Volume 81, Issue 4 pp. 494-500
ORIGINAL ARTICLE
Full Access

Calcium antagonists, diltiazem and nifedipine, protect broilers against low temperature-induced pulmonary hypertension and pulmonary vascular remodeling

Ying YANG

Ying YANG

State Key Laboratory of Animal Nutrition, College of Animal Science and Technology, and

These authors contributed equally to this work.

Search for more papers by this author
Mingyu GAO

Mingyu GAO

Department of Basic Veterinary Medicine, College of Veterinary Medicine, China Agricultural University (CAU), Beijing, China

These authors contributed equally to this work.

Search for more papers by this author
Yuming GUO

Yuming GUO

State Key Laboratory of Animal Nutrition, College of Animal Science and Technology, and

Search for more papers by this author
Jian QIAO

Corresponding Author

Jian QIAO

Department of Basic Veterinary Medicine, College of Veterinary Medicine, China Agricultural University (CAU), Beijing, China

Jian Qiao, China Agricultural University, Beijing, 100193, China. (Email: [email protected]) Search for more papers by this author
First published: 21 July 2010
Citations: 10

ABSTRACT

This study was designed to determine whether calcium antagonists, diltiazem and nifedipine, can depress low temperature-induced pulmonary hypertension (PH) in broilers (also known as ascites) and to characterize their efficacy on hemodynamics and pulmonary artery function. Chicks were randomly allocated into six experimental groups and orally administered with vehicle, 5.0 mg/kg body weight (BW)/12 h nifedipine or 15.0 mg/kg BW/12 h diltiazem from 16 to 43 days of age under low temperature. The mean pulmonary arterial pressure (mPAP), the ascites heart index (AHI), the erythrocyte packed cell volume (PCV) and the relative percentage of medial pulmonary artery thickness were examined on days 29, 36 and 43. The data showed that administration of diltiazem protected broilers from low temperature-induced pulmonary hypertension and vascular remodeling. Although nifedipine prevented mPAP from increasing during the early stage, it did not suppress the development of PH during the late stage and did not keep heart rate (HR), PCV, AHI and the thickness of pulmonary small artery smooth muscle layer at the normal levels. Taken together, our results showed that diltiazem can effectively prevent low temperature-induced pulmonary hypertension in broilers with fewer side-effects and may be a potential compound for the prevention of this disease in poultry industry.

INTRODUCTION

Ascites syndrome (also known as pulmonary hypertension syndrome), one of the common end points of a number of pathogenies, is a metabolic disease in broilers characterized by pulmonary hypertension (PH)-induced right ventricular hypertrophy and failure (Huchzermeyer & De Ruyck 1986; Wideman et al. 1995; Currie 1999). These pathogenies may be due to a high basal metabolic rate originating from intensive genetic selection of broilers for rapid growth, long-term exposure to extreme environmental conditions and nutritional levels, such as low temperature, high altitude or high energy diets, which lead to an increased requirement for oxygen at tissue and cell levels. The resultant hypoxia in turn triggers an increased cardiac output and a consequential elevation in pulmonary arterial pressure, leading to right ventricular hypertrophy and pressure overload in the right ventricle, and sequentially results in right ventricular failure, liver congestion and ultimate ascites (Currie 1999). Regardless of the etiology, prolonged pulmonary vasoconstriction has been suggested to be an important pathogenetic factor in all forms of PH. Therefore, pulmonary vasodilation is a major strategy for the treatment of this disease, and vasodilator drugs have been proposed as an alternative or an adjunct approach to oxygen supplementation to attenuate the PH and prevent ascites in broilers for years (Yang et al. 2005, 2007). Evidence indicated that alteration of calcium homeostasis contributed to the development and perpetuation of PH through destroying the balance between vasodilating and vasoconstrictive forces, and proliferative and anti-proliferative forces (Enkvetchakul et al. 1995; Wideman et al. 1995; Currie 1999).

Application of calcium antagonists alone, or in combination with other agents is one of the effective ways to understand whether the calcium signaling is involved in these pathophysiological phenomena, such as various types of hypertension and cardiovascular diseases in mammals including humans. Attention was initially drawn to calcium antagonists because of their vasodilating properties, which might prevent vascular smooth muscle cells, cardiac muscle cells and conductive tissues from proliferating by inhibiting their trans-membrane influx of calcium ions (Cubeddu et al. 1986; Armas-Padilla et al. 2000). Calcium antagonists have been extensively investigated and are being used in treatment for hypertension, angina, and certain cardiac diseases in humans; however, the pharmacological reversal of hypoxic PH with a calcium antagonist has not been comprehensively evaluated in poultry affected with chronic pulmonary arterial hypertension.

In this study, nifedipine and diltiazem, two L-type Ca2+ channel antagonists were used in chicks to understand the actions of calcium signaling in the development of pulmonary arterial hypertension and to characterize their preventive efficacy on hemodynamics and pulmonary artery function of broilers with pulmonary hypertension syndrome induced by low temperature.

MATERIALS AND METHODS

Animal preparation and treatment

One-day-old male broiler chicks (Arbor Acre) of a commercial strain were maintained in an environmental chamber with continuous lighting at temperature of 28–30°C with free access to water and a commercial type broiler starter diet (23% crude protein, metabolizable energy = 13.4 megajoule/kg from 1 to 15 days of age) and then a grower diet (20% crude protein, metabolizable energy = 13.4 megajoule/kg from 16 to 43 days of age). At 16 days of age, birds were randomly allocated into six experimental groups, including one control group and one placebo group at normal temperature (28°C), four groups in low temperature (15–18°C) treated with placebo, slow-release nifedipine (5.0 mg/kg body weight (BW)/12h), slow-release diltiazem (diltiazem hydrochloride, 15.0 mg/kg BW/12 h) or without treatment. Birds were orally administered with nifedipine or diltiazem from 16 to 43 days of age, and equal volumes of placebo were orally given to broilers at the same time-interval in two placebo groups. Birds were weighed every other day to determine the doses of the two calcium antagonists to be administered. Birds for time-interval sampling were randomly selected from each group on days 29, 36 and 43. All animal protocols were approved by the Institutional Animal Care and Use Committee of the China Agricultural University.

Measurement of pulmonary arterial pressure and femoral arterial pressure

At 29, 36 and 43 days of age, heart rate (HR, beat/min, bpm), pulmonary arterial systolic pressure (mmHg) and diastolic pressure (mmHg) of broilers were measured using a right cardiac catheter based on a modified method (Guthrie et al. 1987), and simultaneously, another polyethylene plastic catheter was placed in the femoral artery to monitor systemic pressure just before sacrificing. Pressure signals were monitored by a host computer of Biopac Systems (Biopac Inc., Goleta, CA, USA) through sensors. The sensors were placed at the same level as the birds' hearts.

Measurement of erythrocyte packed cell volume and right ventricular hypertrophy

After the in vivo measurements were completed, erythrocyte packed cell volume (PCV, %) and the ascites heart index (AHI, %), ratio of the weight of right ventricle to the weight of the whole ventricle, were measured as previously reported (Yang et al. 2007).

Determination of pulmonary artery remodeling

Pulmonary artery remodeling was determined as previously described (Yang et al. 2007). Briefly, after transparency treatment by dimethylbenzene, the lung tissue segments with the thickness of 0.5 cm adjacent to the bronchi were embedded in paraffin and routinely processed into sections of 5 µm in depth followed by Weigert–van Gieson staining for elastin. Small pulmonary arterioles with external diameters in the ranges of 20–50, 50–100 and 100–200 µm were studied using an automatic image analyzer (BH2; Olympus, Tokyo, Japan) with the advanced software (Motic 3.0: Motic China Group Co., LTD, Xiamen, China). Ten average regions of cross-section were chosen. The adventitia and the lumen diameter were measured, following which the relative median thickness (%) and relative median area (%) were recorded and analyzed. The relative median thickness and the relative median area of pulmonary arterioles with different cut angles and conditions of either contractile or relaxation were computed from the above measurements according to the methods of Barth (Barth et al. 1993) and Wang (Wang et al. 2001). For each bird, 5–10 pulmonary median muscularized arteries with external diameters ranging from 20 to 50, 50–100 µm and from 100 to 200 µm were measured. The analysis was performed in a blind fashion by two experienced investigators.

Statistical analysis

Comparisons between groups on days 29, 36 or 43 were performed using one-way analysis of variance (ANOVA) followed by the Duncan test. Differences were considered statistically significant at the level of P < 0.05, and values are means ± SD The statistical analysis was performed with the software of SPSS 11.0 for Windows (SPSS Inc, Chicago, IL, US).

RESULTS

Femoral arterial pressure, pulmonary arterial pressure and heart rate

Group means for femoral arterial pressure (FAP) and pulmonary arterial pressure (PAP) are shown in Table 1. Three-week low temperature treatment had no significant effect on FAP, but it significantly increased PAP of broilers during the later stage of the experiment compared with that of control (Pall < 0.05). The HR of untreated broilers at low temperature was higher than the control birds on days 29 and 36 (Pall < 0.05); however, it was significantly lower than that of control on day 43 (P < 0.05).

Table 1. Mean femoral artery pressure (mFAP), mean pulmonary artery pressure (mPAP), heart rate (HR), erythrocyte packed cell volume (PCV) and ascites heart index (AHI) of broilers in different groups and investigated intervals, respectively. Values are expressed as mean ± SD
Mean arterial pressure = Diastolic pressure + 1/3 (Systolic pressure – Diastolic pressure)
Treatment Normal temperature Low temperature
Item Age (day) Control (n = 20) Placebo (n = 20) Control (n = 20) Placebo (n = 20) Nifedipine (5mg/kg/day) (n = 20) Diltiazem (15mg/kg/day) (n = 20)
mFAP (mmHg) 29 124 ± 16.3 127 ± 11.4 128 ± 16.6 126 ± 16.8 122 ± 14.7 124 ± 12.5
36 129 ± 12.2a 131 ± 9.1a 134 ± 15.0a 131 ± 12.3a 113 ± 12.3b 127 ± 11.7a
43 134 ± 10.4a 131 ± 11.4a 134 ± 12.9a 132 ± 10.1a 107 ± 27.9b 128 ± 9.7a
mPAP (mmHg) 29 24.6 ± 3.89ab 24.1 ± 3.61a 26.1 ± 4.05ab 26.6 ± 3.04b 24.8 ± 3.29ab 24.9 ± 3.04ab
36 26.6 ± 2.55a 26.7 ± 2.87a 35.0 ± 3.05b 34.1 ± 3.37b 26.1 ± 2.85a 26.4 ± 3.17a
43 29.1 ± 4.00a 30.6 ± 3.06a 38.0 ± 3.16b 36.9 ± 3.38b 34.0 ± 4.09c 29.0 ± 2.94a
HR (beat/min) 29 390 ± 17.9a 389 ± 14.3a 402 ± 15.3b 404 ± 15.5bc 412 ± 16.3c 394 ± 16.5ab
36 395 ± 25.5a 397 ± 14.0a 420 ± 15.2b 417 ± 14.8b 413 ± 15.5b 400 ± 13.1a
43 400 ± 12.8a 399 ± 12.7a 382 ± 18.0b 387 ± 15.6b 416 ± 16.0c 394 ± 22.4a
PCV (%) 29 35.1 ± 2.40a 34.8 ± 4.49a 39.8 ± 3.33b 39.2 ± 4.14b 33.8 ± 3.73a 34.4 ± 3.59a
36 32.3 ± 3.51a 31.6 ± 3.36a 40.8 ± 4.53b 38.8 ± 3.86bc 36.7 ± 3.77c 33.4 ± 3.46a
43 34.9 ± 4.21a 34.7 ± 3.91a 42.5 ± 5.47b 41.7 ± 5.88b 38.1 ± 4.00c 35.3 ± 5.90ac
AHI (%) 29 21.0 ± 3.59 21.4 ± 2.96 22.5 ± 3.33 22.9 ± 2.85 22.5 ± 3.57 22.6 ± 2.84
36 21.4 ± 3.17a 22.3 ± 2.67a 28.6 ± 3.50b 27.2 ± 3.47b 24.4 ± 2.54c 22.4 ± 2.95a
43 23.4 ± 3.78a 24.4 ± 4.36a 37.7 ± 4.24b 38.9 ± 3.46b 37.4 ± 6.42b 24.8 ± 5.71a
  • a,b,c means values within the same row with the different letters in their superscripts differs significantly (P < 0.05).

After 3 weeks of oral administration of nifedipine, mean PAPs on days 29 and 36 were at the same level with those of control, but was significantly higher than that of control on day 43 (P < 0.05). Mean FAPs were significantly lower than those in the control (P < 0.05) and HR of broilers was constantly higher than those of the control (P < 0.05). Although mean FAP tended to be lower in the diltiazem group compared with that of the control on day 43, no significant difference was observed. Meanwhile, mean PAP in the diltiazem treatment group was kept at the same level as that of the control. Diltiazem did not significantly affect HR throughout the experiment.

Packed cell volume and ascites heart index

Low temperature significantly increased the PCV and AHI of broilers compared with control, especially on day 43 (Pall < 0.05) (Table 1). Both nifedipine and diltiazem maintain the PCV at the same level as that of controls at normal temperature on days 29 and 36. However, on day 43 the PCV and AHI of broilers treated with nifedipine were significantly higher than those in controls at normal temperature, while those in the diltiazem-treated group showed no statistical difference compared with that of controls at normal temperature.

Pulmonary artery remodeling

The relative median thickness and relative median area of pulmonary arterioles are shown in Table 2. There was no difference in external diameters of pulmonary arteries between groups, either ranging 20 ≤Φ≤ 50, 50 ≤Φ≤ 100 µm or 100 ≤Φ≤ 200 µm on days 29, 36 and 43. The relative median thicknesses and relative median areas were significantly greater in the two non-pharmaceutically treated groups at low temperature than that at normal temperature on days 36 and 43, respectively (Pall < 0.05) (Table 2, Fig. 1A,C). No significant difference on relative median thickness and relative median area was observed between diltiazem-treated birds and that of controls, whereas the two parameters were significantly greater in nifedipine-treated groups at low temperature than at normal temperature within the three ranges on days 36 and 43, respectively (Pall < 0.05).

Table 2. Relative medial thickness (RMT) and relative medial area (RMA) of broilers with pulmonary arteriol external diameter ranging from 20 to 50, 50–100 µm and 100–200 µm in the different groups and investigated intervals. Values are mean of 15 birds with 5–10 arteriols each and expressed as mean ± SD
Treatment Normal temperature Low temperature
Item Φ(µm) Age (day) Control (n = 15) Placebo (n = 15) Control (n = 15) Placebo (n = 15) Nifedipine (5mg/kg/day) (n = 15) Diltiazem (15mg/kg/day) (n = 15)
RMT (%) 20∼50 29 30.6 ± 4.23 31.5 ± 4.17 34.6 ± 4.46 35.5 ± 2.51 31.5 ± 2.47 29.5 ± 3.27
36 33.3 ± 3.92a 32.7 ± 3.69a 38.5 ± 2.59b 38.3 ± 3.79b 35.2 ± 4.31ab 32.8 ± 3.83a
43 32.5 ± 4.08a 31.9 ± 4.36a 40.3 ± 4.26b 41.2 ± 5.15b 37.6 ± 3.84c 33.5 ± 4.21a
50∼100 29 24.4 ± 3.02 24.9 ± 2.13 26.7 ± 2.98 27.6 ± 2.41 24.8 ± 1.90 24.1 ± 1.74
36 25.7 ± 2.69a 26.2 ± 1.88a 30.0 ± 2.87b 30.7 ± 3.02b 30.8 ± 2.64b 25.4 ± 2.13a
43 26.9 ± 2.95a 27.1 ± 2.98a 32.2 ± 4.11b 33.7 ± 3.85b 31.9 ± 3.08b 25.9 ± 1.89a
100∼200 29 18.8 ± 1.42 19.1 ± 1.76 20.5 ± 1.57 21.7 ± 3.11 21.7 ± 3.28 18.9 ± 1.61
36 19.1 ± 2.48a 19.4 ± 2.11a 23.4 ± 3.65b 22.6 ± 3.14ab 23.7 ± 2.28b 20.4 ± 3.95a
43 19.7 ± 1.91a 20.5 ± 2.33ac 24.6 ± 2.58b 23.1 ± 2.08b 24.5 ± 3.04bc 20.8 ± 1.79a
RMA (%) 20∼50 29 49.2 ± 5.62 48.1 ± 4.48 53.7 ± 6.43 55.6 ± 4.95 52.4 ± 6.45 50.5 ± 4.57
36 50.4 ± 3.25a 49.4 ± 3.57a 56.6 ± 5.38b 57.3 ± 6.36b 53.6 ± 4.31a 48.8 ± 4.58a
43 48.7 ± 3.67a 50.3 ± 4.33a 57.7 ± 6.25b 57.6 ± 4.69b 56.6 ± 5.34b 48.6 ± 3.36a
50∼100 29 51.6 ± 4.95 52.3 ± 5.32 57.6 ± 5.67 56.4 ± 5.14 55.1 ± 4.78 50.5 ± 4.69a
36 54.8 ± 5.11a 54.1 ± 5.32a 66.1 ± 5.71b 66.5 ± 6.28b 64.9 ± 5.74b 53.9 ± 4.61a
43 53.7 ± 4.61a 54.3 ± 5.14a 66.7 ± 4.85b 67.5 ± 5.29b 64.1 ± 5.66b 52.8 ± 4.47a
100∼200 29 35.1 ± 2.78 36.3 ± 2.29 38.7 ± 3.19 39.5 ± 3.86 39.1 ± 3.07 35.4 ± 2.32
36 35.4 ± 2.66a 36.4 ± 2.35a 44.8 ± 2.96b 43.9 ± 3.46b 40.5 ± 3.73b 34.2 ± 3.08a
43 36.8 ± 3.21a 37.5 ± 2.18a 46.4 ± 3.56b 47.7 ± 4.69b 41.6 ± 3.79b 34.6 ± 3.38a
  • a,b,c means values within the same row with the different letters in their superscripts differs significantly (P < 0.05).
Details are in the caption following the image

Microphotograph of pulmonary artery structure of broilers with pulmonary arteriol external diameter ranging from 20 to 50 µm (Weight & Van Gieson ×100). (A) Section of a pulmonary arteriol of a broiler at 43 days old at normal temperature. (B) Section of a pulmonary arteriol of a broiler treated with placebo for 28 days (at 43 days of age) in normal temperature. (C) Section of a pulmonary arteriol of a broiler at 43 days old at low temperature. (D) Section of a pulmonary arteriol of a broiler treated with placebo for 28 days (at 43 days old) in low temperature. (E) Section of a pulmonary arteriol of a broiler orally treated with nifedipine (5 mg/kg body weight (BW)/12h) for 28 days (at 43 days old) at low temperature. (F) Section of a pulmonary arteriol of a broiler orally treated with diltiazem (15 mg/kg BW/12h) for 28 days (at 43 days old) at low temperature.

DISCUSSION

The method of low temperature-induced pulmonary hypertension symptoms in broilers has been confirmed (Wideman et al. 1995). As expected, in this study, low temperature did cause the development of the pathologic process, which is represented by PH, increased PCV and right ventricular hypertrophy, as well as the thickening of the pulmonary small artery smooth muscle layer (Enkvetchakul et al. 1995; Wideman et al. 1995). Under a low temperature environment, broilers' demands for oxygen were physiologically increased to produce more energy and heat to maintain body temperature at a normal level, leading to a relative hypoxia status. Hypoxia has been confirmed as an important pathophysiological stimulus for erythrocyte number increment accompanied by PH (Maxwell et al. 1987). Additionally, with long-term hypoxia and the development of PH, right ventricular compensatory hypertrophy forms, which explains the increase in AHI (Wideman & Tackett 2000).

The main finding of the study is that, except for a slight decrease in femoral arterial pressure by nifedipine on day 43, no difference was observed in other parameters between diltiazem-treated broilers (15.0 mg/kg BW/12 h) and those of controls throughout the experiment. This result showed that diltiazem prevented PH and PCV from increasing, induced lower temperature, and arrested right ventricular hypertrophy and pulmonary arterial vascular remodeling. Although nifedipine prevented pulmonary arterial pressure from increasing during the early stage of the experiment (29 and 36 days old), it did not suppress the development of PH and pulmonary artery remodeling during the later stage (43 days old), and did not keep HR, PCV and AHI at the same levels as those of controls. It should be noted that nifedipine and diltiazem, as well as verapamil which was used in our previous study (Yang et al. 2007), all belong to L-type Ca2+ channel blockers. They display differences in preventing low temperature-induced pulmonary arterial hypertension in broilers. Diltiazem and verapamil attenuate pulmonary hypertension, whereas the nifedipine has little effect on pulmonary hypertension and displays some side-effects on the growth and performance of broilers. This is in consistent with the observations obtained from humans, and suggests that this may be due to the differences in molecular mechanisms. Further study is needed to explain these observations.

The predominant viewpoint on pulmonary hypertension are based on the principle that hypoxia inhibits an outward potassium current, leading to membrane depolarization and calcium entry through the voltage-dependent calcium channels (Post et al. 1992; Hong et al. 2004). The elevation of intracellular Ca2+ concentration ([Ca2+]i) in pulmonary small artery smooth muscle cells activates the contractile system in the smooth muscle, causing cell contraction and proliferation, consequently resulting in the thickening of the pulmonary small artery smooth muscle layer and eventually PH (Platoshyn et al. 2000; Wamhoff et al. 2006). It has been proven that the voltage-dependent L-type Ca2+ channel antagonists could inhibit the rise of [Ca2+]i. Two decades after their first introduction in treatment for primary PH (Camerini et al. 1980; Kambara et al. 1981), diltiazem and nifedipine have been reported to be effective in various experimental models and human clinics (Gassner et al. 1990; Antezana et al. 1998; Islam et al. 1999; Cockrill et al. 2001).

In the present study, oral nifedipine with 5.0 mg/kg BW/12 h increased the broilers' HR and depressed PH at the same time. However, in our preliminary experiment, several doses of nifedipine were tested and the results showed that those doses lower than 5.0 mg/kg BW/12 h did not prevent PH, while doses higher than 5.0 mg/kg BW/12 h prevented PH, but it depressed FAP and boosted HR greatly. In addition, we found that nifedipine could not arrest the pulmonary artery smooth muscle remodeling at several doses (2.5, 5.0, 8.0, 10.0 mg/kg BW/12 h, data not shown). These data indicated that although nifedipine may be a useful agent to reduce pulmonary resistance in PH, this effect was less pronounced on PAP than FAP in broilers. Several other studies also demonstrated that the effects of nifedipine were often partial, temporary and sometimes complicated by changes in cardiac output (Clozel et al. 1987; Agostoni et al. 1989; Davidson et al. 1989). It is suggested that nifedipine may be particularly useful in low temperature-induced PH in broilers when vasoconstriction is the main determinant factor of PH as in primary PH in humans, and nifedipine cannot arrest the PH caused by pulmonary artery wall remodeling developed in the later stage (Rich & Brundage 1987).

Diltiazem was reported to reduce pulmonary arterial pressure in recurrent PH associated with pulmonary hypoplasia (Islam et al. 1999) and two clinical studies had also shown a reduction of PH in patients after diltiazem administration (Kambara et al. 1981; Crevey et al. 1982). Our results on diltiazem confirmed that the appropriate dose of diltiazem might depress PH in broilers, maintain HR at normal levels (no HR reduction), arrest right ventricular hypertrophy (expressed by AHI) and prevent pulmonary artery wall remodeling in the later stages (Fig. 1C). No mortality or serious morbidity was associated with the drug testing. Moreover, the results also provided another beneficial action of diltiazem in preventing PH syndrome, which is represented by a higher erythrocyte packed cell volume and right ventricular hypertrophy (Julian 1993). In this case, diltiazem is preferable to nifedipine in preventing low temperature-induced PH in broilers as it has no systemic side-effects, and is more selective for the pulmonary vasculature than for the systemic one.

The mechanism of calcium antagonists preventing PH in broilers is the same as in mammals, that is, calcium antagonists block the intracellular signal pathway, which interdicts the activation of the contractile action on pulmonary vessels, keeping normal pulmonary function (Mohanty et al. 1987; Wang et al. 2005). Therefore, calcium antagonists prevent PH and avoid increasing erythrocyte production pathologically. The preventive effects of L-type Ca2+ channel antagonists suggest that calcium signals are associated with PH in broilers, which leads to the development of ascites. Calcium antagonists, such as diltiazem, when administered at appropriate doses, may cause substantial reductions in pulmonary arterial pressure and prevent the occurrence of pulmonary arterial hypertension syndrome in broilers.

ACKNOWLEDGMENTS

This work was supported by Scientific Research Start-up Foundation of China Agricultural University (NO. 2006001), Ministry of Sci & Tech., P. R. China (NO. 2006BAD12B07) and National Natural Science foundation of China (NO. 30700576).

      The full text of this article hosted at iucr.org is unavailable due to technical difficulties.