T. Inagaki et al.

mer’s disease. Prag ；・481-495.et al. Receptors for

>ping and neuron昌l

13: 21-30. oa L et al. Bram

vascular dementia.

41-151. ~ R et al. Structural

,r light-chain gene.

Is and cort1cotrop1c

>Sin l出 ibitor(and

:of. 1962; 19: 293-

.an unna可 trypsin

iver. LφSci. 1992;

Changes in the Ratio of Urina工yα1-Microglobulin to Ulinastatin

Levels in Patien七S wi七hPsychiatric Diseases

Takuj i 工nagaki, Tadahiro Shikimi, Akihiko Fujimoto, Hi工oshi

Ishino, Hideki Okunishi, and Shuji Takaori

Department of Psych工atry (TI, AF, HI) and Department of

Pha工macology (TS' HO' ST)' Sh工mane Medical Uni ve工sity F 工zumo,

Shimane 693, Japan

1

I n七工oduc七工on

Alpha-1-Mic工oglobulin （αlM, 3 lkDa) and ulinastatin安（UT, 40kDa)

in human u工ine are derived f工om a similar precurso工 protein in

the liver (Kaumeyer e七 al 1986). Hitherto, the physiologically

relevant functions of these substances have not been cla工ified.

The amino acid sequence of UT is no七 only highly homologous to

the inhibitory domain of amyloid 戸－p工otr~in precursor associated

with senile plaque or amyloid angiopathy ( Ki七agu七i et al 1988;

Ponte et al 1988; Tanzi et al 1988), bu七 aUT-like immunoreactive

substance with trypsin inhibi七ory activities is also found in

brain sites related to memory and lea工ning ( Shikimi et al 19 9 3）・

The relationships of u工inaryαlM andl UT levels have been

p工eviously found to differ between the dementia and aged-matched

con七工ol q工oups. In addition, diffe工ent patte工ns of αlM/UT ratio

in relation to the severity of dementia and dura七ion of the

disease have been displayed between Alzheime工－type senile

dementia and vascular dementia patients (Inagaki et al 1995). In

the present study, we studied further whE~ther or not the relation

between urinary levels of αlM and UT differed in patients with

psychoses.

合 The generic name of this human U工inary trypsin inhibi七or has

been recently changed from ur工nastat工n to ul工nas七at工n, as the

substance is found not only in urine but also in serum.

2

Methods

Twelve pa七ien七S with schizoph工enia and 19 pa七ients with mood

disorde工（ 10, pat工ents with b工polar affective disorde工； 9I

patients with major dep工ession) we工e diagnosed according to the

DSM-III-R (American Psychiat工ic Association 1987）・ Patients ove工

60 yea工s of age we工e excluded in this study. The durations of

schizophrenia and mood diso工der in patients were 10. 4 ± 2. 3 and

6 . 1 ± 1. 8 years ( mean 士 S.E.M), respec七ively. The respective

severities of schizophrenia and mood disorder patients were 43 +

6 (Brief Psych工atryRat工ng Scale, Overall and Gorham 1962) and 24

± 4 points (Hamilton Depression Rating Scale, Hamilton 1960)(mean

±S.E.M）・ Patien七S with bipolar affective disorde工（ 6 patients

with and 4 w工thout 1工th工um treatment for at least 4 months)

indicated neither manic nor dep工ess工ve E~pisodes for at leas七 30

days. All the patien七s wi七h majo工 dep工essionwe工e 七工eatedwith

antidep工essants. As age-matched cont工ols, 18 subjects withou七

any presen七 0工 pas七 h工story of psychia七工工c diseases we工e

included. None of the subjects indica七ed hepa七ic／工enal

dysfunctions or o七her malignan七 diseases. 工nfo工medconsent was

obtained from all subjects befo工ethe study.

Spontaneously collected urine (from the subjects) was

centrifuged a七 1,000 x g for 10 min at 4°C 七o remove debris and

amo工phous salts P工ior 七o storage at -50°C until assay. u工ina工Y

levels of αlM and UT were measured with EL工SA, whereby

galactosidase-labelled goat anti-rabbit IgG was interacted

3

(Biot工in 工nternational, Dublin) with rabb工七 anti－αlM 工gG (Dako

Co. , Copenhagen) and r abb工t ant工ーUT IgG (Sh工k工mi et al 19 9 O ) ,

工espectively. StandardαlM (Dako Co・F Copenhagen) and UT we工e

used as references. C工eatin工ne levels工nthe ur工newe工emeasured

with a photometric assay based on the Jaffe 工eaction (Wako

C工eatinine kit, Wako Chem工cals, Osaka）・ U工ina工y levels of αlM

and UT we工e expressed as μg /mg creatinine to co工工ect for the

dilution rate of urine samples.

Linear regression analyses were evaluated by us工ng the Sta七

View IV system (Abacus Concept，工nc., CA）・ Statistical

significance was verif工ed by the factor工al AN OVA followed by

Scheffe『s post hoc test.

statistically significant.

Results

Values where p<O. 05 were considered

In the bipolar affective disorder g工oup,, significant differences

in the age and urinary levels of creatinine FαlM and UT, and

αlM/UT 工atio we工e no七 detected between the li thium-t工ea七ed and

untreated patients. The lithium-treated and untreated patients

were thus pooled as a bipolar affect工ve d工sorder group. 工n the

Table 1, age and levels of creatinine, o.lM and UT did not diffe工

4

significantly among the cont工ol, schizoph工enia, mood diso工der

q工oups and 工ts subd工V工ded q工oups (b工pola工 affective disorder and

major dep工ession）・ The αlM/UT ratio of the mood diso工dergroup,

but not the sch工zophren工a group, was s.ignif icantly higher than

七ha七 of the cont工ols (P<O. 01）・ The bipolar affective disorder

patients, bu七 notthe ma〕ordepress工oncases, contr工butedto this

change.

Relating u工工nary levels of UT and αlM, a positive correlation

was obse工ved in the control, schizophrenia and mood diso工der

q工oups, acco工dingly (Fig. 1）・ A similar tendency was manifested

by the bipolar affective diso工der and major dep工ession groups

(Fig. 2）・ Although the regression slope of the schizoph工enia

cases was not S七atis七工cally d工ffe工ent f工om that of cont工ols, a

significan七 difference was no七ed between the mood disorde工 and

control g工oups (P<O. 05). Between the sch工zoph工enia and mood

disorder g工oups，七he regression slope was significantly ( P<O. 05)

differen七． Fur七hermore, the 工egress.ion slope of the majo工

depression q工oup, but not the bipola工 affective diso工der group,

was statistically ( P<O. 05) d工ffe工ent f工om those of con七工ols and

schizophrenia q工oups. 工n the schizoph工enia and mood diso工der

groups, neither the sco工es f工ompsychiat工ic ra七ing scales (B工ief

Psychiatric Rat工ng Scale and Hamil ton Dep工ession Rating Scale)

nor du工ations of disease correlated well with the U工inarylevels

of αlM O工 UT, orαlM/UT ratio.

5

Discussion

Comparing 七he schizophrenia and control q工oups, the工e were no

significant differences in the 工egression slopes, urinary levels

of αlM and UT, and αlM/UT 工atio. F'u工七hermore, neithe工 the

scores of Br工ef Psychiatric Ra七工ng Scale nor durations of 七he

diseases co工工elated well W工th 七he u工工nary levels of αlM or UT.

These findings sugges七七hat the urinary levels of αlM and UT and

αlM/UT 工atio are not influenced by the pathophysiology of

schizophrenia.

Comparing the mood disorder and control g工oups, the regression

slope of the former differed from that of controls. Within the

mood diso工der category, the slope of 七he maj o工 dep工ession group

was especially significant ( P<O. 0 5) . Furthe工mo工eF 七he αlM/UT

ratio of mood disorder was significantly ( P<O.01) higher than

that of controls. Within the mood diso工de工 catego工Y, the αlM/UT

工atio of the bipolar affective diso工der group was particularly

significan七 (P<0.01). These results suggest that the

工elationships between the urinary levels of αlM and UT in mood

disorder as a whole differ from those of cont工ols.

Comparing the mood diso工dercategory (and inclusive of bipolar

affec七ive diso工de工 and major depression) with schizophrenia

patients, the regression slopes of the mood disorde工 as a whole

6

and major dep工essionq工oupsdiffered f工omtha七 of schizophrenia.

工n our p工evious study (Inagaki e七 al 1995), a positive

co工relation between urina工y levels of αlM and UT has been

obse工ved in dementia patients. The 工espective regression slopes

of dementia q工oups (Alzhe工mer-type and vascular-type) we工e

significantly d工fferent from 七hat of non-dernen七ed con七工ols in

spite of insignificant difference in the αlM/UT ratio.

Therefore, the relat工on between ur工nary levels of αlM and UT

differed among the sch工zophren工a, mood d工so工der as a whole and

dementia pa七工ents.

The αlM/UT ratio of bipolar affec七ive disorder group was

significantly higher than 七hat of cont工ols. However, li thiurn

medication ( 6 0 毛 in the group we工e medicated with lithium) did

not affect the change. This was advocated by insignif ican七

differences in 七he urinary levels of creatinine FαlM and UT and

αlM/UT ratio between lithium－七工eated and untreated pa七ients.

With regard to diet, variations in the urinary levels of αlM and

UT and αlM/UT 工atio were not obse工ved. 工n addition, even in

patients wi七h uro工ogical diseases F七he regression slope rela七ing

to u工ina工y levels of αlM and UT remained statistically

insignificant f工omthat of heal七hysubjects (Shikirni et al 1994）・

Howeve工， thereg工essionslope of七hemood diso工de工 q工oupdiffered

from tha七 ofcont工ols. As none of the p工esentsubjects indicated

7

any renal dysfunctions F 工enalfactors cont工ibuting七ochanges in

the 工elation between urina工Y levels ofαlM and UT in mood

disorde工 patientsmay be discounted.

Glucoco工ticoid-responsive elements (Diarra-Mehrpour et al

1990) and interleukin 6 sequence motifs (Vetr and Gebhard 1990)

have been found in a gene of the αlM- and UT-de工ived precursor

protein. In addition, glucocorticoid-associated fluctuations of

UT in urine, in which plasma levels of glucocorticoids are

工eportedlyaffected by cytok工nes (Spolsky et al 1987; Hermus and

Sweep 19 9 0) , have been observed ( Faa工vang 1962). The urinary

levels of αlM, UT and αlM/UT ratio are also influenced by the

forms (free and complexed) of αlM and UT ex工st工ng工n plasma （αlM

and UT complexes respect工vely ex工st 工n IgA-complexed (Demars e七

al 1989) and IgG-complexed forms (Hochst工asser et al 1981)),

since only the free forms of these substances are found in u工ine

(Jonsson-Berling e七 al 1989; Demars et al 1989）・ Inthe e七iology

of psychosoc工ald工seasesF 七he 工mmunesystem has been focused as a

key facto工（Sm工th 1991）・ Changes 工n th12 system工c immune system,

neu工oimmune mechanisms in the cent工al nervous system and/or

cytokine concent工ations at pe工工pheral S工tes 工n psychoses may be

associated with the va工iable工elationships between u工inarylevels

of αlM and UT. Although七heexact mechanism(s) and physiological

significance of the αlM/UT 工atiowa工工antfu工therstudies, our

8

p工esen七 findings sugges七七hat a change in the relationship

be七ween Y

E

a

n

－－ 工u

levels of αlM and UT 工efleets the

psychoneu工olog工cald工ffe工ences unde工ly工nsJthe various psychiatric

diseases in human.

9

Acknowledgements

The authors thank Dr. T. Sakura工（KonanHosp工tal, Shimane) and

colleagues 工n the Department of Psych工atry, Sh工mane Medical

university for providing 七he urine samples of subjects for this

study. Standard UT was a k工nd q工ft from Moch工da Pharmaceutical

Co・F Tokyo.

10

References

Ame工icanPsychiatr工C Associa七工on (1987): Diagnostic and

Sta七isticalManual of Mental Disorde工S' 3工ded, rev.

Washington DC: American Psychiatric Association.

Demars DD, Katzmann JA, K工ml工ngerTK, Calore JD, Tracy RP (1989):

Simultaneous measurement of total and工gA-conjugatedα1-

microglobulin by a combined immunoenzyme/immunoradiometric

assay technique. Cl工nChem 35:766-772.

Diarra-Meh工pourM, Bourguignon J, Sesbolie R, Salier J-P,

Leveillard T, Ma工七inJ-P (1990): St工uctu工alanalysis of the

human inter－α－trypsin inhibitor light-chain gene. Eu工 J

Biochem 191:131-139.

Faa工vangHJ (1962): The influence of glucocorticoids and

co工ticotropichormone on output of human u工ina工y trypsin

inhibitor (and hyaluronidase inhibi七O工） . Acta Pharmacol

Toxicol 19:293-304.

Hamilton M (1960): A 工atingscale for dep工ession. J Neu工ol

Neu工osu工gPsych工a七ry23:56-61.

He工musAR, Sweep CG (1990): Cy七okines and the hypothalamic-

pitui七ary-ad工enal axis. J Ste工oidBiochem Mol Biol 37:867-

871.

Hochst工asserK, Schonberger OL, Lempart K, Metzge工 M (1981):

Kunitz-type p工oteinase inhibito工S derived by limited

11

一三 也一一 一一一 一」

proteolysis of the inte工ーα－trypsininhibitor. VI. Detec七ionof

a complex between immunoglobul工n G and the inhibi七0工y active

par七 of the inte工ーα－tryps工n 工nh工bitor. Hoppe-Seyle工’ S z

Physiol Chem 362:1363-1367.

Inagaki T, Shikimi T F 工shinoH, Okunishi H, Takaori s (1995):

Changes in七heratio of urina工y α1-microglobulin to

ulinastatin levels in patients with Alzheimer-type and

vascular dement工a. Psych工a工y Cl工nNeurosci 49:287-290.

Jonsson-Be工工工ngB-M, Ohlsson K, Rosengren M (1989):

Radioimmunological quantitation of the u工ina工y trypsin

inhibi七orin normal blood and urine. Biol Chem Hoppe-Seyle工

370:1157-1161.

Kaumeyer JF, Polaz z工J, Kotick MP (1986): The mRNA fo工 a

protease inhibi七orrelated七o the H工－30domain of in七er－αー

trypsin inhibito工 alsoencodes α－1-microglobulin (protein HC).

Nucleic Acids Res 14:7839-7850.

Kitaguti N, Takahashi Y, Tokushima Y, Shiojiri s, Ito N (1988):

Novel p工ecursorof Alzhe工me工Ts d工seaseamylo工d protein shows

p工oteaseinhibito工Y activity ・ Natu工e (London) 331:530-532.

Ove工allJE, Gorham DR (1962): The b工iefpsychiat工ic 工atingscale.

Psychol Rep 10:799-812.

Ponte P, Gonzalez-Dewhitt P, Schilling J, Miller J, Hsu D,

G工eenbe工q B, Dav工S K, Wallace w, L工eberburg 工， FullerF,

12

... ι

Cordell B ( 1988): A new A4 amyloid mRNA. contains a domain

homologus to se工工neprote工nase工nhib工tors. Nature (London)

331:525-527.

Sapolsky R, R工V工erc, Yamamoto G, Plotsky P, Vale W (1987):

Interleukin-1 stimula七esthe secretion of hypo七halamic

corticotropin-releasing factor. Science 238:522-524 (1987)

Shikimi T, Suzuki S, Takahashi M, Kaneto H (1990): Sandwich

enzyme-immunoassay of human urina工y trypsin inhibitor

(u工inastatin) and u工工nastatin-like immunoreactive substance in

mouse urine. Scand J Clin Lab Invest 50:1- 8.

Shikimi T, Wessel T, Joh TH, Takahashi M, Kaneto H, Hattori K,

Takaori s (1993): Demonst工ationof a human u工inarytrypsin

inhibito工（urinastatin) -like substance in the murine b工ain.

Brain Res 616:230-235

Shikimi T, Himeno Y, Shigeno K, Gonda T, Ishibe T, Ha七七0工iK,

Takao工iS (1994): Relationships between ulinastatin and

alpha-1-mic工oglobulin in human urine. Clin Chim Ac七a 227:

195-200.

Smith RS (1991): The immune system is a key fac七O工 inthe

etiology of psychosoc工ald工sease. Med Hypoth 34:49-57.

Tanzi RE, Mcclatchey AI, Lampe工tiED, Villa-Koma工offL, Gusella

JF, Neve RL (1988): Protease 工nh工b工七0工 domain encoded by an

amyloid p工oteinp工ecursormRNA associated with Alzheimer『 S

disease. Nature (London) 331:528-530.

Vetr H, Gebhard W (1990): Structure of the human α1-mic工0-

globulin-bikunin gene. Biol Chem Hoppe-Seyler 371:1185-1196.

13

二 ι一一一一－＝＝二1

Table 1. Numbe工F sex, age, levels of creatinine F α1-mic工0-

globulin （αlM) and ulinastatin (UT), and αlM/UT 工atio in u工ine

of subjects

Mood disorder

Normal Schizophrenia

Control Bipolar Depression Total

Number 18 12 10 9 19

of Subjects

Sex 10/8 6/6 5/5 5/4 10/9

( M／円

Age 40 ± 2 34 ± 3 42 ± 3 33 ± 4 38 ± 3

(years)

Cr 1.37 ± 0.16 0.97 ± 0.24 0.96士 0.17 1.37 ± 0.12 1.15 ± 0.11

(mg/ml)

α1M 6.87±0.81 6.56±0.83 9.77± 1.13 6.90± 1.21 8.41 ± 0.87

(μg/mg Cr)

UT 10.34 ± 1.33 7.75 ± 1.39 6.76 ± 1.20 5.22 ± 0.77 6.03 ± 0.73

(μg/mg Cr)

Ratio 0.75 ± 0.09 1.16士 0.21 1.65 ± 0.17a) 1.36士 0.15 1.51±0.12a)

（α1M/Uη

Values in age and urinary levels of creatinine (Cr）， αlM, UT and

αlM/UT 工atio are represented as mean ± s. E .M.. Abbreviations:

n

o

F

・工

工

s

a

S

1ょ

e

O

E

PιDι

・14

e

B

d

bipolar affective disorde工； Depress工on, 工0

・1Ja

m

Superscript a）工ep工esents values (where P<0.01)

significantly different from those of cont工ols.

14

二 ι一一一＝＝二

Con trot

ii •

,s

..... u en 10

E Cb ユ：ε

Schizophrenia

／／

15

υ 白 10E Cb ユ主。5I

－

．

Mood disorder

．・〆

－

． 、

Y2 o.』，x+ 2.aJ rx: 0.67

( n:118, P<0.01)

d s

同 0.59'/ y,,, 0.95X + 2.65

ロ 0.60 (na12, P<1:l.05) (n= 19, P<D.001)

10 20 10 20 10

UT (μg/mg Cr) UT (μg/mg Cr) UT（ドダmgCr)

Fig.l. Correla七ionbe七weenurinary levels of ulina S七a七in (UT) and

α1-microglobulin （αlM) in age-ma七chedcon七rols and pa七ien七swi七h

e i七her schizophrenia or mood disorde工・s. The correl a七ion

coefficien七 and number of subjec七s are represen七edwi七h r and n

in paren七heses, respec七ively. Cr represents crea七ininein urine.

15

20

Bipolar ~ression 15「 15 ． ． ． ． 、

~ ，，．、、lo-。10υ10 ワ30) E ． t ユ

． 冶） ． ユ ． ／・。、ー” ． 、、，d

2 Y= 0.76X + 4.61 ．

Y=1.30X+0.11 γー・ 5 。 ．

「コ 0.81「ご 0.82

(n=10, Pぐ0.01)( n=9, P<0.01)

。 。10 20 10 20

UT (μg/mg Cr) U了（μg/mgCr)

F工9・2. Correla七ionbe七weenurinary levels of ulinas七a七in (UT) and

α1-microglobulin （αlM) in pa七ien七s wi七h bipolar affec七ive

disorder (Bipolar) or major depression (Depression ）・ The

correla七ioncoefficien七 and number of subjec七s are represen七ed

wi七h r and n in paren七heses, respec七iveユy.

crea七ininein urine.

16

Cr represen七S