MATHIAS JOHN,t HARTMUT HOPP, NORBERT. FUHR,. MARKUS VAN ...... Buttery, L D., McCarthy, A., Springall, D. R., Sullivan, M. H.. Elder, M. C.,. Michel, T.
Expression placenta
of inducible of women
nitric
oxide
synthase
with gestational
diabetes
GILBERT SCHONFELDER, MATHIAS JOHN,t HARTMUT HOPP, MARKUS VAN DER GIET,* and MARTIN PAUL” *
Department
of Clinical Pharmacology,t
Department
Gynecology, Benjamin Franklin University Hospital,
Gestational diabetes is one of the most prevalent medical complications of pregnancy and causes increased fetal wastage. Investigation of placentas from diabetic mothers indicate chronic disturbances in intervillous circulation, dilatation of capillaries, and a relatively immature villous structure. Abnormal levels of nitric oxide (NO) may contribute to maternal disorders such as the pathogenesis of diabetic vascular complications. In the normal placenta NO is generated only by endotheial NOS, which apparently serves to regulate vascular tone in the fetoplacental circulation. In contrast, studies have reported the absence of inducible nitric oxide synthase (iNOS) in human placentas under normal conditions. The aim of our study was to investigate whether iNOS is expressed in placentas from patients with gestational diabetes. Reverse transcription-polymerase chain reaction and Western blot analysis demonstrated iNOS mRNA and protein expression in placental tissue only from patients with gestational diabetes. Inununohistochemistry localized iNOS staining to endotheial cells and trophoblasts. We conclude that iNOS can be expressed in human placenta. Its expression might play an important role in placental pathophysiology.-Schonfelder, G.,John, M., Hopp, H., Fuhr, N., van der Giet, M., Paul, M. Expression of inducible nitric oxide synthase in placenta of women with gestational diabetes. F4SEBJ. 10,777-784 (1996)
ABSTRACT
Key Words: human
placenta
RT-PCR
GESTATIONAL DIABETES IS ONE OF THE MOST prevalent medical complications of pregnancy. Between 1 and 5% of pregnant women display abnormalities in glucose metabolism. In Germany, however, only 10% of these women are diagnosed with gestational diabetes by the current risk-screening measures (1). Gestational diabetes causes increased fetal wastage such as macrosomia, fetal retardation, edema, and multiple metabolic effects. Investigation of placentas from diabetic mothers demonstrate plethora, chorangiosis, edema, and a relatively immature villous structure (2-7), which lead to chronic disturbances in intervillous circulation and dilatation of capillanes (8, 9).
0892.6638/96/001
0-0775/$01 .50. © FASEB
in
of Gastroenterology,
NORBERT
and lDepartment
FUHR, of Obstetrics
and
Freie Universitht Berlin, 12200 Berlin, Germany
As a vasodilator, nitric oxide (NO)2 is involved in regulation of vascular tone. In the fetoplacental circulation, NO apparently serves to maintain basal tone as well to attenuate the action of vasoconstrictors (10). However, abnormal levels of NO may contribute to maternal disorders (11). Furthermore, increased blood flow and vascular permeability in early diabetes have been associated with increased NO generation via inducible nitric oxide synthase (iNOS; E.C. 1.14.13.39) (12). Characteristically, vasodilatation and increased blood flow are early vascular responses to acute hyperglycemia and increased insulin levels (12, 13). From previous studies, it has been hypothesized that the generation of NO by interleukin-13 (IL-1) -induced nitric oxide synthase could participate in cell damage in early insulin-dependent diabetes mellitus (type 1 diabetes) and in apoptosis (14-17). Corbett and McDaniel (18) demonstrated that the combinations of tumor necrosis factor (TNF) and lipopolysaccharide (LPS) as well as TNF and ‘y-interferoti (y-IFN) also induced the expression of iNOS in intact islets, but there did not appear to be a direct effect on or a cells. Furthermore, their findings provide the first evidence that a limited number of nonendocrine cells can release sufficient quantities of IL-i in islets to induce iNOS expression and inhibit the function of the cell, which is selectively destroyed during the development of autoimmune diabetes. Therefore, it is possible that NO may play a significant role in the pathogenesis of type 1 diabetes and its complications (19). The endothelial NOS (eNOS) has been characterized in human and rat placentas both in vivo and in vitro under physiological conditions. eNOS was localized to syncytiotrophoblasts and to the endothelium of stem villous yes-
ITO whom
correspondence
should
be addressed,
at: Department
of
Clinical Pharmacology, Universithtsklinikum Benjamin Franklin, Hindenburgdamm 30, 12200 Berlin, Germany. 2Abbreviations: IL-1f3, interleukin 1J3; NO, nitric oxide; iNOS, inducible nitric oxide synthase; eNOS, endothelial nitric oxide synthase; RT, reverse transcription; DEPC. diethyl pyrocarbonate; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; y-IFN, y-interferon; LPS, lipopolysaccharide; SDS, sodium dodecyl sulfate; type 1 diabetes, insulin-dependent diabetes mellitus (IDDM); TNF, tumor necrosis factor; BSA, bovine serum albumin; TBS, Tris-buffered saline; ‘VFBS, Tween-20 Iris-buffered saline; PBS, phosphate-buffered saline; PCR, polymerase chain reaction.
777
TABLE
1. Maternal
characteristics
Case Age
(years)
Week of gestation Height (cm) Prepregnancy weight (kg) Maternal weight at delivery (kg) Infant birth weight (g) Blood pressure at the time of delivery (mmHg) Blood glucose over 100 mg/dl Pathological 75 g oGTT’ HbAlc (%) Diet (30-35 kcal/kg ideal body weight)
Insulin treatment Infection at the time of delivery
IDDM
of patients
with
gestational
diabetes#{176}
ig
2g
3g
4g
Sg
#{244}g
7g
8g
26
28
29
30
33
34
29
33
37 167 66 75 4120
38 170 65 73 2700
38 164 61 86 4000
39 157 63
37 168 123 111 2740
39 157 56 70 3680
37 173 63 75 3080
40 162 86 97 4310
120/80 + + nd.
120/80 + + 3.2
140/80 + + 4.4
+ + 5.4
135/40 + + 4.3
110/80 + + nd.
110/70 + + nd.
120/80 + + 5.1
+ +
+
+
+ +
+ +
+
+
+ +
1
3
2
74
2370 100/70
+
Preeclatnpsia
+
+ +
Antihypertensive
Treatment (Dihydralazin) Total number of gestations
2
‘ml.: not determined. The threshold for HbAlc
1
1
was 6%, IDDM: insulin-dependent
sels (20-27) by immunohistochemistry, in situ hybridization, and determination of calcium-dependent NOS activity (28, 29). In contrast, studies have reported the absence of iNOS mRNA and enzyme activity in healthy human placentas (24, 29). In diseased placentas, however, the expression of iNOS has not been investigated in detail. Morphological and biochemical studies have been conducted in an attempt to describe the pathology and the variations in human placenta. However, little is known regarding the molecular basis of the development and evolution of the pathology or about the variation of hemodynamic parameters and contractility of smooth muscle cells. The cellular interactions within placental vessels in the fetomaternal circulation under diabetic conditions, therefore, remain unclear. Because NO may participate in counterregulation of chorangiosis, proliferation of syncytiotrophoblasts, and dilatation of capillaries, the aim of our study was to investigate whether iNOS is expressed in human placenta of patients with gestational diabetes.
METHODS Cestational diabetes was diagnosed by routine blood glucose sampling during outpatient care throughout the pregnancy by using hemo-cue b-glticose system, which uses the glucose-dehydrogenase method in a microcuvette system. For blood glucose values higher than 120 mgldl, a 75 g oral glucose tolerance test (oGT’F) was performed for evaluation of gestational diabetes according to the National Diabetes Data Group criteria (30). Test results in which two or more values were elevated were considered abnormal. Patients received a comprehensive education program, instructions for daily multiple self-monitoring of capillary whole-blood glucose levels, and an individual dietary regimen with 30-35 kcallkg ideal body weight in combination with or without insulin treatment (Table 1). As soon as the patient was hospitalized, diurnal
77R
May
199h
2
Tho
FASFR
3
diabetes niellitus.
and nocturnal measurement of fasting, pre-, and postprandial blood glucose levels was performed. Patients were treated to attain fasting blood glucose levels between 3.3 and 4.4 mmol/l and a 2 h postprandial value between 5.6 and 6.6 mmol/l. If blood glucose levels could not be maintained normoglycemic by diet, appropriate insulin treatment was administered. After elective caesarean section or spontaneous uncomplicated vaginal delivery, placentas from healthy women (n 10) and from patients with gestational diabetes (n8) were obtained (Table 1 and Table 2) from the Department of Obstetrics of the Benjamin Franklin Hospital, rinsed three times with 0.9% saline, and immediately frozen in liquid nitrogen. Placental tissue was microscopically and macroscopically investigated by a pathologist, and placentas with signs of inflammation were excluded from investigation. Tissue collection was performed following the standard guidelines set by the ethics committee of the Benjamin Franklin University Hospital, Freie Universii#{228}t Berlin.
Isolation of total RNA and reverse-transcription-polymchain reaction (RT-PCR)
erase
Total cellular RNA was isolated from placental tissues by using TRIZOL Reagent (Gibco BRL, Eggenstein, Germany) according to the manufacturer’s instructions. To prevent DNA contamination, RNA samples were exhaustively
digested
with
DNase
I (Boehnnger,
Mannheim,
Germany).
RNA integrity was checked by gel electrophoresis in a 2% agarose gel with subsequent ethidium bromide staining. Iota! RNA concentration was estimated by spectrophotometty. For RT-PCR analysis, first-strand cDNA was synthesized from 1 JIg of total RNA in a 20 j.tl reaction solution containing 4 l.tl 5x RT-buffer (Gibco BRL), 2.5 pM random hexamer primers, 0.5 mM dNTPs, 8.4 J.tl diethyl pyrocarbonate (DEPC)treated distilled water, 2.5 mM dithiothreitol, 4 U RNasin (Promega, Serva, Heidelberg, Germany), and 200 U MMLV-Reverse Transcriptase (Gibco BRL) at 21#{176}C for 10 mm, followed by 37#{176}C for 1 h and 94#{176}C for 5 mm. Control PCR assays were performed to check cDNA amount, as well as RT efficiency and cDNA quality, using specific primers to human glyceraldehyde-3-phosphate dehydrogenase (GAPDH), 5’GACCCC1TCATTGACCTCAA-3’ (sense; position: 163-182), and 5’-
CATGGACTGTGGTCATGAGC-3’
(antisense;
position:
579-598).
Reactions were carried out in a total volume of 50 .il containing 5 il of each RT sample, 5 j.tl lOx buffer (Gibco BRL), 2 mM MgCI2, 0.4 mM dNTP, 0.1 JtM of each specific primer (Table 3), DEPC-treated water, and 2.5 U Taq polymerase (Gibco BRL). The PCR temperature profile
Ie,iirnal
er”m.iccm
rco
rr
Al
TABLE 2. Maternal
characteristics
of healthy
Case Age (years) Week of gestation Height (cm) Prepregnancy weight (kg) Maternal weight at delivery (kg) Infant birth weight (g) Blood pressure at the time of delivery (mmHg) Blood glucose over 100 mg/dl Infection at the time of delivery Total number of gestations
women
1 34 38 172 61 68 3230
3 32 41 161 86 88
4 25 39 167 45 57
3230
3250
130/80 3
2 30 39 172 65 76 2880
120/80 -
130/80 2
-
1
7
8
9
10
3700
6 40 37 14.6 61 68 3140
30 41 171 63 75 4150
37 41 170 63 77 4530
25 41 172 75 81 3800
23 41 170 67 78 3970
130/80
110/80
110/70
120/80
120/70
-
-
-
-
-
130/80 -
115/80 -
3
1
-
-
-
-
5
2
consisted of 26 cycles at 94#{176}C for 30 s, 60#{176}C for 1 mm, and 72#{176}C for 30 s, followed by an additional 6 mm extension period at 72#{176}C. PCR for iNOS was carried out in a total volume of 50 JIl containing 5 J.il of RT in 50 JJJ lOx buffer (Gibco BRL), 1.5 mM MgCI2, 0.4 mM dNTPs, 0.1
using
bovine
serum
albumin
(BSA)
as a reference
JIm thick) of placental tissue from healthy women and patients with gestational diabetes were cut at -25#{176}C with a Reichert Jung ciyostat (Heidelberg, Germany), mounted onto chrome alum-gelatinized glass slides, air-dried for 1 h, pretreated with acetone for 10 mm at -25#{176}C, and, finally, air-dried for 10 mm. Section-mounted slides were rinsed twice in phosphate-buffered saline (PBS), pH 7.6, at 37#{176}C and incubated for 1 h at room temperature with IgG2a-monoclonal antibodies against iNOS (AFFINITI Research) at a concentration of 2.5-10 JIg/mI or for 3 h with polyclonal antibodies against iNOS (BlOMOL Feinchemikalien, Hamburg, Germany) at a 1:400 dilution carried out in PBS. Sections were rinsed twice more for 10 mm and immunoreactivity was visualized using TRITC-labeled swine anti-rabbit lgC (1:40; DAKO Diagnoslika GmhH, Hamburg, Germany). For screening of tissue macrophages. section-mounted slides were double-incubated for 1 h at room temperature with monoclonal antibodies against human macrophages (clone: Ber-MAC3; DAKO Diagnostika GmbH) at a 1:30 dilution and with polyclonal antibodies against iNOS (BIOMOL Feinchemikalien) at a 1:400 dilution. Sections were rinsed twice for 10 mm and reactivity was demonstrated using TRITC-labeled swine antirabbit IgG (1:40; DAKO Diagnostika GmbH) and FITC-conjugated affinipure goat-anti-mouse lgG (1:20; Jackson LmmunoResearch Laboratories Inc., West Grove, Pa.). Control reactions for immunostaining were carried out following the same procedures as for immunoincubation, substituting PBS or normal mouse serum for monoclonal iNOS- and anti-macrophages (clone: BerMAC3) antibodies. Tissue sections from human tonsils were used as a positive control for the monoclonal antibodies against human macrophages (clone: Ber-MAC3).
stand-
ard.
SDS-PAGE
and Western
blotting
Total cellular
placental homogenates containing 10 Jig protein as well and LPS-activated mouse macrophages as positive controls were diluted (1:4) in sodium dodecyl sulfate (SDS) sample buffer according to Laemmli (32) (0.5 M Tris-HCI pH 6.8, 10% (v/v) glycerol, 10% (w/v) SDS, 5% (v/v) (-mercaptoethanol, 0.05% (w/v) bromphenol blue) and heated at 95#{176}C for 4 mm. Products were separated by SDS-PAGE using 10% gels and then electroblotted to nitrocellulose membranes using a wet blot transfer method at constant voltage as lysate
from y-IFN-
TABLE 3. INOS expression at mRNA and p rotein levels in placent
Case iNOS,PCR iNOS,protemn
as
2
Cryostat sections (4
preparation
(31)
1
Immunohistochemistry
Placental tissue samples were thawed in 3 ml ice-cold washing buffer (50 mM Tris, 2 mM EDTA, 150 mM NaC1, 0.5 mM D’VF, pH 7.5) per 1 g tissue wet weight and homogenized on ice in 2.5 ml homogenization buffer (50 mM Tris pH 7.4, 0.2 mM EDTA, 0.5 mM Dl’!’, 1 j.tM pepstatin A, 100 JIM phenylmethylsulfonyl fluoride, 2 jiM leupeptmn) per 1 g tissue wet weight using a Potter-Elvehjem Teflon glass tissue homogenizer (10 strokes; B. Braun, Melsungen, Germany). All subsequent procedures were performed at 4#{176}C. Homogenates were centrifuged at 15,000 x g for 30 mm. The resulting supernatants were retained for further experiments. Protein concentrations were detennined in triplicate according to Bradford
1
settings (100 V. 1 h). The quality of protein blots was determined by Ponceau S staining of nitrocellulose. BSA (5% w/v) -blocked nitrocellulose membranes (1 h, room temperature) were washed twice in 0.1% Tween-2O Tris-buffered saline (1TBS), pH 7.5, for 5 mm, rinsed twice for 5 mm in Iris-buffered saline (TBS), pH 7.5, and subsequently incubated overnight with lgG2a-monoclonal antibody against iNOS (AFFINETI Research, Nottingham, U.K.) at a 1:500 dilution. After washing in TTBS and TBS, blots were incubated for 1 h with anti-mouse IgGconjugated alkaline-phosphatase antibody at a 1:5000 dilution (Dakopatts, Hamburg, Germany). Blots were finally rinsed with TTBS and TBS, and immune complexes were visualized by staining for alkalinephosphatase activity.
tM of each specific primer, DEPC-treated water, and 2.5 U Taq polymerase (Gibco BRL). The PCR temperature profile for iNOS cDNA consisted of 33 cycles at 94#{176}C for 30 s, 61#{176}C for 1 mm, and 72#{176}C for 30 s, followed by an additional 6 mm extension period at 72#{176}C using the following primers: 5’-GCAGAATGTGACCATCATGG-3’ (sense; position: 1490-1509) and 5’-ACAACCTT’GGIGTI’GAAGGC-3’ (antisense; position: 1915-18)6). Control reactions for RT and PCR were performed by using water instead of mRNA or cDNA in reaction mixtures. The presence and size of the obtained PCR products were analyzed on ethidium bromide-stained agarose gels (2%). The specificity of the obtained iNOS PCR product was checked by digesting the amplificate with two specific restriction enzymes (Stu I and Ava I).
Protein
5 20 38 168 59 73
of h ealthy women a nd patient
1 -
2
3
4
Healthy 5 6
7
8
9
10
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
ig + +
s with
2g + +
gestat
3g + +
ional diabetes
a
Cestational diabetes 6g 4g Sg + + + + + +
7g + +
8g + +
‘Numbers correspond to Tables 1 and 2.
‘NOS EXPRESSION IN PLACENTAS DURING
GESTATIONAL
DIABETES
779
RESULTS RT-PCR
kDa
A
2
1
3
4
A
B
5
205
analysis
RT-PCR
analysis revealed expression of iNOS mRNA in placental tissue obtained from patients with gestational diabetes (lanes 2-5; Fig. lA). In contrast, iNOS mRNA was not detectable in placentas from healthy women (lanes 6-9; Fig. 1A). Figure 1A represents a typical analysis of PCR products by gel electrophoresis, resulting in bands of the predicted size of 426 bp in lanes 2-5 (gestational diabetes) but not in lanes 6-9 (controls). Therefore, iNOS mRNA appears to be expressed exclusively in placentas of patients with gestational diabetes. Restriction analysis of the iNOS PCR product by Stu I or Ava I revealed fragments of the predicted size of 200 and 226 bp for Stu I, and 321 and 105 bp for Ava I. Thus, the identity of the expected iNOS fragment was confirmed. RT-PCR analysis of GAPDH mRNA expression in pla-
116 97 66
45
B bp
1353 1078 872 603
310 281,271 234
1
2
3
4
5
6
7
8
9
10
Figure 2. A) Western blot analysis of iNOS from placentas of healthy women (lanes 1-4) and a patient with gestational diabetes (lane 5). INOS protein expression was found only in the placenta of a patient with gestational diabetes (lane 5). In lane 2 we could detect a protein band of 159 kDa, perhaps corresponding to unspecific staining of an unknown protein. B) Western blot analysis of protein from placenta of patient with gestational diabetes of the predicted size of 130 ±1 kDa (lane A) and y-IFN- and LPS-treated mouse macrophages (lane B).
1 Figure
2
3
1. A) Representative
PCR using
primer
against
iNOS
4
5
6
gel electrophoresis cDNA.
Lane
7 of products
1: X174/l-IAE
8
9
from RIIII digested
DNA size marker. Lanes 2-5 show RT-PCR analysis of iNOS mRNA expression in placental tissue from patients with gestations! diabetes; lanes 6-9, of healthy women. mNOS mnRNA expression was detected only in placentas from patients with gestational diabetes (lanes 2-5). B) RT-PCR analysis of GAPDH mnRNA expression in placental tissue from patients with gestational diabetes (lanes 2-5) and from healthy women (lanes 6-9). Lane 1: X174/HAE Ill-digested DNA size marker. PCR analysis revealed single bands of the predicted size of 435 bp with nearly identical intensity.
780
May
1996
cental tissue from patients with gestational diabetes (lanes 2-5, Fig. 1B) and healthy women (lanes 6-9; Fig. 1B) resulted in single bands of the predicted size of 435 bp with nearly identical intensity. Amplification of genomic DNA at the expected size of 1029 bp was not detected in either RT sample. Control reactions revealed no cDNA or genomic DNA contamination in RT or PCR (data not shown). RT-generated cDNA was not contaminated with genomic DNA. Therefore, the possibility of
The FASEB Journal
SCHONFELDER
EF AL.
gle iNOS protein bands of identical mass of 130 ±1 kDa (Fig. 2B).
apparent
molecular
Immunohistochemistry Immunostaining for human macrophages (clone: BerMAC3) (33) was performed with cryosectioris of placental tissue and revealed no immunoreactivity in placental tissue in either group of women, even though immunoreactivity could be detected in activated macrophages of the human tonsils as positive control reactions. Immunostaining for iNOS was performed with cryosections of placental tissue from healthy women (Fig. 3) and from patients with gestational diabetes (Fig. 4) using two different antibodies against iNOS. Both antibodies revealed identical cell-specific immunoreactivity, but polyclonal antibodies demonstrated more intense immunoreactivity. Polyclonal and monoclonal antibodies against iNOS revealed no immunoreactivity in the placental tissue of healthy women (Fig. 3). In contrast, immunoreactivity was localized to placental villi of patients with gestational diabetes with intense immunoreactivity in trophoblasts of placental stem villous (Fig. 4; using polyclonal antibody against
Figure 3.
iNOS immunoreactivity in placentas of healthy women. No immunoreaction against iNOS was detected in trophoblasts () (A: intervillous
space;
400x).
false positive iNOS mRNA detection from genomic DNA can be excluded. Furthermore, compared to similar levels of the housekeeping gene GAPDH, the levels of iNOS gene expression appeared to vary among placentas from patients with gestational diabetes (Fig. 1A, B). Western
blot analysis
The presence of iNOS protein expression in patients with gestational diabetes was investigated by Western blot analysis. A representative Western blot analysis of protein from placentas of healthy women (lanes 1-3) and from one patient with gestational diabetes (lane 4) is shown in Fig. 2A using monoclonal antibodies against iNOS. A single band at the predicted molecular mass of 130 ±1 kDa was detected only in protein samples from patients with gestational diabetes, but not from healthy women (Table 3). In healthy women we could only detect a single band of 159 kDa in lane 2 of Fig. 2A, perhaps corresponding to unspecific staining of an unknown protein. Furthermore, Western blot analysis of placental protein from patients with gestational diabetes and control of yIFN- and LPS-treated mouse macrophages revealed sin-
iNOS EXPRESSION IN PLACENTAS DURING
GESTATIONAL
DIABETES
Figure 4. iNOS immunoreactivity in placental tissue from patient with gestational diabetes. Intense iNOS inlmunofluorescence was shown in trophoblasts (A) of the stem villus from human placenta villus in patient with gestational diabetes (l000x).
781
ble for NO generation in the human placental villous tree (20-27). Measurement of NOS enzyme activity in human primordial placenta has demonstrated only eNOS activity (29). The major finding in the present study is that human placental villi appear to express the calcium-independent iNOS under pathophysiological conditions (Table 3). In agreement with our results, previous studies have failed to show a signal for iNOS gene expression either by in situ hybridization or by PCR analysis in human placentas under physiological conditions (22, 24). In contrast to findings in the normal placenta, we showed iNOS expression in placental tissue of patients with gestational diabetes at the mRNA level by RT-PCR, and at the protein level by Western blotting and immunohistochemistry. The antibody used for iNOS detection by Western blot analysis in our study detected a single band of 130 ±1 kDa in placental proteins of patients with gestalional diabetes. Therefore, there was no detectable cross-reactivity with eNOS, which would have yielded a band corresponding to 135 kDa in protein samples from women of both groups. Because other studies failed to demonstrate neuronal NOS expression (bNOS) on mRNA and protein level in
Figure
5. Representive
staining
for iNOS immunoreactivity
in placental
dmabees. iNOS immunoreactivity was detected only in trophoblasts (A) of human placental villi from patient with gestatmonal diabetes (: mntervillous space). No staining of cells in placental stroma was found by double incubation using antibodies against iNOS and human macrophages, clone: Ber-MAC3 (400x). tissue
from
patient
with
gestational
iNOS). Neither placental stroma nor Hofbauer cells stained for iNOS by double incubation with antibodies against iNOS and human macrophages (Fig. 5; using polyclonal antibodies against iNOS and monoclonal antibodies against human macrophages, clone: Ber-MAC3). Furthermore, iNOS immunoreactivity was observed in the endothelium of fetal blood vessels in the region of the chorionic plate (Fig. 6; polyclonal antibody against iNOS). No immunofluorescence was detected in control reactions (data not shown).
DISCUSSION Because the human placenta lacks local autonomic vascular control, circulating vasoactive factors are essential for the hemodynamic control of placental function. For example, NO appears to play an important role in maintaining blood pressure in the human placental villous vascular tree (10, 34). Previous studies have shown that different eNOS-containing cellular sources are responsi-
782
May
1996
6. iNOS immunoreactivity
Figure
gestational diabetes. close
The FASEB Journal
to
the
chorionic
in placental
Endothelium (A)
of
tissue
fetal blood
from
vessels
plate showed iNOS immunoreactivity
patient
with
in stem villus (l000x).
SCHONFELDER
Er AL.
human placentas of healthy women (21, 22), we can exclude a cross-reactivity with other NOS isoforms. Thus, the single band corresponding to a 159 kDa protein in lane 2 of Fig. 2A, may represent unspecific staining of an unknown protein. To localize iNOS protein expression at a cellular level, we applied immunocytochemistry. In our mRNA studies, we were able to exclude competition between the endogenous iNOS RT amplificate and the genomic iNOS amplificate in cDNA by using oligonucleotides for PCR spanning introns in genomic DNA and using Dnase I digestion. Intense immunofluorescence was seen in the syncytiotrophoblast of placental stem villi, as well of the intermediate placental villi and the endothelium of fetal blood vessels, using monoclonal and polyclonal antibodies against iNOS. We detected iNOS at the molecular and cellular level in the placental tissue of patients with gestational diabetes. The expression of iNOS was not due to expression in human macrophages, because immunohistochemistry by double incubation using a polyclonal antibody against iNOS and a monoclonal antibody against human macrophages (33), clone: Ber-MAC3, excluded mononuclear cell infiltraUon in tissue sections of the present study. Thus, we can exclude any contribution to our results of iNOS mRNA expressed by stimulated macrophages. Furthermore, cells in placental stroma did not stain for iNOS. The expression of iNOS in human macrophages has been a matter of debate, and mRNA expression has been observed in human macrophages only after heat shock treatment or stimulation with LPSIIFN-y, or after coculture with certain human tumor cell lines (35, 36). iNOS Protein expression was not detected in any of these studies. Just recently, Moncada reported for the first time iNOS gene expression and enzyme activity in human macrophages in vitro after stimulation with interleukin-4 and CD23 receptor stimulation with antibodies (oral report; XIIIth International Congress of Nephrology 1995, Madrid, Spain, July 2-6, 1995). Previous studies have shown that IL-13 induced NO generation by iNOS is an important mediator of 3 cell damage in early type 1 diabetes and an activator of programmed cell death (14-18). Thus, the contribution of IL-13 to iNOS gene and protein expression during gestational diabetes is of interest. The expression and secretion of IL-13 in cytotrophoblasts of human placenta has been reported (37-39). It is postulated that IL-13 modulates placental development, and we attempted to investigate whether iNOS expression was dependent on IL-1 expression. In current studies, no correlation was found between IL-13 mRNA and iNOS mRNA expression (unpublished data). However, it still remains for further studies to clarify whether the expression of IL-1-protein leads to the expression of iNOS in placentas during gestational diabetes. In future studies we will investigate whether there is any correlation of the differentiated iNOS expression patterns with placental pathology, week of gestation, plasma glucose or insulin levels, and treat-
INOS
EXPRESSION
IN PLACENTAS
DURING
GESTATIONAL
DIABETES
ment of patients. Our findings of iNOS expression in the fetomaternal compartment of the placenta may be of interest in understanding the regulation of blood pressure under pathophysiological conditions. The syncytiotrophoblast cells layer the intervillous space of human placenta and might regulate the intervillous blood flow by secretion of vasoactive substances. In addition, vasoactive substances secreted by the endothelium are, despite of lack of autonomic innervation of the placenta, important for the regulation of placental blood flow. During gestational diabetes, chronic disturbances in intervillous circulation and dilatation of capillaries (8, 9) are occurring in the placenta. We suggest that these disturbances might be caused by increased NO release by iNOS. Furthermore, because it has been established that NO generation by iNOS has cytotoxic effects and can contribute to programmed cell death, we speculate that iNOS may participate in the regulation against chorangiosis and relative immature villous structure. In contrast to the findings of Conrad, Garvey, and Kukor (22, 24, 29), who observed that only eNOS is expressed in human placenta and that it contributes to the control of blood flow in the placenta circulation, we suggest that in pathophysiological conditions iNOS expression can be induced in human placenta and might play an important role in pathological influence on blood flow and cellular defense. The authors wish to thank Drs. Julie Yu, Olivier Costerousse, and Thorsten Raff for critical discussions. The study was supported by a grant from the Deutsche Forschungsgemeinschaft to Martin Paul.
REFERENCES 1.
Hopp, H., Vollert, W., Ragosch, V., Pritze, W., Ehert, A., Entezami. M., and Weitzel, H. (995) Prevention of neonatal risk by general screening for diabetes in pregnancy, intensive diagnosis and subsequent therapy. Geburtsh. Frauenheilkd.
2.
3. 4. 5. 6.
7.
8.
9. 10.
11.
55,28-31
Thomsen, K., and Lieschke, C. (1958)
Untersuchungen zur Placentamorphologie bei Diabetes mellitus. Acta En&,crinol. Copenh. 29,602-614 Burstein, R., Berns, A. W., Hirata, Y., and Blumenthal, H. 1. (196.3) A comparative histo- and immunopathological study of the placenta in diabetes metlitus and in emytliroblastosis fetal is. Am. J. Obstet. Gynecol. 86, 66-76 Vogel, M. (1967) Plakopathia diabetics. VirchowsArch. (Pathol. Anat.) 343, 51-63 Fox, H. (1969) Pathology of the placenta in maternal diabetes mellitus. Obste:. Gynecol. 34, 792-798 Haust, M. D. (1981) Maternal diabetes mellitus-Effects on the fetus and placenta. In Perinatal Disea.ses (Naye, R. L., Kissane, J. M., and Kaufman, N., eds) pp. 201-285, Williams & Wilkins, Baltimore Benirschke, K., and Kaufman, P. (1990) Diabetes during pregnancy. In Pathology of the Human Placenta (Benirschke, K.. and Kaufman, P., eds) 2nd Ed, pp. 475-479, Springer-Verlag. New York Emmrich, P., and MSlzer, C. (1968) Zur Morphologic der Plazenta bei Uberiragung. Pat ho!. etIicrobiol. 32, 285-302 Mohr. H. (1950) Zur Histologic dci- ubertragenen Plazenta. Zbl. Gynoek. 72, 1530-1535 Myatt, L., Brewer, A.. and Brockman. D. E. (1991) The action of nitric oxide in the perfused human fetal-placental circulation. Am. J. Olaiet. Gynecol. 164,687-692
Nagy, C. (1993) Late complications of gestational diabetes-maternal effects. Gynaekol. 115, 450-453 Sharma, K., Danoff,T. M., DePiero, A., and Ziyadeh, F. N. (1995) Enhanced expression of inducible nitric oxide synmhase in murine macrophages and glomerular mesangial cells by elevated glucose levels: possible mediation via protein kinase C. Biochem. Biophys. Res. Comniun. 207, 80-88 Bat-on. A. D. (1994) Hemodynamic actions of insulin. Am.). Physiol. 267, E187-E202
Zentralbl.
12.
13.
783
14.
15.
16.
17.
18.
19.
20.
21.
Lukic, M. L., Stosic-Grujicic, S., Ostojic, N., Chan, W. L., and Liew, F. Y. (1991) Inhibition of nitric oxide generation affects the induction of diabetes by streptozocin in mice. Biochem. Biophys. Res. Commun. 178,913-920 KoIb, H., Kiesel, U., Km-oncke,K. D., and Kolb-Bachofen, V. (1991) Suppression of low dose streptozotocin induced diabetes in mice by administration of a nitric oxide synthase inhibitor. Ljfe Sci. 49, PL213-PL217 Amikarcrona, M., Dypbukt, J. M., Brune, B., and Nicotera, P. (1994) Interleukin-1 beta-induced nitric oxide production activates apoptosis in pancreatic RINmSF cells. Exp. Cell Res. 213, 172-177 Cetkovic-Cvrlje, M., and Eizink, D. L. (1994) TNF-alpha and IFN-gamma potentiate the deleterious effects of IL-I beta on mouse pancreatic islets mainly via generation of nitric oxide. Cytokine 6,399-406 Cot-ben, J. A., and McDaniel, M. L. (1995) Intraislet release of interleukin 1 inhibits cell function by inducing cell expression of inducible nitric oxide synthase. J. Exp. Med. 181, 559-568 Lindsay, R. M., Smith, W., Rossiter, S. P., McIntyre, M. A., Williams, B. C. and Baird, J. D. (1995) N-omega-nitm-L-arginine methyl ester reduces the incidence of IDDM in BB/E rats. Diabetes 44, 365-368 Eis, A. 1., Brockman, D. E., Pollock, J. S., and Myatt, L. (1995) Immunohistochemical localization of endothelial nitric oxide synthase in human villous and extravillous trophoblast populations and expression during syncytiotrophoblast formation in vitro. Placenta 16, 113-126 Myalt, L., Brockman, D. F., Eis, A. L., and Pollock, J. S. (1993) Immunohistochemical localization of nitric oxide synthase in the human placenta. Placenta
22.
23.
24.
25.
26.
784
28.
29.
1996
Kristoffersen, F. K., and Ulvestad, E. (1994) Nitric oxide synthase in human placenta. Acta Pat/to!. Microbiol. Immunol. Scand. 102, 509-513 Myatt, L, Brockman, D. E., Langdon, G., and Pollock, J. S. (1993) Constitutive calcium-dependent isoform of nitric oxide synthase in the human placental villous vascular tree. Placenta 14, 373-383 Kukor, Z., and Toth, M. (1994) Ca(2 + )-dependent and Ca(2+ )-independent NO-synthesizing activities of human primordial placenta. Acta Physiol. Hung.
82,313-319
30.
Classification and diagnosis of diabetes mellitus and other categories of glucose intolerance. National Diabetes Data Croup (1979) Diabetes 28,
31.
Bradford, M. M. (1976) A rapid and sensitive method for quantitation of microgramm quantities of protein utilizing the principle of protein-dye
1039-1057
binding.
32. 33.
Anal. Bloc/tern.
35.
72,248-254
Laemmli, U. K. (1970) Cleavage of structural proteins during assembly of the head of bacteriophage 14. Nature (London) 227, 680-685 Holcberg, C., Kossenjans, W., Brewer, A., Miodovnik, M., and Myatt, L. (1995) The action of two natriuretic peptides (atrial natriuretic peptide and brain natriuretic peptide) in the human placental vasculature. Am.). Obstet. Gynecol.
34.
14,487-495
Conrad, K. P., Vill, M., McGuire, P.C., Dail, W. G., and Davis, A. K. (1993) Expression of nitric oxide synthase by syncytiotrophoblast in human placental villi. FASEB). 7, 1269-1276 Buttery, L D., McCarthy, A., Springall, D. R., Sullivan, M. H.. Elder, M. C., Michel, T., and Polak, J. M. (1994) Endothelial nitric oxide synthase in the human placenta: regional distribution and proposed regulatory role at the feto-maternal interface. Placenta 15, 257-265 Garvey,E.P.,TuttIe,J.V.,Covington,K.,Merrill,B.M.,Wood,E.R.,Baylis, S. A., and Charles, I. C. (1994) Purification and characterization of the constitutive nitric oxide synthase from human placenta. Arch. Biochem. Biophys. 311,235-241 Dikranian, K., Trosheva, M., Nikolov, S., and Bodin, P. (1994) Nitric oxide synthase (NOS) in the human umbilical cord vessels. An immunohistochemical study. Acta Histochem. 96, 145-153 Springall. D. R., Riveros-Moreno, V., Buttery, L, Subum, A., Bishop, A. E., Merrett, NI., Moncada, S., and Polak, J. M. (1992) Immunological detection of nitric oxide synthase(s) in human tissues using heterologous antibodies suggesting different isofonns. Hi tochemistry 98, 259-266
May
27.
172,71-77
Backe, E., Schwarting, R., Genies, J., Ernst, M., and Stein, H. (1991) Ber-MAC3: new monoclonal antibody that defines human monocytelmacrophage differentiation antigen.). Clin. Pat/mo!. 44,936-945 Kroncke, K. D., Fehsel, K., Alsdorif, K., and KoIb-Bachofen, V. (1994) Expression of inducible NO-synthase iii human monocytes. Immunbiology 19,267-268
36.
Reiling, N., Ulmer, A. J., Duchrow, M., Ernst, M., Had, H. D., and Hauschildt, S. (1994) Nitric oxide synthase: mRNA expression of different isoforms in human monocytes/macrophages. Eur. ). Immunol. 24,
37.
Hu, X. L, Yang, Y., and Hunt, J. S. (1992) Differential distribution of interleukin-1 alpha and interleukin-1 beta proteins in human placentas.). Reprod. Immunol. 22, 257-268 Harty, J. R., and Kauma, S. W. (1992) lnterleukin-1 bets stimulates colonystimulating factor-i production in placental villous core mesenchymal cells. ). Clin. End ocrinol. Metab. 75,947-950 Librach, C. L., Feigenbaum, S. L., Bass, K. E., Cui, 1. Y., Verastas, N., Sadovsky, Y.. Quigley, J. P.. French, D. L., and Fisher. S. J. (1994) Interleukin-1 beta regulates human cytotrophoblast metalloproteinase activity and invasion in vitro.). Rio!. C/tern. 269, 17125-17131
1941-1944
38.
39.
The FASEB Journal
Received for publication November 20. 1995. Accepted for pub! ication February 6. 1996.
SCHONFELDER
Er AL.
