Small intestinal biopsies are frequently taken as part of the diagnostic workup in children investigated for diarrhea, abdominal pain or failure to thrive. Routinely, these biopsies are sent for histology only. However, mucosal morphology does not invariably reflect mucosal function, and it is thus also valuable to extract information regarding the transport function of the recovered tissue. In several previous articles, we have described a miniaturized Ussing chamber in which transmucosal potential difference (Pd), and total tissue resistance (Rt) can be measured. With this setup, it is also possible to investigate the secretory response by stimulating the biopsies with known secretagogues such as prostaglandins and acetylcholine (1,2).
In intestinal diseases, mucosal damage is often associated with increased permeability. In the absence of overt mucosal damage (cell destruction), epithelial permeability is mainly maintained by the tight junctions, complex structures forming junctions between the cells. It was formerly believed that the tight junctions were just organized “pores” between the cells, but it is now clear that they are made up of complex proteins containing regulatory elements (3). The standard way of measuring intestinal permeability in adults is to make the patient ingest a permeability marker, which is then recovered in the urine. Depending on the size of the marker, one can draw conclusions not only about overall permeability changes but also estimate “apparent pore size.”
In the pediatric setting, this type of setup is not very useful, and there is a need for alternative techniques. Because taking intestinal biopsy samples is routinely done in the diagnostic workup of failure to thrive and diarrhea of unknown origin, it would be of value if permeability data could be extracted from the recovered tissue. The aim of the present paper was to evaluate measurement of paracellular electrical resistance as an indirect marker for solute permeability. Because it has previously been shown that intestinal permeability is increased in children with partial or total atrophy due to celiac disease (4–6), we used a patient material of this type to test our hypothesis. Paracellular electrical resistance (Rp) was measured by square wave current analysis, a technique based on the epithelial layer behaving as an electrical capacitance in parallel with an electrical resistance (ie, the tight junctions) (7). Rp was compared in controls and patients with partial and subtotal villous atrophy. To characterize the underlying epithelial transport mechanisms, we also included a short series of patients with cystic fibrosis (CF), a disease due to dysfunction of the chloride channel cystic fibrosis transmembrane conductance regulator (8), the channel normally responsible for active chloride secretion.
PATIENTS AND METHODS
Intestinal biopsies were obtained from 66 consecutive patients with CD in 3 different phases: acute disease, after a period of gluten-free diet or after gluten challenge. Twenty-one biopsies showing partial villous atrophy were obtained from patients with a median age of 41 months (range, 14–98 months). Seventeen biopsies showed subtotal villous atrophy, and the median age in this group of patients was 17 months (range, 9–44 months). Twenty-eight biopsies were obtained from children with CD in clinical remission (median age, 30 months; range, 17–114 months). They had been on a gluten-free diet for 12.5 months (range, 11.5–18 months). All of the biopsies in the latter group showed normal morphology. After gluten challenge, all of these children were later shown to have CD according to the European Society for Pediatric Gastroenterology, Hepatology and Nutrition criteria (9).
Biopsies with normal morphology were obtained from 78 children, median age 34 months (range, 9–184 months), and were used as controls. These patients were recruited as relatives to patients with CD or investigated due to failure to thrive, diarrhea, vomiting, constipation or abdominal pain. All of the patients had mild symptoms, and none had malnutrition. Two of the children had lactose intolerance, and all of the others were subsequently found to be healthy.
Five CF patients, ages 1 month to 27 years, were also included; 2 were homozygous for Δ F508, and 3 were compound heterozygotes, the other mutations being 3659 del C, 994 del 9 and 1249-5A→6, respectively. Biopsies from 4 of these patients showed normal morphology. In 1 patient, no biopsy was analyzed due to lack of material.
The biopsies were performed using a Watson pediatric biopsy capsule under fluoroscopic control and after sedation intravenously with midazolam 0.1 to 0.2 mg/kg body weight. Metoclopramide 0.3 mg/kg body weight was given intravenously to facilitate passage of the capsule. One half of the biopsy was used for routine histological examination. The biopsies were classified in 3 groups: normal morphology, partial villous atrophy or subtotal villous atrophy. Subtotal villous atrophy was defined as completely flat mucosa with no visible villi. Partial villous atrophy included some detectable villi with reduced height. Typical crypt hyperplasia and intraepithelial lymphocytes were seen in all biopsies with mucosal atrophy.
The results of the electrophysiological examinations in some of these patients have been reported previously (1,2,10).
The study was approved by the Ethics Committee of the Faculty of Medicine at Göteborg University, and informed consent was obtained from the parents.
The biopsy specimens were immediately placed in ice-cold Krebs-bicarbonate solution (122 mmol/L NaCl; 1.2 mmol/L MgCl2; 3.5 mmol/L KCl; 1.2 mmol/L KH3PO4; 23 mmol/L NaHCO3; 2.0 mmol/L CaCl2) gassed with 95% O2/5% CO2 and divided in 2 under a stereomicroscope within 5 to 10 minutes. One half of the biopsy was taken for histological examination, and the other was mounted in a modified Ussing chamber and incubated at 37°C in oxygenated Krebs-bicarbonate solution. The serosal solution (10 mL) contained 10−2 mol/L glucose, and the mucosal solution (10 mL) contained 10−2 mol/L mannitol. The exposed area was 0.03 cm2.
The test solutions used in these experiments were prostaglandin E2 (PGE2) 10−5 mol/L, dibutyryl cyclic adenosine monophosphate (cAMP) 10−3 mol/L and acetylcholine (ACh) 10−3 mol/L (Sigma Chemicals Co, St Louis, MO). When several drugs were used in the same experiment, the order of the secretagogues was as mentioned above. In a few experiments, the order was changed, but no difference was seen in the secretory response.
At the end of the experiments, glucose 10−2 mol/L was administered on the mucosal side to measure sodium uptake via sodium-glucose cotransport. Biopsies with a glucose response of less than 0.2 mV were excluded.
Estimation of Subepithelial and Epithelial Resistance, Measurement of Transmural Potential Difference and Calculation of Epithelial Current
Potential difference was measured with a pair of matched calomel electrodes (Radiometer, Denmark) connected to the mucosal and serosal chambers via Ringer-agar bridges. Potential difference was continuously registered on an Xt-recorder (Potentiometer Schreiber RE 571, BBC GOERZ, Austria) using a specially constructed amplifier. Significant electrode drift was excluded by comparing the 0-mV registration before the tissue was mounted to the 0-mV registration after it had been removed. Fluid resistance was measured before the tissue was mounted.
The method has previously been described in detail (7,11,12). It is based on the assumption that the epithelium acts as an electrical circuit consisting of a current generator parallel with a resistance (Rp) and a capacitance (Cp). Between the parallel circuit and the serosal agar bridge, there is also a series resistance (Rs). If a square-pulse is fed through this circuit, the corresponding potential change (registered on an oscilloscope) will be distorted. The distorted part of the square-pulse is due to the RpCp element, which has an exponential voltage/time curve. The rapid rise of the compound potential curve will be due to Rs (Fig. 1).
Square-pulses (10 μA, 5 ms, 5 Hz) from a square-pulse generator were intermittently fed via a resistor through the chamber for 30 seconds. The resulting potential curve, registered by the calomel electrodes, was displayed on an oscilloscope (Hung Chang 5502 20 MHz, Korea). Subepithelial resistance (Rs) and Rp were determined by defining the starting point of the distorted part of the curve. Total epithelial resistance (Rp) was calculated by adding Rs and Rp, that is, by measuring the total height of the potential curve including the distorted part. The epithelial current (Iep) was calculated as Pd divided by Rp.
Median values and ranges are given if not otherwise stated. Wilcoxon signed rank test was used for paired observations. The Mann-Whitney U test and the Kruskal-Wallis test were used for independent observations. The significance level was set at P ≤ 0.05.
Epithelial resistance was measured in different parts of the duodenum, but no significant differences in Rp or total resistance were observed related to localization of the biopsy in controls and patients with CD nor in relation to patients' age. Medians for Rp in controls were 2.4 (range, 1–6.9) Ω × cm2 (n = 8) in the most proximal part and 2.5 (range, 1.8–16.2) Ω × cm2 (n = 9) in the most distal part of the duodenum. The Rt values were 12.8 (range, 10.2–16.2) Ω × cm2 and 12.3 (range, 9.6–33.8) Ω × cm2, respectively.
During basal conditions, Rp was significantly lower in biopsies with mucosal atrophy than in controls. No difference was measured between partial and subtotal villous atrophy. However, total Rt was significantly smaller in biopsies with subtotal atrophy, compared with biopsies with partial atrophy (Table 1).
There was no significant difference in Rp between controls and patients on a gluten-free diet. Total resistance was numerically lower, albeit not significantly so, in treated patients with CD (P = 0.057).
During the experiments, Rt decreased gradually. This was seen in all biopsies except those with subtotal atrophy. The Rp remained unchanged in all biopsies during the experiment except biopsies from patients on gluten-free diet, where the resistance decreased. The secretagogues in the concentrations used did not affect Rp or Rt in any biopsy.
Elevated basal Iep was found in patients with partial villous atrophy, compared with controls, and this difference was not seen in biopsies from patients treated with gluten-free diet (Table 1). The basal Iep did not differ between controls and patients with subtotal villous atrophy.
Significantly lower Rp and total resistance were found in duodenal biopsies from the CF patients, compared with controls (Table 1). The basal Iep values were also significantly lower in the biopsies from CF patients, compared with controls. This difference was only apparent when using the square wave current technique. Calculation of “traditional” short-circuit current from Pd and total resistance generated values, which did not differ significantly from controls. When glucose was administered on the mucosal side to stimulate Na+ absorption in the biopsies from CF patients, no effect was seen on Rp or total resistance, but Iep increased significantly (Table 2).
The key finding of this study is that measurement of Rp in duodenal biopsies reflected the permeability disturbances in CD and CF previously reported in studies using permeability markers (4,5,13,14).
Before discussing the implications of our findings, we want to stress that electrical resistance (or rather conductance) is by no means equivalent to solute permeability. The tight junctions are known to be exceedingly complex regulated structures (3), and it is not to be expected that subtle changes in their geometry or charge distribution will necessarily be reflected by changes in their electrical resistance. On the other hand, if this is indeed the case, as suggested by the present data, measurement of electrical resistance is technically simple if biopsy material is available, as is usually the case. The resistance decrease should therefore be regarded as an indirect marker for an underlying structural abnormality, rather than as an etiological mechanism in itself.
The epithelium in acute CD is more leaky than in normal mucosa (6). It has been shown that the epithelial barrier function is disturbed by structural modifications of the tight junctions (15–17). Epithelium from patients with acute CD also exhibits more rapid cell turnover. This leads to impaired cell differentiation by more secretory crypt cells, leading to a more permeable epithelium. Both these effects, changes in tight junction structure and changes in mucosal cell composition, may affect the relative magnitude of Rp to total resistance. In our study, the mean Rp value in percent of the total Rt was 17% in controls, 11% in partial villous atrophy and 16% in total villous atrophy. Other authors who have studied Rp found the Rp to be 42% in human normal mucosa, 35% in human CD (18), 19% in rat jejunum (19), 54% in partly stripped rat ileum (20), 30% in stripped and 20% in nonstripped rat jejunum (7). If this ratio changes, it will markedly influence the magnitude of the Iep.
The epithelial barrier dysfunction was only partly recovered in CD patients treated by gluten-free diet (16). Schulzke et al. (18) also described a remaining decrease in Rp in treated patients. In contrast to that study, we did not observe any differences in Rp or Iep between controls and patients on the gluten-free diet. On the other hand, we found significantly lower Rs in dietary-treated patients with CD compared with controls, thus explaining the difference in current reported earlier (2).
The absolute Rps observed in this study were lower than those previously reported in the literature. This can be due to many different factors, for example, the mounting procedure, the construction of the chamber, the size of the biopsies, species differences, etc. Edge damage also lowers Pd and Rt; the smaller the orifice, the larger the impact of this factor. Actual leakage across the tissue is a more serious problem, but no glucose leakage was seen indicating that significant damage is unlikely. Our Rs values (ie, total resistance minus Rp) were somewhat higher than previously reported (7,18–20). The reason for this is unclear. It may because of the type of capsule used, the mode of suction pressure application yielding more or less submucosal tissue in the biopsy or other factors. Subepithelial resistance, however, only played a minor role in the key data in this investigation.
Patients with CF were found to have reduced Rp, which is in accordance with previous marker studies (8). A markedly reduced basal Iep was also found in the biopsies from these patients. This is actually to be expected if there is a dysfunction of the cystic fibrosis transmembrane conductance regulator, which contributes significantly to basal Pd in the absence of luminal glucose. The dissociation between Rp changes and Iep in CD and CF supports the specificity of the data obtained with the current technique.
In conclusion, we have described a new technique for measuring Rp in pediatric biopsies taken from proximal small intestine and report changes in electrical resistance consistent with those reported in studies concerning conventional whole-gut oral permeability markers. The technique offers the possibility of combining functional and morphological analysis of pediatric biopsies and may thereby provide a new tool for improved diagnosis in this patient group.
The authors thank Drs Marie Krantz, Henry Ascher and Audur Gudjonsdottir for the generous support in providing part of the intestinal biopsy material during routine examinations.
This work was supported by grants from the Swedish Medical Research Council (1995), the Gothenburg Mansonic Order and the Göteborg Medical Society.
1. Reims A, Redfors S, Hemlin M, et al
. Electrogenic ion transport along the human duodenum in childhood. Scand J Gastroenterol 1997; 32:894–899.
2. Reims A, Redfors S, Ascher H, et al
. Electrogenic ion transport in duodenal biopsies from children with celiac disease
. Scand J Gastroenterol 2002; 37:43–50.
3. Liu Z, Li N, Neu J. Tight junctions, leaky intestines, and pediatric diseases. Acta Paediatr 2005; 94:386–393.
4. Stenhammar L, Falth-Magnusson K, Jansson G, et al
. Intestinal permeability to inert sugars and different-sized polyethylenglycols in children with celiac disease
. J Pediatr Gastroenterol Nutr 1989; 9:281–289.
5. Bjarnason I, Peters TJ. In vitro
determination of small intestinal permeability: demonstration of a persistent defect in patients with celiac disease
. Gut 1984; 25:145–150.
6. Lavö B, Knutson L, Lööf L, et al
. Signs of increased leakage over the jejunal mucosa during gliadin challenge of patients with celiac disease
. Gut 1990; 31:153–157.
7. Hemlin M, Jodal M, Lundgren O, et al
. The importance of the subepithelial resistance for the electrical properties of the rat jejunum in vitro
. Acta Physiol Scand 1988; 134:79–88.
8. McAuley DF, Elborn JS. Cystic fibrosis: basic science. Paediatr Respir Rev 2000; 1:93–100.
9. Walker-Smith JA, Guandalini S, Schmitz J, et al
. Revised criteria for diagnosis of celiac disease
. Arch Dis Child 1990; 65:909–911.
10. Hallberg K, Reims A, Strandvik B. Electrogenic ion transport in duodenum, an aid in cystic fibrosis diagnosis. Scand J Gastroenterol 2000; 35:1106–1109.
11. Theorell T. Applications of square wave analysis to bioelectric studies. Acta Physiol Scand 1946; 12:235–254.
12. Gebhardt U. Exakte kompensation des lösungswiderstandes bei anwendung der voltage-clamp methode an darmepithelien. Med Thesis, Homburg/Saar, Germany; 1979.
13. Hallberg K, Grzegorczyk A, Larson G, et al
. Intestinal permeability in cystic fibrosis in relation to genotype. J Pediatr Gastroenterol Nutr 1997; 25:290–295.
14. Dalzell AM, Freestone NS, Billington D, et al
. Small intestinal permeability and orocaecal transit time in cystic fibrosis. Arch Dis Child 1990; 65:585–588.
15. Madara JL, Trier J-S. Structural abnormalities of jejunal epithelial cell membranes in celiac sprue. Lab Invest 1980; 43:254–261.
16. Schulzke J-D, Bentzel CJ, Schulzke I, et al
. Epithelial tight junction structure in the jejunum of children with acute and treated celiac sprue. Pediatr Res 1998; 43:435–441.
17. Clemente MG, De Virgiliis S, Kang JS, et al
. Early effects of gliadin on enterocyte intracellular signalling involved in intestinal barrier function. Gut 2003; 52:218–223.
18. Schulzke J-D, Schulzke I, Fromm M, et al
. Epithelial barrier and ion transport in celiac sprue: electrical measurements on intestinal aspiration biopsy specimens. Gut 1995; 37:777–782.
19. Fromm M, Schulzke J-D, Hegel U. Epithelial and subepithelial contributions to transmural electrical resistance of intact rat jejunum, in vitro. Pflugers Arch 1985; 405:400–402.
20. Gitter AH, Fromm M, Schulzke J-D. Impedance analysis for the determination of epithelial and subepithelial resistance in intestinal tissues. J Biochem Biophys Methods 1998; 37:35–46.