We have found that in pigmented arrestin knockout mice with
defective rhodopsin shut-off and prolonged
photoresponse,
14 photoreceptors were progressively lost
when the animals were maintained in cyclic light. The fact that the
degeneration was prevented when the knockout mice were reared in the
dark indicates that the excessive signal flow was light mediated.
When the pigmented arrestin knockout mice were exposed to constant
light, photoreceptor degeneration was markedly accelerated. The degree
of light-induced damage in the pigmented arrestin knockout mice
(Fig. 3) was almost identical with that seen in albino
mice.
15 18 Thus, the arrestin knockout results in a change
in susceptibility of the retina to constant light that apparently
negates the high level of protection normally afforded by eye
pigmentation.
13 19 20 The normal-pigmented control mice
were undamaged for up to 3 weeks of constant light, as expected from
results in previous studies in which similarly light-exposed pigmented
mice showed no degeneration for up to 18 weeks
12 or 23
weeks.
13
Another significant difference between light damage in the arrestin
knockout mice and normal albino mice is that the arrestin knockout mice
show a greater sensitivity to light in the inferior hemisphere
(Fig. 3) , whereas normal albino mice are more severely damaged in the
superior hemisphere of the eye.
15 21
One explanation of the much greater susceptibility of the arrestin
knockout mice to excessive light and the reversal in hemispheric
sensitivity may lie in different degeneration mechanisms from those
seen in normal albino animals usually used in constant light
experiments. The main damaging agent in the nonpigmented albino eye is
thought to be reactive oxygen species.
22 23 24 25 26 27 However, it
is unlikely that significant levels of free radicals were generated
from the amount of light irradiating the retinas in the pigmented
arrestin knockout mice, given that normal pigmented mice show no damage
with up to 23 weeks of similar constant light exposure.
13 Nevertheless, the amount of light entering the pigmented eye should be
sufficient to generate a signal flow that might be matched only by
bright-light exposures when normal shut-off is in place. Clearly,
direct experimental evidence is needed to ascertain the levels of
reactive oxygen species in the arrestin knockout mice, but our findings
suggest that the arrestin mouse model can allow for a functional
dissection of two molecular bases of pathogenesis: constitutive signal
flow and free radical generation.
Certain experimental results appear to be in conflict with the
equivalent-light hypothesis. For example, transgenic
mice
28 and rats
29 overexpressing rhodopsin
that cannot be properly turned off by phosphorylation (Ser334ter) show
photoreceptor cell loss independent of light exposure.
30 Overexpression of Lys296Glu in photoreceptors of transgenic mice also
causes retinal degeneration that is apparently not related to elevated
rhodopsin activity, because it is inactivated by arrestin
binding.
31 It should be pointed out that these animal
models were generated by a gene-addition technique in which the
transgene is expressed in addition to the endogenous wild-type
rhodopsin. Importantly, it has been observed that rhodopsin overdosage,
itself, can be detrimental to photoreceptors.
32 The
carboxyl terminal of rhodopsin, furthermore, has been implicated in
vectorial transport of rhodopsin in photoreceptors
30 33 and polarized MDCK cells.
34 Deletion of this domain can be
expected to disrupt rhodopsin transport and adversely affect the health
of photoreceptors through a mechanism that is unrelated to
phototransduction. These confounding variables therefore interfere with
the proper testing of the equivalent-light hypothesis. In the arrestin
knockout mice used in the present study, the only perturbation to the
system was the removal of this capping protein, leading to a defined
defect in phototransduction shut-off. Our results therefore provide
strong support to the notion that constitutive signal flow is a
stimulus for photoreceptor cell death. In other mice with a clearly
defined defect in phototransduction shut-off—that is, in rhodopsin
kinase knockout mice—constitutive signal flow appears to be a stimulus
for photoreceptor cell death.
35
It has been clearly shown in the normal rat retina that the superior
hemisphere is damaged more severely by excessive light than the
inferior hemisphere, regardless of the pigmentation type or direction
of the light source.
19 36 Thus, some undefined intrinsic
difference exists in the two hemispheres of the rat retina in the
response to constant light, and a similar increased susceptibility of
the superior hemisphere exists in the mouse retina.
15 21 The significantly increased susceptibility of the inferior hemisphere
to constant light in the arrestin knockout mice also suggests that
asymmetry exists in the substrate for the degeneration. This remains to
be identified.
It is thought that cone photoreceptors are lost as a consequence of rod
cell death.
37 38 Because of this dependency of cones on
rod survival, both daytime vision and nighttime vision are eventually
lost, even when the primary defect lies in the rod photoreceptors. We
provide evidence that rod photoreceptors die from constitutive signal
flow that is light induced. The progression of this cell death may
eventually lead to cone loss and subsequently to total blindness, as is
evidenced in some patients with diagnosed Oguchi disease. However, we
have now found that photoreceptor cell death can be prevented by
removing the light stimulus in arrestin knockout mice. Our results
therefore provide an incentive for restricting light exposure in those
patients who have retinal disorders arising from constitutive signal
flow.
There is accumulating evidence that photoreceptors undergoing inherited
and age-related retinal degenerations may, in general, be more
susceptible to the damaging effects of excessive
light.
39 40 41 42 The arrestin knockout mice, as far as we are
aware, are the most sensitive to the damaging effects of light of any
of the rodent models tested and are the first pigmented model to show
progressive retinal degeneration due simply to cyclic light exposure.
This underscores the notion that patients with mutations leading to
nonfunctional arrestin and rhodopsin kinase, such as Oguchi disease,
should avoid excessive light exposure.