Determination of antipituitary antibody in patients with endocrine disorders by enzyme-linked immunosorbent assay and Western blot analysis

ORIGINAL ARTICLES Determination of antipituitary antibody in patients with endocrine disorders by enzyme-linked immunosorbent assay and Western blot analysis SHIGEKI YABE,TSUGIYASU KANDA, MINA HIROKAWA, SHIZUE HASUMI, MIZUHO OSADA, YUKIHITO FUKUMURA,AND ISAO KOBAYASHI MAEBASHI, JAPAN

The identification of pituitary antigens recognized by human antipituitary antibodies (APAs) is important in evaluating the pathophysiology of multiendocrine disorders linked to autoimmune factors. However, there is no convenient method for the quantitative analysis of circulating APAs. This study reports the development of an enzyme-linked immunosorbent assay (ELISA) for the detection of APAs. APAs were measured by ELISA and confirmed by Western blot analysis in sera from patients with endocrine disorders. APAs were detected frequently in patients with autoimmune thyroiditis, insulin-dependent diabetes mellitus (IDDM), or pituitary dwarfism. Circulating APAs were d e t e c t e d in 18% of patients with autoimmune thyroiditis. Confirmation by Western blot revealed positivity for APAs in the serum of 36% of patients with Hashimoto disease and in 29% of patients with Graves disease. Notably, 39% of patients with IDDM were also positive for APAs by ELISA. The identification of APAs by ELISA may be useful in evaluating autoimmune mechanisms involved in patients with multiendocrine disorders. (J Lab Clin Med 1998;132:2531)

Abbreviations: APA = Antipituitary antibody; BSA = bovine serum albumin; ELISA= enzymelinked immunosorbent assay; GH = growth hormone; IDDM = insulin-dependent diabetes melIitus; IgG = immunoglobulin G; NIDDM = non-insulin-dependent diabetes meliitus; PBS= phosphate-buffered saline solution; RT= room temperature; SDS= sodium dodecylsulfate

A

variety of autoantibodies have been detected in the sera of patients with autoimmune diseases. Such autoantibodies are classified as either organ-specific or organ-nonspecific and may From the Department of l~aboratory Medicine and Clinical Laboratory Center, Gunma University School of Medicine. Supported in part by a grant-in-aid (#04454545, I.K.) for Scientific Research from the Ministry of Education, Scienceand Culture, Japan. Submitted for publication Sept. 13, 1996; revision submitted March 13, /998; acceptedMarch 18, 1998. Reprint requests: Isao Kobayashi, MD, Department of Laboratory Medicine and Clinical Laboratory Center, Gunma University School of Medicine, 3-39-15, Showa-machi,Maebashi 371-8511, Japan. Copyright © 1998 by Mosby, Inc. 0022-2143/98 $5.00 + 0 5/1/90358

be involved in the pathogenesis of the related diseases. Although APAs are thought to be involved in the pathogenesis of autoimmune multiendocrine disorders, little information is available on the specific pituitary antigens they recognize. APAs have been identified in patients with adrenocorticotropic hormone deficiency, 1 panhypopituitarism, 2 empty sella syndrome,3, 4 tumor o f the pituitary, 4 Hashimoto thyroiditis, 5 Graves disease, 6 and IDDM, 7 suggesting a pathophysiologic role for APAs in these disorders. This study describes an E L I S A 8 that was developed for the detection of APAs. The presence of APAs in patients with endocrine disorders was evaluated by E L I S A and in same cases confirmed by Western blot analysis. 9 25

26

Yabe et al.

J Lab Clin Med July 1998

4.00

NEG -e- POS ....._-.4

3.00" o ffl ~ 2.00. ffl m < 1.00 -

~

o.oo.//~

A

10

0 rat

O I

O

0

i

i

25 50 100 antigen concentration ( # g/mL)

pituitary

200

4.00

-g

3.00 -

POS

o ~ 2.00. o3 < 1.00-

ooo-,

B

, 25

.

50

~

/

1oo

200

Dilution

~,

?

?

400

800

1600

rate

4.00 NEG A v

A

3.00 Q

2.00-

1.00 -

0.00- ~,/ C

500

~

1000

~

2000 D i l u t i o n

4000

8000

16000

rate

Fig. 1. A, ELISA APA assay showing binding of APA-positive serum (POS) and APA-negative serum (NEG) to rat pituitary antigen in concentrations of 0 to 200 I.tg/ml. B, ELISA APA assay comparing dilutions (1:25 to 1:1600) of APA-positive serum (POS) and APA-negative serum (NEG). Rat pituitary antigen was coated at the concentration of 25 ~g/ml. C, Optimal proportions of peroxidase-conjugated anti-human IgG rabbit polyclonal antibody. METHODS Subjects. Serum samples were obtained from 358 Japan-

women, aged 21 to 60 years). Written informed consent was obtained from each subject.

ese patients. The diagnoses included Graves disease (143), Hashimoto thyroiditis (76), IDDM (64), NIDDM (38), pituitary adenoma (12), and pituitary dwarfism (25). Control sera were collected from 219 healthy volunteers (108 men and 111

Preparation of rat pituitary antigens. Rat pituitary glands were obtained (RKL, Gilbertsville, Penn) and homogenized in 3 ml of buffer (0.25 mol/L sucrose, 0.1 mrnol/L ethylelediamine tetraacetic acid, 3 mmol/L Tris-HC1 buffer, pH 7.4) by a Poly-

J Lab Clin Med Volume 132, Number 1

tron homogenizer (Brinkmann Instruments, Westbury, N.Y.) and centrifuged at 10,000 g at 4 ° C for 10 minutes. The resultant supernatant was used as the source of pituitary antigens for subsequent steps. The protein concentration and protein standard were determined by protein assay kit (Bio-Rad, Hercules, Calif.). . . . Enzyme immunosorbent assay. Microtiter plates (Nunclmmuno Module, A/S Nunc, Roskilde, Denmark) were coated with 100 gl rat pituitary antigen (supematant diluted to 25 gg/ml in 0.05 mol/L sodium bicarbonate buffer, pH 9.6) in each well by incubation overnight at 4 ° C. The plates were washed with 0.05% PBS-Tween 20 and blocked with 300 gl of 3% BSA in PBS at 37 ° C for 30 minutes. After washing with 0.05% PBSTWeen 20, each well received 100 gl of sample sera or APApositive control sera and the plate was incubated with shaking at RT for 2 hours. Plates were then washed five times with PBS. Peroxidase-labeled rabbit anti-human IgG polyclonal antibody (Sigma, St. Louis, Mo.) was added, 100 gl to each well, and allowed to incubate at RT for 1 hour. After washing with PBS, 100 gl of substrate solution (0.1 mol/L sodium acetate-citrate containing 0.006% H202 and 0.2 mg/ml 3,3",5,5"-tetramethylbenzidine dihydrochloride, pH 5.5) was added to each well, followed by incubation at RT for 30 minutes. The colorimetric reaction was stopped by 100 btl of 0.5 N H2SO4. The absorbance at 450 nm was measured with Behring ELISA processor III (Behring Werke AG, Marburg, Germany), and the cutoff index was calculated. Assays were typically performed in duplicate. Western blot analysis. SDS polyacrilamide gel electrophoresis was performed according to the method of Laemmli. 10 In brief, 50 gl of rat pituitary antigen (approximately 200 gg/protein) was mixed with 50 gl sample buffer (0.1 dithiothreitol, 2% SDS, 15% glycerol, 0.006% bromophenol blue, 5% 2- mercaptoethanol, and 0.08 mol/L Tris-HC1,pH 6.8), heated at 96 ° C for 1 minute, and electrophoresed in gradient 4% to 20% polyacrylamide gel (TEFCO, Tokyo, Japan) for 1.5 hour at 18 mA (nmning buffer: 25 mmol/L Tris, 192 mmol/L glycine, 0.1% SDS). Separated proteins were transferred to a polyvinyfidene difluoride membrane (FEFCO) by means of a semidry blotting apparatus (Bio-Rad Laboratories, Richmond, CA) at 6 V for 1 hour. Nonspecific binding sites were blocked by incubation of the membrane overnight at 4 ° C with Block Ace (Dainippon Inc, Osaka, Japan). The protein-loaded membrane was washed three limes in washing buffer (5% nonfat milk, 0.05% Tween-20 in PBS, pH 7.2). The membrane was then incubated in 1:101 diluted patient serum in dilution buffer (5% nonfat milk, 3% BSA in PBS, pH 7.2) at RT for 2 hours and washed in washing buffer. It was then incubated in 1:500 diluted biotinylated rabbit anti-human IgG polyclonal antibodies (Sigma) in dilution buffer at RT for 1 hour, and washed in washing buffer for 5 minutes three times. The membrane was then incubated in 1:50 diluted streptoavidin-biotin complex peroxidase (Dakopatts, Clostmp, Denmark) in 3% BSA in PBS, pH 7.2, at RT for 30 minutes and washed in washing buffer for 5 minutes three times. The labeledbands were revealed by chemiluminescence using POD Immunostein Set (Wako thare Chemicals, Kyoto, Japan) for 5 minutes. Statistical analysis. The data are expressed as mean _+SD, unless otherwise indicated. The statistical analysis was performed by unpaired Student's t-test or one-way analysis of variance. The percent positive analysis was performed by X2 test.

Yabe et al,

27

Table I. R e p r o d u c i b i l i t y o f ELISA Cutoff index (mean _+SD)

Sample Intraassay (n = 10) Sample 1 Sample 2 Sample 3 Sample 4 Interassay (n = 7) Sample 1 Sample 2 Sample 3

0.974 7.925 4.505 3.300

Coefficient of variation (%)

+ 0.094 +_0.418 + 0.193 _+0.134

9.640 5.274 4.280 5.820

7.707 _+0.743 4.060 + 0.306 2.067 +_0.193

9.600 7.540 9.370

Table II. P e r c e n t suppression o f E L I S A - d e t e r m i n e d APAs a f t e r a d d i t i o n o f various rat tissues a n d h u m a n g r o w t h h o r m o n e ( m e a n + SD; N = 5) Added element Stomach Liver Pancreas Thyroid GH

Inhibition (%) 1.1 0.8 4.7 7.5 89.8

+ 4.0 _+9.8 _+4.0 _+2.8 _+4.3

RESULTS Validation of reagents

Coating antigen, For each ELISA, the minimum level of detection was defined as the m i n i m u m amount o f antibody required for the serum to have at least one point (10 gg/ml) within the linear range of the APA-positive serum concentration-response curve (Fig. 1, A). This was estimated to be 25 btg/ml for the turning point. Reactivity of APA-containing serum. Plates were coated at 25 btg/ml with rat pituitary antigen and incubated with serum containing A P A at dilutions from 1:25 to 1:1600. The difference between the absorbance of APApositive serum and that o f A P A : n e g a t i v e serum was clear at dilutions from 1:100 to 1:400 (see Fig. 1, B).

Reactivity of peroxidase-conjugatod anti-human IgG rabbit polyclonal antibody. Plates were coated with rat pituitary antigen (25 gg/ml) and incubated with peroxidase-conjugated anti-human IgG rabbit polyclonal antibody (Sigma) at dilutions from 1:500 to 1:16,000. The difference between the absorbance of antibody-positive and that of antibody-negative samples was clear at dilutions from 1:2500 to 1:10,000 (see Fig. 1, C). Assessment of efficiency

Linearity of diluted samples. T w o A P A - p o s i t i v e samples and one APA-negative sample were tested. At dilutions from 1:50 to 1:6400, the absorbance was linear between 1.0 and 2.0 at 450 nm (data not shown). Reproducibility. To assess the intraassay and interassay variability, i n d e p e n d e n t repeats of seven samples

28

J Lab Clin Med July 1998

Yabe et al,

Table III. Growth h o r m o n e reactivity after i n c u b a t i o n with sera positive for APA Sample no.

1¢

2§

3§

GH added (ng/ml)

Observed value (ng/ml)*

0 3.57 7.79 0 3.57 7.79

0.45 3.94 8.26 1.5 4.39 7.52

0 3.57 7.79

0.13 3.41 6.60

Expected value (ng/ml)

Recovery (%)t

4.02 8.24

98.0 100.2

5.07 9.29

86.6 80.9

3.48 7.92

98.0 83.3

*Mean of two replicates. tCalculated as (observed value + expected value) x 100. *APA-negative sample. §APA-positivesample.

from two different sera were examined by ELISA. The mean cutoff index, standard deviation, and coefficient of variation were determined (Table I). Interfering experiment. Bilirubin C (209 rag/L), bilirubin F (191 mg/L), chyle (1950 FD), hemoglobin (4.7 gin/L), ascorbic acid (5 gin/L), and rheumatoid factor (600 IU/ml) were examined as contaminating interfering substances (data not shown). The presence of these substances did not affect the results of APA analysis. Absorption of nonspecific antibody. Acetone-treated rat liver powder (Sigma) was mixed (10 mg/ml) with 3 APA-positive and 19 APA-negative samples (verified by Western blot) and incubated for 1 hour at 37 ° C. APA-positive and APA-negative sera were clearly differentiated in the presence of rat liver nonspecific antibodies. The methods of absorption were described as follows. Various tissues including stomach, liver, thyroid, and pancreas were homogenized and centrifuged at 10,000 g at 4 ° C for 10 minutes. The resultant supernatant was used as the source of tissue antigens. These antigens were added to sera from APA-positive patients at 1:51 dilution in PBS. The ratio of tissue antigen solution to patient sera was 1:1. Human GH (100 gg/ml, CHEMICON, Inc. Temcula, CA) was added in a similar manner. The extract of rat stomach, liver, thyroid gland, and pancreas did not cross-react with this APA. In contrast, APA apparently cross-reacted with human GH, as shown in Table II. Therefore, the APA was considered to be a specific antibody against pituitary gland. Recovery experiment of APA-positive sera on human GH. Samples of human GH (0 ng/ml, 35.7 ng/ml, 77.9

ng/ml) were mixed with sera of three patients at the dilution of 1:10. The incubation was performed for 12 hours at 4 ° C. The recovery rate was calculated as follows: Recovery rate (%) = (Observed GH value in ng/ml + Expected GH value in ng/ml) x 100 The experiments were done in duplicate. GH was

measured by Ab Bead HGH Immunoradiometric Assay kit (Eiken Chemical Co., Ltd., Tokyo, Japan). The presence of APA-positive sera reduced the amount of GH that was detected (Table III). Variations in patients with endocrine disorders Cutoff value in healthy volunteers. Among 219 healthy

volunteers, APA values ranged from 0.33 to 6.54 (mean _+SD; 1.57 +_0.99). Values did not differ by sex. Western blot analysis of 21 healthy volunteers with values >2.5 revealed bands that were positive for APA in 10 subjects (47.6%). Excluding those 10 APA-positive subjects, the distribution of values was from 0.33 to 4.05 (1.42 _+0.64). Values >2.7 (mean + SD) were considered as positive for APA (Fig. 2). Autoimmune thyroiditis. Of the 76 patients with Hashimoto thyroiditis, 12 (15.8%) were positive for APA, as were 28 of 143 patients with Graves disease (19.6%; p < 0.05 versus healthy volunteers). Confirmation by Western blot analysis revealed APA-positive bands in 4 (36.4%) of 11 patients with Hashimoto disease and in 6 (28.6%) of 21 with Graves disease (Fig. 3; see Fig. 2). Antinuclear antibody was detected in 11 (20.4%) of 54 patients with Hashimoto disease and 11 (20.4%) of 54 patients with Graves disease. Three (5.6%) of 54 patients with Hashimoto disease and 11 (20.4%) of 54 patients with Graves disease were positive for both APA and antinuclear antibody. Diabetes mellitus. APAs were detected in 2 (5.3%) of 38 patients with NIDDM and 25 (39.1%) of 64 patients with IDDM (p < 0.001 versus healthy volunteers). Confirmation by Western blot analysis revealed APAs in 1 of 2 patients with NIDDM and 16 (64.0%) of 25 patients with IDDM (see Figs. 2 and 3). Pituitary gland disorders. APAs were detected in 2 (8.0%) of 25 patients with pituitary dwarfism but in none of 12 patients with pituitary adenoma. Western blot analysis showed no positive bands for APA in sam-

J Lab Clin Med Volume 132, Number 1

Yabe et al.

29

8.0

6.0

.=

d

......i .......

v

<~ EL

4.0

<

< 2.0 03 m uJ

0.0

Healthy Hashimoto's Graves' volunteers disease disease (n=219) (n=76) (n=143)

NIDDM

IDDM

(n=38)

(n=64)

Pituitary adenoma (n=l 2)

Pituitary dwarfism (n=25)

Fig. 2. Distribution of ELISA-determined APAs in sera from healthy volunteers (n = 219) and from patients with Hashimoto thyroiditis (n = 76), Graves disease (n = 143), NIDDM (n = 38), IDDM (n = 64), pituitary adenoma (n = 12), and pituitary dwarfism (n = 25). The statistical significance in Hashimoto thyroiditis, Graves disease, and IDDM was p < 0.1, p < 0.01, and p < 0.001, respectively, against healthy volunteers: The dotted lines at SD, bar indicates mean value.

ples from patients with pituitary disorders (see Figs. 2 and 3). DISCUSSION

APAs were detected in control sera significantly less often than in sera from patients with autoimmune thyroiditis, IDDM, or pituitary dwarfism. The presence of circulating APAs suggests that there exists an autoimmune regulation of thyroid hormone and insulin secretion via the hypothalamic-pituitary-thyroid axis. Absorption of nonspecific antibodies did not interfere with analysis of APAs. The significant reduction of GH values observed after mixing with APA-positive serum suggests that specific epitopes of APA may be found on GH and related hormones in the pituitary gland. Detection of APAs in patients with endocrine disorders by ELISA was consistent with results of previous immunofluorescence studies.2-5,11,12 The frequency of antinuclear antibody was low (<6%) in patients with autoimmune thyroiditis, suggesting that the antigens that produce APAs have organ-specific epitopes. The thyroid gland is regulated by pituitary hormones such as thyroid-stimulating hormone and adrenocorticotropic hormone. 12 In Graves disease, APAs may interact with regulatory hormones in the pituitary gland, resulting in the development of hyperthroidism. Hashimoto thyroiditis appears to result from organ-specific autoimmunity. 13 Although the pituitary is not the target organ in thyroiditis, the hypothalamus-pituitary-thyroid axis completely regulates synthesis and secretion of thyroid hormones. Investigation of the relationship between APAs and thyroid function has been initiated.

22kDa

Fig. 3. Western blot analysis of rat pituitary antigen on a 4% to 20% SDS polyacrlamide gel with sera from 10 patients. Lanes 1 and 2, NIDDM; lanes 3 and 4, IDDM; lanes 5 and 6, Hashimoto thyroiditis; lanes 7 and 8, Graves disease; lanes 9 and 10, pituitary dwarfism.

Results of this study are consistent with previous reports of an increased frequency of APAs in the sera of patients with IDDM. 7,14 Sugiura et al. 12 reported that 24 of 81 patients with IDDM demonstrated APAs by indirect immunofluorescent techniques. All patients with NIDDM had negative APA. In contrast, the present study showed the presence of APA-positive serum in 5.3% of 38 patients with NIDDM. In the observation of clinical course or severity of disease, we previously reported that the urinary levels of C-peptide were significantly lower in patients with NIDDM compared with APA-negative ones. 14 A recent report showed that

30

J Lab Clin Med July 1998

Yabe et al,

APAs were positive in 7% of patients with lymphocytic adenohypophysitis with panhypopituitarism. 15 However, there was no evidence of clinical significance in patients with positive APAs. We suggest that patients with positive APAs may represent the autoimmune state. The significance of detectable levels of APAs in normal subjects was not clear, although the autoimmune state may be invaluable. Hypopituitarism has been associated with diabetes, 12 and virus-infected cells have been observed in the pancreas as well as in the pituitary gland of diabetic mice infected with reovirns. 16 These findings suggest that these organs share a common antigenicity, suggesting a link between the pituitary gland and the secretion of insulin by the pancreas. Tartaglia et al. 17 identified a high-affinity receptor for leptin in the hypothalamus. Leptin is secreted by adipocytes and is an important circulating signal for weight control. Recently proposed mechanisms of weight control postulate that the hypothalamic receptor receives the leptin signal and controls energy expenditure through the hypothalamus-pituitary-thyroid axis by regulating the secretion of insulin and glucocorticoids, 18 hormones that stimulate leptin output. 19 APAs may play a role in the pathogenesis of the onset of I D D M and of insulin deficiency in patients with NIDDM. 14 In the present study, APAs were detected in a low percentage (3/28, or 8.0%) of patients with pituitary dwarfism and in none of 12 patients with pituitary adenoma. Maxjorie et al. 20 reported that 45% of patients with pituitary disorders such as pituitary adenoma and empty sella syndrome had APAs such as anti-GH antibody, antithyroid-stimulating hormone antibody, and anti-adrenocorticotropic hormone antibody. Also, Bottazzo et al. 11 reported the existence of autoantibodies to anterior pituitary cells secreting GH, luteinizing hormone, and follicle-stimulating hormone. Our previous report mentioned that a 22-kD band of APA was detected as a soluble cytoplasma in GH-related protein. 9 Crock et al. 2~ reported the detection of APAs by immunoblotting with human pituitary tissues as antigens. They showed that a 45-kD pituitary-specific membrane protein was identified as an autoantigen in 1 of 19 patients with idiopathic GH deficiency and empty sella syndrome, and a 43-kD membrane protein in pituitary and brain was identified as an autoantigen in another patient with idiopathic GH deficiency and in 1 of 14 patients with secondary GH deficiency. 21 These data suggest that the target antigen of APA is a protein growth factor. We mentioned more detail about Western blotting and showed the results in Fig. 3. The concordance between the positive APA findings with E L I S A and Western blotting were investigated. In normal subjects (n = 39), there was 100% concordance. All of the ELISA-nega-

tive subjects (n = 36) tested negative by Western blotting, and the other 3 subjects were both positive. Moreover, in patients with endocrine disorders, approximately 90% of the results were coincident. The E L I S A and Western blot analysis techniques described in this study were useful in detecting APAs in serum. APAs were detected in relatively few patients with certain pituitary disorders but were detected frequently in patients with autoimmune thyroiditis or IDDM. Results suggest a role for APA in autoimmune disorders involving the hypothalamus-pituitary-thyroid axis. REFERENCES

1. Sauter NP, Ton R, Mclaughlin CD, Dyess EM, Kritzman J, Lechan RM. Isolated ACTH deficiency associated an autoantibody to a corticotroph antigen that is not ACTH or other propiomelanocortin-derived peptides. J Clin Endocrinol Metab 1990;70:1391-7. 2. Ozawa Y, Shishiba Y. Recovery from lymphocytic hypophysitis associated with painless thyroiditis: clinical implications of circulating antipituitary antibodies. Acta Endocrinol (Copenh) 1993;128:493-8. 3. Komatsu M, Kondo T, Yamauchi K, et al. Antipituitary antibodies in patients with primary empty sella syndrome. J Clin Endocrinol Metab 1988;67:633-8. 4. Mau M, Phillips TM, Ratner RE. Presence of anti-pituitary hormone antibodies in patients with empty sella syndrome and pituitary tumors. Clin Endocrinol (Oxf) 1993;38:495-500. 5. Kobayashi I, Inukai T, Takahashi M, et al. Anterior pituitary cell antibodies detected in Hashimoto's thyroiditis and Graves' disease. Endocrinol Japan 1988;35:705-8. 6. Hansen BL, Hagedus L, Hansen GN, Hagen C, Hansen JM, Hoier MM. Pituitary-cell antibody diversity in sera from patients with untreated Graves' disease. Autoimmunity 1989;5:49-57. 7. Mirakian R, Cudworth AG, Bottazzo GF, Richardson CA, Doniach D. Autoimmunity to anterior pituitary cells and the pathogenesis of insulin-dependent diabetes mellitus. Lancet 1982;2:755-9. 8. Engwall E, Perlman R Enzyme-linked immunosorbent assay (ELISA): Quantitative assay of immunoglobulin G. Immunochemistry 1971;8:871-4. 9. Yabe S, Murakami H, Kobayashi I, et al. Western blot analysis of rat pituitary antigens recognized by human pituitary antibodies. Endocrine J 1995;42:115-9. 10. Laemmli UK. Cleavage of structaral proteins during the assembly of the head of bacteriophage T4. Nature 1970;227:680-5. 11. Bottazzo GF, Poupland A, Florin-Christensen A, Doniach D. Autoantibodies to prolactin-seereting cells of human pituitary. Lancet 1975;2:97-101. 12. Sugiura M, Hashimoto A, Shizawa M, et al. Heterogeneity of anterior pituitary cell antibodies detected in insulin dependent diabetes mellitus and adrenocorticotropic hormone deficiency. Diabetes Res 1986;3:111-4. 13. Schloot N, Eisenbarth GS. Isohormonal therapy of endocrine autoimmunity. Immunol Today 1995;16:289-93. 14. Kobayashi T, Yabe S, Kikuchi T, Kanda T, Kobayashi I. Presence of anti-pituitary antibodies and GAD antibodies in NIDDM and IDDM. Diabetes Care 1997;20:864-6. 15. Hashimoto K, Takao T, Makino S. Lymphocytic adernohy-

J Lab Clin M e d Volume 132, N u m b e r 1

pophysitis and lymphocytic infundibuloneurohypophysitis. Endocrine J 1997;44:1-10. 16. Onodera T, Toniolo A, Notkins AL. Virus-induced diabetes mellitus XX: Polyendocrinopathy and autoimmunity. J Exp Med 1981;153:1457-73. 17. Tartaglia LA, Demski M, Weng X, et al. Identification and expression cloning of a leptin receptor, OB-R. Cell 1995;83:1263-71. 18. Scott J. New chapter for the fat controller. Nature 1996; 379:113-4.

Y a b e et al.

31

19. Jeanrenaud FR, Jeanrenaud B. Obesity, leptin, and the brain. N Engl J Med 1996;334:324-5. 20. Majorie M, Terry M, Ranter PRE. Presence of anti-pituitary hormone antibodies in patients with empty sella syndrome and pituitary tumors. Clin Endoclinol 1993;38:495-500. 21. Crock P, Salvi M, Wall J, Guyda H; Detection of anti-pituitary autoantibodies by immunoblotting. J Immmunol Methods 1993;162:31-40.