Intracellular neurophysiological analysis reveals alterations in excitation in striatal neurons in aged rats

Intracellular neurophysiological analysis reveals alterations in excitation in striatal neurons in aged rats

Brain Research, 494 (1989) 215-226 Elsevier 215 BRES 14700 Intracellular neurophysiological analysis reveals alterations in excitation in striatal ...

930KB Sizes 1 Downloads 73 Views

Brain Research, 494 (1989) 215-226 Elsevier

215

BRES 14700

Intracellular neurophysiological analysis reveals alterations in excitation in striatal neurons in aged rats C. Cepeda, J.P. Walsh, C.D. Hull, N.A. Buchwald and M.S. Levine Mental Retardation Research Center, University of California at Los Angeles, Los Angeles, CA 90024 (U.S.A.) (Accepted 17 January 1989) Key words: Aging; Striatum; Reduced excitation; Intracellular recording; Elevated excitatory threshold

Intracellular recordings were used to characterize the physiological changes underlying decreases in excitation observed in striatal neurons during the aging process. Rats were divided into 3 age groups: young (3-5 months), middle-aged (10-12 months) and aged (>24 months). All experiments were performed in urethane-anesthetized rats. Recordings were obtained from 33 neurons in young, 17 in middle-aged and 20 in aged rats. When identified by intracellular injections of Lucifer yellow all recorded neurons were medium-sized spiny cells. Resting membrane potentials were at least -40 mV and action potentials greater than 35 mV. Postsynaptic responses were evoked by stimulation of frontal cortex. In all recorded neurons, regardless of age, excitatory postsynaptic potentials (EPSPs) could be evoked. However, the threshold currents for eliciting both EPSPs and synaptically driven action potentials were significantly higher in neurons obtained from aged rats than those recorded in the other two groups. Other changes in excitation in aged striatal neurons consisted of absence of spontaneously occurring EPSPs, higher current to induce firing by intraceUular injections of depolarizing current and an inability of orthodromically induced action potentials to follow paired stimulation pulses to the cortex at short interpulse intervals. These data were interpreted to indicate that a combination of changes in synaptic connectivity and in membrane properties underlie the decreases in excitation. Together with our previous findings obtained from aged cats these results indicate that decreased neuronal excitability is a major effect of aging in the striatum.

INTRODUCTION

experiments that used extracellular recording techniques 23'24. Consequently, the decrease in the

O u r l a b o r a t o r y has investigated age-related alterations in the caudate nucleus of cats 18-2°'22-24.

p r o p o r t i o n of initially excitatory responses (or increase in the p r o p o r t i o n of initially inhibitory re-

Electrophysiologically, caudate neurons exhibited decreases in indices of excitation in animals over 10 years of age 23'24. These indices included lower s p o n t a n e o u s firing rates and fewer initial excitatory responses to activation of monosynaptic inputs, F u r t h e r , in aged animals thresholds for evoking excitatory responses were higher than thresholds for evoking inhibitory responses. This difference did not occur in 1- to 3-year-old cats. Concomitant with these i m p a i r m e n t s , morphological changes also occurred 19'2°. These consisted of decreases in synaptic density, spine density and loss of distal dendritic segments, The neurophysiological findings were based on

sponses) when m o n o s y n a p t i c inputs were activated may be explained in m o r e than one way. First, in aged cats cells may have displayed excitatory postsynaptic potentials (EPSPs) that were subthreshold for action potential generation. Second, initially inhibitory postsynaptic potentials ( I P S P s ) m a y have occurred in aged cats. Third, the biophysical characteristics (e.g. input resistance, m e m b r a n e potential) inherent to caudate neurons m a y have b e e n altered by the aging process. These alternatives are not necessarily mutually exclusive and m a y be acting in concert to decrease the excitability of caudate neurons. In o r d e r to delimit the relative contribution of these possibilities the present investigation was

Correspondence: M.S. Levine, Mental Retardation Research Center, University of California at Los Angeles, 760 Westwood Plaza, Los Angeles, CA 90024, U.S.A. 0006-8993/89/$03.50 © 1989 Elsevier Science Publishers B.V. (Biomedical Division)

216 undertaken. Intracellular recordings were performed in rats to determine how the characteristics of EPSPs and IPSPs were altered by the aging process. Rats were chosen instead of cats for both scientific and practical reasons. Using the rat as an alternative model would allow our results to be generalized to another species. Many of the characteristics of striatal electrophysiology are similar in carnivores and rodents 4~13'34. Of importance for the present study is that activation of monosynaptic inputs to striatal neurons produces initially excitatory responses in young adult animals of both species 4'1~'34. These potentials are typically followed by long duration inhibitory responses to produce the characteristic EPSP-IPSP sequence. From a practical point of view the rat was also the species of choice for these experiments. The yield of successful impalements in individual intracellular recording experiments in vivo tends to be low. Thus, many aged animals would have to be used in order to assess a large population of neurons. Since we do not have an unlimited supply of aged cats, using such animals for intracellular experiments would not be cost effective. MATERIALS AND METHODS Three age groups of male rats were used: young, 3-5 months (n = 16); middle-aged, 10-12 months (n = 11); and aged, >24 months (n = 7). The young group consisted of both Sprague-Dawley and Fisher 344 rats. Sprague-Dawley's were obtained from Bantin and Kingman in Fremont, CA and Fisher 344 animals were obtained from the breeding colony at the National Institute of Aging. Since there were no differences in the results obtained from animals of either strain, data were pooled. The middle-aged rats were Sprague-Dawleys while the aged group consisted of only Fisher 344 rats. Before surgery all animals were anesthetized with intraperitonealinjectionsofurethane(1.1-1.3g/kg), Each animal was given two supplementary injections (0.2 g/kg), 3 and 6 h after the initial injection, Electrocardiogram and body temperature were continuously monitored. Using these procedures, depth of anesthesia appeared to be consistent across age groups. Animals were placed in a standard stereotaxic headholder. After craniotomy, cisternal drain-

age was performed to reduce brain pulsation. Stimulating electrodes consisted of one pair of stainless-steel wires (0.5 mm bare tip) implanted in the frontal cortex 1.0-1.5 mm below the surface of the brain (intertip distance was 1.0 mm). One tip was placed in the medial aspect of the lateral agranular cortex (close to the medial agranular field) and the other in the lateral agranular cortex ~. Resistance of stimulating electrodes was 9.92 kg2. This value did not vary among the 3 age groups. Stimuli consisted of single rectangular pulses (0.1 ms duration, intensities ranging from 100 ,uA to 4 mA). Intracellular recordings were obtained from potassium acetatefilled micropipettes with impedances between 30 and 60 Mg2. Intraceilular signals were amplified with a WPI preamplifier (WPI M-4A) using an active bridge circuit. Potentials were monitored on an oscilloscope and recorded on FM tape for subsequent computer analysis. After impaling a neuron its response to stimulation of cortex was assessed. For each cell threshold currents for evoking both postsynaptic potentials and action potentials were determined by stimulating the cortex with a series of ascending intensities. For quantification, the amplitude and duration of the postsynaptic responses (EPSPs and IPSPs) were measured from averaged records (3-5 trials) at twice threshold for each neuron. In addition, the resting membrane potential (RMP), the amplitude of the action potential (evoked on the EPSP), the latency of the first spike induced by cortical stimulation, and the presence or absence of post-inhibitory rebound were also determined. In several cells input resistance was determined by plotting the amplitude of the potential obtained from intracellular injections of 50 ms hyperpolarizing pulses at different intensities. In several experiments recorded neurons were identified by intracellular injections of Lucifer yellow 35. Recording electrode tips were filled with a 5% solution of Lucifer yellow dissolved in distilled water. The remainder of the electrode shank was filled with a0.1 M solution of lithium chloride. After recordings were completed from each neuron, Lucifer yellow was injected (2-5 nA hyperpolarizing current for 2-5 min). Neuronal evoked response measurements were pooled for all neurons within each age group. Data

217

TABLE I Quantified electrophysiologicalparameters

Values are means + S.E.M. Percentages between parentheses indicate percentage change from 3- to 5-month-old group. Age

Resting membrane potential (mV) Action potential amplitude (mV) Threshold intensity to evoke EPSPs (mA) Threshold intensity to evoke action potentials (mA) Duration of evoked EPSPs (ms) AmplitudeofevokedEPSPs(mV) Threshold amplitude of EPSP for action potential initiation (mV) Threshold membrane potential for action potential initiation (mV)

3-5 Months

10-12 Months

24 Months

-52.9 + 2.1 39.8 + 2.2 0.75 + 0.09 1.30 + 0.09 40.3 + 1.9 7.4+0.7

-47.6 + 1.9 38.3 + 1.6 0.83 + 0.16 (11%) 1.45 + 0.17 (11%) 39.5 + 2.3 (-2%) 6.6+0.9 (-11%)

-57.5 + 3.1 38.6 + 7.2 1.56 + 0.13 (108%) 2.30 + 0.35 (77%) 36.7 + 2.4 (-9%) 5.6+ 1.0 (-24%)

7.8 + 0.9

7.7 + 1.1

(-1%)

12.5 _+2.3

(60%)

-42.2 + 3.0

-42.3 + 2.5

(2%)

-46.9 + 2.9

(14%)

were c o m p a r e d among age groups by a p p r o p r i a t e analyses of variance and N e w m a n - K e u l s multiple comparisons for average values and X2-tests for distributional measures. Pearson p r o d u c t - m o m e n t coefficients were used for correlation analysis, A t the end of the recording experiment all animals were sacrificed with an overdose of p e n t o b a r b i t a l and the brain perfused with 10% formaldehyde, Serial coronal sections (80 ktm) were obtained to verify the location of recording and stimulating electrodes. The rodent does not have a distinct caudate nucleus and the caudate and putamen form a single structure interdigitated by fibers of the internal capsule. In the present study we will refer to all r e c o r d e d cells as striatal neurons.

The impalements lasted from 5 to 40 min in all age groups. T h e r e were no particular difficulties in recording from neurons in the aged rats c o m p a r e d to the o t h e r groups. The average R M P s and spike amplitudes were not significantly different in the three age groups (Table I). H o w e v e r , examination of distributions (Fig. 3) revealed that p r o p o r t i o n a t e l y m o r e neurons were h y p e r p o l a r i z e d (-60 to - 7 0 m V RMPs) in the aged animals. Regardless of age, recorded cells displayed few s p o n t a n e o u s action potentials or were silent. In all age groups rapid

3 Months

RESULTS The most important finding of the present experiment was that intracellular recordings from striatal neurons in aged rats revealed that the threshold currents for evoking both EPSPs and synaptically driven action potentials were considerably elevated compared to values obtained from young and middle-aged animals under similar conditions, Intracellular recordings were obtained from 70 neurons localized in the dorsal striatum (33 in young, 17 in m i d d l e - a g e d and 20 in aged rats). Histological reconstructions indicated that neurons were located in the same region of the striatum in the three age groups. Localization of stimulation sites was also similar in the three groups,

\

~,

Shin CX

f k ..~j ~._.~ / R~P- 5 5 m y ~_~._~_~f~ ]~omv A s,,~cx loo.... Fig. 1. Intracellular response of a striatal neuron to frontal cortex stimulation in a 3-month-old rat. Top: a single response. Bottom: average of 3 responses. This response consists of an initial EPSP followed by a long lasting low amplitude IPSP. An excitatory rebound usually follows the IPSP.

218 TABLE II

and aged groups p r o p o r t i o n a t e l y fewer neurons

Percentage occurrence of different types of responses

displayed r e b o u n d excitations (Table I1). Initial IPSP responses were never observed. Fig. 2 compares responses of striatai neurons from

Age

EPSP EPSP-IPSP EPSP-IPSPrebound

3-5

10-12

24

Months

Months

Months

a young and an aged rat to stimulation of the cortex at a series of ascending current intensities. Both cells

27 9 64

18 29 53

25 20 55

had RMPs o f - 6 0 inV. H i g h e r stimulus intensities were required to evoke excitatory responses in the neuron obtained from the aged rat. In this neuron,

firing rates were an indication of cell damage. Fig. 1 shows the typical response e v o k e d in striatal neurons when cortical inputs were activated. This response consisted of an initial EPSP with an e v o k e d action potential followed by a longer duration but smaller amplitude IPSP. In many cases a r e b o u n d excitation followed the IPSP. The basic components of this response were consistent across the three age groups (Table II). H o w e v e r , in both middle-aged

an orthodromically activated spike could only be evoked at an extremely high intensity (4 m A ) . Quantitative analyses were p e r f o r m e d on data obtained from 55 neurons, 25 from young, 13 from middle-aged and 17 from aged rats. In both middleaged and aged rats the threshold current necessary to evoke the EPSP was higher than in young rats ( F = 14.6, df = 2/52, P < 0.01) (Table I). N e w m a n - K e u l s analysis indicated that only the differences in threshold current between aged and young and aged and middle-aged groups were statistically significant. The

5 Months ~

~

24 Months 05mA

A

1

A

Stim CX A

I0 mV 20 msec

A Shin CX

Fig. 2. Comparison of threshold currents necessary to evoke EPSPs in striatal neurons in a young (3 month) and an aged animal (24 months). Both cells had resting membrane potentials of-60 inV. Higher stimulus intensities were required to evoke responses in the neuron obtained from the aged rat. In this neuron, a spike could only be evoked at an intensity of 4 mA.

219 70

80

60 •J ~< £

50

4o 30

i

FTYOUNG I MIDDLE-AGED ~AGED

20 lO o

70

l nl°H,m -50

so

<

30

lO

n -70

-60

60

~

~YOUNG I MIDDLE-AGED ~AGED

40_IFt

~

-40

~ ~< o

FJI

o

-80

0-4

4-8

MEMBRANE POTENTb~L(mY)

8-12

EPSP AMPLITUDE(mY)

Fig. 3. Distributions of resting membrane potential values. Although there were no significant differences in average values for this measure, in the aged rats more neurons tended to have slightly higher resting membrane potentials.

60 ,~ S ~-

significantly elevated in the aged animals compared to the young and middle-aged groups (F = 11.2, df

Z

~

50

r'-7yOUNG

I MIDDLE-AGED ~AGED

40

20 10

60

|

o r7 J

O

50

40

30 oz

20

r--1Y O U N G I MIDDLE-AGED

O- 10

Ill

~AGED

]I [71

10

~

0


2-4

4-.6

6-8

10-20

20-30

30-40

40-50

)50

Fig. 5. Distributions of EPSP amplitude and duration. Proportionately more neurons in the aged group had smaller EPSPamplitudes(<4mV).Proportionatelymoreneuronsin the aged group had shorter duration EPSPs (<20 ms) as well. I~

[-]

8-1

>1

= 2/42, P < 0.01). Again the difference between young and middle-aged rats was not statistically significant. Not all neurons in aged rats had high threshold currents for e v o k i n g E P S P s and action potentials. In both m i d d l e - a g e d and aged groups a p o p u l a t i o n of

60 50

r-qYOUNG

I MIDDLE-AGED ~AGED

40 ]il

330

~ll

20

10 0

~ EPSP DURATION(rnsec)

EPSP THRESHOLD (rnA)

~-

n. >16

12-16

<.5

~ .5-1

~ -1.~ 1.5-2 2-2.5 >2.~

SPIKE THRESHOLD (mA)

Fig. 4. Distributions of EPSP and spike thresholds. Approximately 45% of aged neurons had elevated thresholds (>1 mA) for EPSP generation. More than 40% of the neurons in the aged group had elevated thresholds to evoke action potentials (>2.0 mA).

neurons had threshold values in the same range as those of young rats. Proportionately, the population of neurons with high threshold currents increased with age. Distributions of the proportion of neurons in each threshold category were constructed (Fig. 4). In aged rats more than 45% of the neurons had elevated thresholds (>1.0 mA). In both middle-aged and young animals thresholds for evoking EPSPs were never greater than 1.0 mA. With increasing age proportionately fewer neurons had thresholds below 0.5 m A . T h e difference b e t w e e n the d i s t r i b u t i o n o b t a i n e d from aged rats was statistically significant

from that obtained from young animals (X2= 18.5, df = 5, P < 0.0028). Similarly, in aged rats more

220

•• RMP-SSmV ~ r

10 Months

A

Slim {X

24 Monlhs

RMP - 60 mV

RMP60mV ~"~-~%'~ A

}lOmV 20 msec

Slim CX

Fig. 6. Ability of striatal neurons to follow high-frequency stimulation. This capacity was tested with paired pulses with different time intervals between each pulse (from 20 to 200 ms, in 10 ms steps). All neurons in young animals could follow twin pulses at intervals greater than 20 ms. Only one (out of 4) neuron in aged rats could follow this frequency of stimulation. Top: neuron in a young rat responding at a 20 ms interpulse interval. Bottom: neuron in an aged rat showing a spike failure aboveat 50 mStheinterpulsethresholdforinterval" Cortical stimulation was 1.5× spike generation in both neurons.

24 Months

.~.¢~,,~,~ f v RIWP -60mY

than 40% of the neurons had elevated threshold currents ( > 2 . 0 m A ) for o r t h o d r o m i c elicitation of action potentials (Fig. 4). As is a p p a r e n t from the distributions, with increasing age a population of neurons with thresholds b e y o n d 2.0 m A emerges. Differences among the distributions obtained from aged rats were significantly different from those

\ ~

_

] ]~omv 20 msec

h,~

o b t a i n e d from young and middle-aged animals (X2 = 14.7, df = 5, P < 0.01 for comparison between young and aged; X2 = 9.2, df = 3, P < 0.02 for

Fig. 7. The ability of striatal neurons to generate action potentials was directly tested by intracellular injections of positive current. Top: in a 10-month-old rat the neuron generated two action potentials with 0.6 nA for 50 ms. Bottom: in a 24-month-old rat 3.0 nA induced a single action potential at the onset of current injection. Spike at offset is current transient. RMPs were the same for both neurons.

comparison between middle-aged and aged). A l t h o u g h there was a decrease in average amplitude of the e v o k e d EPSP in the aged group the differences among these values when assessed at twice threshold current were not statistically significant (Table I). EPSP amplitudes were measured in all age groups in the same range of m e m b r a n e potentials to account for voltage d e p e n d e n t changes in E P S P amplitude. W h e n distributions of EPSP amplitude were constructed (Fig. 5) it was a p p a r e n t that p r o p o r t i o n a t e l y m o r e neurons in aged rats had lower amplitude EPSPs. F o r statistical analysis dis-

tributions of E P S P a m p l i t u d e were divided into two categories using 4 mV as a cutoff score. The difference b e t w e e n distributions o b t a i n e d from young and aged rats was statistically significant (X2 = 4.35, df = 1, P < 0.05). The E P S P amplitude necessary to reach threshold for action potential initiation was significantly h i g h e r in aged rats comp a r e d to animals in the o t h e r two groups ( F = 3.73, df = 2/25, P < 0.05) (Table I). W h e n this value was a d d e d to the R M P for each neuron, the resulting

221 TABLE III

5 Months

lntercorrelations of measures

RMP, resting membrane potential; ETH, EPSP threshold; SPKT, spike threshold; EAMP, EPSP amplitude; EDUR, EPSP duration; SPKL, spike latency. - 6 0 mV

24 Months

~

RMP

ETH

SPKT

EAMP

0.42 -0.15 -0.54* -0.21 -0.40

0.52* -0.11 0.14 -0.24

-0.12 -0.11 -0.12

0.70* 0.17

10-12 Months ETH SPKT EAMP EDUR SPKL

-0.04 -0.21 -0.27 -0.37 -0.37

0.57* 0.14 0.07 0.26

0.40 0.51" 0.74*

24 Months ETH SPKT EAMP EDUR SPKL

0.36 -0.20 -0.11 -0.23 0.07

0.76* -0.76* -0.52* 0.67*

-0.75* -0.41 0.78*

EDUR

3-5 Months

l IOmV

ETH SPKT EAMP EDUR SPKL

0.21

- 7 0 mV 2 sec

Fig. 8. Spontaneous EPSPs in striatal neurons. Most neurons in young rats with resting membrane potentials of at least -60 mV displayed spontaneous depolarizations (5-10 mV in amplitude) (top). In neurons in aged rats these potentials were seldom observed (bottom).

m e m b r a n e potential was the threshold for spike initiation (Table I). The differences among these values were not statistically significant because neurons in aged animals tended to have more hyperpolarized RMPs. The average duration of the EPSP decreased slightly with age. This change was not statistically significant when assessed at stimulation intensities of twice threshold, however. When distributions were constructed (Fig. 5) it was apparent that there was an increase in the proportion of neurons with short duration EPSPs. The differences among the 3 distributions were not statistically significant. The latency of the first spike evoked by cortical stimulation was slightly increased in the aged animals. This increase was also not statistically significant, No consistent differences were found in the amplitude or duration of the IPSP among the 3 groups, The ability of a neuron to generate a series of action potentials when it is challenged with high frequency stimulation is one index of its excitability, We tested this capacity by applying twin pulses to the cortex at intensities 1.5× threshold current for spike elicitation (Fig. 6). The delays between the two stimulus pairs varied from 20 to 200 ms (in 10 ms steps). All the cells tested in young rats (n --- 4) could follow paired pulses at interpulse intervals greater than 20 ms with little decrement in the amplitude of the second spike. In contrast, in 80% (4/5) of the

0.89* -0.02 0.27

0.46 -0.95*

0.18

* Correlation coefficient significantly different from 0, P < 0.05.

neurons tested in aged rats a second action potential could not be generated at interpulse intervals less than 80 ms. The amplitude of the second action potential was also markedly decreased in aged animals. The second EPSP, however, did not show a parallel decrement implying further that the capacity for action potential generation is altered in aged striatal neurons. A n o t h e r test of excitability is to determine the effect of depolarizing intracellular current injections (Fig. 7). Depolarizing pulses (50 ms duration, 0.5-1.2 n A intensity) were sufficient to evoke single or multiple spikes in young animals. In contrast, in aged neurons the current necessary to evoke even a single action potential was consistently higher (range 1-3 nA). Furthermore, in the aged animals single stimulation pulses rarely evoked more than a single action potential while in young rats repetitive firing often occurred. When tested with sustained depolarizing current, cells in young and middle-aged animals were able to generate a series of action

222 potentials when the membrane potential was held at -30 to -35 mV. In neurons obtained from aged animals it was extremely difficult to depolarize them more than a few millivolts. Even when neurons in aged rats could be depolarized (up to -30 mV) they often did not generate multiple action potentials, In several instances we obtained an approximate calculation of input resistance. In young and middleaged animals input resistance of striatal neurons varied between 10.0-18.5 M(2. These values are similar to those reported previously 14'35. The measurements in the aged group were within these limits (11.5-14.9 Mg2). In some cases (n = 15) in the young and middle-aged groups, especially with long lasting impalements of cells with at least -60 mV RMP spontaneous EPSPs occurred (5-10 mV in amplirude) (Fig. 8). These spontaneous depolarizations were never accompanied by action potentials. In the aged group, these spontaneous potentials were not observed, Measures of resting membrane potential, EPSP threshold current, spike threshold current, EPSP amplitude, EPSP duration and spike latency were intercorrelated to determine if there were differential relationships among these measures in each age group. Pearson product-moment correlations were calculated separately for each age group (Table III). In all groups there were a number of significant correlations. However, in the 24-month-old rats there were twice as many significant correlation coefficients than in the other two age groups. This finding probably is a function of the increased variation and more heterogeneous population of neurons encountered in this group. In addition, the statistically significant correlation coefficients were generally higher. For example in 3- to 5- and 10- to 12-month-old rats, EPSP and spike threshold were positively correlated (r = 0.52 and r = 0.57 for each group, respectively). In the 24-month-old rats this correlation was substantially higher (r = 0.76). Resting membrane potential correlated negatively with EPSP amplitude in young rats only, indicating that neurons with high resting membrane potentials (negative values) had high amplitude EPSPs. In young and middle-aged rats, EPSP threshold current correlated positively only with spike threshold current indicating that neurons with high EPSP thresh-

olds also had high thresholds for producing spikes. In contrast, in the aged rats this variable also correlated significantly and negatively with EPSP amplitude and EPSP duration indicating that neurons that had higher threshold currents also had lower amplitude and shorter duration responses. EPSP threshold current in this group was positively correlated with spike latency indicating that neurons with higher thresholds had longer latencies to produce spikes. EPSP amplitude and EPSP duration were positively correlated in young and middle-aged but not aged rats indicating that in the former groups larger EPSPs also were of longer duration. Six cells in the young and middle-aged groups were successfully filled with Lucifer yellow. All recovered cells were medium-sized spiny neurons. One neuron was filled in the aged group. This neuron was also a medium-sized spiny cell. It had fewer spines on distal dendritic segments and the diameter of the dendrites appeared to be decreased compared to diameters obtained from neurons in young rats. Finally, accumulations of autofluorescent lipofuscin surrounded the soma. DISCUSSION The present study demonstrates that activation of cortical inputs produces initial EPSPs in striatal neurons in aged rats. The threshold currents for evoking the EPSPs and orthodromic action potentials in these neurons is markedly elevated compared to values obtained in young rats. A number of other age-related changes were observed in this experiment. All of these involved decreased excitation and included decreased ability to follow paired pulses at short interpulse intervals, decreased ability of depolarizing current injections to evoke action potentials, decreased proportions of rebound excitations and decreased spontaneously occurring depolarizations. The present findings in conjunction with our previous studies 23"24 provide evidence that striatal neurons in rodents and carnivores display similar electrophysiological changes during aging. In previous studies in which extracellular recordings were obtained from aged cats, we demonstrated decreases in excitation in striatal neurons 23'24. These were expressed mainly by decreases in the proportion of initially excitatory responses evoked by cortical and

223 substantia nigra stimulation 23. In the second study we demonstrated that there was a decrease in the proportion of responsive striatal neurons to somatosensory stimulation 24. When neurons in aged cats were responsive, receptive fields were larger indicating less precise ability of such cells to encode somatosensory information. These two extracellular studies, however, could not elucidate whether changes in excitation were due to the inability of the cells to generate EPSPs, increased occurrence of initial IPSPs or inability of the cell to generate action potentials. The present findings using intracellular recordings in aged rats demonstrated that initial EPSPs occurred. In every neuron recorded, the initial response was an evoked EPSP. In no case was an initial IPSP observed. The extracellular recordings in our previous work could not detect EPSPs subthreshold for action potential generation. The present intracellular study demonstrates that increased proportions of initially inhibitory responses in extracellular recordings are not related to the occurrence of initial IPSPs but rather to an increase in the threshold for EPSP and action potential generation in aged animals, Cells with greater RMPs were found more frequently in the aged group. The increase in RMP could reflect an intrinsic property of aging of striatal neurons, a loss of tonic excitatory influences or some combination of the two. The decrease in spontaneous EPSPs in the aged rats is evidence for loss of excitatory inputs, probably from cortex or thalamus, The average evoked EPSP amplitude was also slightly decreased in this group. Since, in general, at increased RMP the amplitude of the EPSP tends to increase 36, the possibility that the difference in EPSP amplitude is only a voltage dependent effect can be ruled out. The slight increase in the proportion of cells that were hyperpolarized in the aged population may partially account for the observed increase in the threshold current to evoke action potentials, Decreases in the occurrence of spontaneous EPSPs in aged striatal neurons probably underlie the decreases in spontaneous action potential generation we reported previously 23"24. Decrements in spontaneous activity have been observed, using extracellular recordings, in cerebellum 29.3°, locus coeruleus 2~, hippocampus 1"2 and striatum 33, but not in the frontal cortex of rats 33. The result that in all the

ceils EPSPs could still be elicited suggests that inputs remained functional. At suprathreshoid stimulation intensities, differences in EPSP amplitude also occurred. These age-related alterations in EPSP threshold and amplitude could be related to changes in the corticostriatal pathway. These changes appear separate from alterations in striatal neuron action potential threshold since the value of the membrane potential of striatal neurons for initiating action potentials did not change markedly. The more hyperpolarized RMPs may thus reflect a loss of tonic afferent excitatory input. Thresholds for spontaneous or induced activation of cortical neurons may have increased as evidenced by the larger stimulus intensity required to evoke excitatory synaptic input. The present experiments were not directly designed to determine the relative contributions of agerelated changes in cortical neurons versus the changes in striatal neurons. Subsequent studies will be performed to address these points. In our previous studies we demonstrated morphological substrates for the electrophysiological alterations in aged cats 19"2°. Many of these changes are apparent in the ubiquitous medium-sized spiny neuron in the striatum. When recorded striatal neurons are identified by intracellular injection of horseradish peroxidase or Lucifer yellow, virtually all cells are medium-sized spiny neurons 15,21,35. It is thus likely, that most recorded striatal cells are this type of neuron. When cells were identified in the present experiment, they were all medium-sized spiny neurons. It is probable that the morphological changes we have observed in this type of cell using other techniques are valid indicators of morphological events occurring in the present experiments. In both middle-aged and aged cats there is a decrease in the density of spines on distal dendritic segments 1s.19. Furthermore, in cats over 11 years of age there is a loss of dendritic segments 19. Ultrastructurally, there is a decrease in synaptic density and an increase in synaptic apposition or contact length in both middleaged and aged cats 2°. A majority of the inputs synapse on the spines of these spiny neurons 32. Consequently, these morphological changes suggest that there should be decreases in the number of synaptic contacts underlying the generation of EPSPs. However, it should be pointed out that in our previous uitrastructural study 2° no attempt was

224 made to determine which types of synapses were affected by the aging process. It is possible that the density of synapses of intrinsic origin was also altered by the aging process as well as the density of synapses of extrinsic origin. The implication of the present study is that the alterations in the striatum contribute significantly to changes in excitability during aging. As indicated above it is possible however, that corticostriatal or other input neurons are also affected. The present findings cannot determine whether a change in one structure is causally related to changes in the other. There is considerable morphological evidence that at least cortical neurons display degenerative alterations during the aging process 31. There is little information on electrophysiological changes in these cells during aging, however 3~. It is apparent that the structural changes occurring during aging appear to be much more dramatic than the functional changes. This may indicate that compensatory mechanisms counteract morphological alterations 5,v'9"m. In our morphological studies at least two examples of compensatory mechanisms have been found. Electron microscopic analysis of synaptic changes revealed that a significantly higher number of enlarged synapses occur in aged striatal cells of cats 2°. Such synapses may provide a mechanism to offset the synaptic loss that also occurs. In middle-aged cats, dendritic elongation and increases in the number of dendritic branches on striatal spiny neurons occur TM. Again this mechanism may be compensating for synaptic loss that is also present in cats of this age. We have stressed the role of synaptic connectivity alterations as underlying the observed age-related decreases in excitation. However, intrinsic cellular mechanisms may also be affected. Higher currents were necessary to elicit action potentials with the injection of depolarizing pulses into the cell indicating that the capacity for spike generation may be altered. In contrast, membrane potential at threshold for initiating action potentials did not change. It was also very difficult to depolarize cells in the aged rats with positive current. In some cases it was virtually impossible to elicit action potentials in this way in aged animals. Another index of decreased excitation was that the majority of aged cells was not able to follow high frequency cortical stimulation,

Failures were observed at time intervals which were 2 or 3 times longer than in young neurons. This observation indicates that accommodation to repeated stimulation is altered in aged animals. The ability of an aged neuron to respond to iterative information is decreased. Previously, we observed that the ability of caudate neurons to follow high frequencies of stimulation is also compromised in old cats 23. Similar observations have been made with regard to long-term potentiation in the hippocampus of aged rats ~v. Taken together these findings indicate impairments in the mechanisms that generate the action potentials. It has been shown that outward rectification in striatal cells occurs in the depolarizing direction 14. It is possible that with age there is an accentuation of this rectifying mechanism making neurons less likely to depolarize. A recent theory on the cellular mechanisms of aging 1~, postulates that in old animals there is a perturbation in calcium buffering mechanisms. Landfield et al. j7 found an increase in the duration of calcium spikes in hippocampal slices of aged animals. While an increase in outward rectification would tend to oppose actively generated calcium potentials, an imbalance in the calcium buffering system could cause an increase in calcium-activated potassium currents, which in turn would prevent cell depolarization, as has occurred in the present study. There is still much debate about the pre- and postsynaptic effects of calcium irabalance ~ and this hypothesis awaits more fundamental data. The input resistance measured with hyperpolarizing current pulses was not significantly altered in the aged group. Studying motoneurons in aged cats Chase and Morales found an increase in input resistance 3'6"27. These findings indicate that changes in input resistance may not be generalized to all areas of the brain during aging. More intracellular studies are needed to characterize functional changes in passive and active electrical properties during aging. There are several variables that may have affected the outcome of the present study. First, aged rats may have been differentially sensitive to the anesthetic. Several factors argue against a simple interaction between age and anesthetic. Similar studies using different anesthetics and different species have come to the same conclusion as the present

225 experiment 16'23'24'33.We have also observed some of the same types of electrophysiological changes in unanesthetized animals further arguing that differential sensitivity to the anesthetic is not a crucial variable 24. Finally, the result that not all neurons in a nucleus are equally affected by the aging process (see below) argues against differential sensitivity to the anesthetic. Another variable concerns the two different strains of rat used in the present study, Although there were no strain differences in electrophysiology that could be detected between the young groups, there may have been strain differences that occurred between middle-aged and aged rats of the two strains. Again, this seems unlikely because many of the effects we observed can be detected across species, anesthetic and experimental conditions and appear to be quite robust, It is now well established that at the neuronal level not all neurons are equally affected by the aging process 5'9"1°'19"25"26. Our findings on the striatum in the aging rat and cat serve to underscore this idea. A continuum appears to exist in which certain neurons are more affected by age than are other neurons. The intercorrelations calculated in the present experiment are supportiveofthisidea. More than twice as many correlations were statistically significant in the aged neurons than in neurons obtained from the other two groups indicating that age changes the electrophysiological attributes of populations of neurons. These findings suggest that the neurons that are most affected by the aging process form one population, the neurons least affected by the process probably form another population with a set of distinctly different values. In between, is a population of neurons that is undergoing a change. Data obtained from the middle-aged rats are also supportive of this notion. On most of the measures, results from this group were similar to

REFERENCES 1 Barnes, C.A., Memory deficits associated with senescence: a neurophysioiogy and behavioral study in the rats, J. Cornp. Physiol. Psychol., 93 (1979) 74-104. 2 Barnes, C.A. and McNaughton, B.L., Neurophysiological comparison of dendritic cable properties in adolescent middle-aged and senescent rats, Exp. Aging Res., 5 (1979) 195-206. 3 Boxer, P.A., Morales, ER. and Chase, M.H., Alteration of group 1A-motoneuron monosynaptic EPSPs in aged

results from young animals. However, a population of neurons with altered electrophysiological responses was emerging. Similar findings and conclusions have been reached for hippocampal neurons 5' 9,10, spinal cord motoneurons 3'6"27 and neocortical neurons 31. We recently proposed a model to account for the neurophysiological changes observed in aged striatal neurons TM. In young adult animals intracellular recording experiments indicate that most striatal neurons respond to activation of their monosynaptic inputs with an initial EPSP which is usually followed by an I P S P 4'12'13'15. This IPSP is thought to be generated, at least in part, by mutual inhibition from neighboring striatal neurons ~3'21. In aged animals we hypothesized that there were at least two populations of neurons. One population still receives most of its excitatory inputs while the other receives reduced excitatory input. When recorded from, neurons in this latter population should display higher thresholds for EPSP and action potential generation and reduced amplitude EPSPs. The results of the present study provide evidence that some of these changes occur. We also hypothesized that IPSPs should be reduced in duration and amplitude in aged striatal neurons because some of the surrounding neurons are not activated and cannot inhibit their neighbors. The present data do not appear to support this hypothesis. There was little consistent change in amplitude or duration of IPSPs. Age-related alterations in inhibitory events in the striatum will require further study to uncover their underlying mechanisms. ACKNOWLEDGEMENTS Supported by USPHS A G 7462, A G 1558 and HD 5958.

cats, Exp. Neurol., 100 (1988) 583-595. 4 Buchwald, N.A., Price, D.D., Vernon, L. and Hull, C.D., Caudate intracellular responses to thalamic and cortical inputs, Exp. Neurol., 38 (1973) 311-323. 5 Buell, S.J. and Coleman, P.D., Quantitative evidence for selective dendritic growth in normal human aging but not in senile dementia, Brain Research, 24 (1981) 23-41. 6 Chase, M.H., Morales, ER., Boxer, P.A. and Fung, S.J., Aging of motoneurons and synaptic processes in the cat, Exp. Neurol., 90 (1985) 471-478. 7 Coleman, P.D. and Flood, D.G., Dendritic proliferation in

226

8

9

10

11 12

13

14

15

16

17 18

19

20

21

22

the aging brain as a compensatory repair mechanism, Prog. Brain Res., 70 (1986) 267-277. Donoghue, J.P. and Wise, S.E, The motor cortex of the rat: cytoarchitecture and microstimulation mapping, J. Comp. Neurol., 212 (19821 76-88. Flood, D.G., Buell, S.J., Defiore, C.H., Horowitz, G.J. and Coleman, P.D., Age-related dendritic growth in dentate gyrus of human brain is followed by regression in the 'oldest old', Brain Research, 345 (1985)366-368. Flood, D.G., Guarnaccia, M. and Coleman, P.D., Dendritic extent in human CA 2/3 hippocampal pyramidal neurons in normal aging and senile dementia, Soc. Neurosci. Abstr., 12 (1986) 272. Gibson, G.E. and Peterson, C., Calcium and the aging nervous system, Neurobiol. Aging, 8 (1987) 329-343. Hull, C.D., Bernardi, G. and Buchwald, N.A., Intracellular responses of caudate neurons to brain stem stimulation, Brain Research, 22 (1970) 163-179. Hull, C.D., Bernardi, G., Price, D.D. and Buchwald, N.A., Intracellular response of caudate neurons to ternporally and spatially combined stimuli, Exp. Neurol., 38 (19731 324-336. Kita, T., Kita, H. and Kitai, S.T., Passive electrical membrane properties of rat neostriatal neurons in an in vitro slice preparation, Brain Research, 300 (1984) 129139. Kitai, S.T., Kocsis, J.D., Preston, R.J. and Sugimori, M., Monosynaptic inputs to caudate neurons identified by intracellular injection of horseradish peroxidase, Brain Research, 109 (1976) 601-606. Lamour, Y., Dutar, P. and Jobert, A., Septo-hippocampal neurons: altered properties in the aged rat, Brain Research, 416 (1987) 277-282. Landfield, P.W., Increased calcium-current hypothesis of brain aging, Neurobiol. Aging, 8 (1987) 346-347. Levine, M.S., Neurophysiological and morphological alterations in caudate neurons in aged cats, Ann. N.Y. Acad. Sci., 515 (1988) 314-328. Levifie, M.S., Adinolfi, A.M., Fisher, R.S., Hull, C.D., Buchwald, N.A. and McAllister, J.P., Quantitative morphology of medium-sized caudate spiny neurons in aged cats, Neurobiol. Aging, 7 (1986) 277-286. Levine, M.S., Adinolfi, A.M., Fisher, R.S., Hull, C.D., Guthrie, D. and Buchwald, N.A., Ultrastructural alteration in caudate nucleus in aged cats, Brain Research, 440 (1988) 267-279. Levine, M.S., Fisher, R.S., Hull, C.D. and Buchwald, N.A., Postnatal development of identified medium-sized caudate spiny neurons in the cat, Dev. Brain Res., 24 (1986) 47-62. Levine, M.S., Lloyd, R.L., Fisher, R.S., Hull, C.D. and Buchwald, N.A., Sensory, motor and cognitive alterations

in aged cats, Neurobiol. Aging. 8 (1987) 253-263. 23 Levine, M.S., Lloyd, R.L., Hull, C.D., Fisher, R.S. and Buchwald, N.A., Neurophysiological alterations in caudate neurons in aged cats, Brain Research. 401 (1987)213-23(I. 24 Levine, M.S., Schneider, J.S., Lloyd, R.L., Hull, C.D. and Buchwald. N.A., Aging reduces somatosensory responsiveness of caudate neurons in the awake cat, Brain Research, 405 (1987) 389-394. 25 McNeill, T.H., Koek, L.. Brown, S. and Rafols, J., Age-correlated dendritic changes in medium spiny striatal neurons of the C57BL/6NNIA mouse, Soc. Neurosci. Abstr., 11 (1985) 896. 26 McNeill, T.H., Brown, S.A., Rafols, J.A. and Shoulson, I., Atrophy of medium spiny striatal dendrites in advanced Parkinson's disease, Brain Research, 455 (1988) 148-152. 27 Morales, F.R., Boxer, P.A., Fung, S.J. and Chase, M.H., Basic electrophysiological properties of spinal cord motoneurons during old age in the cat, J. Neurophysiol., 58 (1987) 180-194. 28 Olpe, H.R. and Steinmann, M.W., Age-related decline in the activity of noradrenergic neurons of the rat locus coeruleus, Brain Research, 251 (1982) 174-176. 29 Rogers, J., Silver, M.A., Shoemaker, W.J. and Bloom, EE., Senescent changes in a neurobiological model system: cerebellar Purkinje cell electrophysiology and correlative anatomy, Neurobiol. Aging, 1 (1980) 3-11. 30 Rogers, J., Zornetzer, S.F. and Bloom, F.E., Senescent pathology of cerebellum: Purkinje neurons and their parallel fiber afferents, Neurobiol. Aging, 2 (1981) 15-25. 31 Scheibel, M.E., Lindsay, R.D., Tomiyasu, V. and Scheibel, A.B., Progressive dendritic changes in aging human cortex, Exp. Neurol., 47 (1975) 392-403. 32 Somogyi, P., Bolam, J.P. and Smith, A.D., Monosynaptic cortical input and local axon collaterals of identified striatonigral neurons. A light and electron microscopic study using the Golgi-peroxidase transport-degeneration procedure, J. Comp. Neurol., 195 (1981) 567-584. 33 Stern, W.C., Pugh, W.W. and Morgane, P.J., Single unit activity in frontal cortex of young and old rats, Neurobiol. Aging, 6 (1985) 245-248. 34 Vandermaelen, C.P. and Kitai, S.T., Intracellular analysis of synaptic potentials in rat neostriatum following stimulation of cerebral cortex thalamus and substantia nigra, Brain Res. Bull., 5 (1980) 725-733. 35 Walsh, J.P., Zhou, EC., Hull, C.D., Fisher, R.S., Levine, M.S. and Buchwald, N.A., Physiological and morphological characterization of striatal neurons transplanted into the striatum of adult rats, Synapse, 2 (1988) 37-44. 36 Wilson, C.H., Postsynaptic potentials evoked in spiny neostriatal projection neurons by stimulation of ipsilateral and contralateral neocortex, Brain Research, 367 (1986) 201-213.