Analysis of kinesin-2 function in photoreceptor cells using synchronous Cre-loxP knockout of Kif3a with RHO-Cre - PubMed (original) (raw)

Analysis of kinesin-2 function in photoreceptor cells using synchronous Cre-loxP knockout of Kif3a with RHO-Cre

David Jimeno et al. Invest Ophthalmol Vis Sci. 2006 Nov.

Abstract

Purpose: To determine the relationship between the presence of kinesin-2 and photoreceptor cell viability and opsin transport, by generating RHO-Cre transgenic mice and breeding them to mice with a floxed kinesin-2 motor gene.

Methods: Different lines of RHO-Cre transgenic mice were generated and characterized by transgene expression, histology, and electrophysiology. Mice from one line, showing uniform transgene expression, were crossed with Kif3a(flox)/Kif3a(flox) mice. The time courses of photoreceptor Cre expression, KIF3A loss, ectopic opsin accumulation, and photoreceptor cell death were determined by Western blot analysis and microscopy.

Results: One of the RHO-Cre lines effected synchronous expression of Cre and thus uniform excision of Kif3a(flox) in rod photoreceptors across the retina. After the neonatal production of CRE and the initiation of KIF3A loss, ectopic accumulation of opsin was detected by postnatal day (P)7, and ensuing photoreceptor cell death was evident after P10 and almost complete by P28. Of importance, the photoreceptor cilium formed normally, and the disc membranes of the nascent outer segment remained normal until P10.

Conclusions: The RHO-Cre-8 mice provide an improved tool for studying gene ablation in rod photoreceptor cells. Regarding kinesin-2 function in photoreceptor cells, the relatively precise timing of events after CRE excision of Kif3a(flox) allows us to conclude that ectopic opsin is a primary cellular lesion of KIF3A loss, consistent with the hypothesis that opsin is a cargo of kinesin-2. Moreover, it demonstrates that KIF3A loss results in very rapid photoreceptor cell degeneration.

PubMed Disclaimer

Figures

FIGURE 1

FIGURE 1

RHO-Cre transgene and its expression in the retina. (A) The components of the RHO-Cre transgene. (B–E) The distribution of Cre-recombinase activity in the retinas of _RHO-Cre_-8 (B, D), _RHO-Cre_-16 (C), and control (lacking RHO-Cre) (E) mice was evaluated by X-Gal staining of the retinas, after crossing with Gt(ROSA)26Sortm1Sor reporter mice. Retinal wholemounts are from the double transgenic mice at ages 1 month (B, C) and 7 days (D, E) of age. (F–H) The cellular location of Cre-recombinase activity in the retinas of _RHO-Cre_-8 (F) and _RHO-Cre_-16 (G, H) mice was evaluated by X-Gal staining of retinal sections of double-transgenic mice at 1 month of age. Staining of the outer nuclear layer (ONL) was uniform throughout the retina in _RHO-Cre_-8 mice, but patchy in the _RHO-Cre_-16 mice (G, H). (I–K) Frozen sections of retinas from 1-month-old _RHO-Cre_-8 mice were immunolabeled with antibodies against CRE (red) and cone opsin (green), showing that CRE is expressed only in rod photoreceptors. The CRE signal was detected in the nuclei of rod photoreceptor cells, but it does not overlap with the cone opsin signal (e.g., arrows). The red signal in the sclera and choroid of (I) and (K) is due to autofluorescence and is present in control sections stained without primary antibody (not shown). Scale bar: (F–K) 25 _μ_m.

FIGURE 2

FIGURE 2

(A) Retinal function of RHO-Cre-8 mice. Rod and cone functions in RHO-Cre-8 mice were evaluated by ERG at the ages indicated. Data are presented as the percentage of the normal control at each time point. At least three mice of each genotype were studied at each age. The means ± SD of each type of response are plotted**. (B)** Photoreceptor cell loss in RHO-Cre-8 mice. The retinal structure of RHO-Cre-8 mice at different ages was evaluated by histology. At 4 weeks of age, the outer nuclear layer (ONL) and outer segments (OS) were normal. At later ages, the ONL and OS layer were both reduced in thickness. Scale bar, 25 _μ_m.

FIGURE 3

FIGURE 3

Western blot analysis of retinal proteins, immunolabeled with antibodies against CRE (A, B) and KIF3A (C). Each lane was loaded with 25% (A, C) or 20% (B) of lysate, containing all retinal proteins. The retinas were from Kif3aflox/Kif3aflox;RHO-Cre-8 mice (A) and RHO-Cre-8 mice (B) of different ages. The decrease in CRE after P14 in Kif3aflox/Kif3aflox;RHO-Cre-8 mice was due to the loss of photoreceptor cells in these retinas. Note that, in RHO-Cre-8 mice, CRE continued to increase beyond this age. (C) Retinas from either Kif3aflox/Kif3aflox (C) or Kif3aflox/Kif3aflox; RHO-Cre-8 (M) mice of different ages, except that those shown in the two rightmost panels were from a 2-month-old rd1 mouse retina (rd1) and its age-matched control (C).

FIGURE 4

FIGURE 4

Changes in levels of photoreceptor CRE and KIF3A and number of photoreceptor cells in Kif3aflox/Kif3aflox;RHO-Cre-8 retinas. Because CRE was present only in rod photoreceptor cells, its levels were determined directly from retinal lysates. They are shown relative to the maximum level of CRE measured in the Kif3aflox/Kif3aflox;RHO-Cre-8 retinas, which was at P14 (CRE cf. P14). KIF3A levels were determined from measurements of retinal lysates, adjusted by subtracting the estimated contribution from nonphotoreceptor cells. This contribution (˜56%) was determined from the amount of KIF3A in a 2-month-old rd1 mutant mouse retina (at this age, the rd1 retina has no photoreceptor cells) relative to that in a 2-month-old control mouse retina. KIF3A levels and photoreceptor cell numbers are shown as a proportion of that measured in Kif3aflox/Kif3aflox (no RHO-Cre-8) retinas of same-aged littermates. CRE and KIF3A was quan-tified by densitometry from immunolabeled Western blots, such as those shown in Figure 3, where each lane was loaded with the same proportion of total retinal lysate. Photoreceptor nuclei were counted in the areas 500 _μ_m both sides of the optic nerve head. Arrowhead: time point at which mislocalization of opsin was first evident (P7).

FIGURE 5

FIGURE 5

Opsin mislocalization in Kif3aflox/Kif3aflox;RHO-Cre-8 retinas. (A–E) Light micrographs of semithin sections of retinas, immunolabeled with opsin antibody and peroxidase secondary antibody. Opsin was localized normally in the outer segments of the control retinas: P7 (A) and P14 (C), Kif3aflox/Kif3aflox without CRE; P28 (E), RHO-_Cre_-8 only. It was distributed throughout the photoreceptor cells in the mutant P7 (B) and P14 (D) Kif3aflox/Kif3aflox;RHO-_Cre_-8 retinas. Note that at P7 the outer segments were still very small, so that, at the magnification shown in (A) and (B), immunolabeling of the outer segment was barely evident (A, arrows: labeled outer segments). The important point is that some opsin was detected throughout the photoreceptor nuclear layer in the mutant (e.g., B, arrowheads), but not in the control (A). (F–H) Electron micrographs of rod photoreceptor cells, immunogold labeled with opsin antibody. (F) Micrograph showing immunogold label accumulated in the inner segment, beneath the connecting cilium (CC) and basal body (BB) of a photoreceptor from a P7 Kif3aflox/Kif3aflox;RHO-_Cre_-8 retina. (G) Control photoreceptor cell from a P10 animal. IS, inner segment. (H) Mutant photoreceptor cell from a P10 Kif3aflox/Kif3aflox;RHO-_Cre_-8 animal. In the mutant retinas (F, H), photoreceptor cilia and the nascent outer segments had formed and were still intact at P10 (H), despite the defective distribution of opsin. Scale bars: (A–E) 25 _μ_m; (F), 300 nm; (G, H) 1 _μ_m.

FIGURE 6

FIGURE 6

Photoreceptor degeneration due to the loss of KIF3A. Light micrographs of retinas from a control P28 mouse (A) and from P7 (B), P14 (C), and P28 (D)Kif3aflox/Kif3aflox;RHO-_Cre_-8 mice. Sections were prepared from blocks embedded in resin. (E) A cone photoreceptor synaptic terminal (pedicle) from P28 Kif3aflox/Kif3aflox;RHO-_Cre_-8 retina. Arrowheads: synaptic ribbons; in contrast to rod synaptic terminals (spherules), cone pedicles have more than one synaptic ribbon. Scale bar: (A–D) 25 _μ_m; (E) 500 nm.

References

    1. Chen S, Kadomatsu K, Kondo M. Effects of flanking genes on the phenotypes of mice deficient in basigin/CD147. Biochem Biophys Res Commun. 2004;324:147. - PubMed
    1. van den Hurk JA, Hendriks W, van de Pol DJ. Mouse choroi-deremia gene mutation causes photoreceptor cell degeneration and is not transmitted through the female germline. Hum Mol Genet. 1997;6:851–888. - PubMed
    1. Le Y, Sauer B. Conditional gene knockout using cre recombinase. Methods Mol Biol. 2000;136:477–485. - PubMed
    1. Nagy A. Cre recombinase: the universal reagent for genome tailoring. Genesis . 2000;26:99–109. - PubMed
    1. Marszalek JR, Liu X, Roberts EA. Genetic evidence for selective transport of opsin and arrestin by kinesin-II in mammalian photoreceptors. Cell. 2000;102:175–187. - PubMed

Publication types

MeSH terms

Substances

LinkOut - more resources