Radiation induces acute alterations in neuronal function

Abstract

Every year, nearly 200,000 patients undergo radiation for brain tumors. For both patients and caregivers the most distressing adverse effect is impaired cognition. Efforts to protect against this debilitating effect have suffered from inadequate understanding of the cellular mechanisms of radiation damage. In the past it was accepted that radiation-induced normal tissue injury resulted from a progressive reduction in the survival of clonogenic cells. Moreover, because radiation-induced brain dysfunction is believed to evolve over months to years, most studies have focused on late changes in brain parenchyma. However, clinically, acute changes in cognition are also observed. Because neurons are fully differentiated post-mitotic cells, little information exists on the acute effects of radiation on synaptic function. The purpose of our study was to assess the potential acute effects of radiation on neuronal function utilizing ex vivo hippocampal brain slices. The cellular localization and functional status of excitatory and inhibitory neurotransmitter receptors was identified by immunoblotting. Electrophysiological recordings were obtained both for populations of neuronal cells and individual neurons. In the dentate gyrus region of isolated ex vivo slices, radiation led to early decreases in tyrosine phosphorylation and removal of excitatory N-methyl-D-aspartate receptors (NMDARs) from the cell surface while simultaneously increasing the surface expression of inhibitory gamma-aminobutyric acid receptors (GABA(A)Rs). These alterations in cellular localization corresponded with altered synaptic responses and inhibition of long-term potentiation. The non-competitive NMDAR antagonist memantine blocked these radiation-induced alterations in cellular distribution. These findings demonstrate acute effects of radiation on neuronal cells within isolated brain slices and open new avenues for study.

Figure 1. Radiation induces early alterations in NMDAR phosphorylation and internalization of NR1 and NR2A subunits.

(A) Immunoprecipitation of NR2A subunits and semi-quantitative western blotting analysis of NR2A expression and phospho-tyrosine performed on dentate samples collected 30 minutes following a dose of 10 Gy radiation revealed diminished tyrosine phosphorylation. (76.7±4.9% of control, *p = 0.005, n = 7, data presented as mean ± s.e.m.) (B) No alteration in tyrosine phosphorylation at residue 1472 of the NR2B subunit was detected by western blotting from homogenates of dentate slices. (100.5±5.8% of control, p = 0.51, n = 6) (C) AMPA receptor phosphorylation status was assayed by semi-quantitative western blot analysis using phosphorylation site-specific antibodies. (GluR1; phospho-serine831 101.9±5.9% of control, phospho-serine845 102.1±6.3%, GluR2; phospho-serine880 97.6±3.6%, p > 0.4 for each, n = 8) (D) Sham and radiated dentate slices were incubated at 4°C in cross-linking reagent and levels of internal receptors quantified by western blot analysis of lower molecular weight bands corresponding to intact subunits. (NR1, 33.3±9.3% increase in intracellular, NR2A, 28.1±8.6%, *p = 0.004 and 0.005 respectively, students paired t-test with post-hoc Bonferroni correction; NR2B −4.1±7.9%, GluR1 −3.4±9.5%, GluR2/3 −3.8±4.8%, p > 0.3 for each, n = 11) (E) Irradiated and sham cultured slices were collected and incubated at 4°C in membrane-impermeable biotin. Surface receptors were separated by pull down with neutravidin beads and resolved by western blot. (NR1, 24.5±4.8% decrease in surface, NR2A, 23.4±5.9%, *p = 0.01 and 0.03 respectively, n = 10; NR2B 7.1±9.6%, GluR1 4.9±8.7%, p>0.25, n = 10; GluR2/3 2.97±8.8%, p = 0.89, n = 5).

Figure 2. Radiation increases surface expression of GABA_A Beta 2 containing receptors.

(A) Membrane-impermeable cross-linking studies, performed in dentate slices, revealed a decrease in the intracellular pool of Beta2 subunits without a detectable change in the localization of Beta3 containing receptors. (Beta2, 41±7.5% decrease in the intracellular pool of receptor, *p = 0.003, n = 9, Beta3, 0.2±4.8%, p = 0.6, n = 12) (B) Biotinylation of whole cultured hippocampal slices confirmed increased surface retention of GABA receptors containing Beta 2 subunits. (Beta2, 40.2±9.4%, *p = 0.0008, n = 9, Beta3, −5.78±10.16% n = 6, p = 0.9).

Figure 3. Receptor trafficking is observed following therapeutic doses.

Dentate slices were incubated at 4°C in cross-linking reagent 30 minutes following sham, 2 Gy or 5 Gy radiation and internal receptors quantified by western blot. (NR1, 24.4±6.6% and 28.9±7.9% increase in intracellular 2 and 5 Gy respectively, NR2A, 30.7±13.9% and 33.4±12.6%, Beta2, −22.9±9.8% and −19.3±6.8%, *p≤0.05, n = 4 animals).

Figure 4. Radiation is not associated with caspase 3 activation.

Whole hippocampal slice cultures were subject to sham, 10 Gy or staurosporine (Stau) and harvested at 1.5, 6 and 12 hours. Western blotting with anti-caspase 3 did not reveal evidence of caspase cleavage following radiation (12-hour results shown). Staurosporine treatment induced caspase cleavage and appearance of a lower molecular weight band.

Figure 5. Radiation-induced NMDAR endocytosis is tyrosine phosphatase dependent and differential trafficking of NMDA and GABA receptors inhibited by memantine.

(A) Prior to radiation slices were incubated in the tyrosine phosphatase inhibitor bpV(phen) (10 µM). Pretreatment was sufficient to prevent internalization of NMDA subunits, but had no effect on trafficking of Beta 2 subunits of the GABA_A receptor (NR1, −2.7±6.7% and 29.5±5.0% change internal 10 Gy with drug and 10 Gy respectively, NR2A, −9.1±3.3% and 22.4±2.6% Beta2, −27.1±2.3% and −24.3±2.2%, *p≤0.05, **p = 0.006, n = 4). (B) Pre-incubation in the non-competitive NMDAR antagonist memantine (50µM) prevented both NMDAR internalization and GABA surface retention (NR1, −2.1±2.0% and 25.7±5.0% change internal 10 Gy with drug and 10 Gy respectively, NR2A, −5.4±6.4% and 35.6±10.5%, Beta2, 4.7±8.4% and −27.5±5.3%, *p≤0.05 two-tailed Student’s t-tests, n = 4).

Figure 6. Radiation induces early alterations in synaptic function.

(A) Whole cell recordings were obtained from dentate slices subjected to 10 Gy or sham. To allow for comparison between control (Con) and radiated slices (Rad), NMDA responses were normalized to AMPA responses. Initial recordings indicated radiation did not significantly change the amplitude of NMDA currents (p = 0.349, Student’s t-test; sham n = 7; rad n = 8). (B) However, NR1/NR2A responses obtained in the presence of the NR2B antagonist Ro25-6981, were significantly diminished in amplitude in radiated slices (*p<0.001, con n = 7; rad n = 8). (C) While average channel activation time (τ_act, the time, measured in msec to activate 200 pA current) did not differ between in control and radiated slices [F(3,32) = 0.361, p>0.782, one-way ANOVA], the average channel deactivation time (τ_deact, msec to inactivate 20 pA current) was significantly increased in NMDAR currents measured from radiated slices [F(3,32) = 13.751, **p<0.001, one-way ANOVA]. Ro25-6981 (RO) did not significantly alter τ_act or τ_deact of NMDA EPSCs in either control or irradiated slices. (D) GABA_A responses were normalized to AMPA responses. 10 Gy significantly increased the GABA_A/AMPA ratio (**p = 0.001, con n = 6; rad n = 4). (E) In order to evaluate for potential changes in synaptic plasticity, baseline field EPSP recordings were obtained in control or radiated slices and LTP was induced by tetanic stimulation (HFS at time  = 0). In comparison to control (n = 4), LTP was significantly attenuated in radiated slices (n = 5, p = 0.02, 30 minutes post HFS).