Penn A A, Wong R O, Shatz C J
Department of Molecular and Cell Biology, University of California, Berkeley 94720.
J Neurosci. 1994 Jun;14(6):3805-15. doi: 10.1523/JNEUROSCI.14-06-03805.1994.
During the first 3 weeks of postnatal development in the ferret retina, cells in the ganglion cell layer spontaneously generate waves of electrical activity that travel across the retina in the absence of mature photoreceptors (Meister et al., 1991; Wong et al., 1993). Since few chemical synapses are present at the earliest stages when waves are present, we have explored whether gap junctions could act to correlate the activity of cells in the immature ganglion cell layer. Retinal ganglion cells in a living in vitro preparation from postnatal day 1 (P1) to P45 were intracellularly injected with the tracer Neurobiotin and the fluorescent dye Lucifer yellow, molecules that are known to pass through gap junctions. Lucifer yellow consistently filled only the injected cell, whereas Neurobiotin filled not only the injected cell but also passed to a constellation of neighboring cells. Coupling revealed by Neurobiotin is seen as early as P1, but, at this stage, it was not possible to identify the various morphological types of cells that were coupled. Thereafter, alpha ganglion cells showed homologous coupling to other alpha cells and to both conventionally placed and displaced amacrine cells. Likewise, gamma ganglion cells appeared coupled to other gamma cells and to amacrine cells. However, beta ganglion cells never showed tracer coupling in the neonatal or in adult retinas. The percentage of alpha and gamma cells that were coupled to other cells increased progressively with age. By the end of the third postnatal week, the pattern of Neurobiotin coupling in the ferret retina was adult-like, with virtually every injected alpha cell showing tracer coupling. Our observations suggest that intercellular junctions able to pass Neurobiotin are present in the inner plexiform layer during the period when the firing of retinal ganglion cells is highly correlated. Such junctions could contribute to synchronization of the activity of subsets of neighboring ganglion cells during development, but it cannot be the sole mediator of this activity because beta cells, which also participate in the correlated activity, showed no coupling at any stage. In addition, the continued presence of coupling in the adult retina implies that other changes in retinal circuitry are likely to contribute to the disappearance of the waves.
在雪貂视网膜出生后发育的前三周，神经节细胞层中的细胞会自发产生电活动波，这些电活动波在没有成熟光感受器的情况下在视网膜上传播（迈斯特等人，1991年；王等人，1993年）。由于在电活动波出现的最早阶段几乎没有化学突触，我们探究了缝隙连接是否能够使未成熟神经节细胞层中的细胞活动产生关联。从出生后第1天（P1）到第45天的活体视网膜体外制备物中的视网膜神经节细胞被细胞内注射了示踪剂神经生物素和荧光染料路西法黄，已知这两种分子能通过缝隙连接。路西法黄始终只填充注射的细胞，而神经生物素不仅填充注射的细胞，还传递到一群相邻细胞。神经生物素显示的耦合最早在P1时就能看到，但在这个阶段，无法识别被耦合的各种形态类型的细胞。此后，α神经节细胞显示出与其他α细胞以及常规位置和移位的无长突细胞的同源耦合。同样，γ神经节细胞似乎与其他γ细胞以及无长突细胞耦合。然而，β神经节细胞在新生或成年视网膜中从未显示出示踪剂耦合。与其他细胞耦合的α和γ细胞的百分比随年龄逐渐增加。到出生后第三周结束时，雪貂视网膜中神经生物素耦合的模式已类似成年期，几乎每个注射的α细胞都显示出示踪剂耦合。我们的观察结果表明，在视网膜神经节细胞放电高度相关的时期，能够通过神经生物素的细胞间连接存在于内网状层中。这种连接可能有助于发育过程中相邻神经节细胞亚群活动的同步，但它不可能是这种活动的唯一介导因素，因为同样参与相关活动的β细胞在任何阶段都没有显示出耦合。此外，成年视网膜中耦合的持续存在意味着视网膜回路中的其他变化可能导致电活动波的消失。
