Within the retina, bipolar cells act as the signal couriers between the photoreceptors that react to light stimuli and the ganglion cells that carry these signals out of the eye and into the cortex. Bipolar cells are so-called because they have two polar extensions that protrude from opposite ends of the soma. One of these extensions extends towards the photoreceptors (connecting to either multiple rods or a single cone) while the other delivers the processed signal to the dendritic arbors of the ganglion cells.
There are at least nine morphological types of cone bipolar (CB) and one type of rod bipolar (RB) cells in the mammalian retina. They have distinctive morphology from amacrine cells and ganglion cells, characterized by varicose axon terminals in the IPL.
Visual Response Function
Bipolar cells receive upstream innervations from the retina's photoreceptors. Although photoreceptors respond to light by hyperpolarizing, bipolar cells can translate this signal in either a sign-conserving or sign-inverting fashion.
Unlike other neurons, bipolar cells do not transmit signals by way of action potentials. They instead make use of a potential gradient that can be modulated by the connecting horizontal and amacrine cells.
Bipolar cells have distinctive morphology from ACs and GCs in the IPL, characterized by the varicose axon terminals. The different types of bipolar cells differ in their dendritic branching pattern, the number of cones contacted, and the stratification level of their axons in the IPL. A type of bipolar cells tile up the entire space of the strata, with little overlapping region to each other.
Within the retina, bipolar cell bodies (somas) are located within the the inner nuclear layer (INL). The cells' dendrites project to the outer plexiform layer, where they receive signals from the photoreceptors and horizontal cells. The cells' axons project to the inner plexiform layer, where they synapse with amacrine and ganglion cells.
Within rabbit retinas, bipolar cells were found to be 41% of all inner nuclear layer cells. Subsequent protein kinase C staining showing that rod and cone bipolar cells were 10% and 31% of the total INL cells, respectively.
This ratio is not constant across all mammal species. In rabbits, the rod to cone bipolar cell ratio is approximately 50 to 1 but within monkeys the ratio is closer to 12.5 to 1.
All bipolar cells share the same general morphological shape: a cell body with two projections that extend in opposite directions. The specific length and arborizations of the dendrites is a factor that is used in classifying the nine different cone bipolar cell and one rod bipolar cell subtypes (image at right).
Bipolar cells make synaptic connections with photoreceptors, as well as amacrine, horizontal, and ganglion cells.
The ON and OFF center circuits within the retina are a product of the either sign-conserving or sign-reversing synapse that the bipolar cell shares with its paired photoreceptor. Sign-conserving synapses result in an OFF center while sign-reversing synapses produce an ON center. The surround portion of the center/surround functionality is dependent upon the aggregate signals from surrounding horizontal and amacrine cells.
Rod bipolar cells express protein kinase C, which is not the case with cone bipolar cells. Antibodies against this molecule can be used to determine the ratio population of bipolar cells that synapse with rods or cones.
Bipolar cells have been known since at least 1894 by Santiago Ramón y Cajal, and possibly back to 1887, as he says of Ferruccio Tartuferi's Sulla anatomia della retina (Archivio per le science mediche, Vol. XI. No. 16. p. 335. 1887): "[Tartuferi] succeeded, above all, to detect the true morphology of bipolar cells in the inner nuclear layer."
Open questions / status / relevance to EyeWire
In eyewire, the primary focus is to catalog the connections made between ganglion, bipolar, and amacrine cells. Understanding these synaptic inputs will allow a better comprehension of how retinal processing is able to create emergent ganglion cell functions, such as direction-selective cells, motion-selective cells, and even general center-surround inhibition circuits. Is it hypothesized that these properties do not only arise once the signals reach the ganglion cells, but rather begin to form within the upstream signals exchanged within the outer plexiform layer (OPL) where the bipolar cells synapse with horizontal cells and photoreceptors. Compiling information on the bipolar cells connected to a given direction-selective ganglion cells may give a clearer picture of how these computations are developed and transmitted.
- H. Wassle; C. Puller; F. Muller; S. Haverkamp. (2009) "Cone Contacts, Mosaics, and Territories of Bipolar Cells in the Mouse Retina" J. Neurosci. '29' (1):106-117 doi:10.1523/JNEUROSCI.4442-08.2009
- Santiago Ramón y Cajal The Retina of the Vertebrates [Die Retina der Wirbelthiere] 1894