In this study, we examined the consequences of p27 gene inactivation for maintenance of adult lingual taste buds. To date, p27 is the only member of the Cip/Kip family of Cdk inhibitors known to be produced by taste cells . Given the important anti-proliferative role played by this cell cycle regulator in other sensory and neural systems, and the effects on tissue organization that result from disruption of its corresponding gene [58, 64–71], we predicted that absence of p27 would have consequences for taste cell number and/or taste bud size. Our experiments, however, show that loss of a functional p27 gene produces little in the way of overall structural changes in the taste epithelium. This lack of an overt phenotype, nevertheless, belies substantial alterations in taste cell turnover.
One alteration we see is a greater proliferation of precursors in taste epithelium of p27
mice. BrdU incorporation studies reveal a 2-fold increase in the number of S phase-labeled cells within mutant taste buds during the first week after BrdU injection. Evidence to date has generally supported the view that taste cell precursors arise from progenitors that lie outside of the taste bud and enter the bud in an immature state, as basal cells, to undergo maturation [[1, 2, 5, 29]; however, see ]. Within this framework, our data indicate that, in mutants, more precursor cells are entering taste buds per unit time. This interpretation is consistent with results obtained in other systems indicating that p27 participates in regulating the timing of cell cycle exit, and, in this role, determines in part the number of cells that will go on to differentiate [51, 54, 72]. Clearly, p27 is not the only factor important for cell number determination, as cell cycle exit and differentiation do continue to take place in its absence. Recent in vitro studies of mRNA expression have identified other regulators that are present in the taste epithelium and thus are candidates for a role in contributing to, or partially substituting for p27 in, supporting these processes [46, 73].
In parallel with an increase in the number of taste cell precursors incorporated in p27
taste buds, we observed an elevated level of cell death in mutant buds. This was initially evident as a decline of BrdU-immunopositive cells from twice as many as in wild-type buds at day 7, to an equal number at day 15 post-injection. It was also confirmed by direct measurement of significantly greater numbers of apoptotic bodies in mutant taste buds. Thus, the fact that there are no overall changes in taste bud size and cell number with this mutation could be explained by a coincident change of equal magnitude in these two processes, which would be expected to counteract each other. Although measurements of apoptotic bodies in mutants indicate only a 33% increase in the steady state number of dying cells, compared to a 100% increase in the number of BrdU-labeled cells, direct comparison of the two measures is not necessarily warranted. Apoptotic bodies represent only those cells caught in the brief final stages of apoptosis, which are typically removed quickly from the tissue; newly generated, BrdU-labeled cells, on the other hand, persist for much longer. Our counts, therefore, very likely underestimate the amount of ongoing cell death in taste buds, as well as the relative change in this process in mutants. Together, our findings imply that removal of p27 from the system results in a state of chronically high cell flux through the taste buds.
Increased apoptosis in p27
taste buds can be interpreted as a homeostatic response to an increased input of precursors into buds ("push"), due to an enhanced production of cells by taste progenitors. In this view, as long as the cell death machinery is intact, taste buds are capable of autonomously regulating their cell numbers, independently of progenitor proliferation rate. However, loss of p27 could additionally or alternatively be influencing cell death directly. In differentiating mouse embryonic stem cells in vitro, deletion of p27 increases the number of cells that undergo apoptosis, indicating that this Cdk inhibitor normally has a survival function in these cells . If p27 plays a similar role in the taste system, a null mutation could promote cell death and induce the requirement for enhanced generation of replacement cells to maintain the taste bud ("pull"). In either case, our results suggest that taste buds may be subject to growth compensation, and that there are global mechanisms at work to regulate the size of taste buds within strict parameters [75, 76].
Our experiments combining BrdU birth-dating with immunolabeling for p27 define more precisely the association of p27 up-regulation with the initiation of phenotypic maturation in taste cells. In the newly generated population, p27 is expressed strongly after about three days post-injection. Others have reported that the period between 2 and 5 days post-injection sees the onset of expression of various proteins associated with phenotypic and functional maturity, including the signaling molecules gustducin [77–79] and PLCβ2 [80, 81], the synapse-related SNAP-25 , cytokeratins 8 and 18 [35, 82], and T1R3 . Peak expression of these molecules has been reported to occur at about 6-7 days post-injection [77, 80]; in our experiments, the percentage of BrdU-labeled cells that were immunoreactive for p27 continued to increase to day 7.
Interestingly, however, while high levels of p27 expression are temporally associated with cell maturation, the protein is present in less than 50% of BrdU-labeled cells at any time. Theoretically, this limited expression could reflect restriction to a particular stage of cell maturity, or to a discrete cell type (or both). In the present studies, we observed p27 co-expressed along with either PLCβ2 or SNAP-25, proteins which are mostly present in receptor and presynaptic cells, respectively [16, 19, 26]. On the other hand, we could not verify co-labeling for p27 in NTPDase2-positive cells, which are likely to be Type I cells . The simplest interpretation of these data is that p27 expression is restricted to and maintained in maturing and mature cells in the bud, excluding the glia-like supporting cell population. Curiously, this is distinctly different from the pattern observed in the peripheral visual and auditory systems. Here, although p27 is expressed in nearly all cells at the time of differentiation, including sensory cell precursors, maintained expression of p27 in adults is limited to the mature supporting/glial populations [58, 65, 66, 68, 83, 84].
The reason for sustained, high expression of p27 in certain groups of taste cells and not others is unclear. There is evidence in developing neuronal systems that different Cdk inhibitors may time cell cycle exit in distinct subsets of progenitor cells that give rise to different cell types [58, 85]. This raises the possibility that mature non-p27-expressing taste cells arise from a progenitor population in which a different Cdk inhibitor performs this function. The Cip/Kip family member p21 is also expressed in mouse taste buds, but p21-labeled cells represent only a small fraction of taste cells , and thus seem unlikely to correspond to the large population of non-p27-expressing cells we see in the bud. A possible role for other Cdk inhibitors (i.e., the INK4 family) in this regard requires further investigation. Also, in non-gustatory sensory systems, the cells that continue to express p27 in adulthood generally retain the ability to re-enter the cell cycle and proliferate in response to injury or other pathological conditions, a process which includes down-regulation of p27 [67, 86]. While mature taste cells are generally considered to be post-mitotic, evidence for a small fraction of cycling cells within buds has recently been reported , suggesting that in some circumstances certain taste cells may retain this ability. Thus, it is possible that one role of this Cdk inhibitor is to generally restrict cell cycle reentry in functionally mature cells. However, we found little evidence for ectopic cell division within taste buds of p27-null animals, again suggesting that if p27 plays such a role, it is not acting alone.
While p27 is clearly important in taste bud turnover, our data indicate that it does not have a direct role in determining taste cell fate. Triple-label immunohistochemical studies show that lack of p27 does not alter the proportion of cells that label for each of three functional cell type marker proteins. This finding is in agreement with the general consensus in developing systems that the role of p27 is restricted to regulating cell number and not cell differentiation [58, 65, 85]. In our studies, the dispersed CV papilla preparation proved a reliable method for phenotypic evaluation of taste cells by immunolabeling. In quantifying the differential labeling, we found that the variability was low within groups as well as between the two groups of animals, and that the proportions of differentially-labeled cell types we found were broadly within the ranges that would be expected, based on data from intact mouse CV . Importantly, we did not find any double-labeled cells in these preparations, indicating that the antibodies we used labeled distinct taste cell populations.
Overall, these studies demonstrate, for the first time, that the developmentally important cell number regulator p27 also plays a role in the regenerative cell system of adult mouse taste buds. Given its demonstrated importance in several developing sensory systems, it seems likely that p27 might function in the developing taste epithelium as well. However, lack of the protein from birth does not ultimately alter the number or organization of taste papillae, buds or cells. Nonetheless, because in adults the protein seems largely to regulate the dynamics of taste cell life cycles, a similar effect may be evident as alterations in developmental timing and, consequently, sequencing of events leading to a functional peripheral taste system. Defining the role of p27 in the development of this system is the object of our continuing studies. Finally, the implications of chronic increase in cell flux through the taste bud for gustatory function, and for maintenance of the taste epithelium in aging animals, are unclear. Resolution of these questions also awaits future study.