Control of neuronal size homeostasis by trophic factor-mediated coupling of protein degradation to protein synthesis

Abstract

We demonstrate that NGF couples the rate of degradation of long-lived proteins in sympathetic neurons to the rate of protein synthesis. Inhibiting protein synthesis rate by a specific percentage caused an almost equivalent percentage reduction in the degradation rate of long-lived proteins, indicating nearly 1:1 coupling between the two processes. The rate of degradation of short-lived proteins was unaffected by suppressing protein synthesis. Included in the pool of proteins that had increased half-lives when protein synthesis was inhibited were actin and tubulin. Both of these proteins, which had half-lives of several days, exhibited no degradation over a 3-d period when protein synthesis was completely suppressed. The half-lives of seven other long-lived proteins were quantified and found to increase by 84-225% when protein synthesis was completely blocked. Degradation-synthesis coupling protected cells from protein loss during periods of decreased synthesis. The rate of protein synthesis greatly decreased and coupling between degradation and synthesis was lost after removal of NGF. Uncoupling resulted in net loss of cellular protein and somatic atrophy. We propose that coupling the rate of protein degradation to that of protein synthesis is a fundamental mechanism by which neurotrophic factors maintain homeostatic control of neuronal size and perhaps growth.

Figure 1

Degradation of total cellular protein in sympathetic neurons. (A) Biphasic time-course of loss of TCA-precipitable counts in cultures maintained in medium containing NGF and labeled with a 1-h hot-pulse/1-h cold-chase protocol. (B) Monophasic time-course of loss of TCA-precipitable counts in cultures labeled with a 24-h hot-pulse/6-h cold-chase protocol and then maintained in medium containing or lacking NGF. This pulse-chase procedure resulted in enriched labeling of long-lived proteins. Inset shows TCA-precipitable counts appearing in the medium over 30 h after NGF-deprivation. Open squares in the inset show data from cultures deprived of NGF and maintained in a viable state by 1 μg/ml CHX. CHX prevented loss of TCA-precipitable counts into the culture medium. The other symbols in the inset have the same meaning as in the main graph. Acid-precipitable counts that disappeared from cells were found in the culture medium as acid-soluble counts suggesting complete degradation of proteins to amino acids and the release of the amino acids from cells (not shown). (C) CHX (1 μg/ ml) did not affect the more rapid component of degradation in the presence or absence of NGF. However, the slower component of degradation was blocked by CHX treatment in the presence, but not absence, of NGF. Dotted line is the same as the fitted line in Fig. 1 A. (D) CHX (1 μg/ml) almost completely blocked degradation of the long-lived pool of proteins over the period shown in the presence of NGF, but not in its absence. Dotted line is the same as the fitted line in Fig. 1 B. Data are shown as percentage of TCA-precipitable counts measured at the initial time-point (T ₀). n = 19–50 cultures from 2 to 5 platings for each data point in A and C. In B and D, n = 23–40 cultures from 3 to 4 separate platings for each data point except for the inset where n = 4 cultures from a single plating for each data point. Lines are the sums of best least-squares fits of two exponential equations to the data in A and C and single exponential equations in B and D. The line in the inset is not fitted to the data. The curve fit in A was started at 6 h because of a lag phase before degradation started. Since, in the presence of CHX (C), this lag phase was absent, it probably resulted from a cytoplasmic pool of labeled amino acids that continued to be incorporated into proteins after extracellular washout of the label.

Figure 2

Effect of suppressing protein synthesis with CHX (1 μg/ml) on degradation of individual protein species in sympathetic neurons. Autoradiogram showing the time-course of decay of individual proteins. CHX inhibited degradation when NGF was present in the culture medium but not when it was absent. Molecular weight (kD) of protein bands is indicated to the right. Actin decay was completely blocked over a 72-h period when CHX and NGF were both present in the culture medium. In CHX-maintained cultures deprived of NGF, degradation was approximately the same as in control cells maintained in NGF without CHX. (C) Decay of tubulin was also completely blocked by CHX treatment of cultures maintained in medium containing NGF. As with actin, in the absence of NGF decay was approximately the same as in control cells maintained in NGF without CHX. (D) Degradation of a protein(s) with a molecular weight of 100 kD was suppressed, but not completely blocked, by the presence of CHX in NGF-maintained cultures. Again, when NGF was absent, CHX did not affect degradation of this protein(s). We have previously identified the two prominent bands in A as being actin and tubulin (not shown). Each band shown may contain more than one protein. For example, the tubulin band contains both α and β tubulin. n = 3–9 from 2 to 3 separate platings for each data point. For quantification of degradation, the density of all protein bands was normalized to their densities at T₀ on the same autoradiogram.

Figure 3

Dose- and time-dependent suppression of rate of protein synthesis by inhibitors of translation and transcription and by NGF withdrawal. (A) Exposure of neurons to different concentrations of CHX for 4 h or 72 h resulted in similar dose/response curves for suppression of protein synthetic rate. (B) A similar rapid, sustained suppression of rate of protein synthesis was seen with the translational inhibitor ANIS. (C) The transcriptional inhibitor ACT D caused a slowly developing suppression of synthesis. (D) Time-course of decrease of protein synthetic rate after withdrawal of NGF. n = 11–36 cultures from 2 to 4 platings for each data set except for NGF withdrawal data where n = 6 to 7 for each data point from a single plating. Curves are best least-squares fits of logistic equations to the data.

Figure 4

Dose-dependent suppression of degradation of long-lived proteins by inhibitors of translation and transcription. (A) CHX effect on protein degradation. (B) ANIS effect on protein degradation. (C) ACT D effect on protein degradation. (D) Rate of degradation of long-lived proteins was a linear function of protein synthesis rate. Suppression of protein synthesis rate for 72 h caused an almost equivalent percentage reduction in protein degradation rate over the same period. The rate constant of degradation (k _d) was, thus, similarly related to the rate constant of protein synthesis (k _s). The larger deviations between CHX and ANIS at higher concentrations of each (i.e., lower protein synthesis and degradation) were because of ANIS being somewhat less effective at suppressing protein degradation at concentrations that completely suppressed protein synthesis than was CHX (Figs. 3, A and B, and Fig. 4 B). This difference may have been caused by slight ANIS toxicity since, at concentrations only slightly higher than those shown here, ANIS caused substantial toxic effects. CHX did not have apparent toxic effects at any concentration tested. Another inhibitor of protein synthesis, puromycin, was highly toxic to cells (data not shown). Neurons in A–C were labeled as in Fig. 1, B and D and TCA-precipitable counts remaining were measured 72 h after the initial time point. n = 12–31 cultures from 2 to 3 separate platings for each data point in A–C. Curves in A and B are best least squares fits of logistic equations to the data. Protein synthesis data in D are 72 h data from Fig. 3, A and B and degradation data are from Fig. 4, A and B. Lines are linear regressions of CHX and ANIS data or the combination of the two data sets together (solid line).

Figure 5

The somatic atrophy preceding apoptosis induced by NGF withdrawal appears to be caused by decreased total cellular protein. (A) Phase contrast photomicrographs illustrating atrophy after NGF-deprivation. Left, control cells maintained in culture medium containing NGF; middle, cells maintained in NGF and exposed for 30 h to CHX (1 μg/ml) had soma diameters similar to those of control cells; right, cells deprived of NGF and maintained in 1 μg/ml of CHX for 30 h had shrunken somas. (B) Time-course of changes in average soma diameter of neurons receiving the indicated in treatments. Inset shows calculated cell volumes at different time-points; calculations assumed a roughly spherical soma. (C) Total protein content of cultures receiving the indicated treatments for 30 h. Asterisks, significant difference from protein content measured at the initial time-point. Data are representative of three separate experiments. Approximately 20 phase-contrast photomicrographs of randomly chosen fields of view were taken for each experimental treatment and time-point in part B. Diameters of phase-bright somas in these photographs were measured. At the 24 h and 30 h time points, many cells in NGF-deprived cultures had died and remained as debris in the culture dish. At these time points only diameters of cells that remained alive (phase bright) were measured. None of the cells deprived of NGF and maintained in CHX fragmented or died during the period shown. n = 50 from a single plating for each point in B. n = 10–16 from a single plating for each condition in C. All treatments started 5 d after plating. Bar, 10 μm.

Figure 5

The somatic atrophy preceding apoptosis induced by NGF withdrawal appears to be caused by decreased total cellular protein. (A) Phase contrast photomicrographs illustrating atrophy after NGF-deprivation. Left, control cells maintained in culture medium containing NGF; middle, cells maintained in NGF and exposed for 30 h to CHX (1 μg/ml) had soma diameters similar to those of control cells; right, cells deprived of NGF and maintained in 1 μg/ml of CHX for 30 h had shrunken somas. (B) Time-course of changes in average soma diameter of neurons receiving the indicated in treatments. Inset shows calculated cell volumes at different time-points; calculations assumed a roughly spherical soma. (C) Total protein content of cultures receiving the indicated treatments for 30 h. Asterisks, significant difference from protein content measured at the initial time-point. Data are representative of three separate experiments. Approximately 20 phase-contrast photomicrographs of randomly chosen fields of view were taken for each experimental treatment and time-point in part B. Diameters of phase-bright somas in these photographs were measured. At the 24 h and 30 h time points, many cells in NGF-deprived cultures had died and remained as debris in the culture dish. At these time points only diameters of cells that remained alive (phase bright) were measured. None of the cells deprived of NGF and maintained in CHX fragmented or died during the period shown. n = 50 from a single plating for each point in B. n = 10–16 from a single plating for each condition in C. All treatments started 5 d after plating. Bar, 10 μm.

Figure 5

The somatic atrophy preceding apoptosis induced by NGF withdrawal appears to be caused by decreased total cellular protein. (A) Phase contrast photomicrographs illustrating atrophy after NGF-deprivation. Left, control cells maintained in culture medium containing NGF; middle, cells maintained in NGF and exposed for 30 h to CHX (1 μg/ml) had soma diameters similar to those of control cells; right, cells deprived of NGF and maintained in 1 μg/ml of CHX for 30 h had shrunken somas. (B) Time-course of changes in average soma diameter of neurons receiving the indicated in treatments. Inset shows calculated cell volumes at different time-points; calculations assumed a roughly spherical soma. (C) Total protein content of cultures receiving the indicated treatments for 30 h. Asterisks, significant difference from protein content measured at the initial time-point. Data are representative of three separate experiments. Approximately 20 phase-contrast photomicrographs of randomly chosen fields of view were taken for each experimental treatment and time-point in part B. Diameters of phase-bright somas in these photographs were measured. At the 24 h and 30 h time points, many cells in NGF-deprived cultures had died and remained as debris in the culture dish. At these time points only diameters of cells that remained alive (phase bright) were measured. None of the cells deprived of NGF and maintained in CHX fragmented or died during the period shown. n = 50 from a single plating for each point in B. n = 10–16 from a single plating for each condition in C. All treatments started 5 d after plating. Bar, 10 μm.

Figure 6

Summary models. (A) Schematic diagram of possible relationship between protein degradation/synthesis coupling, size homeostasis, and growth homeostasis. (B) Hypothetical model for control of protein degradation by a mediator that is made short-lived by the presence of NGF.