Plasma beta-endorphin and natural killer cell activity in major depression: A preliminary study

PsychiclrryResearch,43: I 11- 119 Elsevier

111

Plasma Beta-Endorphin and Natural Killer Cell Activity Major Depression: A Preliminary Study Denis F. Darko, Michael

R. Irwin, S. Craig Risch, and J. Christian

in

Gillin

Received August 28, 1991; revised version received January 13, 1992; accepted May 9, 1992. Abstract. Low concentrations of P-endorphin have been found to enhance human natural killer (NK) cell activity in vitro. Both /J-endorphin and NK activity are changed by clinical depression. To evaluate whether circulating concentrations of /3-endorphin have a role in the in vivo modulation of cellular immunity in humans, we measured plasma fi-endorphin and NK cell activity in 14 depressed patients and 14 age-matched control subjects. In the depressed patients, both plasma fi-endorphin and NK cell activity were reduced to 76% and 57%, respectively, of the mean levels in the control subjects. In addition, P-endorphin showed a significant positive correlation with lytic units of NK cell activity in the combined group of all subjects and in the patient group (p = 0.04), but not in the control group. The study supports the hypothesis that circulating endorphin is correlated with NK cell activity in vivo. This correlation may be higher in the depressed patient group. Key Words. Affective disorder, opiate peptides, cellular immunity. Opiate peptides are likely to have a role in the modulation of immune function. Radiolabeling techniques have demonstrated specific opiate binding on lymphocytes (Mehrishi and Mills, 1983; Ausiello and Roda, 1984), and opiate peptides appear to have both inhibitory and enhancing effects on the immune system. For example, in vitro studies have demonstrated that very low, physiological concentrations of /Iendorphin (10 nM) produce a naloxone-reversible enhancement of natural killer (NK) cell activity (Mandler et al., 1986). Natural killer cells are a subset of lymphocytes that are important in the immune response to certain herpeivirus and cytomegalovirus infections. These cells re_spond rapidly to viral challenge and mount both a proliferative and a cytolytic response several days before a more specific T-cell response can be mobilized (Habu et al., 1984; Bukowski et al., 1985; Biron et al., 1989; Ritz, 1989).

Denis F. Darko, M.D., is Director, Mental Health Research Clinic, U.S. Department of Veterans Affairs Medical Center, San Diego, and Assistant Adjunct Professor, Department of Psychiatry, University of California at San Diego (UCSD) School of Medicine. Michael R. Irwin, M.D., is Associate Director, Alcohol Research Center, U.S. Department of Veterans Affairs Medical Center, San Diego, and Assistant Professor in Residence, Department of Psychiatry, UCSD School of Medicine S. Craig Risch, M.D., is Professor, Department of Psychiatry, Emory University School of Medicine, Atlanta, GA. J. Christian Gillin, M.D., is Director, National Institute of Mental Health Mental Health Clinical Research Center, UCSD, and Professor, Department of Psychiatry, UCSD School of Medicine. (Reprint requests to Dr. D.F. Darko, VAMC, 116A, 3350 La Jolla Village Dr., San Diego, CA 92161, USA.) 0165-1781/92/$05.00 @ 1992 Elsevier Scientific Publishers Ireland Ltd.

112 The release of /!l-endorphin might also have a role in the modulation of NK cell activity in vivo. In human subjects, exercise-induced enhancement of NK cytotoxicity is completely antagonized by the pre-administration of the opiate antagonist naloxone (Fiatarone et al., 1988). Similarly in rats, the acute peripheral release of P-endorphin was positively correlated with splenic NK activity after forced immobilization stress (Irwin and Hauger, 1988). Finally, pre-administration of antiserum to corticotropin releasing factor inhibits footshock-induced release of anterior pituitary peptides and produces a larger stress-induced suppression of splenic NK cytotoxicity than that in animals receiving footshock alone (Irwin et al., 1990~). Despite these clinical and preclinical data that suggest a role for P-endorphin in the enhancement of NK cell activity, few if any studies in human subjects have examined the relationship between circulating concentrations of fi-endorphin and levels of NK activity. In addition, both NK cell activity (Irwin et al., 1987, 1990~~; Irwin and Gillin, 1987) and plasmap-endorphin (Risch, 1982; Gispen-de-Wied et al., 1987) have been reported to be abnormal in patients with major depression. If a relationship exists between NK cell activity and plasma fi-endorphin, a further question is whether this relationship is altered in depressed patients. In this study, resting levels of plasma P-endorphin and NK activity were measured in control subjects and depressed patients to test the hypotheses that circulating concentrations of P-endorphin are positively correlated with natural cytotoxicity in humans, and that depressed mood affects this correlational relationship.

Methods Subjects. The sample comprised 28 subjects-14 patients and 14 control subjects. The patients were recruited for admission to the National Institute of Mental Health Clinical Research Center (MHCRC) from patients presenting to the San Diego Department of Veterans Affairs Medical Center (VAMC) Mental Health Clinic or Psychiatric Emergency Service. The patients were hospitalized of clinical necessity on a voluntary basis on the MHCRC (University of California, San Diego) in the VAMC. All patients studied were admitted to the MHCRC, were 20 to 65 years of age, and met diagnostic screening criteria. All patients completed a Schedule for Affective Disorders and Schizophrenia (SADS) interview (Endicott and Spitzer, 1978) conducted by a Research Fellow in psychiatry. Research diagnoses were made in a consensus meeting of three psychiatrists, who used the DSM-III-R (American Psychiatric Association, 1987) nosologic system. All patients included had a DSMIII-R diagnosis of major depression. (Six of the patients had a history of probable or possible hypomanic episode, and may therefore meet DSM-III-R criteria for bipolar disorder not otherwise specified [bipolar II].) Subjects who did not have a diagnosis of major depression were excluded. The severity of depression was measured by the objective 21-item Hamilton Rating Scale for Depression (HRSD; Endicott et al., 1981) and the subjective Beck Depression Inventory (BDI; Beck et al., 1979), a self-report scale. The physical health of all patients was assessed by medical history, physical examination, complete blood count, CHEM 20 blood chemistry profile, measures of serum cholesterol, triglycerides, triiodothyronine, and thyroxin, rapid plasma reagin test for syphilis, urinalysis, and electrocardiogram. Those with any medical histories or conditions that could have affected the study. such as rheumatoid arthritis or congestive heart failure, were excluded. Some of the subjects involved in this study also volunteered for other studies, which have been published (Dark0 et al., 19880, 19886, 1988c, 1988d, 1989a, 1989b, 1991; Irwin et al., 1987, 1990a, 1990b; Irwin and Gillin, 1987). While all patients had a history of previous treatment with medication, they were hospitalized during the study and had been medication free for at least 14 days before their admission to

113

the study. This 2-week medication-free period is considered long enough to negate any effect of medication on the assays. None of the subjects had been taking the long half-life antidepressant fluoxetine (Prozac). The 14 control subjects were medically and psychiatrically healthy. They were recruited and paid for their participation in the study. Control subjects were not admitted to the hospital, and had no current or past history of mental illness, as determined by SADS interview. The physical health of all controls was assessed by medical history and complete blood count. Mononuclear Cell Collection for Assay of Natural Killer Cell Activity. For the assessment of NK cell activity, 20 ml of venous blood were drawn by standard clinical phlebotomy into heparinized tubes between 830 and lo:30 a.m. (Subjects were not recumbent before the first blood draw.) To obtain mononuclear cells, blood was layered onto FicollHypaque density gradient medium, and centrifuged to separate the monocyte layer. The peripheral blood mononuclear cells (PBMC) were suspended in fully supplemented RPM1 1640 with 10% fetal bovine serum at a concentration of 2 X 106 cells per milliliter. Lymphocyte studies were carried out with the freshly drawn blood samples by a laboratory worker who did not know whether samples were from patients or controls. NK cell activity was assayed with a fixed number of purified lymphocytes titrated for cytolytic activity against 5000 5iCr-labeled K562 target cells. The release of 5tCr was determined in triplicate across four effector:target cell ratios (40: 1, 20: 1, 10: 1, and 5: l), and the following formula was used to express the data as the percentage of specific cytotoxicity: CPM experimental Ye SICr release =

- CPM SR

CPM TR - CPM SR

where CPM equals counts per minute, SR equals spontaneous release (target cells and medium, 10% or less for all experiments), and TR equals total release (target cells in HCl). Data for NK activity are presented as the number of lytic units in 106 cells. Lytic unit values reflect the relative cytotoxicity of the lymphocyte preparations and may be more accurate than comparative chromium release data, which may be influenced by the percent of lysis (Pross and Maroun, 1984). Lytic units were obtained by fitting scale families of curves to our data. In this case, each curve was fit separately to the modified exponential family of equations using the nonlinear least squares program of the SAS computer package (Bloom and Korn, 1983). One lytic unit was defined as the number of effector cells killing 20% of the target cells, calculated according to the equation y = A (1 - e-Kx) where y = functional chromium release, x = lymphocyte-target cell ratio, K = a constant proportional to effector cell number and equal to the negative slope of the target cell survival curve obtained by plotting in (A - JJ) vs. x, and A = maximum cell-mediated lysis (Pross and Maroun, 1984). P-Endorphin Assay. After the blood sample for the assay of NK cell activity was obtained, additional venous blood samples were drawn into heparinized tubes for measurement of circulating levels of P-endorphin. A butterfly needle was placed in a forearm vein, and three or four samples were drawn at 15-minute intervals over the next hour. Since /3-endorphin is secreted in a pulsatile manner, the level of /3-endorphin used in the analyses is the average of the three or four separate blood samples. The plasma was collected and frozen at -80 “C until thawed for the /I-endorphin radioimmunoassay, measuring P-endorphin-like-immunoreactivity, which we refer to as the P-endorphin level within the context of this article. Plasma fi-endorphin-like immunoreactivity was quantitated using a solid phase 1251two-site immunoradiometric assay (IRMA) produced by Nichols Institute. All samples were assayed in duplicate and within the same assay. The intra-assay coefficient of variation was 4.4%. The sensitivity of the assay is 10 pg/ml with a 16% cross-reactivity to human P-lipotropin. Statistical Analysis. The 14 patients and 14 control subjects are used in the analyses. There are some occasional missing data (but probably not enough to affect the results, especially

‘(z aiqB.l. pue E %A) slaaal uydlopua-d ~I!M pw (z +hd pw z alqeL) sl!un zyA1 ~I!M palEla.I.IOD 61aA!Jt38au SBM aJo3s asaH aq$ ‘aldures lelol ayl u! ‘a.IoI_ulaq~m~ ‘(SN ‘9’0 > d ‘L= u ‘9z.o = 1) dnoAl ~OJJUOSayl u! IOU lnq @co*0 = d ‘11= u ‘~9.0 = A f 1 %A) dnoh? wayed passaldap aql u! pun03 SBM 6$!~,rxo~o~L~ put? uydlopua-d uaawaq uoyalaJJo3 .It?l!tu!s v .(z alqel) ICl~agx IlaD ‘>IN 30 sl!un y61 ~I!M paJt?lallos d11uex3!u8!s SBM uydlopua-d ecuseld ‘aldwxs IEJOJ ayl UI .dnoA?! ycwa U~J!M pue aldures p?lol ayl u! d~!ayx? cyIc[ ~J!M palelar.Io3 alaM uydlopua-Q3o sanlelz ‘&y~x IlaD )IN pue uydlopua-d3o s[arza[ 8uy?p-tx!:, uaaMlaq d!qsuoyvla~ aq) alenlaAa 01

IeJnieu 40 si!un 3!11(1 palaln3le3 = swn z!til ‘Jai!l!ll!w Jad slueJ6oa!d u! passaJdxa ‘saldwas Jno) u! u!qdJopua-d etuseld a6eJaAe = uydlopug-d .(aJnSeatU aA!pa!qnS) hOP.IaAUl uO!SSaJdaa y3ag = IQa .(aJnSeaUI aA!g3a!qO) uo!ssaJdaa 101ale39 6ugeu UO)l!UJeH= QSuH ‘(KIaJa#!p Awe3!guG!s ale sdnoJ6 0~ au) 10 sa3ue!JeA al(i uaqM 4 azm!JeA aleJedaS asn at.? :s,l a3ue!JeA aleledas ale S,#J Ja6aW!Uou ‘s,~ ame!JeA palood aJe S,#J Ja6aiw) luopaaJ) 10 saaJ6ap =]p ‘a@N

162 1

>

6O’t 1

92

SN ‘ 1’0 6000’0 >

91

SN ‘90’0

6000’0

IP

d

9z

SN ‘6’0

E9’16&‘16Z‘L St”8 z I’O-

I

S’PL 9’Pl Z Z Zl

as

P’EZ E’EZ 1 1 9P

u-w

wa!qns

L’C 1 1‘0 1 6 6 Pl

as

E.&l L’9 1 oz zz

svun

3!tii

U!qdJOpua-d a1035 WXS

9P

lag

aStlH (4

a6v

uea((y

swaged passaldaa

lOJW’3

sw!qns low03 p 1 fiue swaged passaAdap p 1 u! Ayngcte llact sally lemleu pue ‘u!qdropua-$j eurseld ‘Agganas uo!ssaJdap We uo sisal J dno.6uaannlaq 40 sllnsau - 1 alqel ‘( 1 a[qcL ‘90.0 = d) pa!pnls walqns phenol Papua) ]eyl aDuaJa33!p I?-dno.@

30 dnoJ% lpxus aql u! axwx3!u8!s ~e~!~sye~s loquo~ aql ~I!M pa.wdwoD SB dno&i passald

-ap aq3 U! OhLg Lq pa3npaJ SvM h!A!l~e )IN ‘(L861 ‘UIll’f) ‘4661 . . pw U!MJI :90fjfjI ‘~861 “p la ~‘~11) wodal sno!Aa.id ~I!M JuaJs!suo3 j 1 alqe1‘ 1'0= d) axIex3$!!!S 1123 -!~sge)s yeaI IOU p!p aDuala33!p s!y~ -(Ohpz 30 uoynpal e) dnoti lo.r~uo~ aql u! pun03 ley~ 30 Ogj9~ SEM dnol8 passaldap aq$ u! uyd.Iopua-fj 30 [ahal surse~d ucatu ayl .passaldap 1ClaJaAas ue peq ~30 alox Ia$J B yl!~ qi!~ luayzd ayl ‘p 010 u10.13 sluaged aq~103 sa103s ~I!M sdnol8 0~1 ayl ‘p 01 0 ~0.13 aq) 103 sa.Io3s I#!M ‘( 1 aIqE1) ‘~9 01 sz u10.13 ‘slo_wo~ ayl u! .w1!tu!s a.w sdno.@ OMI ayl

01 klaleJaporu 118a.IaM waged aqL’l1 30 a103s aSaH luaged ayi :zz 30 a1ox Iafl t! peq 9 30 a1ox C[saH ue wacqns 10.1~~03 aql103 asoqi pue ‘8~ 01 p ~013 Bu+w ‘(1 alqel) saIo3s Ia u! luala33Ip dIpay.nxu osp2 aJaM slol~uo~ aqi 103 asoql put? ‘SE 019 1x0.13%u@n?~ siuayed salox asaH u! luala33!p lCIpay_nxu aJaM sdno.Cl ayL tp9 01 zz u10.13a% u! pafluel sluayed ayL .( I alqel) a%

~I!M ‘sasr(leue 5alqv!.nx snonuyuo3

3atyea d paly-oml

p2uoy2la~o3 aq3 -103 pasn aJaM (1) sluay33aoD uoyv~a~.to~ uoslead JO axnzg!u%!s sdno.%uaamlaq wopaay

JO saadap

u! sa3ualaJJg

aql ssasse 01 pasn aJaM sisal I pal!el-om$ p~epuel~ x~ep %u!syu asay I3aUal (Jj9) ‘(si3a[qns 10132103ayl u! am e~sp %!ss!w aql JO ISOUI a3u!s

Table 2. Significant correlations in the total group of subjects P-endorphin with lytic units Lytic units with HRSD score /3-endorphin with HRSD score

f

n

P

0.45 -0.59

18

0.05

17

0.01

-0.43

26

0.03

Note. r = Pearson correlation coefficient n = number of subjects involved in the correlation. p = two-tailed significance value. HRSD = objective Hamilton Rating Scale for Depression. P-Endorphin = average plasma fi-endorphin in four samples, expressed in picograms per milliliter. Lytic units = calculated lytic units of natural killer cell activity across the range of four killertarget cell ratios.

Fig. 1. Relationship between plasma &endorphin and natural killer (NK) cell activity in depressed patients h c

401 l

IO 20 U-Endorphin in

0 Patient data: r = 0.63, n = 11, p

= 0.038

30

40

pg/ml

(two-tailed). The regression

equatron (line) IS y = 0.72x - 0.45.

Fig. 2. Relationship between HRSD score and NK activity in the total group of subjects _”

z ‘2 3 2 r 5 z 5 .z Y Y Y

40

I

Patients 0 Controls l

0 l

10-

0

IO

Hamilton

20 Depression

30 Score

40

HRSD = Hamilton Rating Scale for Depression. NK = natural killer cell. Data on all subjects: r = -0.59, n = (two-tailed). The regression equation (line) is y = 0.64x + 27.85.

i 7, p

= 0.01

116

Fig. 3. Relationship between HRSD score and /3-endorphin in the total group of subjects

6OP

0

l Patients 0 Controls

IO 20 Hamilton Depression

HRSD = Hamilton Rating Scale for Depression. regression equation (line) is y = -0.44x + 26.11.

30 Score

40

Data on all subjects: r = -0.43, n = 26, p = 0.03 (two-tailed). The

Discussion This study evaluated the hypotheses that circulating concentrations of P-endorphin play a role in vivo in the modulation of NK cell activity in human subjects, and that depression affects this modulation. In this study of 14 patients and 14 control subjects, a tendency existed for lytic units of NK cell activity to be lower between groups, consistent with previous reports (Irwin et al., 1987, 1990a, 1990b; Irwin and Gillin, 1987). Basal plasma levels of P-endorphin were also reduced in the depressed subjects. As others have reported, however, this baseline difference was not significantly significant between groups (Meador-Woodruff et al., 1987; Rupprecht et al., 1988; Maes et al., 1990). Finally, NK cell activity was positively correlated with the mean level of plasma P-endorphin in the entire group of 28 subjects and also with that in the depressed patient group. Together, these data support the hypothesis that decreased circulating concentrations of P-endorphin correlate with reduced NK activity, at least within the depressed subjects, similar to the in vitro activity of opioid peptides on cytotoxic activity. Severity of depression also appears to play a role. HRSD scores correlated significantly with both lytic units of NK cell activity and levels of P-endorphin. The patients were all moderately to severely depressed, as assessed by either BDI or HRSD score. Though one patient had a low HRSD score while another had a low BDI score, in each patient, the other scale showed at least moderate depression. Before the study, we did not hypothesize whether objective or subjective measures of depression would correlate more significantly with NK cell activity and fi-endorphin levels. The significant correlations we found with the objective measure would support further testing of a hypothesis that the HRSD score may be correlated with immune and neuroendocrine measures. Our control subjects had HRSD scores in the range of 0 to 4, with BDI scores in the same range. These scores are consistent with depression scores in control subjects in our previous studies (Dark0 et al., 1988a, 19886, 1988c, 1988d, 1989a, 1989b, 1991), and represent individuals without depression.

117

Due to technical problems, the NK assay achieved values for only seven control subjects; this is a small number for statistical analysis. Though it is not uncommon to have fewer representative healthy control subjects than experimental subjects in comparisons, the small number of controls may have contributed to the lack of a significant correlation of /3-endorphin and cytotoxicity in the control group. Elevated basal plasma levels of /3-endorphin and elevated secretion of fi-endorphin in response to cholinergic stimulation in depressed patients have been described by some (Risch, 1982; Gispen-de-Wied et al., 1987) but not by all (Young et al., 1990) investigators. Decreased total &lipotropin/P-endorphin secretory responses after infusions of corticotropin releasing factor have been found in depressed patients (Young et al., 1990). Nonsuppression of plasma /3-endorphin after dexamethasone in depressed patients in whom dexamethasone also did not suppress cortisol has been described, even when the baseline /?-endorphin level was not elevated (MeadorWoodruff et al., 1987; Rupprecht et al., 1988; Maes et al., 1990). As can be seen from this literature, baseline /3-endorphin levels are not consistently altered in depressed patients. Given this background, the negative correlation between HRSD scores and P-endorphin levels is not necessarily unusual. Many hormones and other peripheral modulators may affect NK cells, and their net effect would determine the activity of NK cells in vivo. The current study is limited to an examination of the effect of /3-endorphin within this hormonal milieu and to a comparison of this effect in depressed patients and control subjects. It is not necessarily the case that the circulating level of /3-endorphin in normal individuals would by itself support normal levels of NK cell activity. There are numerous reports in the literature on the duality of the response patterns of leukocytes to /3-endorphin. The dose-response curve for /3-endorphin on NK cell activity is reported to have an arched shape. Further, the conclusion of this study must be cautiously interpreted due to the small sample size and the risk of type II error. The limited power of the analyses may possibly explain the lack of significant group differences in NK activity in the present study. Previous studies have found no difference in NK cell numbers in comparisons of depressed patients with control subjects (Dark0 et al., 1988~) with significant differences being found in NK cell activity (Irwin et al., 1987, 1990a, 19906; Irwin and Gillin, 1987). However, other preliminary work has suggested that NK cell activity in depression may be related to a reduction of NK cell numbers in the periphery (Evans et al., 1990). Nevertheless, the correlation between mean levels of fl-endorphin and NK activity suggests an association between circulating concentrations of P-endorphin and NK activity in depression that warrants further study by ourselves and others in a larger group of subjects. In planning future work, investigators should also include an illness-morbidity measure since /?-endorphin, NK activity, and morbidity have been linked in other studies (Levy et al., 1991). Acknowledgments. This study was supported by National Institute of Mental Health (NIMH) grants MH-42762 (D.F.D., FIRST Award) and MH-44275 (M.R.I., FIRST Award), a VA Merit Review grant (M.R.I.), NIMH Mental Health Clinical Research Center grant MH-30914, a grant from the Georgia Department of Human Resources to the Department of Psychiatry, Emory University School of Medicine, and the Research Service of the U.S. Department of Veterans Affairs Medical Center, San Diego.

118

References American Psychiatric Association. DSM-III-R: Diagnostic and Statistical Manual of Mental Disorders. 3rd ed., revised. Washington, DC: American Psychiatric Press, 1987. Ausiello, C.M., and Roda, L.C. Leuenkephalin binding to cultured human T lymphocytes. Cell Biology International Report, 8:353-362, 1984. Beck, A.T.; Rush, A.J.; Shaw, B.F.; and Emery, G. Cognitive Therapyfor Depression. New York: Guilford Press, 1979. Biron, C.A.; Byron, KS.; and Sullivan, J.L. Severe herpes virus infections in an adolescent without natural killer cells. New England Journal of Medicine, 320:!’ 2-1735, 1989. Bloom, E.T., and Korn, E.L. Qualification of natural cytotoxicity by human lymphocyte subpopulations isolated by density: Heterogeneity of the effector cells. Journal of Immunological Methoak, 58:323-335, 1983. Bukowski, J.F.; Warner, J.F.; Dennert, G.; and Welsh, R.M. Adoptive transfer studies demonstrating the antiviral effect of natural killer cells in vivo. Journal of Experimental Medicine, 16 1:40-52, 1985. Darko, D.F.; Gillin, J.C.; Risch, S.C.; Bulloch, K.; Golshan, S.; and Baird, SM. Peripheral white blood cells and the HPA axis neurohormones in major depression. International Journal of Neuroscience, 45( l-2):153-159, 1989a. Darko, D.F.; Gillin, J.C.; Risch, S.C.; Bulloch, K.; Golshan, S.; Tasevska, Z.; and Hamburger, R.N. Immune cells and the hypothalamic-pituitary axis in major depression. Psychiatry Research, 25:173-179, 1988a. Darko, D.F.; Gillin, J.C.; Risch, S.C.; Bulloch, K.; Golshan, S.; Tasevska, Z.; and Hamburger, R.N. Mitogen stimulated lymphocyte proliferation and the HP axis in major depression. Biological Psychiatry, 26:145-155, 1989b. Darko, D.F.; Lucas, A.H.; Gillin, J.C.; Risch, S.C.; Golshan, S.; Hamburger, R.N.; Silverman, M.B.; and Janowsky, D.S. Cellular immunity and the hypothalamic-pituitary axis in major affective disorder: A preliminary study. Psychiatry Research, 25:1-9, 19886. Darko, D.F.; Lucas, A.H.; Risch, S.C.; Gillin, J.C.; Hamburger, R.N.; and Janowsky, D.S. Age, cellular immunity and the HP axis in major depression. Progress in Neuropsychopharmacology and Biological Psychiatry, 121713-720, 1988~. Darko, D.F.; Rose, J.; Gillin, J.C.; Golshan, S.; and Baird, S.M. Neutrophilia and lymphopenia in major mood disorder. Psychiatry Research, 25:243-25 1, 1988d. Darko, D.F.; Wilson, N.W.; Golshan, S.; and Gillin, J.C. A critical appraisal of mitogen induced lymphocyte proliferation in depression. American Journal of Psychiatry, 148:337344,199l. Endicott, J.; Cohen, J.; Nee, J.; Fleiss, J.; and Sarantakos, S. Hamilton Depression Rating Scale. Archives of General Psychiatry, 38:98-103, 1981. Endicott, J., and Spitzer, R.L. A Diagnostic Interview: The Schedule for Affective Disorders and Schizophrenia. Archives of General Psychiatry, 35:837-844, 1978. Evans, D.; Petitto, J.; Folds, J.; Pedersen, C.; Golden, R.; Corrigan, M.; Haggerty, J.; Duryea, N.; Blair, N.; Tamul, K.; Murphy, C.; Silva, S.; and Ozer, H. Altered natural killer cell populations in major depression: Relationship to cytotoxicity and severity of depression. American College of Neuropharmacology 1990 Annual Meeting, Abstracts of Posters and Panels, 1990. p. 147. Fiatarone, M.A.; Morley, J.E.; Bloom, E.T.; Benton, D.; Makinodan, T.; and Solomon, G.F. Endogenous opioids and the exercise-induced augmentation of natural killer cell activity. Journal of Laboratory and Clinical Medicine, 112:544-552, 1988. Gispen-de-Wied, C.C.; Westenberg, H.G.; Thijssen, J.H.; and van Ree, J.M. The dexamethasone and cortisol suppression test in depression: Beta-endorphin as a useful marker. Psychoneuroendocrinology, 12:355-66, 1987. Habu, S.; Akamatsu, K.; Tamaoki, N.; and Okumura, K. In vivo significance of NK cells on resistance against virus (HSV-1) infections in mice. Journal of Immunology, 133:2743-2747, 1984.

119 Irwin, M.; Caldwell, C.; Smith, T.L.; Brown, S.; Schuckit, M.A.; and Gillin, J.C. Major depressive disorder, alcoholism, and reduced natural killer cell cytotoxicity: Role of severity of depressive symptoms and alcohol consumption. Archives of General Psychiatry, 47:713719, 1990a. Irwin, M.R., and Gillin, J.C. Impaired natural killer cell activity among depressed patients. Psychiatry Research, 20:181-182, 1987. Irwin, M.R., and Hauger, R.L. Adaptation to chronic stress: Temporal pattern of immune and neuroendocrine correlates. Neuropsychopharmacolgy, 1:239-243, 1988. Irwin, M.R.; Patterson, T.; Smith, T.L.; Caldwell, C.; Brown, S.A.; Gillin, J.C.; and Grant, I. Reduction of immune function in life stress and depression. Biological Psychiatry, 27122-30, 1990b. Irwin, M.R.; Smith, T.L.; and Gillin, J.C. Low natural killer cytotoxicity in major depression. Life Sciences, 41:2127-2133, 1987. Irwin, M.R.; Vale, W.; and Rivier, C. Central corticotropin-releasing factor mediates the suppressive effect of stress on natural killer cytotoxicity. Endocrinology, 126:2837-2844, 199oc. Levy, S.M.; Fernstrom, J.; Herberman, R.B.; Whiteside, T.; Lee, J.; Ward, M.; and Massoudi, M. Persistently low natural killer cell activity and circulating levels of plasma beta-endorphin: Risk factors for infectious disease. Life Sciences, 48:107-l 16, 1991. Maes, M.; Jacobs, M.P.; Suy, E.; Leclercq, C.; Christiaens, F.; and Raus, J. An augmented escape of /3-endorphins to suppression by dexamethasone in severely depressed patients. Journal of Affective Disorders, 18: 149-56, 1990. Mandler, R.N.; Biddison, W.E.; Mandler, R.; and Sarrate, S.A. Beta-endorphin augments the cytolytic activity and interferon production of natural killer cells. Journal of Immunology, 136:934-939, 1986. Meador-Woodruff, J.H.; Haskett, R.F.; Grunhaus, L.; Akil, H.; Watson, S.J.; and Greden, J.F. Postdexamethasone plasma cortisol and beta-endorphin levels in depression: Relationship to severity of illness. Biological Psychiatry, 22: 1137-l 150, 1987. Mehrishi, J.N., and Mills, I.H. Opiate receptors on lymphocytes and platelets in man. Clinical Immunology and Immunopathology, 271240-249, 1983. Pross, H.F., and Maroun J.A. The standardization of NK cell assays for use in studies of biological response modifiers. Journal of Immunological Methods, 68:235-249, 1984. Risch, S.C. A.E. Bennett Award paper: Beta-endorphin hypersecretion in depression: Possible cholinergic mechanisms. Biological Psychiatry, 17:1071-1079, 1982. Ritz, J. The role of natural killer cells in immune surveillance. New England Journal of Medicine, 320: 1748-1749, 1989. Rupprecht, R.; Barocka, A.; Beck, G.; Schrell, U.; and Pichl, J. Pre- and postdexamethasone plasma ACTH and beta-endorphin levels in endogenous and nonendogenous depression. Biological Psychiatry, 23:531-535, 1988. Young, E.A.; Watson, S.J.; Kotun, J.; Haskett, R.F.; Grunhaus, L.; Murphy-Weinberg, V.; Vale, W.; Rivier, J.; and Akil, H. Beta-lipotropin/ beta-endorphin response to low-dose ovine corticotropin releasing factor in endogenous depression: Preliminary studies. Archives of General Psychiatry, 47~449-57, 1990.