Pediatric and Developmental Pathology

, Volume 7, Issue 5, pp 433–442

Maternal Smoking, Intrauterine Growth Restriction, and Placental Apoptosis

  • Christina Vogt Isaksen
  • Rigmor Austgulen
  • Lisa Chedwick
  • Pal Romundstad
  • Lars Vatten
  • Catherine Craven†
Original article

DOI: 10.1007/s10024-004-0105-1

Cite this article as:
Isaksen, C., Austgulen, R., Chedwick, L. et al. Pediatr Dev Pathol (2004) 7: 433. doi:10.1007/s10024-004-0105-1

Abstract

Pregnant women who smoke are at greater risk of delivering a growth-restricted infant than nonsmoking mothers. We wanted to see if apoptosis could be involved in the mechanisms behind smoke-induced growth restriction, and our aim was to compare apoptosis in the placenta of smoking mothers giving birth to growth-restricted infants and nonsmoking mothers with infants of appropriate weight. The project was conducted at the Magee—Womens Hospital and Magee—Womens Research Institute, University of Pittsburgh, PA. Histological sections from 20 placentas were selected from smoking mothers who had given birth to small-for-gestational-age infants (birth weight ≤ 2 SD). The controls were gestational-age matched nonsmoking mothers with infants having appropriate-for-gestational-age weight. The TUNEL method was used to demonstrate DNA fragmentation in nuclei, and a monoclonal antibody M30, specific for a neo-epitope on cytokeratin 18, was used to identify apoptotic epithelial cells. The positive nuclei (TUNEL) and positive cells (M30-positive cytoplasm) were counted blindly both in villous tissue and in decidual/basal plate tissue. M30-positive cells in villous tissues were significantly increased in placentas from smoking mothers compared to nonsmoking mothers. When evaluated by the TUNEL method, the difference between the two groups of women was not significant. Our study shows that apoptosis was increased in the placentas of smoking mothers with growth-restricted infants. The difference between the two groups was mainly in the syncytiotrophoblast layer and in connection with perivillous fibrin deposition. Cigarette smoke with reduction in blood flow has previously been shown to increase apoptosis, and it is possible that this could be one of the mechanisms playing a role in the growth restriction.

apoptosisgrowth restrictionplacentapregnancysmoking

INTRODUCTION

Pregnant women who smoke are at greater risk of giving birth to a growth-restricted infant, than nonsmoking women [1]. They present a higher rate of perinatal deaths, in addition to an increased risk of spontaneous abortion and preterm delivery [2,3]. Women who smoke are also at risk for placentae previa, abruptio placentae, and premature rupture of the membranes [1]. Babies of smokers weigh on the average 150–250 g less than nonsmokers’ babies. This smoking-related weight reduction is independent of other maternal and infant factors that influence birth weight [1]. The mechanisms leading to intrauterine growth restriction (IUGR) in infants of smoking mothers are not clear. Nicotine and cotinine, a metabolic product of nicotine, both cross the placenta, and a direct toxic action of the components of cigarette smoke on the fetus is one of the most accepted theories explaining the adverse effects of smoking, though not clearly understood [3]. Nicotine is also a potent vasoconstrictor and it is believed that smoking affects intrauterine vessels, with an ensuing constriction contributing to diminished blood flow to the placenta [4]. A direct correlation between fetal oxygen consumption and birth weight has been demonstrated in experimental studies [5]. The fetal growth restriction associated with maternal smoking might therefore be related to chronic hypoxia.

Hypoxia has been hypothesized to induce apoptosis [6]. At the same time, hypoxia has been shown to induce resistance to apoptosis in several cell types, and a significant decrease in neutrophil apoptosis under hypoxic conditions has been demonstrated [7]. Both nicotine and cotinine have been related to the inhibition of apoptosis in both normal and tumor cell cultures [8,9]. Maternal smoking also appears to reduce apoptosis in villous trophoblasts [3], though studies have also shown that nicotine has little effect on cytotrophoblast apoptosis [10].

The proliferation rate of cytotrophoblasts is inversely related to oxygen tension [11]. Since hypoxia has been suggested as the main pathogenic factor in smoking, it is remarkable that no increase in proliferation rate has been observed with maternal smoking [12]. In placentas from mothers who smoked during pregnancy, there is a decreased number of cytotrophoblasts. Recent research has demonstrated that nicotine has a significant negative impact on mitosis, reducing the number of cytotrophoblasts [10]. The placental transfer of amino acids is also impaired in smoking mothers, which may have implications for intrauterine growth restriction [13].

Extensive morphological changes have been observed in placentas of smoking mothers, and oxygenation and the passage of nutrients may be limited by these changes. Since oxygen tension is associated with apoptosis [14], it is reasonable to hypothesize an association between smoking, breaks in the trophoblast covering of the villi with deposition of fibrin, and placental apoptosis [15,16].

Few studies on maternal smoking and placental apoptosis have been conducted [3,10,17]. Some controversy exists as to the effect of smoking on apoptosis [3,9,10,17,18,19,20]. Maternal smoking and fetal growth has been extensively studied [1,2,21,22,23,24], though the correlation between intrauterine growth-retarded infants in smoking mothers and placental apoptosis has not been elucidated.

In order to see if intrauterine growth restriction caused by maternal smoking influenced placental apoptosis, we conducted a retrospective study comparing apoptosis in the placenta of smoking mothers giving birth to small-for-gestational-age (SGA) infants and nonsmoking mothers with infants of appropriate-for-gestational-age (AGA) weight.

METHODS

The project was conducted at the Magee—Womens Hospital and Magee—Womens Research Institute, University of Pittsburgh. The Institutional Review Board at the hospital approved the project.

Tissue specimens

Cases and controls were selected using the following criteria:

Cases

Twenty placentas from smoking mothers having given birth to an infant with SGA birth weight ≤ 2 SD, as assessed by standard weight percentiles [25].

Controls

Twenty placentas from nonsmoking mothers having given birth to an infant with AGA (10th–90th percentile) birth weight.

We could have used other control groups, for example, smokers with normal-weight infants or nonsmokers with growth-restricted infants. In the first category, placental changes are less likely to be expected, and in the second, other reasons for the growth restriction could confound the results.

Previous studies have shown that apoptosis increases towards term [26,27]. The controls and cases were therefore matched for gestational age.

Information was obtained by using the database at the Department of Pathology. All reports were from the period 1996 to 2000, and supplementary clinical information was pulled from the hospital records.

Placentas from women with preeclampsia, gestational hypertension, diabetes, infection, or any other serious illness were excluded from the study. Placentas from infants with serious or multiple developmental anomalies, and placentas with chorioamnionitis or extensive villitis, were also excluded. Clinical data on both groups are given in Table 1. The results of the morphological examination of placental tissue are shown in Table 2.
Table 1.

Clinical information

Smokers with SGA infants

Nonsmokers with AGA infants

No.

Parity

Clinical information

Smoking (n = cig.)

GA (wk)

Birth weight (g)

Parity

Clinical information

GA (wk)

Birth weight (g)

1

G1P0

Severe IUGR

20

33.4

1164

G5P3

Placenta previae

33.0

2412

2

G3P2

IUGR

20

38.0

1915

G3P2

Normal pregnancy

37.2

3108

3

G3P2

Asthma

10

41.0

2505

G3P2

Placenta previae

41.1

3833

4

G5P4

IUGR

+

36.0

1406

G3P1

PROM

36.0

2398

5

G2P1

Tetralogy of Fallot

20

37.0

2359

G3P2

Placenta previae

37.0

2844

6

G3P1

IUGR, Rh–

20

39.0

2727

G2P1

Normal pregnancy

39.0

3089

7

G1P0

IUGR, asthma

20

38.0

2130

G3P1

Placenta previae

37.1

3461

8

G3P2

IUGR, asthma

15

36.5

1743

G1P0

PROM

35.0

2395

9

G2P1

IUGR

20

38.6

2361

G2P0

Maternal fever

40.1

3120

10

G2P0

IUGR

8

38.0

2351

G2P0

Normal pregnancy

38.6

2951

11

G3P2

IUGR, Rh–, drug abuse

15

35.6

1728

G2P1

Fetal distress

35.0

2345

12

G1P0

PROM, fetal distress

20

29.1

668

G4P2

Cerclage PROM

28.6

1462

13

G1P0

Drug abuse

40

38.4

2228

G5P2

Normal pregnancy

39.0

3640

14

G2P0

IUGR

10

38.0

2183

G1P0

Normal pregnancy

38.0

3246

15

G2P1

IUGR

40

38.0

2138

G2P1

PROM

38.0

3510

16

G2P1

IUGR, fetal distress

15

35.0

1212

G2P1

Placenta previae

35.0

2276

17

G1P0

IUGR

+

37.3

2196

G2P1

Placenta previae

37.0

3012

18

G1P0

IUGR

10

37.0

2257

G1P0

Placenta previae

37.0

2979

19

G2P0

Severe IUGR

10

25.1

791

G2P0

Placenta previae

26.0

850

20

G2P0

Asthma

20

36.4

1626

G3P2

Placenta previae

36.4

3093

SGA, small-for-gestational-age; AGA, appropriate-for-gestational-age; G, gravida; P, para; cig., cigarettes; GA, gestational age; wk, weeks; PROM, premature rupture of the membranes; IUGR, intrauterine growth restriction.

Table 2.

Placental examination

Smokers with SGA infants

Nonsmokers with AGA infants

No.

Placental weight (g)

Weight percentile

Morphological findings

Placental weight (g)

Weight percentile

Morphological findings

1

200

<10

AVM, increased perivillous fibrin 20%, increased decidual fibrin

370

25–50

NVM, mild chorionic hemosiderosis

2

220

<10

NVM, increased perivillous fibrin <5%

470

50–75

NVM, normal morphology

3

400

25–50

NVM, subchorial intervillous thrombi, intervillous fibrin

290

<10

NVM, meconium macrophages

4

260

<10

NVM, meconium macrophages

340

10–25

NVM, partial necrosis of amnion

5

275

<10

NVM, villous sclerosis and hemorragic endovasculitis 15%

400

25–50

NVM, subchorionic fibrin, marginal blood clot

6

410

25–50

NVM, villous sclerosis, minimal chorioamnionitis

360

10–25

NVM, retroplacental hematoma <1%, villous sclerosis <5%

7

500

75–90

NVM, subchorionic fibrinous plaques

600

>90

NVM, flattened segment involving 30% of disc

8

290

<10

Multiple infarcts 30%, recent marginal intervillous thrombus

410

50–75

NVM, mild villous edema

9

280

<10

DVM, meconium macrophages

410

25–50

NVM, normal morphology

10

320

<10

NVM, subamniotic hemorrhage, subchorionic fibrinous plaques

400

25–50

NVM, meconium pigment accumulation

11

450

50–75

NVM, mural thrombus chorionic vein, focal chorangiosis <5%

360

10–25

AVM, meconium macrophages

12

145

<10

AVM, multiple infarcts 15%, decidual vasculopathy

235

10–15

NVM, intervillous hemorrhage 10%, chorangiosis

13

290

<10

NVM, normal morphology

490

75–90

NVM, two vessels in umbilical cord

14

460

50–75

NVM, normal morphology

460

50–75

NVM, meconium macrophages

15

440

50–75

NVM, mild perivillous fibrin depostion

500

75–90

NVM, normal morphology

16

250

<10

NVM, recent central infarcts <10%

340

10–25

NVM, subchorionic fibrinous plaques

17

320

<10

NVM, focal membranous edema

600

>90

NVM, circummarginate placenta

18

375

25–50

NVM, marginal perivillous fibrin, multiple infarcts <10%

370

25–50

Intervillous thrombi, marginal blood with focal abruption

19

150

<10

AVM, villous infarctions 50%, focal placental abruption

200

25–50

NVM, maternal blood clot with focal placental abruption

20

370

25–50

NVM, fetal thrombotic vasculopathy 25%

345

10–25

NVM, normal morphology

SGA, small-for-gestational-age; AGA, appropriate-for-gestational-age; AVM, advanced villous maturation; NVM, normal villous maturation; DVM, delayed villous maturation; %, percent of affected placental volume.

Two formalin-fixed and paraffin-embedded blocks were retrieved from each placenta, one from the periphery and one central. Five micron-thick sections were cut from each block. In order to evaluate apoptosis, both the TUNEL method and immunohistochemistry with a M30 antibody were performed.

TUNEL method

A DNA fragmentation kit was used for the TUNEL (terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate [dUTP] nick end labeling) staining (Oncogene Research Products, Boston, MA). The nick end labeling refers to staining of the DNA 3’ hydroxy groups generated as a result of endonuclease activity at linker DNA regions between nucleosomes. The commercially given instructions were followed. TUNEL-positive nuclei stained dark brown, and when they also looked pyknotic, they were regarded as apoptotic. Sections from human lymph node were included in all batches as positive controls. Sections from placental tissue, to which addition of enzyme was replaced by distilled water, served as negative controls.

Immunostaining with the monoclonal antibody M30

The monoclonal antibody (mab) M30 (Boehringer Mannheim, Mannheim, Germany), specific for a neo-epitope on cytokeratin 18 (CK18), was used to recognize apoptotic epithelial cells [28,29]. CK 18 is cleaved by caspases early in the apoptotic cascade, thus visualizing apoptosis before DNA fragmentation detectable by the TUNEL method. Immunostaining with M30 was performed as previously described [30]. Sections from an adenocarcinoma of colon were added as positive controls [31]. M30 positivity gives a cytoplasmic staining, most often homogeneous, but sometimes more granular. Cells with pyknotic nuclei and red staining of cytoplasm were regarded as apoptotic. The TUNEL method and the mab M30 immunostaining were performed on serial slides from each placenta, and all the slides were blinded. The coded slides were selected in such a way that the gestational age-matched cases and controls were included in the same batch.

Quantification of apoptosis

TUNEL-positive nuclei and M30-positive cells were counted blindly according to the criteria given above. Twenty fields of villous tissue at 200× magnification and 20 fields of maternal basal plate tissue at 400× magnification were counted on each slide, though on a number of slides there was not sufficient basal plate tissue to count 20 fields. Only fields with at least 80% tissue coverage were evaluated. The localization and origin of the positive cells were taken into consideration and categorized thereafter. When TUNEL-positive nuclei and M30-positive cells in all slides were counted, the code was broken. Cells from the peripheral and central sections were added to each other so that altogether 40 fields of villous tissue from each placenta were evaluated. With M30 immunohistochemistry, the number of fields in decidual/basal plate tissue varied from 4 to 36, with 18 as the median, with interquartile ranges (IQR) corresponding to 25–75% of 12.0–24.0. For the TUNEL method, the number of fields varied from 15 to 40, with 40 as the median, IQR 25–75% of 31.3–40. The number of M30- and TUNEL-positive cells per field was calculated, and the rate of apoptosis in villous and decidual/basal plate tissue is shown in Table 3.
Table 3.

Number of apoptotic cells per field in various placental tissues as assessed by mab M30

Tissue

Group

Mediana

25th

75th

P-value*

Villous tissue

Smokers

0.56

0.48

0.96

0.04

 

Controls

0.35

0.18

0.63

Syncytiotrophoblast layer

Smokers

0.13

0.08

0.20

0.50

 

Controls

0.10

0.05

0.15

Peri/intervillous fibrinoid

Smokers

0.38

0.16

0.73

0.17

 

Controls

0.19

0.08

0.44

Villous stroma

Smokers

0.03

0.01

0.05

0.49

 

Controls

0.04

0.01

0.07

Decidual/basal plate

Smokers

0.61

0.24

0.67

0.10

 

Controls

0.38

0.09

0.66

Anchoring villi

Smokers

0.10

0.04

0.22

0.42

 

Controls

0.10

0.00

0.20

Extravillous trophoblasts

Smokers

0.33

0.13

0.56

0.06

 

Controls

0.23

0.00

0.41

All tissues combined

Smokers

1.10

0.79

1.41

0.10

 

Controls

0.79

0.27

1.40

mab, monoclonal antibody.

aMedian number of apoptotic cells per field counted with 25th and 75th percentiles.

*P-value: Wilcoxon paired-rank correlation test of difference towards the control group.

Statistical analysis

It was calculated that to obtain a statistical power of 90%, study groups of 20 were needed. Sample size calculations were performed on the Number Cruncher Statistical Systems (NCSS) program (Hintze J [2001] Kaysville, UT, USA). The apoptotic rate showed a marked skewed distribution towards zero, therefore median group values were employed with corresponding IQR and application of nonparametric statistics. In order to evaluate differences in apoptotic rate between the matched groups, P-values were calculated by using Wilcoxon signed-rank test for paired data. For unmatched comparison, the Mann–Whitney U-test was applied. Statistical analyses were performed using Statistical Package for Social Services (SPSS) version 11 (SPSS for Windows, SPSS Inc., Chicago, IL, USA).

RESULTS

M30-positive cells showed either a homogenous or granular red cytoplasmic staining (Fig. 1A). Positive syncytiotrophoblasts were either located on the villous surface, as syncytial knots (Fig. 1B), or lying in connection with peri- and intervillous fibrin deposition (Fig. 1C). Like the M30 positivity, TUNEL-positive nuclei were found in the villous syncytiotrophoblast layer (Fig. 2A), in syncytial knots (Fig. 2B), and in connection with fibrin deposition (Fig. 2C).
https://static-content.springer.com/image/art%3A10.1007%2Fs10024-004-0105-1/MediaObjects/fig1.gif
Figure 1.

M30-positive cells. A. Homogenous or granular red cytoplasmic staining. B. Positive syncytiotrophoblasts. C. Positive cells in fibrin.

https://static-content.springer.com/image/art%3A10.1007%2Fs10024-004-0105-1/MediaObjects/fig2.gif
Figure 2.

TUNEL-positive nuclei. A. In villous syncytiotrophoblast layer. B. In syncytial knots. C. In connection with fibrin deposition.

Total cell number in fields of villous tissue (n = 5) was counted in order to calculate the percentage of apoptotic cells. The median number was 840, IQR: 816–833. If 840 represents the total cell number in the evaluated villous field, the total apoptotic rate in villous tissue of controls as evaluated by M30 was estimated as 0.04%, IQR: 0.02–0.07%.

Table 3 shows the number of apoptotic cells per field in various placental tissues as assessed by immunostaining with the M30 antibody. M30 positivity was significantly increased in villous tissue (including syncytiotrophoblasts and peri/intervillous fibrin) in placentas from smoking mothers with SGA infants compared to nonsmoking mothers with normal infants. Generally, the M30 count was higher in the smoking group compared to the control group, though except for total villous tissue, this difference was not significant. There was no significant difference between the two groups of women with regard to apoptosis as evaluated by the TUNEL method. This pertains to all types of tissue counted: syncytiotrophoblasts, fibrin, villous stroma, anchoring villi, and extravillous trophoblasts.

In decidual/basal plate tissue, M30-positive extravillous cytotrophoblasts and TUNEL-positive nuclei were detected. Since M30 positivity occurs in epithelial cells, it was not relevant to compare the decidual cells. There was an increased number of M30-positive cells in extravillous trophoblasts (P = 0.06) but not in anchoring villi in smokers as compared to controls (Table 3).

Morphological alterations in the placenta were more frequent in placentas of smoking mothers (Table 2). Advanced villous maturation was described in one case in the control group and in three cases in the smoking group. Peri/intervillous fibrin deposition or infarcts were found in 10 of the smoking cases, while in none of the control group. Five cases were described as largely normal in the smoking group, while 15 were described as normal in the control group.

DISCUSSION

Two major reasons for the intrauterine growth restriction in infants of smoking mothers have been proposed, namely toxic effects of nicotine and its metabolites, and a vasoconstrictor effect causing diminished blood flow to the placenta with fetal hypoxia [4]. The placenta retains substantial amounts of compounds from cigarette smoke, both in their native forms and in intracellular substrates [32,33]. Cotinine levels in plasma have been shown to correlate with uteroplacental vascular resistance estimated from blood flow velocity measures by Doppler [4,34].

What is then the link to apoptosis? Apoptosis, or programmed cell death, was first described by Kerr et al. [35]. It is a normal physiological form of cell death, that together with mitosis, has the function of controlling cell population [26,36]. The process may be influenced by different stimuli, both physiological and nonphysiological [37].

Nicotine has been related to the inhibition of apoptosis in cell cultures [8,9], while hypoxia has been shown to induce apoptosis [6,38,39]. Enhancement of apoptosis by hypoxia has also been demonstrated in cultured human trophoblasts, while addition of epidermal growth factor (EGF) to the cultures lowers the level of apoptosis [39], suggesting that EGF may protect against apoptosis induced by hypoxia.

Cigarette smoke has also been shown to increase apoptosis in the gastric mucosa of rats, by reducing gastric blood flow, thus decreasing serum epidermal growth factor [19,20]. Could the same mechanisms play a role in the uterus, with a decrease in serum EGF implicated in the intrauterine growth restriction? Other studies have not shown increased apoptosis in placentas of smoking mothers [3], however, these have not taken into account whether the infants were growth restricted or not.

Increased apoptosis has been found in placentas of pregnancies complicated by fetal growth restriction [40]. We also found a significant increase of apoptosis in villous tissue in the placentas of smoking mothers who delivered a growth-restricted infant. This increase could be explained by hypoxia accentuated by reduced blood flow with decrease in EGF. Since EGF is present in amniotic fluid [41], the IUGR found in children of smoking mothers might be a direct effect of decreases in EGF levels either in serum or amniotic fluid. Nicotine has also been shown to have an inhibitory effect on early trophoblast differentiation [10,42], and the increased apoptosis may, in part, be due to a direct toxic effect of the components of cigarette smoke [3].

Apoptosis, as evaluated by the TUNEL method, was not significantly increased as compared to controls. Completely normal controls for the various gestational ages were difficult to find, and altogether nine cases had placenta previae. Some of these placentas had higher TUNEL counts than M30, which we ascribe to the fact that TUNEL staining is not specific for apoptosis, but also stains necrotic cells [43]. It is therefore important to take into account false-positive TUNEL staining. With the M30 antibody, there was little nonspecific staining present, and we recommend the use of this antibody when studying apoptosis in CK18-positive epithelial cells.

In smoking mothers, the placental/fetal ratio is increased and morphological alterations have been demonstrated [42,44]. In our study, infarcts or peri/intervillous fibrin deposition were registered in 50% of cases in the smoking group, with none in the control group. Fibrinoid deposition during gestation occurs at sites of trophoblast discontinuity, it is nonrandom and influenced by the quality of vascular perfusion [16,45,46]. Smoking during pregnancy affects placental blood flow [47,48], therefore, the morphological findings in placentas of smoking mothers do not seem to be haphazard, but related to the effects of cigarette smoke.

During the first trimester, focal defects in floating villi with absence of cytotrophoblast stem cells have been demonstrated, together with an increase in the number of cytotrophoblast columns that fail to reach the uterus or degenerate in the intervillous space, implying a reduction in the number of anchoring villi [10]. A reduced number of Ki-67-positive cytotrophoblasts means that fewer cells are in the S-phase causing a premature depletion of cytotrophoblast stem cell population [10]. We can thus conclude that the cytotrophoblast population in placentas of smoking mothers is reduced. When apoptosis among these cytotrophoblasts is increased, this will further diminish blood flow and oxygenation potential over the fetal–maternal membrane, also decreasing EGF and escalating the nutritional defects to the fetus.

CONCLUSION

Our study shows that apoptosis in fetal villous tissue is increased in the placentas of smoking mothers with SGA infants compared to gestation-matched controls with AGA infants and mothers who do not smoke. Increased peri/intervillous fibrin deposition was found in the smoking group, but not in the control group. The role of hypoxia in the induction of apoptosis has previously been demonstrated. Apoptosis in syncytiotrophoblasts causes discontinuities in the trophoblast layer and deposition of fibrin-type fibrinoid, which further reduces factors implicated in growth.

ACKNOWLEDGMENTS

The authors thank Professor Steinar Tretli for his help concerning calculations of statistical power and study group size.

Grant support was provided by the Research Council of Norway.

Copyright information

© Society for Pediatric Pathology 2004

Authors and Affiliations

  • Christina Vogt Isaksen
    • 1
    • 2
  • Rigmor Austgulen
    • 3
  • Lisa Chedwick
    • 4
  • Pal Romundstad
    • 5
  • Lars Vatten
    • 5
  • Catherine Craven†
  1. 1.Department of Pathology and Medical Genetics/National Center for Fetal MedicineTrondheimNorway
  2. 2.Department of Laboratory MedicineChildren’s and Women’s Health, Faculty of MedicineTrondheimNorway
  3. 3.Institute of Cancer Research and Molecular BiologyNorwegian University of Science and TechnologyTrondheimNorway
  4. 4.Magee—Womens Research InstituteUniversity of PittsburghPittsburghUSA
  5. 5.Institute of Community Medicine and General PracticeNorwegian University of Science and TechnologyTrondheimNorway