliary margin-derived progenitors of the neural retina may also exist in mammals. 

Ectopic expression of Cre and Cre-mediated excision of DNA were observed in some embryonic extraocular tissues, including the mesencephalon, trigeminal nerve ganglion, and dorsal root ganglia, which do not express endogenous TRP1. ¹⁷ In contrast, neither Cre expression nor floxed excision of DNA was seen in the choroid, which expresses endogenous TRP1. ¹⁷ These findings are in keeping with the reported properties of the TRP1 promoter’s activity in several other transgenic mice, ⁷ which, depending on the line, show reporter gene expression in the developing lung, trachea, testis, ectoderm, or forelimb, but not in melanocytes. The discrepancies between the expression patterns of the endogenous TRP1 gene and TRP1-Cre transgene may result from absence of the regulatory sequences and/or positional effects dependent on the integration site. The large variation in the Cre-mediated excision of DNA patterns between our transgenic mouse lines suggests that the latter factor plays a significant role. 

Because it is unlikely that the Cre-mediated excision of DNA observed in extraocular tissues affects eye development and functions, the present TRP1-Cre mouse line should be very useful in studying genes that control the development and pleiotropic physiological functions of the RPE. That TRP1-Cre induces excision of floxed DNA in almost all RPE cells but not in choroidal melanocytes is of interest in the study of RPE-choroid interactions at the molecular level. Note, however, that Cre-mediated excision of DNA in some cells of the neural retina may complicate the interpretation of loss-of-function retinal phenotypes generated with the TRP1-Cre mouse line. The investigator may, however, take advantage of excision of DNA in the ciliary margin of the retina and ciliary pigment epithelium to investigate mechanisms involved in growth and differentiation of the retina, because these tissues may contain retinal stem or progenitor cells. ^{15 16}  

 Supported by funds from the National Center for Scientific Research (CNRS), the National Institute of Health and Medical Research (INSERM), Louis Pasteur University (ULP), the University Hospital of Strasbourg, the College of France, the University Institute of France, the Association for Cancer Research (ARC), the Foundation for Medical Research (FRM), the National League against Cancer, the Human Frontier Science Program, Bristol-Myers Squibb, and European Economic Community (EEC) Contract FAIR-CT97-3320. MMo was supported by fellowships from INSERM and from the Ministry of Education, Culture, Sports, Science, and Technology of Japan.

 Submitted for publication October 29, 2001; accepted December 20, 2001.

 Commercial relationships policy: N.

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Representation of the TRP1-Cre transgene. Black box: the 4.4-kb TRP1 promoter region; open box: the Cre coding region, hatched box: rabbit β-globin exons; dotted box: simian virus (SV)40 polyadenylation signal; arrow: transcription initiation site.

Figure 2.

 

PCR detection (A) and semiquantitative evaluation (B) of Cre-mediated excision of DNA in various organs isolated from 2-month-old TRP1-Cre^(tg/0):RXRα^{L2/ +} mice. The positions of the L2 (200 bp), wild-type (WT; 150 bp), and L⁻ (100 bp) RXRα alleles are indicated. PCR was performed for 40 cycles in (A) and 34, 36, and 38 cycles in (B). c, central retina; p, peripheral retina.

Figure 3.

 

Analysis of expression of Cre in developing and adult TRP1-Cre mice by in situ hybridization. Frontal sections of E10.5 (A), E11.5 (B), and E14.5 (C) embryos. Eye sections (posterior segments) of mice at P2 (D), P12 (E), and adulthood (F). Frontal section of E11.5 (G) and E14.5 (H) embryos. Eye sections (anterior segments) at P2 (I) and P12 (J). Frontal section of E10.5 (K) embryos. Transverse trunk section of an E10.5 embryo (L). Each panel shows dark-field (left) and corresponding bright-field (right) photomicrographs. All sections were prepared from albino mice. RPE, retinal pigment epithelium; L, lens; CB, ciliary body; CM, ciliary margin of the retina; GCL, ganglion cell layer; NBL, neuroblastic layer; INL, inner nuclear layer; ONL, outer nuclear layer; ON, optic nerve; OS, optic stalk; TG, trigeminal nerve ganglion; I, iris; CB, ciliary body; MES, mesencephalon; DRG, dorsal root ganglia. Scale bars: 100 μm.

Table 1.

 

Summary of Cre Expression and Cre-Mediated DNA Excision in TRP1-Cre Mice

	E10.5–E11.5			E14.5			P2			P12			Adult
		a	b	a	b	a	b	a	b	a	b
RPE	+++	+++	+++	+++	++	++	+	++	−	++
Ciliary body and iris	NA	NA	+	+	+	+	+	+	−	+
Neural retina	++*	++*	+, †	+, ‡	−	+, ‡	−	+, ‡	−	+, ‡
Optic stalk or nerve	++	++	+	+	−	+	−	+	−	+
Trigeminal nerve ganglia	+	+	−	+	ND	ND	ND	ND	ND	ND
Dorsal root ganglia	+	+	+	+	ND	ND	ND	ND	ND	ND
Brain	+, §	+, §	−	−	ND	ND	ND	ND	−	−

 The relative levels of Cre expression (lane a) and Cre-mediated excision of floxed DNA (lane b) were estimated by signal analyses of situ hybridization (Fig. 3) and alkaline phosphatase staining, respectively (Fig. 5) , as being either weak (+), medium (++), strong (+++), or not detectable (−). NA, not applicable because the structure has not yet formed; ND, not determined.

*  In ciliary margin.

†  Scattered.

‡  In a gradient, being stronger in the periphery.

§  In mesencephalon only.

Figure 4.

 

Representations of the Z/AP and Z/APΔ transgenes (A) and alkaline phosphatase activity in Z/APΔ mice (B, C). Frontal sections of an E13.5 embryo head and an adult eye from Z/APΔ mice in which the floxed lacZ gene is excised in the germ line cell were stained for AP in (B) and (C), respectively. B, brain; E, eye; T, tongue; GCL, ganglion cell layer; INL, inner nuclear layer; ONL, outer nuclear layer; RPE, retinal pigment epithelium.

Figure 5.

 

Analysis of alkaline phosphatase activity in developing and adult TRP1-Cre:Z/AP double-transgenic mice. Frontal sections through the eye region at E10.5 (A, G) and E14.5 (B). Eye sections (posterior segments) of mice at P2 (C), P12 (D, F, H), and adulthood (E). Eye sections (anterior segments) at P2 (I) and adulthood (J). Frontal sections including the trigeminal nerve ganglion (K) and mesencephalon (L) at E10.5. Transverse trunk section at E10.5 (M). (C, D, E) Central portions near the optic nerve exit point. Abbreviations are defined in Figure 3 . Scale bars: 100 μm.

The authors thank Ian J. Jackson for the TRP1 promoter, Andras Nagy for the Z/AP mice, Philipp Weber for the Z/APΔ mice, and Sylviane Bronner and Valérie Fraulob for help with in situ hybridization.