Institute of Clinical Physiology

> FU Berlin / HU Berlin
 > Charité
  > Departments
   > Inst. Clin. Physiology
    > ResearchTopics
  ! Main idea of our research ! 

 

Tight junctions, channels and carriers  Research topics
- Tight junction proteins: claudin(s), occludin, and tricellulin
- Barrier defect in inflammatory bowel diseases (IBD)
- Cytokines like TNFa affect barrier function

- Epithelial apoptosis causes leaks
- Restitution of epithelial single cell defects
- Direct measurement of paracellular conductance
- Localization of ion channels in crypts and surface epithelium

- Epithelial Na+-channel (ENaC)

- Magnesium

- Physiology of the eye
- Glaucoma
- Retinal degeneration
cartoon Methods
- Short-circuit current (Ussing)
- Conductance scanning
- Impedance spectroscopy
- Patch-clamp
- Molecular biology
- Fluorescence microscopy
- Freeze fracture electron microscopy
cartoon Keywords
- HT-29/B6 confluent colon cell line 
- Inflammatory bowel diseases (IBD)
- Ulcerative colitis and Crohn's disease
- Leak flux diarrhea
- Apoptosis in a colonic epithelium
- Tutorial (A.H. Gitter) Electrical impedance of epithelia
Completed  projects (German text)
 

  Main idea of our research 
Physiologists always ask How does it work?
  • This question cannot be fully answered using classic, e.g. electrophysiological, methods alone without knowing about the molecular structures. On the other hand, molecular biology and biochemistry results usually do not offer conclusions regarding the function of the whole cell or the intact body. In our current projects we therefore combine functional (= mostly electrophysiological) and structural (biochemical, molecular biological, and microscopical) techniques.
  • In order to understand the pathophysiology of a disease it is necessary to know the mechanisms of normal function. For most topics we study the basic mechanisms and - using the same methods - the impaired function in diseased states. Experiments are performed on isolated epithelia and cell cultures. Within this spectrum of approaches we collaborate for many years with the group of Prof. Jörg-Dieter Schulzke (Dept. of Gastroenterology).

Our main research topic is the tight junction:

The intestinal barrier critically depends on the function of the tight junctions. By use of molecular biology and fluorescence microscopy techniques we study the role of the tight junction proteins of the claudin family, occludin, and tricellulin.

In inflammatory bowel diseases (ulcerative colitis and Crohn's disease) cytokines like tumor necrosis factor alpha (TNFa) induce a dramatic increase tight junction permeability. This barrier break-down is a major determinant of the disease, because it causes leak flux diarrhea and allows for unwanted uptake from the lumen. 

For measuring the barrier function we have developed two electrophysiological techniques,
... and the transporters of the cell membrane:

Tight junctions, Kanäle, Carrier
Research topics 

Tight junction proteins: claudin-1 to -24, occludin, and tricellulin.
Barrier function and genomic regulation

Epithelial tight junctions consist of the transmembranal proteins occludin, tricellulin, and 24 different claudins. Common properties are 4 transmembranal domains und 2 extrazellular loops (ECL). ECL1 is thought to determine the paracellular barrier and/or channel function. ECL2 may act as mechanical contact between opposing tight junction proteins. Also a TJ protein is JAM (junctional adhesion molecule), which forms one transmembranal domain.

Occludin
Localization: All epithelia
Function: The function of occludin is still poorly understood. In a collaboration, the Tsukita group and our group have shown that in occludin knock- out mice the tight juncion barrier is unaltered. This mens that occludin either has no intrinsic barrier properties or can be replaced by other components of the tight junction.

In occludin-knockout mice the glandular structure of the stomach exhibited a complete loss of parietal cells and mucus cell hyperplasia, as a result of which acid secretion was virtually abolished. A dramatic change in gastric morphology and secretory function indicates that occludin is involved in gastric epithelial differentiation.

Little is known about the regulatory mechanisms of occludin that influence occludin gene expression. We aimed to identify the sequences essential in cis for genomic regulation of tight junction formation and to investigate their functional role in cytokine-dependent tight junction regulation.
Using genome walking cloning of occludin-specific human genomic DNA sequences, a 1853 bp DNA fragment containing the transcription start point of occludin cDNA sequences was amplified and sequenced. The proinflammatory cytokines, TNFa and interferon g diminished occludin promoter activity alone and even synergistically, suggesting a genomic regulation of alterations of the paracellular barrier. Both cytokines downregulate the expression of occludin, paralleling the barrier disturbance detected electrophysiologically. This could be an important mechanism in gastrointestinal diseases accompanied by barrier defects, for example inflammatory bowel diseases.

Tricellulin
Localization: Three-corner point of epithelia (tTJ)
Function: Lack of tricellulin prevents the development of the epithelial barrier. With the discovery of tricellulin the group of Shoichiro Tsukita has found another type of tight junction protein. Shoichiro Tsukita died on 10. Dec. 2005, a few days before appearance of this landmark paper: Ikenouchi et al., 2005, J. Cell Biol. 171(6): 939-945. [PubMed abstract] [Full text PDF]. We showed that tricellulin tightens the tricellular junction against macromolecules, leaving the total epithelial ion conductance unaltered.

Claudin family
General: Presently, 24 different claudins are known in mammalia. Some claudins are essential for forming the epithelial barrier while others even do the oppsosite, they form paracellular channels.

Claudin-1
Localization: Typical for epithelia with almost impermable tight junctions ("tight epithelia") like distal colon and distal kidney tubule
Function: Paracellular barrier against transepithelial diffusion
Clinical Impact: Claudin-1 is reduced in hereditary mammary cancer. A defect of claudin-1 is present in neonatal sclerotic cholangitis with ichthyosis. Expression of claudin-1 and of claudin-4 was increased in colorectal cancer and pancreatic and ovarian cancers.

Claudin-2
Localization: Typical for epithelia with rather permeable tight junctions ("leaky epithelia") like small intestine and proximal kidney tubule
Function: We were able to show that claudin-2 is responsible for forming a paracellular channel specific for small cations.
Clinical Impact: Increased expression causes impaired barrier function 

Claudin-3
Localization: Typical for tight epithelia
Function: Paracellular barrier
Clinical Impact: Claudin-3 and -4 are receptors for the enterotoxin of Clostridium perfringens.

Claudin-4
Localization: Typical for tight epithelia
Function: Paracellular barrier
Clinical Impact: Claudin-4 is not expressed in the Williams-Beuren-Syndrome.

Claudin-5
Localization: Typical for endothelia
Function: We were able to show that claudin-5 belongs to the barrier-forming claudins and that it is expressed also in some epithelia.
Clinical Impact: Claudin-5 is deleted in patients suffering from velo-cardio-facial syndrome (DiGeorge syndrome). Claudin-5-deficient mice exhibit a barrier defect of the blood-brain barrier.

Claudin-6
Localization: Embryonic epithelia, adipose tissue
Function: Overexpression of claudin 6 results in decreased expression of other claudins and a severe barrier defect

Claudin-7
Localization: Kidney: Distal nephron
Function: Overexpression of claudin-7 decreases the paracellular Cl- conductance and increases the paracellular Na+ conductance in LLC-PK1 cells

Claudin-8
Localization: Kidney: Aldosterone-sensitive distal nephron (ASDN)
Function: Barrier for cations in tight epithelia.

Claudin-10
Localization: Claudin-10 exists in two splice variants. Variant 1: kidney. Variant 2: many epithelia including kidney
Function: Variant 1 is anion selective. Variant 2 is cation selective.

Claudin-11 (= OSP)
Localization: CNS: oligodendrocytes, Testis: Sertoli cells, Ear: organ of Corti, Kidney: prox. Tubule, Henle's loop
Function: Barrier

Claudin-14
Localization: Ear: cochlea hair cells; Kidney: collecting duct
Function: Barrier in cochlea hair cells

Claudin-16  (= paracellin 1)
Localization: Kidney: thick ascending limb of Henle's loop and distal tubule
Function: Claudin-16 facilitates magnesium and calcium transport
Clinical Impact: Mutations of claudin-16 (and claudin-19) are associated with familial hypomagnesemia, together with hypercalciuria and nephrocalcinosis (FHHNC)

Claudin-17
Lokalization: Epidermis

Claudin-18
Lokalization: Lung, Stomach, Oesophagus, Ear: Cochlea, Corti organ, Stria vascularis marginal cells

Claudin-19
Lokalization: Kidney: Distal Nephron, Nerve: Schwann cells
Function: Barrier

Claudin-20
Lokalization: Eye: Retina pigment epithelium

Claudin-21

Claudin-22

Claudin-23
Lokalization: Placenta, Stomach, colon tumors

Claudin-24

 

Supported by:
  • German Research Society (DFG) Research Unit (FOR 721) "Molecular Structure and Function of the Tight Junction". Coordinator: Fromm
  • German Research Society (DFG), personal grant (Fr 652/4) "Regulation and functional properties of tight junction and channel proteins in colon epithelium"
  • German Research Society (DFG), personal grant (GU 447/11-1) "Barrier- and channel properties of claudins in the renal diluting segment and distal nephron"

The Tight Junctions
Localization: Charité events & clubs in Berlin
Function: Live featuring classic rock
Clinical Impact: Animation of acoustically irradiated hominids

Barrier defect in inflammatory bowel diseases (IBD)

Many diseases of the intestines are caused by impaired epithelial absorption or secretion and by impaired epithelial barrier function.

The pathogenesis of the ulcerative colitis and Crohn's disease is still unknown. A typical symptom in both inflammatory bowel diseases is chronic diarrhea. We investigate the transport and barrier function of the intestine in vitro using two electrophysiological techniques, impedance spectroscopy and conductance scanning.

If the ion permeability is critically increased under pathological conditions a leak flux diarrhea occurs. This type of diarrhea is caused by massive fluxes of solutes and water from the blood into the gut lumen.

Regarding immunological mechanisms, an intact epithelial barrier keeps luminal bacteria, toxins, and antigens away from the subepithelial tissues. It is discussed, whether an impaired intestinal barrier allows for increased uptake of bacteria, toxins, and antigens which then will support the inflammation process.

Bacterial translocation through the intestinal wall has been studied under defined in vitro conditions in our lab.

 

Cytokines like TNFa affect barrier function

Cytokines like tumor necrosis factor alpha (TNFa), interleukins and interferons act as mediators of inflammation. In inflammatory bowel diseases (and in HIV infection) their local concentrations increase. We study the action of cytokines on transport and barrier function of human intestine and cell cultures originating from human colon (HT-29/B6).

Epithelial apoptosis causes leaks

Key word: Epithelial apoptosis

Current opinion assumes epithelial integrity during spontaneous apoptotic cell death. We measured, for the first time, the local conductances associated with apoptoses and show leaks of 50 nS (mean) in human intestinal epithelium. Measurements were performed employing the conductance scanning technique on confluent HT-29/B6 monolayers.

The results disprove the dogma that isolated cell apoptosis occurs without affecting the epithelial cell permeability barrier. After induction of apoptosis by tumor necrosis factor a (TNFa ) the apoptotic leaks were dramatically enhanced: not only increased the frequency 3fold, but the mean conductance increased 12fold to 600 nS. Thus apoptosis accounted for about half of the TNFa -induced permeability increase while the other half is caused by degradation of tight junctions in non-apoptotic areas.

Hence, spontaneous and, more so, induced apoptosis hollow out the intestinal barrier and may facilitate loss of solutes and uptake of noxious agents.

Restitution of epithelial single cell defects

Since the barrier function of an epithelium relies on its complete integrity, which is challenged by the loss of cells, by injury or apoptosis, rapid repair mechanisms are important. Already well known is the repair process of large epithelial wounds (restitution). The gap is closed by the immigration of intact adjacent cells within hours. Deep wounds require also cell proliferation, which takes several days. Little is known, however, about the - much faster - closure of single cell defects. Purely morphological studies cannot assess the functional leak that opens with the hiatus. Our new conductance scanning technique allows to measure the leak. Thus we investigate, for instance, the effects of inflammatory mediators on the restitution of single cell defects in epithelia of the intestine.

Supported by DFG-GRK 276/1-99 "Signal recognition and -translation"

Localization of ion channels in crypts and surface epithelium

The conductance scanning technique in medium resolution allowed to measure in crypt and surface epithelium of the colon the conductivities for Na+, K+, Cl–, which are defined by specific apical channels.

We demonstrated that cAMP-induced Cl– secretion is localized not only - as assumed so far - in the crypts, but to a similar amount also in the surface epithelium:

The aldosterone-simulated electrogenic Na+ absorption is restricted to the surface epithelium:

Regarding the localization of the K+ secretion it is crucial how it is stimulated: The cAMP-induced K+ secretion is localized in crypts as well as in surface epithelium, while after stimulation by aldosterone K+ secretion is induced only in the surface epithelium:

Epithelial sodium channel (ENaC)

The amiloride- blockable epithelial Na+ channel (ENaC) is an important protein for the regulation of the Na+ balance in vertebrates. It is a limiting factor for the absorption of Na+ in several organs. ENaC is regulated in epithelia of the distal kidney tubule and the distal colon by aldosterone and other corticosteroids.

The epithelial sodium channel (ENaC) was cloned a few years ago from intestinal epithelium of the rat [Canessa et al. 1993, Nature; Lingueglia et al. 1993, FEBS Lett.; Canessa et al. 1994, Nature]. The channel-forming protein consists of two a-, one b- and one g-subunit [Kosari et al. 1998, J. Biol. Chem.; Firsov et al. 1998, EMBO J.].

ENaC is regulated by aldosterone on the genomic level. The effect of aldosterone can be divided in 3 phases of: a latent phase, a phase of increasing Na+ transport, and a phase starting after 4 hours of stimulation of the basolateral Na+/K+-ATPase [Benos et al. 1995, J. Membr. Biol.]. The main component is the regulation of the epithelial Na+ channel: a blockade of ENaC by the diuretic amiloride stops the electrogenic Na+ transport (and the effect of aldosterone) completely [Benos 1982, Am. J. Physiol.].

We investigated electrogenic Na+ transport and, in identical tissues, mRNA expression of ENaC subunits in early (EDC) and late (LDC) distal colon of the rat. In both segments 8 h in vitro incubation with 3 nM aldosterone enhanced b- and gENaC mRNA and induced Na+ transport. Na+ transport was ten times higher in LDC than in EDC.

Although it is known, that aldosterone regulates the electrogenic Na+ absorption via a transcriptional mechanism [Verrey & Beron 1996, NIPS], it was unresolved so far, whether aldosterone affects the electrogenic Na+ transport directly by de novo synthesis of channel proteins or indirectly by induction of regulatory proteins. We found that aENaC mRNA was unchanged in EDC, whereas it decreased in LDC. In LDC, b- and gENaC mRNA was induced 1 h after aldosterone addition whereas Na+ transport became apparent >1 h later. Down-regulation of aENaC does not take part in the acute regulation because it started after a lag time.

It was thought so far that the "early effect" of aldosterone is not based on genomic regulation, because ENaC transcription was reported to start later than Na+ transport. However, time correlation of b- and gENaC induction and Na+ transport stimulation suggests that the early aldosterone effect on Na+ absorption in distal colon may be due to transcriptional upregulation of b- and gENaC expression.

ENaC is induced by the mineralocorticoid rezeptor (MR) as well as by the glucocorticoid rezeptor (GR). ENaC expression is synergized by butyrate and by tumor necrosis factor-alpha (TNF-alpha).

In inflammatory bowel diseases (IBD) ENaC expression and thus function is reduced. This effect is mediated by pro-inflammatory cytokines like interleukin-1beta and tumor necrosis factor-alpha:

Supported by the Deutsche Forschungsgemeinschaft (DFG Fr 652/4-3)

Earlier papers on epithelial Na+ transport

Magnesium
cartoon
  Methods 

Isolated but "living" gastrointestinal epithelia and epithelial cell cultures can be functionally characterized regarding their transport and barrier functions by electrophysiological methods.

Short circuit current (Ussing)

Old fashioned, but still powerful in vitro technique for determination of active ion transport (short circuit current), ion permeability (conductance or resistance, respectively) and radioisotope fluxes. Three computerized setups consisting of 6-8 chambers each. Details:

This technique is applied also in the Lab course in Physiology for 2nd year medical students. Tissues are distal colons obtained from a rural rabbit slaughtery close to Berlin. We have published this lab course:

We have developed 2 more refined techniques which allow for discrimination of local conductances within the epithelial tissue:
- Conductance scanning for the measurement of horizontal conductance distribution,
- Transmural impedance for the measurement of vertical conductance distribution.

Conductance scanning

In order to describe quantitatively the heterogenous horizontal distribution of ion permeability in an epithelium, we developed a new electophysiological method. It was named conductance scanning, because it aims at the spatial resolution of epithelial conductivity by means of a scanning microelectrode probe.

The method is based upon the measurement of local differences in current density that are recorded with microelectrodes in the electrolyte solution above the mucosal surface of the flat epithelium, while a defined clamp current is passed through the tissue. Using mathematical models the distribution of epithelial conductivity is derived from the distribution of supraepithelial current density.

We have developed this method for 3 applications which differ reganding their spatial resolution and their mathematical models:

One-path impedance spectroscopy (1PI)

Key word: Electrical impedance of epithelia

Discrimination of epithelial and subepithelial ion conductance in intact gastrointestinal epithelia in vitro. The conductance of the pure epithelium is frequency-dependent, however that of the tissues underlying the intestinal epithelium is not (in the frequency range applied). This is due to the fact that the subepithelial tissues have much higher conductivity due the lack of tight junctions. This allows to locate intestinal permeability changes to the epithelium or the subepithelium. Own development. Two  setups.

Two-path impedance spectroscopy (2PI)

This is a refined technique which is employed to determine paracellular and transcellular resistance in epithelial confluent celllayers. A specific perturbation of one of the two pathways allows for recording two data sets. Para- and transcellular resistance is calculated after combinig the two data sets and fluxes of a suitable paracellular marker.

Patch-Clamp

Measurement of ion specificity, conductance, voltage dependency and open probability of cell membrane channels. Very common electrophysiological method, thus no detailed explanation here.

Note: The Berlin Patch Clamp Colloquium, a regular seminar of  patch clampers throughout Berlin, is organized by our former post-doc Friederike Stumpff.

Molecular biology

In most projects we alter epithelial transport or barrier properties in an electrophysiological experiment, and then use these functionally altered tissues for molecular biology.

Our current projects cover two main topics:
1.  The epithelial sodium channel (ENaC)

2.  Tight junction proteins (occludin and claudin)

For the investigation of structural and functional characteristics we expose epithelial cell layers to proinflammatory cytokines under the standardised conditions of in vitro cell culture. During the incubation period the alteration of the transepithelial resistance is monitored. Subsequently, RNA and proteins are isolated from the cells, separated electrophoretically in a gel matrix and analyzed in Northern and Western blots with gene-specific probes or antisera against tight junction proteins.

Changes in quantity of the biomolecules permit conclusion on the regulation of gene expression and the role of the individual tight junction proteins. Beside classical hybridizing procedures we apply modern molecular biology methods for a structural and functional analysis of paracellular barrier and epithelial transport. With the genome walking technique it is possible to isolate regulatory sequences structurally linked to the gene-of-interest.

After nucleic acid sequencing, reporter gene analyses are performed to characterize these sequences functionally. Motives essential for transcription factor binding can be limited by targeted mutations. Thus, conclusions on the signal transduction cascades connected with the regulation of gene expression can be drawn. This is of importance for the development of improved therapeutic approaches of inflammatory bowel diseases.

Fluorescence microscopy

Many molecules can be stained with an immunofluorescence dye and then detected or localized within the cell by confocal laser scanning microscopy (CLSM). Example: By Western blotting Claudin-1 was found in the membrane fraction of MDCK-C11 cells, which is surprising, because MDCK-C11 form a rather leaky epithelium, while claudin-1 is typical for tight epithelia. However, claudin-1 was not colocalized with occludin within the tight junction but was found below the tight junction. Therefore it does not contribute to sealing properties of the tight junction (see Fig., from Amasheh et al., 2002, J Cell Sci.):

 

Freeze fracture electron microscopy of the tight junction

Permeability of tight junctions are not determined solely by their molecular composition, but also - as known for many years - by the ultrastructural arrangement, expansion, and continuity of tight junction strands. The molecular composition and the ultrastructure are related to each other. Technically, tissues are freeze fractured,  vaporized, analyzed electron microscopically. If the fracture occurs inside the lateral cell membrane often a tight junction meshwork becomme visible. In a morphometric analysis, tight junctions then are analyzed regarding the number of horizontally arranged strands, its extension, and its linearity.