Differential expression of laminin α chains during murine tooth development

Differential Expression of Laminin a Chains DuringMurine Tooth DevelopmentKATRIINA SALMIVIRTA,1 LYDIA M. SOROKIN,2 AND PETER EKBLOM1*1Department of Animal Physiology, Uppsala University, Uppsala, Sweden2Department of Experimental Medicine, University of Erlangen, Erlangen, Germany

ABSTRACT Basement membranes of the de-veloping tooth have been previously shown tocontain laminins, but the nature of the lamininshave not been described. We here studied thedistribution of five different laminin a chainsduring tooth development. We show that bothepithelial and mesenchymal cells produce lami-nin a chains. The mRNAs of three laminin achains, a1, a2, and a4, were expressed in the toothmesenchyme, whereas two, the a3 and a5 chainmRNAs, were found in epithelium. Drasticchanges in the expression patterns of the twoepithelial chains were found during develop-ment. The a5 mRNA was widely expressed intooth epithelia, and the corresponding proteinwas evenly distributed along the tooth basementmembrane throughout embryonic development.This suggests a role for a5 as a major laminin achain in tooth basement membrane during embry-onic stages. The subsequent disappearance of a5and the drastic increase in a3A mRNA expressionduring terminal ameloblast differentiation andenamel secretion suggest that a3A acts as animportant chain in the enamel matrix after degra-dation of tooth basement membrane. These stud-ies show that laminin networks in tooth epitheliaform as a result of epithelial–mesenchymal inter-actions and that the molecular composition of thelaminin networks varies drastically during devel-opment of tooth. Dev. Dyn. 1997;210:206–215.r 1997 Wiley-Liss, Inc.

Key words: laminin; epithelial–mesenchymal in-teractions; tooth

INTRODUCTION

The developing tooth is an excellent model organ tostudy the molecular mechanisms of morphogenesis.Tooth development is regulated by sequential andreciprocal interactions between the neural crest–derived mesenchyme and the oral ectoderm (Thesleffand Hurmerinta, 1981; Ruch, 1987; Lumsden et al.,1988). Odontoblasts, which secrete the extracellularmatrix of dentine, originate from the mesenchymaldental papilla cells adjacent to basement membrane(BM), whereas ameloblasts synthesizing enamel origi-nate from the inner dental epithelial (ide) cells (Thesleff

et al., 1981). The precise molecular mechanisms mediat-ing interactions between epithelium and mesenchymeare not known. However, BM components have beenshown to play important roles in these regulatoryevents (Adams and Watt, 1993). BMs are specializedextracellular matrices mainly consisting of variousforms of laminin, type IV collagen, perlecan, and nido-gen (Timpl, 1996). Previous immunohistochemical stud-ies have suggested that type IV collagen and lamininmolecules are evenly distributed along the dental BMsduring early mouse embryonic development and disap-pear when BM is degraded by enzymes such as type IVcollagenase (Thesleff et al., 1981; Lesot et al., 1981;Sahlberg et al., 1992; Kjoelby et al., 1994). However,this fairly static view of a molecularly homogenous BMduring several stages of tooth development was basedon the use of antibodies reacting with laminin and typeIV collagen chains found in many isoforms.

It has gradually become apparent that there areseveral collagen IV and laminin isoforms. We thereforehere studied the expression of five different laminin achains during tooth development. Laminins consist ofdifferent subsets of a, b, and g chains. So far five a,three b, and two g chains assembled to more than tendifferent isoforms have been identified. Most lamininsare found as combinations of b1g1 or b2g1 chains withone of the five a chains. By contrast, laminin-5 has aunique chain composition, consisting of a3b3g2 (Timpl,1996). Some information about expression of different achains is available from non-dental tissues. Laminin a1chain has a restricted epithelial expression patternboth in adult and embryonic tissues (Klein et al., 1988,1990; Ekblom et al., 1990; Kadoya et al., 1995). Lami-nin a2 chain has been found in placenta, striatedmuscle, peripheral nerve, and thymus (Leivo and En-gvall, 1988; Ehrig et al., 1990; Chang et al., 1993;Bernier et al., 1995). Due to alternative splicing, lami-nin a3 has two isoforms, a3A and a3B (Galliano et al.,1995). These two isoforms of laminin a3, as well aslaminin a4 chain, have been shown to have fairlyrestricted expression patterns (Galliano et al., 1995;

Grant sponsor: Cancerfonden.*Correspondence to: Peter Ekblom, Department of Animal Physiol-

ogy, Uppsala University, Biomedical Center, Box 596, S-75124 Uppsala,Sweden. E-mail: [email protected]

Received 26 May 1997; Accepted 3 July 1997

DEVELOPMENTAL DYNAMICS 210:206–215 (1997)

r 1997 WILEY-LISS, INC.

Iivanainen et al., 1995). a3 has been found in the mostdifferentiated cells at the top of the intestinal villi(Orian-Rousseau et al., 1996), as well as in functionalameloblasts of tooth (Aberdam et al., 1994a; Galliano etal., 1995). Laminin a4 has been shown to be expressedat high levels in fetal kidney and lung but is also foundin a subset of endothelial cells (Iivanainen et al., 1995;Frieser et al., 1997). In contrast to the somewhat

restricted expression pattern of the other a chains, amore widespread distribution in adult tissues wassuggested for the a5 chain (Miner et al., 1995), andrecently it was shown that a5 chain in the embryo isproduced by epithelial cells (Durbeej et al., 1996).

Due to the increased number of identified laminin achains and their isoforms, the role of individual achains for BM assembly during tooth development is at

Fig. 1. Laminin a1 mRNA expression during tooth development. A, C,and E are darkfield and B, D, and F are brightfield illumination. A,B: AtE13, a strong expression is detected in condensed dental mesenchyme.C,D: At E17, tooth development is at the bell stage and expression isdetected in dental papilla as well as in mesenchymal dental sac cells. E,F:

At P1, expression is detected in polarized odontoblasts, while dentalpapilla no longer shows expression. However, the secretory odontoblastsare negative (arrow). de, dental epithelium; dm, dental mesenchyme; dp,dental papilla; ds, dental sac; po, polarized odontoblasts; M1, first molar;M2, second molar. Bars 5 200 µm.

207LAMININ a CHAINS IN DEVELOPING TOOTH

present obscure. In the present study we have thereforeused immunohistochemistry and in situ hybridizationto clarify the distribution of laminin a1–5 chains duringembryonic mouse tooth development.

RESULTS

Laminin a1 mRNA expression was detected in thedental mesenchyme throughout the embryonic toothdevelopment. Early in development (embryonic day 13[E13]), an intense expression was seen in the condenseddental mesenchyme (Fig. 1A,B). Later (E14–18), a1mRNA expression was seen in dental papilla derivedfrom the condensed mesenchyme, as well as in mesen-chymal dental sac cells surrounding the tooth germ.The adjacent non-dental mesenchyme showed weakerexpression (Fig. 1C,D). At postnatal day 1 (P1), theexpression was lost from the dental papilla includingthe functional odontoblasts secreting dentine. Instead,the expression became localized to polarized odonto-blasts underneath the secretory odontoblasts (Fig. 1E,F).Unexpectedly, we failed to detect laminin a1 by immu-nohistochemistry in tissue sections from embryonic

mouse tooth with antibody recognizing the E3 fragmentof laminin a1 chain. However, in immunostaining of anE13 tooth grown in organ culture for 7 days, a strongexpression was detected in the dental papilla as well asin the dental BM (Fig. 2).

Laminin a2 mRNA was also detected in mesenchy-mal cells during tooth development. At earlier stages,the mesenchymal expression was seen around theepithelial bud. However, the intensity of a2 mRNAexpression appeared weaker than the expression of a1mRNA. At later stages (E15–18), a2 mRNA expressionbecame stronger in dental sac cells than in the cells ofdental papilla, where the expression was more re-stricted to polarized odontoblasts. No expression of a2mRNA was detected in functional, dentine-secretingodontoblasts. In addition, strong expression of a2 mRNAwas detected in tongue muscle (Fig. 3). By immunofluo-rescence, laminin a2 was detected at tooth germ BMbefore E15. Later (E15–18), the staining was lost fromthe dental BM (i.e., the BM between ide and the dentalpapilla mesenchyme), but remained strong in the BMbetween outer dental epithelium (ode) and dental sacmesenchyme, thus corresponding to the strong mRNAexpression detected in the dental sac cells by in situhybridization. An intense staining was also seen inmuscular and hair follicle BMs (Fig. 4).

The expression of laminin a3 mRNA was studied byin situ hybridization using two oligonucleotide probesspecific for the A and B isoforms of a3 mRNA, as well asby Northern blot analysis. The expression pattern oflaminin a3A mRNA was found to be regulated in adevelopmental stage-specific manner. At stages E13–E18, the expression was detected only in the oralepithelium and in the epithelial stalk of the tooth germ(Fig. 5A,B). At P1, when a subset of tooth epithelial cellshave differentiated into enamel-secreting ameloblasts,laminin a3A expression become very intense in thesefunctional cells (Fig. 5C,D). Moreover, a drastic in-crease in laminin a3A mRNA (5.5 kb) expression fromembryonic stages to P7 was detected by Northernblotting (Fig. 6), coinciding with the appearance ofsecretory ameloblasts. In contrast, the expression pat-tern of laminin a3B mRNA was highly restricted duringodontogenesis. In fact, we could detect a3B clearly onlyin the enamel knot cells in E14 tooth (Fig. 5E,F), incontrast to a previous report which suggested a3BmRNA to be expressed also in ameloblasts (Galliano etal., 1995). In overexposed films, a weak band correspond-ing to laminin a3B mRNA (8 kb) could be detected bothin embryonic and postnatal tooth RNA (not shown).

Laminin a4 mRNA was intensely expressed in thedental mesenchyme (Fig. 7) as well as in mesenchymesof other organs such as olfactory organ, vibrissa, hairfollicle, salivary gland, lung, and kidney (not shown).The expression was evenly distributed in mesenchymaltissues throughout embryogenesis. However, at P1 thedental papilla lost expression. Moreover, no expression

Fig. 2. Immunohistochemical localization of laminin a1. A: No specificstaining could be detected in mouse first molar tooth at E15. B: In E13mouse first molar tooth cultured for 7 days in vitro; strong expression isseen in the basement membrane (BM) as well as in the dental papilla. de,dental epithelium; dp, dental papilla. Bar 5 100 µm.

208 SALMIVIRTA ET AL.

was detected in the functional odontoblasts. In addi-tion, strong expression was detected in skeletal muscles(Fig. 7).

A strong epithelial expression of laminin a5 mRNAwas observed throughout embryonic tooth developmentby in situ hybridization and Northern blotting. How-ever, the functional ameloblasts at P1 were found to benegative (Fig. 8). In Northern blotting the mRNA (12kb) level was seen to decrease at P7 (Fig. 6), which ismost likely due to the reduction of epithelial stellatereticulum and loss of it during dissection. By immuno-staining of embryonic tissue sections, laminin a5 wasdetected in various BMs including oral and dental BMs.At P1, however, the expression was lost from the dentalBM (Fig. 8).

DISCUSSION

It has been clearly established that epithelial–mesenchymal interactions are major driving forces fordevelopment of many organs (Grobstein, 1967; Saxenand Kohonen, 1969). Basement membranes (BM) arelocated between epithelium and mesenchyme and mightthus be involved in these interactions. Growing evi-

dence shows that BMs, previously seen as static struc-tures, undergo dynamic compositional changes duringorganogenesis and might thus play important roles inthe regulation of morphogenetic processes. Here weshow that each of the five laminin a chains has a uniquedistribution pattern during tooth development and thatrapid changes in the expression pattern of these chainsoccur during different developmental stages.

Laminins have usually been viewed as products ofthose cells that are attached to them, such as theepithelial cells. A major finding in the present studywas that both epithelial and mesenchymal cells pro-duce laminin chains. No less than three laminin chains,the a1, a2, and a4 mRNAs, were expressed in the toothmesenchyme whereas two, the a3 and a5 chain mRNAs,were found in epithelium. A second major finding wasthat a5 mRNA is widely expressed in tooth epithelia,and the corresponding protein is evenly distributedalong the tooth BM throughout embryonic develop-ment. This suggests a role for a5 as the major laminin achain in tooth BM. A third major finding was the drasticincrease in a3A mRNA expression during terminal

Fig. 3. Laminin a2 mRNA expression in developing tooth. A and C aredarkfield and B and D are brightfield illumination. A,B: At E13 theexpression is detected in dental mesenchyme around the epithelial bud.C,D: At E17, in addition to the mesenchymal dental papilla expression, the

mesenchymal dental sac cells show a strong expression. Note alsoexpression in tongue muscle. de; dental epithelium; dm, dental mesen-chyme; dp, dental papilla; ds, dental sac; t, tongue. Bar 5 200 µm.


ameloblast differentiation, suggesting its involvementin cell adhesion during BM degradation and enamelsecretion.

The clear-cut mesenchymal expression of laminin a1chain is notable. In many previous studies, laminin a1chain has been shown to have a highly restricted andtransient expression pattern in embryonic kidney epi-thelium, and in several other embryonic tissues theprotein was noted exclusively in epithelial basementmembranes (Ekblom et al., 1990; Sorokin et al., 1992;Klein et al., 1990). Some laminin a1 chain mRNA hasbeen previously detected in a few embryonic tissues,but the protein was found in the basement membranes(Thomas and Dziadek, 1994; Kadoya et al., 1995). Inthe embryonic tooth, laminin a1 mRNA was intenselyexpressed in the dental mesenchyme throughout embry-onic development. We could also detect laminin a1protein with a monoclonal antibody in the tooth BMand dental papilla in teeth cultured in vitro. However,in tissues taken directly from the embryo, immunohis-tochemical analysis using the same antibody againstthe E3 fragment of laminin a1 chain failed to detect anyprotein. While the reason for this discrepancy is un-clear, it is possible that the accessibility of the E3epitope for the antibody was masked in the tissuestaken directly from embryos. Masking of epitopes hasbeen described for several extracellular matrix proteinsand has been suggested for other laminin chains (Dur-beej et al., 1996). Further studies are needed to clarifythis point.

The other a chains detected in the mesenchyme of thedeveloping tooth, i.e., a2 and a4, have previously beenobserved in several embryonic BMs. Our studies demon-strate that the mRNAs coding for these a chains aredistinctly expressed by the embryonic tooth mesen-chyme. In addition, a2 chain mRNA was intenselyexpressed in the dental sac cells. Results from theimmunohistochemical analysis of a2 expression were inconformity with the detected mRNA patterns. Thesefindings suggest a role for a2 and a4 chains in themesenchymal production of BM proteins. None of themesenchymally expressed laminin a chains (a1, a2,and a4) were expressed in fully differentiated, secretoryodontoblasts. The downregulation of production of theselaminin a chains coincides with the disappearance ofthe dental BM, so this disappearance is not necessarilymainly due to protease activity as previously thought(Sahlberg et al., 1992; Kjoelby et al., 1994).

Laminin a3 and a5 chain mRNAs were both detectedin the dental epithelium. While both A and B isoforms oflaminin a3 mRNA showed restricted and transientexpression patterns, the a5 chain mRNA was widelyexpressed throughout development. a3B mRNA wasexclusively observed in the enamel knot cells, whichrepresent a cluster of non-dividing cells assumed to be aregulatory center in the cuspid formation (Jernvall etal., 1994). This finding may indicate a role for the a3Bisoform in cuspal morphogenesis. Laminin a3A mRNAwas strongly upregulated in the ameloblasts secretingenamel. While the reason for this upregulation is not

Fig. 4. Immunohistochemical localization of laminin a2. A: At E13 andE14.5 (shown), expression is detected in entire dental basement mem-brane (BM). B: At E16, staining has become weaker in dental BMbetween inner dental epithelium (ide) and dental papilla, remaining strongin the BM between outer enamel epithelium (ode) and mesenchymaldental sac cells. C: At E18, no staining is detected in dental BM betweenide and dental papilla, while staining in BM between ode and mesenchy-mal dental sac cells remains strong. Note strong staining in tongue andmasseter muscles. de, dental epithelium; dm, dental mesenchyme; ide,inner dental epithelium; ode, outer dental epithelium; dp, dental papilla;ds, dental sac; t, tongue. Bars 5 100 µm.


known, it is possible that the a3A chain participates incell adhesion events during the period of dental BMdegradation (Sahlberg et al., 1992; Kjoelby et al., 1994)occurring concomitantly with dentine and enamel secre-tion. In contrast to our findings, earlier studies havesuggested expression of a3B in ameloblasts (Galliano etal., 1995). Moreover, the b3 and g2 chains, both compo-nents of laminin-5, have also been shown to be present

in secreting ameloblasts (Aberdam et al., 1994a). Lami-nin-5 has been shown to be involved in maintainingadhesion of the epidermis to the underlying dermis(Aberdam et al., 1994b), and patients with homozygousdeletion in the laminin a3 gene suffer from severeepidermal blistering disease, Herlitz junctional epider-molysis bullosa, which can be accompanied by general-ized enamel hypoplasia (Wright et al., 1993).

Fig. 5. Laminin a3A and a3B mRNA expression during mouse toothdevelopment. A, C, and E are darkfield and B, D, and F are brightfieldillumination. A,B: At E14, expression of laminin a3A mRNA is seen only atthe epithelial stalk of developing tooth and oral epithelium. C,D: At P1, anextensive increase in laminin a3A mRNA expression in functional amelo-

blasts is detected. E,F: Laminin a3B mRNA expression in E14 mousetooth. Expression is restricted to the non-dividing enamel knot cells(arrow). de, dental epithelium; oe, oral epithelium; a, ameloblasts; fa,functional ameloblasts. Bars 5 100 µm (A,B,E,F), 200 µm (C,D).


A homogeneous strong expression of laminin a5chain mRNAwas seen in tooth epithelial sheets through-out several stages of tooth development. The presenceof the protein in the BM was verified with the use of arecently described monoclonal antibody (Sorokin et al.,1997). These findings suggest that a5 could be themajor a chain in embryonic dental basement mem-brane. An interesting detail is that each epithelial cellin the multicellular epithelial cusps produced themRNA, and yet the protein was found only in thebasement membrane. This shows complex and yetunclarified pathways of synthesis of the laminin het-erotrimers. The mechanisms which lead to a selectivepolarized expression of the final polypeptide only in thebasement membrane are completely unknown; ourfindings clearly reveal that the mRNA was not selec-tively made by the cells located most basally as onemight expect based on immunocytochemistry. Althoughthe a5 chain may be a major chain in many adultnon-dental tissues, its expression was downregulatedby advancing development in the tooth. In contrast toa3A, we observed a complete down regulation of a5mRNA in secretory ameloblasts, together with a downregulation of protein in tooth BM at P1.

In conclusion, we have demonstrated a remarkabledifferential expression of five different laminin a chainsduring tooth development. These studies clearly revealthat laminin networks form as a result of epithelial–mesenchymal interactions, each of the two cell typescontributing with synthesis of different chains. More-over, rapid and profound changes in the expressionpattern occur both in mesenchyme and epithelium as

development advances. This raises the possibility thatthe different laminin a chains, by binding to cell surfacereceptors (Ekblom, 1996), contribute to morphogenesisby sending different signals to the developing cells.

EXPERIMENTAL PROCEDURESTissues

Mouse embryos were obtained by mating NMRI mice.The day of the appearance of the vaginal plug wasdesigned as day 0 of embryonic development. Tissuesfrom embryonic day 14 (E14) to postnatal day 1 (P1),were frozen in TissueTek (HistoLab, Miles Inc., Elkhart,IN), cut into 8- to 10-µm cryostat sections, and used forimmunohistochemistry or in situ hybridization.

Antibodies

Undiluted supernatant from rat monoclonal antibody200 (Sorokin et al., 1992) hybridoma culture, was usedto detect the E3 fragment of mouse laminin a1 chain.Undiluted medium from rat monoclonal antibody8G11D10 hybridoma culture (Schuler and Sorokin,1995) was used to detect laminin a2 chain. Laminin a5was detected with undiluted medium from rat monoclo-nal antibody 4G6 hybridoma culture (Sorokin et al.,1997). Cy3y-conjugated affinity purified goat anti-rat IgGpreparations were used as secondary antibodies (JacksonImmunoResearch Laboratories, West Grove, PA).

Immunohistochemistry

Methanol-fixed cryostat sections of embryonic miceheads were stained for immunohistochemical analysis.Non-specific binding sites were blocked by incubatingthe sections with 10% goat serum. After 1 hr incubationwith primary antibody, slides were washed in PBS(3 3 15 min) and incubated with secondary antibody for30–40 min, washed in PBS (3 3 15 min), and mountedwith Vectashield Mounting medium (Vector Laborato-ries). All incubations were made at room temperature.Slides were examined using a Zeiss Axiophot micro-scope. Control sections were incubated with 10% goatserum instead of the primary antibody.

In Situ Hybridization

45–50-mer oligonucleotides (Table 1) were selectedusing OLIGO 4.0y-software and purchased from Phar-macia Biotech Norden (Sollentuna, Sweden). With thesoftware we selected oligonucleotides which do not formstable dimers or hairpin loops. To exclude possiblehomology with other genes, the sequences of selectedoligonucleotides were analyzed against National Cen-ter for Biotechnology Information (NCBI) GenBankSequence database using the BLAST-program (Altschulet al., 1990). Oligonucleotide probes were 38-end labeledwith a35S-dATP, and in situ hybridization was per-formed according to Durbeej et al. (1993). Control sectionswere hybridized with the same amount of labeled probeplus unlabeled probe in 100-fold excess. No specificsignals could be detected in control sections. The slideswere exposed for 2–6 weeks at 14°C, developed, and

Fig. 6. Northern blot analysis of laminin a3A (5.5 kb) and a5 (12 kb)mRNA expression during mouse tooth development in vivo. Total RNAfrom E14, E17, P1, and P7 mouse teeth was used. A drastic increase ina3A expression was seen at P7. Strong a5 mRNA expression is seenthroughout embryonic development, decreasing at P7. The same filter washybridized with a G3PDH cDNA probe to show the amount of RNA loaded ineach lane. The positions of 18 S and 28 S rRNAare indicated to the right.


counterstained with hematoxylin. Photographs weretaken under brightfield and darkfield illumination us-ing a Zeiss Axiophot microscope.

cDNA Probes

To detect laminin a3A and B chain mRNA, a 564-bpcDNA fragment was constructed from E14 mouse lung

total RNA with reverse transcription-PCR using Gene-Amp RNA PCR Kit (Perkin Elmer; Roche MolecularSystems, Inc., Branchburg, NJ). The oligonucleotidesused to obtain the cDNA clone were 58-AGC CAC CTTTTC TTA TGT TGT TT-38, as upper primer, complemen-tary to nucleotides 3985–4007 of mouse laminin a3Achain and nucleotides 6499–6521 of mouse laminin a3B

Fig. 7. Laminin a4 mRNA expression in developing mouse tooth. Aand C are darkfield and B and D are brightfield illumination. A,B: At E14,expression is restricted to the mesenchymal cells surrounding theepithelial tooth bud. C,D: At P1, some expression is seen in mesenchymal

dental sac cells (arrows), while dental papilla no longer shows expression.Strong expression is seen in muscle. de, dental epithelium; dm, dentalmesenchyme; dp, dental papilla; mu, muscle; M1, first molar; M2, secondmolar. Bars 5 200 µm.

TABLE 1. Oligonucleotide Probes Used in In Situ Hybridization

Lamininchain Oligonucleotide probe used in in situ hybridization

Complementarynucleotides in mouse

mRNA (ref.)a1 58-GCC TTC CTT TTA ATG TAT TCT GTC TTG ACT GTG TGC CAC TTG CCA TCA CT-38 8,452–8,501 (Sasaki

et al., 1988)a2 58-CCA ATA GGA GTC GTC GAT GGT CAA GTC TGG AAA CCC TAC TCG GCC AAC-38 6,682–6,729 (Bernier

et al., 1995)a3A 58-GTC GCT GAA GAA ATG CTA CCC TTT GCT GCT CTG AGC CAT ACC CCA-38 100–144 (Galliano

et al., 1995)a3B 58-CTG CCA CCA TGC TGT TGC TGC CAG GGT TGA AGG ACA CAG GGA AAT-38 2,179–2,224 (Galliano

et al., 1995)a4 58-CAT GTG CAC GTT CAG GTA TTC CCC GTT GAC GCT GTG GCC ATG GAC-38 5,418–5,462 (Liu and

Mayne, 1996)a5 58-CAG GGG GCC CAT TGA TCA GCA GCT TCC TCC AGC ATC CTC GGT AGG CAG-38 10,718–10,765 (Miner

et al., 1995)


chain (Galliano et al., 1995), and 58-TCT TTG ACT TTCTTG ATG GTG AC-38, as lower primer, complementaryto nucleotides 4526–4548 of mouse laminin a3A chainand nucleotides 7040–7062 of mouse laminin a3B chain(Galliano et al., 1995). Reverse transcription was car-ried out with 1 µg RNA using random hexamer asprimers. The PCR cycles were 1 min 95°C, 2 min 56°C, 2min 72°C for 30 cycles. The fragment was excised from

a 2% agarose gel, purified using QIAquick gel extrac-tion kit (Quiagen, Chatsworth, CA), and cloned in theTA vector pCRII (InVitrogen, San Diego, CA). To con-firm that the clone represented laminin a3, about 150bp of the insert was sequenced. A 1.1-kb human G3PDHcDNA probe was used to show the amount of RNA

Fig. 8. Laminin a5 expression in developing tooth. A and C aredarkfield and B and D are brightfield illumination of in situ hybridization.A,B: At E14, strong expression of laminin a5 mRNA is seen in all epithelia.C,D: At P1, a5 mRNA is no longer expressed in functional ameloblasts,while its expression remains high in other epithelial structures. E:Immunohistochemical localization of laminin a5 in E14.5 mouse tooth.

Strong expression at all basement membrane (BM) areas is seen. F: AtP1, staining is lost from the dental BM (arrow), remaining strong in otherBM areas. de, dental epithelium; dm, mesenchyme; t, tongue; fa,functional ameloblasts; M1, first molar; M2, second molar. Bars 5 200 µm(A–D), 100 mm (E,F).


loaded (Clontech, Palo Alto, CA). Laminin a5 mRNAwas detected with a previously described 601 bp cDNAfragment (Durbeej et al., 1996).

Northern Blotting

Total RNA was isolated from E14, E17, P1, and P7mouse teeth, fractionated by agarose gel electrophore-sis, transferred to Zeta-Probe GT membrane (Bio-Rad,Hercules, CA), and hybridized with a32P-dCTP–labeledcDNA probes in 0.25 M Na2HPO4 and 7% sodiumdodecyl sulfate (SDS) at 65°C for 18–24 hr. Filter waswashed 2 3 1 hr in 20 mM Na2HPO4, 5% SDS and 2 3 1hr in 20 mM Na2HPO4, 1% SDS at 65°C. Membraneswere exposed to Hyperfilm MP films (Amersham Corp.,Arlington Heights, IL) at 270°C in the presence ofintensifying screens.

REFERENCESAberdam D, Aguzzi A, Baudoin C, Galliano M-F, Ortonne J-P, Me-

neguzzi G. Developmental expression of nicein adhesion protein(laminin-5) subunits suggests multiple morphogenetic roles. CellAdh Comm 1994a;2:115–129.

Aberdam D, Galliano MF, Vailly J, Pulkkinen L, Bonifas J, ChristianoAM, Tryggvason K, Uitto J, Epstein EH, Jr, Ortonne JP, MeneguzziG. Herlitz’s junctional epidermolysis bullosa is linked to mutationsin the gene (LAMC2) for the gamma 2 subunit of nicein/kalinin(LAMININ-5). Nature 1994b;6:299–304.

Adams JC, Watt F. Regulation of development and differentiation bythe extracellular matrix. Development 1993;117:1183–1198.

Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. Basic localalignment search tool. J Mol Biol 1990;215:403–410.

Bernier SM, Utani A, Sugiyama S, Doi T, Polistina C, Yamada Y.Cloning and expression of laminin alpha 2 chain (M-chain) in themouse. Matrix Biol 1995;14:447–455.

Chang AC, Wadsworth S, Coligan JE. Expression of merosin in thethymus and its interaction with thymocytes. J Immunol 1993;151:1789–1801.

Durbeej M, Soderstrom S, Ebendal T, Birchmeier C, Ekblom P.Differential expression of neutrophin receptors during renal develop-ment. Development 1993;119:977–989.

Durbeej M, Fecker L, Hjalt T, Zhang H-Y, Salmivirta K, Klein G, TimplR, Sorokin L, Ebendal T, Ekblom P, Ekblom M. Expression oflaminin a1, a5 and b2 chains during embryogenesis of the kidneyand vasculature. Matrix Biol 1996;15:397–413.

Ehrig K, Leivo I, Argraves WS, Ruoslahti E, Engvall E. Merosin, atissue-specific basement membrane protein, is a laminin-like pro-tein. Proc Natl Acad Sci USA 1990;87:3264–3268.

Ekblom P. Receptors for laminins during epithelial morphogenesis.Curr Opin Cell Biol 1996;8:700–706.

Ekblom M, Klein G, Mugrauer G, Fecker L, Deutzmann L, Timpl R,Ekblom P. Transient and locally restricted expression of laminin Achain mRNA by developing epithelial cells during kidney organogen-esis. Cell 1990;60:337–346.

Frieser M, Nockel H, Pausch F, Roder C, Hahn A, Deutzmann R,Sorokin LM. Cloning of the mouse laminin a4 chain. Expression in asubset of endothelium. Eur J Biochem 1997;246:727–735.

Galliano MF, Aberdam D, Aguzzi A, Ortonne JP, Meneguzzi G. Cloningand complete primary structure of the mouse laminin a3 chain. JBiol Chem 1995;270:21820–21825.

Grobstein C. Mechanisms of organogenetic tissue interaction. NatlCancer Inst Monogr 1967;26:279–299.

Iivanainen A, Sainio K, Sariola H, Tryggvason K. Primary structureand expression of a novel human laminin a4 chain. FEBS Lett1995;365:183–188.

Jernvall J, Kettunen P, Karavanova I, Martin LB, Thesleff I. Evidencefor the role of the enamel knot as a control center in mammalian

tooth cusp formation: non-dividing cells express growth stimulatingFgf-4 gene. Int J Dev Biol 1994;38:463–469.

Kadoya Y, Kadoya K, Durbeej M, Holmvall K, Sorokin L, Ekblom P.Antibodies against domain E3 of laminin-1 and integrin a6 subunitperturb branching epithelial morphogenesis of submandibular gland,but by different modes. J Cell Biol 1995;129:521–534.

Klein G, Langegger M, Timpl R, Ekblom P. Role of laminin A chain inthe development of epithelial cell polarity. Cell 1988;55:331–341.

Klein G, Ekblom M, Fecker L, Timpl R, Ekblom P. Differentialexpression of laminin A and B chains during development ofembryonic mouse organs. Development 1990;110:823–837.

Kjoelby M, Thesleff I, Sahlberg C, Fejerskov O, Josephsen K. Degrada-tion of the dental basement membrane during mouse tooth develop-ment in vitro. Int J Dev Biol 1994;38:455–462.

Leivo I, Engvall E. Merosin, a protein specific for basement mem-branes of Schwann cells, striated muscle, and trophoplast, isexpressed late in nerve and muscle development. Proc Natl Acad SciUSA 1988;85:1544–1548.

Lesot H, Osman M, Ruch JV. Immunofluorescent localization ofcollagens, fibronectin, and laminin during terminal differentiationof odontoblasts. Dev Biol 1981;82:371–381.

Liu J, Mayne R. The complete cDNA coding sequence and tissue-specific expression of the mouse laminin alpha 4 chain. Matrix Biol1996;15:433–437.

Lumsden AGS. Spatial organization of the epithelium and the role ofneural crest cells in the initiation of mammalian tooth germ.Development [Suppl] 1988;103:155–169.

Miner JH, Lewis RM, Sanes JR. Molecular cloning of a novel lamininchain, alpha 5, and widespread expression in adult mouse tissues. JBiol Chem 1995;270:28523–28526.

Orian-Rousseau V, Aberdam D, Fontao L, Chevalier L, Meneguzzi G,Kedinger M, Simon-Assman P. Developmental expression of lami-nin-5 and HD1 in the intestine: epithelial to mesenchymal shift forthe laminin g2 chain subunit deposition. Dev Dyn 1996;206:12–23.

Ruch JV. Determinisms of odontogenesis. Cell Biol Rev 1987;14:1–112.Sahlberg C, Reponen P, Tryggvason K, Thesleff I. Association between

the expression of murine 72kDa type collagenase by odontoblastsand basement membrane degradation during mouse tooth develop-ment. Arch Oral Biol 1992;37:1021–1030.

Sasaki M, Kleinmann HK, Huber H, Deutzmann R, Yamada Y.Laminin, a multidomain protein: the A chain has a unique domainand homology with the basement membrane proteoglycan and thelaminin B chains. J Biol Chem 1988;263:16536–16544.

Saxen L, Kohonen J. Inductive tissue interactions in vertebratemorphogenesis. Int Rev Exp Pathol 1969;8:57–128.

Schuler F, Sorokin L. Expression of laminin isoforms in mousemyogenic cells in vitro and in vivo. J Cell Sci 1995;108:3795–3805.

Sorokin L, Conzelmann S, Ekblom P, Battaglia C, Aumailley M, TimplR. Monoclonal antibodies against laminin A chain of fragment E3and their effects on binding to cell and proteoglycan and on kidneydevelopment. Exp Cell Res 1992;201:137–144.

Sorokin LM, Pausch F, Frieser M, Kroger S, Ohage E, Deutzmann R.Developmental regulation of the laminin a 5 chain suggests a role inepithelial and endothelial cell maturation. Dev Biol 1997 in press.

Thesleff I, Hurmerinta K. Tissue interactions in tooth development.Differentiation 1981;18:75–88.

Thesleff I, Barrach HJ, Foidart JM, Vaheri A, Pratt RM, Martin GR.Changes in the distribution of type IV collagen, laminin, proteogly-can, and fibronectin during mouse tooth development. Dev Biol1981;81:182–192.

Thomas T, Dziadek M. Genes coding for basement membrane glycopro-teins laminin, nidogen and type IV collagen in the embryonic lung:implications for branching morphogenesis. Mech Dev 1994;45:193–201.

Timpl R. Macromolecular organization of basement membranes. CurrOpin Cell Biol 1996;8:618–624.

Wright JT, Johnson LB, Fine JD. Developmental defects of enamel inhuman hereditary epidermolysis bullosa. Arch Oral Biol 1993;38:945–955.


Documents

Differential expression of laminin α chains during murine tooth development