Cell Adhesion Molecule Expression in the Dorsolateral Prefrontal Cortex

BRITISH JOURNAL OF PSYCHIATRY (2002), 181, 129^134

Cell adhesion molecule expression in the met the DSM–IV criteria for dementia, schizophrenia or other psychotic disorders, dorsolateral prefrontal cortex and anterior a manic or hypomanic episode or a mood disorder due to a general medical condi- tion. Subjects were also excluded if they cingulate cortex in major depression in the elderly met the neuropathological criteria for any known cause of dementia (such as Alzhei- ALAN J. THOMAS, I. NICOL FERRIER, RAJESH N. KALARIA, SUE DAVIS mer’s disease, vascular dementia or demen- and JOHN T.O’BRIEN tia with Lewy bodies) or had evidence of any other neurological disorder. Control subjects were known to be capable of living independently and met the same criteria Background Neuroimaging studies The ‘vascular depression’ hypothesis pro- except that they had never suffered a have demonstrated changes in the poses that late-life depression is associated depressive episode. with vascular disease affecting the frontal- All subjects in the depression group had dorsolateral prefrontal cortex (DLPFC) subcortical circuitry, based on associations undergone extensive assessments, including and anterior cingulate cortex (ACC) in between depression and vascular disease a full history, mental state examination, major depression. (Alexopoulos et aletal, 1997) and magnetic physical examination, screening blood tests, resonance imaging studies in major depres- cognitive tests and, in some cases, computed Aims WeinvestigatedWeinvestigatedthe the expression of sion, which have shown an increase in white tomography or magnetic resonance imaging cell adhesion molecules (CAMs) in the matter lesions (WMLs) that may have a scans. All had received standard antidepres- prefrontal cortex in depression. vascular origin (O’Brien et aletal, 1996). The sant treatment regimes, with selective WMLs appear to be most strongly asso- serotonin reuptake inhibitors or tricyclic MethodMethod Immunohistochemistry to ciated with depression when they involve antidepressants singly or often in combina- localise CAMsinCAMs in post-mortemtissuepost-mortem tissue from frontal-subcortical circuits, which recipro- tion with other agents, and 11 had received cally link prefrontal areas (the dorsolateral electroconvulsive therapy. No control sub- 20 subjects with major depression and 20 prefrontal cortex (DLPFC) and the anterior ject had taken any antidepressant or anti- controls, and image analysis to quantify cingulate cortex (ACC)) to the basal ganglia psychotic medication. The case notes on their expression. (Greenwald et aletal, 1998). This is consistent all subjects were examined to see if they with positron emission tomography studies had a history of hypertension sufficient to ResultsResults Wefound significantsignificantincreases increases in depression, which have demonstrated need treatment with antihypertensive in CAMs in the grey matter of the DLPFC hypometabolism in the DLPFC (Bench medication. All subjects had had a full inthein the depression group butnobut no et aletal, 1992) and the ACC (Drevets et aletal,, post-mortem assessment (except one whose comparable differences in the ACC or 1997). This study tested whether ischaemic body was unavailable for autopsy) and the changes occur in the brain in depression in post-mortem delay was recorded. occipital cortex.Inthe white matter there the DLPFC and the ACC, by measuring was a non-significant increase in two cell adhesion molecules (CAMs). The intercellular adhesion molecule-1 in the expression of these CAMs is increased by Tissue DLPFC in the depression group but no ischaemiaischaemia in vitroinvitro (Kim, 1996) and in After death the right hemisphere was fixed human cerebral microvessels in the vicinity in 10% formalin and the brains were dis- increase ininthe the other areas or for vascular of the infarct following ischaemic stroke sected in a standard manner. To obtain tis- celladhesionmolecule-1in anyarea.Paired (Lindsberg(Lindsberg et aletal, 1996).,1996). sue blocks for analysis, tissue was selected tests showed specificity for the DLPFC in from three areas: the DLPFC (Brodmann the depression group only. areas (BA) 9 and 46), the ACC (BA 24) METHOD and the occipital cortex (BA 19 and 39) as Conclusions TheincreaseinCAM a comparison area. The DLPFC blocks expressionintheexpressionin the DLPFC suggests an Subjects were chosen by carefully selecting the coro- inflammatoryreaction andis consistent Brain tissue from 40 subjects was obtained nal slice from each subject to include BA 9 from the Neuropathology Department/ and 46 according to a standard map (Perry, with ischaemia. Newcastle Brain Tissue Bank. Permission 1993). Owing to the variation in humans, Declaration of interest None.None. had been given for post-mortem research, this may have included BA 10 in some sub- and ethical approval was granted for this jects. The ACC block was taken from BA study. Cases consisted of 20 subjects who 24 just rostral to the genu of the corpus had had depression and 20 controls. Sub- callosum or, occasionally, just including jects with depression were included if they the rostral tip of the genu. These blocks were 60 years or over at death and had were chosen to examine the areas identified suffered at least one well-documented as reduced in function in depression in the episode of DSM–IV major depression DLPFC (Bench et aletal, 1992) and ACC (American Psychiatric Association, 1994). (Drevets(Drevets et aletal, 1997). The occipital cortex Subjects were excluded if they had ever block was chosen to determine whether

129129

Downloaded from https://www.cambridge.org/core. 30 Sep 2021 at 08:21:20, subject to the Cambridge Core terms of use. THOMAS ET AL

any changes in the prefrontal areas were lightly counterstained with haematoxylin structures. Interrater and intrarater reliabil- specific for these areas or occurred through- and were examined using a light micro- ity ratings were calculated (from 48 images out the brain. These blocks were embedded scope to check the quality of staining on 16 patients) for the two raters using the in paraffin wax, the duration of fixation of before proceeding to quantitative analysis. intraclass correlation coefficient and the these large blocks was recorded and they coefficient of variation, respectively. were sectioned using a sledge microtome into 10-into10-mmm sections (one per subject) on Quantitative analysis Statistical analysis large slides (36622’’’’). These slides were Analysis was conducted on one section per coded so that all analysis could be carried anatomical area per subject, based on cal- Statistical comparisons were carried out out blind to diagnosis. culations of the variability between fields using SPSS software (Version 9.0). Tests and between different immunocytochem- for normality were conducted and un- ical assays. The mean coefficient of error paired, two-way Student’s tt-tests or-testsor Immunocytochemistry for five fields (from five subjects) was Mann–Whitney tests were used, as appro- Sections were processed for immuno- 6.8% and there was very little difference priate, to compare ICAM-1 and VCAM-1 cytochemical localisation in a standard whether fields were taken from one section expression in the depression and control manner. Briefly, sections were dewaxed per subject (mean coefficient of error¼ groups. To examine whether any changes in xylene, rehydrated and microwaved in 6.6% for ten fields from five subjects) or in ICAM-1 and VCAM-1 expression 0.01% citrate buffer (pH 6.0) to optimise from two sections per subject run in differ- showed specificity for the prefrontal areas, antigen retrieval. They were immersed in ent immunocytochemistry assays (mean pairedpaired tt-tests and Wilcoxon signed rank hydrogen peroxide, blocked with an coefficient of error¼7.3% for ten fields, tests were used, as appropriate, for within- appropriate serum and incubated for 1 h five per section, from five subjects). group comparisons. Secondary analyses at room temperature with the primary Images were captured using a 661010 were conducted to examine possible con- antibody. The primary antibodies used objective lens on a Zeiss Axloplan 2 light founders. These included comparison of were a polyclonal antibody to intercellular microscope coupled to a three-chip CCD collagen IV expression in the two groups adhesion molecule-1 (ICAM-1) (R&D true-colour video camera (JVC KY F55B), and possible effects of treatment and Systems; 1:500 dilution), a polyclonal anti- producing a field size of 185 000 mmmm22 hypertension on CAM expression. Pearson body to vascular cell adhesion molecule-1 (0.185 mm(0.185mm22). For each antibody, five correlation coefficients, tt-tests or Mann– (VCAM-1) (R&D Systems; 1:800 dilution) images selected randomly were captured Whitney tests, and analysis of variance were and a monoclonal antibody to collagen from the grey matter and ten from the used, as appropriate, for these analyses. IV (Sigma; 1:500 dilution). Immuno- white matter on each section from the cytochemistry was carried out using slides DLPFC and the occipital cortex, and RESULTSRESULTS with code numbers to ensure blindedness ten from the grey matter and fifteen from and with a random order of depression the white matter in the ACC. Images were Subjects’ characteristics are given in and control cases. Slides were processed obtained randomly by the operator select- Table 1, which shows no significant differ- together for each primary antibody in ing a field by moving the stage while not ences in age, post-mortem delay or duration order to ensure that the immuno- looking at the section. If the field was of fixation between the two groups. The cytochemistry conditions were the same outside the area of interest (grey or white coefficients of variation were 3.9% for for both groups. The expression of matter) then this was repeated until the rater 1 and 9.8% for rater 2, and the intra- ICAM-1 and VCAM-1 is increased by field included the required tissue. This class correlation coefficient was 0.97. ischaemia (Kim, 1996) in cerebral ensured that each field was selected inde- Figures 1 and 2 show ICAM-1 and endothelial cells in vitroinvitro and studies of pendently of the microscopic appearance VCAM-1 expression in the two groups human ischaemic stroke have also shown of its microvessels. They were analysed and Figures 3 and 4 show ICAM-1 and increased ICAM-1 expression in the micro- blind to diagnosis on a monitor using a VCAM-1 immunoreactivity, respectively. vessels in association with the stroke standard software program (Image Pro lesions (Lindsberg et aletal, 1996). Both Plus, version 4.0; Media Cybernetics). This ICAM-1 and VCAM-1 were therefore involved measuring the area of DAB stain- Comparison of depression chosen as putative markers of cerebral ing, expressed as a percentage of the total and control groups ischaemia. Collagen IV (a basement mem- image area (areal fraction), and then calcu- The ICAM-1 expression was significantly brane protein) is a marker of the density lating a mean score for the grey and white higher in the depression group in both the of the microvascular tree and was meas- matter for each cortical area. The areal grey matter (tt¼2.77, d.f.2.77,d.f.¼38,38, PP¼0.009)0.009) ured to determine whether any differences fraction was not obtained using threshold- and the white matter (tt¼2.28, d.f.2.28,d.f.¼38,38, in ICAM-1 or VCAM-1 reflected altera- ing because this method was found to lead PP¼0.029) in the DLPFC, but not in the tions in the whole vascular tree rather than to inappropriate inclusion of structures ACC (grey matter: tt¼0.86, d.f.0.86,d.f.¼38,38, in ICAM-1 or VCAM-1 expression in the that were not stained with the CAM anti- PP¼0.394; white matter: tt¼1.15, d.f.1.15,d.f.¼38,38, vessels (Kalaria & Hedera, 1995). Appro- bodies. Instead, the dropper method was PP¼0.258). The VCAM-1 expression also priate secondary antibodies were then used, in which the operator selected DAB- was significantly higher in the grey matter applied followed by avidin-biotinylated stained microvessels by eye in each field of the DLPFC (WW¼321,321, PP¼0.017) but not horseradish peroxidase complex (Vector according to their exact red–green–blue in the white matter (WW¼362,362, PP¼0.199).0.199). Laboratories Ltd) and diaminobenzidine colour characteristics; this avoided the In the occipital cortex, ICAM-1 was (DAB) as a chromagen. All sections were thresholding problem of including other significantly increased in depression in the

130130

Downloaded from https://www.cambridge.org/core. 30 Sep 2021 at 08:21:20, subject to the Cambridge Core terms of use. CELL ADHESION MOLECULE EXPRESSION IN DEPRESSION

Ta b l e 1 Characteristics of the depression and control groups

Depression group (nn¼20)Control group (nn¼20)20)Test statistic PP 95% CI

Age, years (mean (s.d.)) 75.0 (7.37) 74.2 (7.46) 0.30 0.770.77 775.45 to 4.05 Gender (M/F) 7/13 7/137/13 Age of onset of depression, years (mean (s.d.)) 63.8 (14.83) Duration of fixation, months (mean (s.d.)) 123.9 (59.68) 115.0 (59.26) 0.47 0.640.64 7746.97 to 29.17 Post-mortem delay, hours (mean (s.d.)) 34.6 (22.72)(22.72)34.6 28.0 (16.18) 1.05 0.30 7719.18 to 6.08 Causes of death Cardiac 787 8 PneumoniaPneumonia 737 3 Carcinoma 141 4 Suicide 202 0 Other respiratory 222 2 Other cause 1111 3322

1. Renal failure. 2. Haematemesis, mesenteric infarction and liver failure.

the DLPFC compared with the ACC ((tt¼3.85,3.85, PP¼0.001) and occipital cortex ((tt¼4.08,4.08, PP¼0.001) but no difference between the ACC and the occipital cortex ((tt¼1.77,1.77, PP¼0.09). In contrast, in grey matter in the control group there were no significant differences between the DLPFC and the ACC (tt¼1.26,1.26, PP¼0.22), the DLPFC and the occipital cortex (tt¼1.71,1.71, PP¼0.10)0.10) or the ACC and the occipital cortex ((tt¼0.02,0.02, PP¼0.98). In the white matter of the control group there was a significant increase in ICAM-1 in the DLPFC compared with the ACC (tt¼2.59,2.59, PP¼0.02)0.02) but no significant differences between the DLPFC and the occipital cortex (tt¼1.76,1.76, PP¼0.10) or between the ACC and the occipital cortex (tt¼1.39,1.39, PP¼0.18). Similarly for VCAM-1, there was a significant increase Fig. 11Fig. Intercellular adhesion molecule-1 (ICAM-1) expression in three cortical areas in major depression. in the grey matter of the depression group in the DLPFC compared with both the grey matter (tt¼2.10, d.f.2.10,d.f.¼38,38, PP¼0.042)0.042) hypothesised a prioriapriori that these measures ACC (ACC(ZZ¼773.44,3.44, PP¼0.001) and the occi- but not in the white matter (tt¼1.55,1.55, would be highly correlated and correcting pital cortex (ZZ¼771.98,1.98, PP¼0.05) but not d.f.d.f.¼38,38, PP¼0.128). There were no signifi- for these would have been too stringent. in the ACC compared with the occipital cant differences in VCAM-1 immuno- cortex (cortex(ZZ¼771.34,1.34, PP¼0.18). In the white reactivity in either the ACC (grey matter: matter of the depression subjects VCAM-1 WW¼397,397, PP¼0.725; white matter: WW¼468,468, Comparison of areas within groups was highly significantly elevated in the PP¼0.123) or the occipital cortex (grey In the grey matter, paired tt-tests showed DLPFC compared with both the ACC matter:matter: WW¼368,368, PP¼0.262; white matter: that ICAM-1 expression in the DLPFC in ((ZZ¼773.58,3.58, PP550.001) and the occipital WW¼4.17,4.17, PP¼0.871). A Bonferroni correc- the depression subjects was highly signifi- cortex (cortex(ZZ¼773.06,3.06, PP¼0.002) but there tion was set at 0.017, based on testing three cantly elevated compared with both the was no difference in VCAM-1 expression areas in each subject, and the increase in the ACC (ACC(tt¼3.05,3.05, PP¼0.01) and the occipital between the ACC and occipital cortex grey matter of the DLPFC remained signif- cortex (cortex(tt¼2.72,2.72, PP¼0.01), whereas there ((ZZ¼771.53,1.53, PP¼0.13). In contrast, in the icant for both ICAM-1 and VCAM-1. Such was no difference between the ACC and control subjects there were no differences corrections were not made for using both the occipital cortex (tt¼1.17,1.17, PP¼0.26).0.26). in VCAM-1 expression in the grey matter ICAM-1 and VCAM-1 or for assessing Similarly in the white matter, there was a between the DLPFC and the ACC grey and white matter, because we had highly significant increase in ICAM-1 in ((ZZ¼771.32,1.32, PP¼0.19), the DLPFC and the

131

Downloaded from https://www.cambridge.org/core. 30 Sep 2021 at 08:21:20, subject to the Cambridge Core terms of use. THOMAS ET AL

PP440.399) or white matter (tt551.43,1.43, d.f.d.f.¼38,38, PP440.16). In the total group studied there was no correlation between age, post- mortem delay or duration of fixation and any of the four significant results (age: rr550.095,0.095, nn¼40,40, PP440.56; post-mortem delay:delay: rr550.14,0.14, nn¼40,40, PP440.37; fixation: rr550.233,0.233, nn¼40,40, pp440.148). Hypertension did not account for the group differences (ad- justedjusted RR22 for contribution of hypertension 550.053) and a post hochocpost comparison of those with and without hypertension (regardless of whether or not they had depression) found no group differences for either ICAM-1 ((tt551.07, d.f.d.f.1.07, ¼30,30, PP440.291) or VCAM-1 ((tt550.78, d.f.d.f.0.78, ¼30,30, PP440.439) in any area for either grey or white matter. In the depression group we found no correlation between age of onset of depres- Fig. 22Fig. Vascular cell adhesion molecule-1 (VCAM-1) expression in three cortical areas in major depression. sion or any of the four significant outcome measures (measures(rr550.224,0.224, nn¼20,20, PP440.343). The group had a late age of onset of depression (see Table 1), with only six subjects having onset before 60 years and only three before 45 years. Using Mann–Whitney tests there was no difference between those who had received (received(nn¼11) and those who had not received (received(nn¼9) electroconvulsive therapy ((WW44108,108, PP440.552), nor was there any difference between those who had (nn¼13)13) and those who had not (nn¼7) received antidepressant treatment within a week of death (death(WW4460,60, PP440.275). Two suicides were included in the depression group and none of the results was changed by re-analysis without their inclusion.

DISCUSSION Key findings Our main findings were of increased ICAM- 1 and VCAM-1 expression in the DLPFC in the depression group. No differences in the expression of either CAM were found in the ACC and only a very modest and non- significant increase in ICAM-1 was found in the grey matter in the occipital cortex. Fig. 33Fig. Intercellular adhesion molecule-1expression (sections have been counterstained with haematoxylin): This increased expression of ICAM-1 and (a) control grey; (b) depression grey; (c) control white; and (d) depression white. Bars represent 50 mmm.m. VCAM-1 showed marked specificity for the Arrows indicate microvessels. DLPFC in the depression group but little specificity in the control group. These results are occipital cortex (ZZ¼770.85,0.85, PP¼0.40) or the ACC and the occipital cortex (ZZ¼770.43,0.43, consistent with the hypothesis that there are ACC and the occipital cortex (ZZ¼770.06,0.06, PP¼0.67).0.67). ischaemia-induced inflammatory changes in PP¼0.96). In the control subjects’ white the prefrontal cortex in depression. matter there was a significant elevation of VCAM-1 in the DLPFC compared with Assessment of potential the ACC ((theACC ZZ¼772.20,2.20, PP¼0.03) and a trend confounders Increase in CAM expression towards significance between the DLPFC There were no significant group differences is consistent with ischaemia and the occipital cortex (ZZ¼771.83,1.83, in the expression of collagen IV in any area We chose to test the vascular depression PP¼0.07) but no difference between the in either grey matter (tt550.85, d.f.d.f.0.85, ¼38,38, hypothesis (Alexopoulos et aletal, 1997) by

132132

Downloaded from https://www.cambridge.org/core. 30 Sep 2021 at 08:21:20, subject to the Cambridge Core terms of use. CELL ADHESION MOLECULE EXPRESSION IN DEPRESSION

The two groups were similar in age, gender, post-mortem delay and duration of tissue fixation, and our analysis of the data found no evidence that the results could be explained by any of these factors. In a previous study (Thomas et aletal, 2001) wewe,2001) found no differences in clinical measures of vascular risk (e.g. hypertension) between the two groups and so the results do not appear to be due to such confounders. Age of onset of depression did not affect the results, but because most of the depression group had a late age of onset we could not fully explore the extent to which increased CAM expression might differ in an early-onset group. We also found no evidence that treatments unique to the depression group (antidepressants and electroconvulsive therapy) could account for the group differences. Because this was a case-note study we have been unable to examine fully all possible confounding factors and it is possible that the up-regulation of the CAMs was due to some other factor or factors that we have been unable to exclude as a con- founder. Increased CAM expression occurs Fig. 44Fig. Expression of vascular cell adhesion molecule-1 (sections have been counterstained with haematoxy- in association with amyloid plaques in lin): (a) control grey; (b) depression grey; (c) control white; and (d) depression white. Bars represent 50 mmm.m. Alzheimer’s disease and after stroke (Kim, Arrows indicate microvessels. 1996), but our neuropathological assessment has excluded such confounders as measuring the expression of ICAM-1 and response causing neuronal damage and sub- explanations. We measured the expression VCAM-1 in the prefrontal cortex because sequent increased glial activity. However, of collagen IV to examine the possibility these two molecules are increased by ischae- the extent to which these findings can be that any differences in CAM expression miamia in vitroinvitro (Kim, 1996) and studies of generalised to late-life depression in general were due to quantitative differences in the human ischaemic stroke have also shown is unclear; our depression subjects were all cerebral endothelium (Kalaria & Hedera, increased ICAM-1 expression in the micro- hospital patients who had had multiple 1995) but because no differences in vessels in association with the stroke lesions episodes of depression. collagen IV were found we have ruled out (Lindsberg(Lindsberg et aletal, 1996). There is increasing this possible explanation as well. evidence that ischaemic brain damage de- velops over a much longer period than was previously believed (Stoll et aletal, 1998) and Other possible explanations Importance of DLPFC our findings are therefore comparable with for the findings in depression in the elderly a chronic reduction in cerebral perfusion Although we carried out the study to test for Our early findings (Thomas et aletal, 2000),2000) to the DLPFC as well as with acute events. evidence of ischaemia, we cannot exclude showed increased ICAM-1 expression in A recent study in the DLPFC suggested that the possibility that depression mediates the DLPFC in depression. We have now elderly people with depression have in- increases in CAM expression by other extended this by replicating the finding creased astrocytes (Miguel-Hidalgo et aletal,, mechanisms, although the weight of evid- with VCAM-1 and found such increased 2000), which would be consistent with our ence associating depression with vascular dis- immunoreactivity to show some specificity finding of increased CAM expression in eases, especially in the elderly (Alexopoulos for the DLPFC that is not found in the the DLPFC because post-ischaemic inflam- et aletal, 1997), favours prefrontal ischaemia as ACC. One possible explanation for this pat- matory changes would increase astrocyte the most likely explanation of the current tern is that the depressive symptomatology activity. Studies in younger patients with de- results. Another possible pathway leading in our subjects was produced by dispropor- pression have reported a reduction in glia in to increased CAM expression could be the tionate dysfunction in the frontal-subcortical both the ACC (Ongur et aletal, 1998) and the increased circulating levels of cytokines circuit linking the DLPFC to the head of DLPFC (Rajkowska et aletal, 1999), which (e.g. interleukin 1 and tumour necrosis fac- the caudate nucleus rather than in the could lead to a lack of glial support for neu- tor alpha), which have also been described circuit linking the ACC to the nucleus rons and increased neuronal vulnerability. in depression (Connor & Leonard, 1998). accumbens (Alexander et aletal, 1986). Alter- In elderly subjects chronic and/or acute These cytokines stimulate the expression natively, because the DLPFC lies in an ischaemia could lead to an inflammatory of CAMs and thus form a possible link. area of the prefrontal cortex between the

133

Downloaded from https://www.cambridge.org/core. 30 Sep 2021 at 08:21:20, subject to the Cambridge Core terms of use. THOMAS ET AL

territories of the anterior and middle cerebral arteries, it appears to be more vulner- CLINICAL IMPLICATIONS able to ischaemia, especially owing to haemodynamic alterations (Chui & Willis, && Inflammatory changes consistent with cerebral ischaemia were associated with 1999). Our results are consistent with this depression in our elderly subjects. because we found a higher CAM immunoreactivity in the DLPFC in both groups, but && These changes showed specificity for the dorsolateral prefrontal cortex, implying significantly more so for subjects with that dysfunction here is particularly associated with depression in the elderly. depression. These two alternatives could be && related because ischaemic damage to the This opens up the possibility for novel (anti-inflammatory and antivascular) DLPFC would produce a characteristic treatments for late-life depression if these findings are confirmed. pattern of depressive symptomatology invol- LIMITATIONS ving retardation, poor concentration and dysexecutive failure. Such a pattern of execu- && Since this was a retrospective post-mortem study we were not able to examine tive dysfunction has been associated with fully all the possible confounders. poor response to antidepressant treatment and with relapse and recurrence in elderly && OurOursubjectswereallelderly(over60yearsold)andtheextenttowhichour subjects were all elderly (over 60 years old) and the extent to which our subjects with depression (Alexopoulos et aletal,, results may apply to younger adults is not clear and needs further investigation. 2000), which suggests that our findings have && clinical importance. From our results we cannot determine causality; that is, whether depression leads Although further investigation is re- to ischaemic changes or vice versa. quired, our findings potentially have major implications for the aetiology and manage- ment of depression in the elderly. If further research shows that post-ischaemic inflammation is involved in late-life depression, ALAN J. THOMAS, MRCPsych, I. NICOL FERRIER, FRCP,RAJESH N. KALARIA, FRCPath, SUE DAVIS, PhD, the use of anti-inflammatory treatments JOHN T.O’BRIEN,DM,WolfsonT.O’BRIEN, DM,Wolfson Research Centre,InstituteCentre, Institute for Ageing and Health, Newcastle General Hospital, (e.g. non-steroidal anti-inflammatory agents Newcastle uponTyne,uponTyne,UK UK or Cox-2 inhibitors) may become indicated Correspondence: Alan J.Thomas,WJ.Thomas,Wolfsonolfson Research Centre,Centre, Institute for Ageing and Health,Health,Newcastle Newcastle to reduce inflammation and prevent further General Hospital, Newcastle uponTyne NE4 6BE,UK.Tel: 0191 256 3323; fax: 0191 219 5051; tissue injury. e-mail: a.j.thomas@@ncl.ac.uk

(First received 8 June 2001, final revision 5 April 2002, accepted 9 April 2002) ACKNOWLEDGEMENTS

WewishtothankAndrewBrownforhelpinprepar- ing the sections for this study and the Stanley Foun- Connor,T. J. & Leonard, B. E. (1998) Depression, adults in major depressive disorder. Biological Psychiatry,, dation, the Mason Medical Research Foundation stress and immunological activation: the role of 48,,861^873. 861^873. and the Newcastle University Hospitals’ Special cytokines in depressive disorders. Life Sciences,, 6262,, Trustees for their financial support. 583^606. O’Brien, J., Desmond, P., Ames, D., et aletal (19 9 6) AA magnetic resonance imaging study of white matter Drevets,W.C., Price, J. L., Simpson, J. R., Jr., et aletal lesions in depression and Alzheimer’s disease. British (19 9 7)7)(19 Subgenual prefrontal cortex abnormalities in REFERENCES Journal of Psychiatry,, 168168, 477^485.,477^485. mood disorders. Nature,, 386386, 824^827.,824^827. Ongur, D., Drevets,W.C. & Price, J. L. (1998) Glial Greenwald, B. S., Kramer-Ginsberg, E., Krishnan, Alexander,G. E., DeLong, M. R. & Strick, P.L. (1986) reduction in the subgenual prefrontal cortex in mood K. R.,K.R., et aletal (19 9 8) Neuroanatomic localization of Parallel organization of functionally segregated circuits disorders. Proceedings of the National Academy of magnetic resonance imaging signal hyperintensities in linking basal ganglia and cortex. Annual Review of Sciences of the United States of America,, 95,13290^13295., 13290^13295. Neuroscience,, 99,,357^381. 357^381. geriatric depression. Stroke,, 29,,613^617. 613^617. Perry, R. H. (1993) Coronal Maps of Brodmann Areas in Alexopoulos, G. S., Meyers, B. S.,Young, R. C., et aletal Howard,C.V.&Reed,M.G.(1998)Unbiased the Human Brain. London: Wolfe. (19 9 7) ‘Vascular depression’ hypothesis. Archives ofofArchives StereologyThree-Dimensional Measurement in Microscopy.. General Psychiatry,, 5454,,915^922. 915^922. Oxford: Bios Scientific Publishers. Rajkowska, G., Miguel-Hidalgo, J. J.,Wei, J., et aletal (1999)(1999) Morphometric evidence for neuronal and glial __ ,, __ ,, __ ,, et aletal (2000) Executive dysfunction and Kalaria, R. N. & Hedera, P. (1995) Differential long-term outcomes of geriatric depression. Archives ofofArchives degeneration of the endothelium and basement prefrontal cell pathology in major depression. Biological General Psychiatry,, 5757,285^290., 285^290. membrane of capillaries in Alzheimer’s disease. PsychiatryPsychiatry,, 4545,1085^1098. Neuroreport,, 66,477^480., 477^480. American Psychiatric Association (1994) Diagnostic Stoll, G., Jander, S. & Schroeter, M. (1998) and Statistical Manual of Mental Disorders (4th edn) Kim, J. S. (1996) Cytokines and adhesion molecules in Inflammation and glial responses in ischemic brain (DSM^IV).Washington, DC: APA. stroke and related diseases. Journal of the Neurological lesions. Progress in Neurobiology,, 5656,149^171.,149^171. Sciences,, 137, 69^78.,69^78. Bench, C. J., Friston, K. J., Brown, R. G., et aletal (19 92)92)(19 Thomas, A. J., Ferrier, I. N., Kalaria, R. N., et aletal The anatomy of melancholia ^ focal abnormalities of Lindsberg, P.P.J., J., Carpen, O., Paetau, A., et aletal (19 9 6) (2000)(2000) Elevation in late-life depression of intercellular cerebral blood flow in major depression. Psychological Endothelial ICAM-1 expression associated with adhesion molecule-1 expression in the dorsolateral MedicineMedicine,, 22, 607^615.,607^615. inflammatory cell response in human ischemic stroke. prefrontal cortex. American Journal of Psychiatry,, 157,, CirculationCirculation,, 9494, 939^945.,939^945. 1682^1689. Chui, H. & Willis, L. (1999) Vascular diseases of the frontal lobes. InThe Human Frontal Lobes (eds B. L. Miller Miguel-Hidalgo, J. J., Baucom, C., Dilley, G., et aletal __ ,, __ ,, __ ,, et aletal (2001)(2001) AAneuropathologicalstudyof neuropathological study of & J. L. Cummings), pp. 370^401.London: Guilford (2000)(2000) Glial fibrillary acidic protein immunoreactivity in vascular factors in late-life depression. Journal ofofJournal Publications. the prefrontal cortex distinguishes younger from older Neurology,NeurosurgeryNeurology, Neurosurgery and Psychiatry,, 7070, 83^87.

134134

Downloaded from https://www.cambridge.org/core. 30 Sep 2021 at 08:21:20, subject to the Cambridge Core terms of use.