Gene Expression Patterns 12 (2012) 68–76

Contents lists available at SciVerse ScienceDirect

Gene Expression Patterns

journal homepage: www.elsevier .com/locate /gep

Stam2 expression pattern during embryo development

Marija Curlin a,⇑, Katarina Kapuralin a, Andres F. Muro b, Francisco E. Baralle b, Kamal Chowdhury c,Srecko Gajovic a

a Croatian Institute for Brain Research, School of Medicine, University of Zagreb, Croatiab International Centre for Genetic Engineering and Biotechnology, Trieste, Italyc Max Planck Institute of Biophysical Chemistry, Department of Molecular Cell Biology, Goettingen, Germany

a r t i c l e i n f o a b s t r a c t

Article history:Received 13 August 2011Received in revised form 19 November 2011Accepted 21 November 2011Available online 28 November 2011

Keywords:STAM2Gene trapMouseEmbryoExpressionX-gal stainingDevelopment

1567-133X/$ - see front matter � 2011 Elsevier B.V.doi:10.1016/j.gep.2011.11.005

⇑ Corresponding author. Address: Croatian InstituteMedicine, University of Zagreb, Salata 12, 10000 Z4566792; fax: +385 1 4596942.

E-mail address: milcic@mef.hr (M. Curlin).

STAM2 is a tyrosine-phosphorylated protein suggested to be involved in cargo selection during endocyticpathway, regulation of exocytosis and intracellular signaling. Gene trap method was used to create viainsertional mutagenesis a mutant mouse line with integration of promoterless bgeo (lacZ–neomycin phos-photransferase fusion) gene in the second intron of Stam2 gene, enabling analysis of its in vivo expressionand function. The inserted b-galactosidase (lacZ) reporter gene was used to reveal Stam2 expression dur-ing development. Stam2 in situ RNA hybridization and immunostaining confirmed the observed b-galac-tosidase activity reflecting high Stam2 expression. The homozygous mutant mice showed no overtphenotypic alterations. Stam2 expression was detected after E9.5 in the gut, notochord, neural tubeand heart. In the nervous system it was located in the floor, roof and basal plates of the developing neuraltube, and in the developing cortex, hippocampus and olfactory bulbs. Toward the end of gestation, Stam2expression appeared in the testis and ovary, lungs, nasal cavity epithelium, kidneys, urogenital sinus,intestine, pancreas, pituitary and adrenal glands, muscles, brown adipose tissue, skin and epitheliumof the tongue and oral cavity.

� 2011 Elsevier B.V. All rights reserved.

Proper cellular response in dynamic environment during devel-opment requires constant monitoring of the incoming signals, theirsubsequent processing and appropriate reaction. An importantmechanism involved in this task is endocytosis, which throughmembrane transport regulates the cell sensitivity, modulates theduration of the signal, and assists in its intracellular processing.Molecules on the cell surface together with those from outsidecould be internalized into early endosomes, and either recycledback to the cell surface or degraded in lysosomes via multivesicularbodies (MVB). The sorting of the right cargo toward these two dif-ferent pathways is not completely understood, but monoubiquiti-nation and deubiquitination processes could play an importantrole (Raiborg et al., 2003; Kirkin and Dikic, 2007; Piper and Luzio,2007).

STAM1 and STAM2 (signal transduction adaptor molecules 1and 2) are considered to be involved in the endosomal sorting ofthe ubiquitinated cargo proteins for trafficking toward the lyso-some (Bache et al., 2003; Mizuno et al., 2003). Together with HRS(hepatocyte growth factor-regulated tyrosine kinase substrate)they form ESCRT-0 complex (Conibear, 2002; Hurley and Emr,

All rights reserved.

for Brain Research, School ofagreb, Croatia. Tel.: +385 1

2006; Williams and Urbé, 2007; Ren et al., 2009), the first of thefour protein complexes referred as ESCRT-0, -I, -II, and -III (i.e.endosomal sorting complexes required for transport). They recog-nize ubiquitinated membrane proteins and direct them toward theMVB pathway (Katzman et al., 2002; Luzio et al., 2009; Raiborg andStenmark, 2009).

Beside its suggested role in the membrane trafficking duringendocytosis, STAM2 is involved in the regulation of exocytosis, asdominant-negative mutants of Stam2 significantly inhibited IgEreceptor (FceRI)-triggered secretory response in RBL-2H3 mastcells (Murai and Kitamura, 2000). In addition, STAM2 seems toconnect these processes with intracellular signaling. STAM2 istyrosine-phosphorylated upon stimulation with a variety of cyto-kines and growth factors (Endo et al., 2000). Through its ITAM(immunoreceptor tyrosine-based activation motif) domain, STAM2is associated with JAK2 and JAK3 tyrosine kinases, and it is in-volved in the regulation of intracellular signal transduction forDNA synthesis and c-myc induction mediated by IL-2 and GM-CSF (Endo et al., 2000; Pandey et al., 2000; Hu et al., 2007).

Loss of function of STAM2 in mouse did not result with anyobvious abnormality (Yamada et al., 2002), while loss of STAM1caused growth retardation in the third week after birth and disap-pearance of hippocampal CA3 pyramidal neurons (Yamada et al.,2001). Double knockout mice for both Stam1 and Stam2 were lethalby embryonic day 11.5 (E11.5), with a defect in ventral folding

M. Curlin et al. / Gene Expression Patterns 12 (2012) 68–76 69

morphogenesis (Yamada et al., 2002) implying that STAMs areindispensable for embryo development.

To explore the possible function of STAM2 during mouse em-bryo development and to relate it to the observed embryo lethality,we analyzed in detail its in vivo expression pattern by tagging itendogenously with LacZ reporter via gene trap mutagenesis meth-od. No phenotypic alterations were detected as a consequence ofthe gene trap mutation. However, the introduced in frame lacZgene allowed us to analyze the developmental expression of Stam2demonstrating high Stam2 expression in the heart and nervous tis-sue, particularly in the differentiating layers of the telencephaliccortex.

1. Results

1.1. Production and characterization of Stam2Gt1Gaj mouse line

The Stam2Gt1Gaj mouse was produced using a large-scale genetrap approach (Gajovic et al., 1998). Gene trap vector used waspKC199bgeo (Thomas et al., 2000) containing the splice acceptorsequence from mouse Hoxc9 gene located upstream of the promot-erless bgeo (fused lacZ and neoR), which generates both b-galacto-sidase reporter and neomycin resistance activities. The vector wasintroduced into mouse embryonic stem (ES) cells by electropora-tion. Chimeric animals were produced from modified ES cellclones, and genes of interest were chosen according to their re-stricted expression pattern in the developing nervous system atE11.5. The expression was visualized by histochemical detectionof b-galactosidase, which was present in the tissue due to theintroduced splice acceptor site located upstream of the lacZ geneand subsequent lacZ transcription driven by endogenous promoterof the modified gene.

As the integration of gene trap vector into mouse ES cell gen-ome is random, the endogenous gene affected by the gene trapmutation had to be identified. The known sequence of the insertedgene trap vector enabled us to amplify and identify the targetedgene by 50 and 30RACE (rapid amplification of cDNA ends) methods(Frohman et al., 1988; Curlin et al., 2002). BLAST search has shownthat the obtained cDNA sequence represents an already knowngene, referred as Stam2 (Genbank NM019667) or Hbp (Hrs bindingprotein; Genbank AB012611; Takata et al., 2000). According to therestricted expression pattern in the developing neural tube andheart, the mouse line with modified Stam2 gene was investigatedfurther. Heterozygous carriers of the mutation were crossed toC57Bl/6NCrl mice, currently for 17 generations, in order to obtaina congenic mouse line that differs from the inbred strain only inthe modified locus.

Sequence analysis of the vector insertion site revealed that itwas integrated in the second intron of Stam2 gene (Fig. 1A). South-ern blot analysis with a vector specific probe excluded the possibil-ity of multiple gene trap vector insertions and confirmed that theused mouse line contained only specific Stam2 mutation (Fig. 1B).As a consequence of the mutation, 86% of the C-terminal wild typeSTAM2 protein was expected to be missing from the mutant pro-tein, leaving only 41 N-terminal amino acid residues (of the totalof 523 amino acids; the product of the first two of total of 14exons) fused to b-galactosidase and neomycin phosphotransferaseprotein encoded by the inserted bgeo gene.

The homozygous carriers of the gene trap mutation were pro-duced by intercrossing heterozygous mice. They were identifiedby Southern blotting with Stam2 specific probe or by PCR(Fig. 1A, C and D). Both heterozygous and homozygous Stam2Gt1Gaj

mice carrying the gene trap mutation were born with the expectedMendelian ratio (v2 test p > 0.1), had no macroscopic abnormalitiesand were viable, fertile and had a normal life span.

1.2. b-Galactosidase activity in the developing mutant mouse embryos

To determine the expression pattern of Stam2 gene, whole-mount histochemical X-gal staining to visualize b-galactosidaseactivity (i.e. to visualize in-frame-inserted lacZ expression pattern)was performed on the mutant and, as a control, on the wild typemouse embryos from E8.5 to E18.5 (Table 1, Fig. 2). Cryosectionsof E16.5 embryos were also prepared and stained by X-gal (Fig. 3B).

1.2.1. Stage E8.5At stage E8.5, no b-galactosidase activity was found.

1.2.2. Stage E9.5The earliest b-galactosidase activity was detected at E9.5, when

it was present in the hindgut (Fig. 2A).

1.2.3. Stage E10.5At E10.5, the X-gal staining was not visible in the gut; however

the b-galactosidase activity was present in the heart, where it per-sisted afterwards throughout gestation (Figs. 2B and C, 3A and B).

1.2.4. Stage E11.5At E11.5, in addition to the heart, b-galactosidase was detected

in the notochord and neural tube (Fig. 2B and D). In the neural tubeit was located in the floor plate and its ventral part correspondingto the basal plates (Fig. 2D).

1.2.5. Stage E12.5The basic pattern of b-galactosidase activity observed at stage

E11.5 was slightly modified at stage 12.5 and later. At 12.5 the b-galactosidase activity was present in the heart and neural tube,but disappeared in the notochord. In the neural tube, the expres-sion was present in the basal plates, floor plate and additionallyin the roof plate. As a background the newly formed choroid plexusshowed diffused positive X-gal staining. This background stainingpersisted afterwards throughout the gestation.

1.2.6. Stage E13.5Together with the expression in the heart and developing spinal

cord, b-galactosidase activity was detected in the floor plate of thebrain stem (medulla oblongata and pons). In the developing spinalcord the expression included not only the floor plate but as wellthe future raphe region. It was not any more present in the basalplates. Low b-galactosidase activity was detected in the testes.Background X-gal staining appeared in the stomach and intestinalmucosa and continued there until the end of gestation.

1.2.7. Stage E14.5At E14.5 b-galactosidase activity was present in the developing

spinal cord (the floor plate, raphe region, the roof plate and weaklyin the gray matter). In the medulla oblongata and pons, the b-galactosidase was present in the floor plate, in the most ventralpart of the raphe and in the gray matter (Figs. 2E, 3A). At this stagea b-galactosidase activity appeared in the telencephalon, i.e. in theintermediate zone of the neopallial cortex, olfactory bulbs, trigem-inal and dorsal root ganglia, pituitary, lungs, nasal cavity epithe-lium, pancreas, urogenital sinus, skin and muscles in the headand neck, tongue, back and limbs. The activity in the testes was vis-ible in the future seminiferous tubules.

1.2.8. Stage E15.5The regions of b-galactosidase activity in E15.5 embryos corre-

sponded to the regions of its activity detected in E14.5 embryos,but the staining was generally stronger. The staining in the heartwas more intensive in the atrium than in ventriculum.

Fig. 1. (A) Schematic diagram of bgeo insertion into intron sequence of Stam2 gene between the second and third exon of Stam2. DNA probes used for the Southern blotexperiments and the primers used for the PCR genotyping were indicated. The insertion site between base 30139010 and 30139009 of the chromosome 2 (GenbankNT_039206) is indicated. (B) Insertion of gene trap vector occurred only once in Stam2 gene. Southern blot analysis shows hybridization of genomic DNA with a vector specificbgeo DNA probe. DNA was cut with restriction enzyme BglI. The 2.9 kb signal represents a mutant Stam2 allele, and is visible in homozygous (hm) and heterozygous (ht), butnot in the wild type (wt) mice. (C) Southern blot detection of homozygous carriers of gene trap modification of Stam2 gene. Southern blot analysis shows hybridization ofgenomic DNA with Stam2 DNA probe. Genomic DNA was cut by restriction enzyme SacI. The 3.5 kb signal represents a mutant Stam2 allele (visible in homozygous (hm) andheterozygous (ht) mice) and the 1.4 kb signal represents a wild type allele (visible in heterozygous (ht) and wild type (wt) mice). (D) PCR-detection of homozygous carriers ofgene trap modification of Stam2. Six hundred bp product is a result of the wild type DNA amplification using LOB31 and LOB5A primers, while 800 bp product corresponds tomutated allele and it is obtained by LOB31 and lac2 primers.

70 M. Curlin et al. / Gene Expression Patterns 12 (2012) 68–76

1.2.9. Stage E16.5At this stage the b-galactosidase activity appeared in the ven-

tricular and marginal zone and cortical plate of the cortex and inthe hippocampus (Figs. 3B, 4A and C). The b-galactosidase activitywas intense in the superficial intermediate zone, and weak in themarginal zone. In the roof plate of the developing spinal cord b-galactosidase activity disappeared, but persisted in the floor plate,raphe and gray matter. The staining in the gray matter, raphe andfloor plate of the brain stem was extended to the ventral ventricu-lar layer of the fourth ventricle. The b-galactosidase activity contin-ued to be present in the heart (equally in the atrium as in theventriculum), testes, olfactory bulbs, trigeminal and dorsal rootganglia, pituitary (both adenohypophysis and neurohypophysis),tongue, lungs, nasal cavity epithelium, pancreas, urogenital sinus,skin and muscles and appeared in the mucosa covering the tongue,kidneys, spleen, medulla of the adrenal glands and brown adiposetissue in the back region (Figs. 2J, 3B and C, 4A).

1.2.10. Stage E17.5At E17.5 the strong b-galactosidase activity in the intermediate

zone extended through the subplate zone and included the corticalplate of the telencephalon (Fig. 2F). The recently formed hippocam-pus showed strong b-galactosidase activity. The activity continuedto be present in the raphe, floor plate and gray matter of the devel-oping spinal cord and brain stem, and in the ventral ventricularlayer of the fourth ventricle. b-Galactosidase activity was strongin all other b-galactosidase positive regions detected in the E16.5embryos and was detected also in the ovaries.

1.2.11. Stage E18.5At E18.5 the b-galactosidase activity occupied the whole region

from the intermediate to the marginal zone (Fig. 2G). It disap-peared from the raphe region and it was weaker in the floor plateof the developing spinal cord and brain stem, but got stronger inother regions of the nervous system that were positive at E17.5(i.e. cortex, hippocampus and olfactory bulbs, gray matter of thespinal cord and pons). It got stronger also in the testes, ovaries,

lungs, kidneys, urogenital sinus (bladder), pituitary and adrenalglands (Fig. 2H, I, L and M). Besides skin and tongue mucosa it ap-peared in the oral cavity mucosa (Fig. 2K).

1.3. b-Galactosidase activity corresponded to Stam2 expression

In order to confirm that the observed b-galactosidase activitypattern indeed reflects the endogenous Stam2 expression, in situRNA hybridization using Stam2 specific probe and X-gal stainingwere compared on the cryo sections of E16.5 embryos. In situRNA hybridization on the wild type embryos showed a weak ubiq-uitous Stam2 expression with distinguished higher expression inthe nervous tissue, lungs, heart, digestive and endocrine system(Fig. 3B, Table 2). X-gal staining was more restricted than in situRNA hybridization (Fig. 3A). The ubiquitous low level of Stam2mRNA was not visible with X-gal staining, and the areas with highmRNA presence were wider than those outlined by X-gal. Forexample, a clear laminar distribution of b-galactosidase activityin the brain cortex was visible, while mRNA distribution in the cor-tex was rather diffused. In situ RNA hybridization revealed highStam2 expression in the intestine which was undetectable due tothe presence of high background by X-gal staining. Also, a weaksignal in the area of the choroid plexus indicated Stam2 expressionin this structure. The Stam2 expression revealed by X-gal stainingmatched the regions of high expression of Stam2 mRNA, i.e. itshowed more restricted pattern than in situ RNA hybridization.

Another verification of observed X-gal staining was obtained byimmunohistochemistry using STAM2 and b-galactosidase antibod-ies on the heterozygous mutant E11.5 and E16.5 embryos. Immuno-histochemistry of both antibodies revealed strong fluorescentsignal in the Stam2 expressing regions identified by X-gal staining(Table 2). However, both were less restricted than X-gal staining,i.e. they demonstrated weak ubiquitous signal that was not visibleby X-gal staining. As well as b-galactosidase activity, the STAM2and b-galactosidase immunostainings were strong in the superficialintermediate zone of the telencephalic cortex in E16.5 embryos, butalso were present throughout the cortex in the ventricular and

Table 1Regional distribution of Stam2 expression assessed by X-gal staining. Only tissues and organs showing lacZ expression are outlined. b – Background staining; ‘‘�’’ – no signal; ‘‘�/+’’, ‘‘+’’, ‘‘++’’ and ‘‘+++’’ – arbitrary estimated intensities of signal strength.

Embryonic day E9.5 E10.5 E11.5 E12.5 E13.5 E14.5 E15.5 E16.5 E17.5 E18.5

Notochord +

Nervous systemSpinal cord (roof plate) � � � + + + + � �Spinal cord (basal plates) � + �/+ � � � � � �Spinal cord (floor plate) � � + + + + + + + �/+Spinal cord (raphe) � + + + + + �Spinal cord (gray matter) �/+ + + + +Oblongata, pons (floor plate) � + + + + + �/+Oblongata, pons (raphe) � � + + + �/+ �Oblongata, pons (gray matter) �/+ + + + +Cortex (ventricular zone) � � �/+ �/+ +Cortex (intermediate zone) + + ++ + +Cortex (cortical plate) �/+ + ++Cortex (marginal zone) �/+ + +Hippocampus � �/+ + +Olfactory bulbs + + + + +Trigeminal ganglia �/+ + + + +Dorsal root ganglia �/+ + + + +Choroid plexus b b b b b

HeartAtrium � + + + + + ++ ++ ++ ++Ventriculum � + + + + + + + + +Spleen �/+ �/+ �/+Lungs � � � � � �/+ + + + +Nasal cavity mucosa �/+ + + + +

Digestive systemHindgut + �Intestine and stomach b b b b b bOral cavity mucosa � � � � +Tongue + + ++ ++ ++Pancreas �/+ �/+ + + +

Urogenital systemKidneys � � � � � � � �/+ + +Urogenital sinus �/+ + + + +Phalus � � + + +

Endocrine systemPituitary �/+ + + + +Adrenal glands (medulla) + + +Testes + + ++ +++ +++ +++Ovaries � + +Skin + + ++ ++ ++Brown adipose tissue �/+ �/+ �/+Muscles �/+ + + + +

M. Curlin et al. / Gene Expression Patterns 12 (2012) 68–76 71

intermediate zones (Fig. 3C and D). Similarly, the STAM2 and b-galactosidase immunostainings were strong in the ventral part ofthe neural tube of E11.5 embryos but also were present throughoutthe whole section of the neural tube. Other organs and tissues didnot show significant difference between strong fluorescence signalsand X-gal staining. The colocalization immunohistochemistry studyof STAM2 protein and b-galactosidase revealed STAM2 and b-galac-tosidase in the same tissues and cells, but the intracellular distribu-tion of those proteins was different. STAM2 was present in the cellbodies and cell processes, but the signal was very intense in punc-tuate structures in the cytoplasm and in some cells in the nucleus(Fig. 3E). The immunohistochemical signal of b-galactosidase washomogenous throughout the cytoplasm and it was not present inthe nucleus. Tiny dots representing accumulations of b-galactosi-dase were visible in some cells as documented previously (Eyerand Peterson, 1994; Letournel et al., 2006).

2. Discussion

The present study represents the first comprehensive mappingof Stam2 expression during mouse embryo development. X-gal

histochemistry of the molecularly tagged Stam2 gene in theStam2Gt1Gaj mouse line allowed analysis of Stam2 expressionpattern in the embryos. The location of gene trap vector withinStam2 gene was confirmed by Southern blot and PCR analyses(Fig. 1B), and the observed lacZ expression pattern wasadditionally verified by in situ RNA hybridization and immunohis-tochemistry (Fig. 3). In situ RNA hybridization and immunohisto-chemistry revealed two levels of Stam2 expression: low andubiquitous vs. high and specific. Regions with strong signal ob-served by in situ RNA hybridization and immunohistochemistryoverlapped the regions that showed b-galactosidase activity. Thelower sensitivity of X-gal histochemistry was expected due to dif-ferent detection levels of the three methods used (mRNA, protein,and enzyme activity). It has been shown that X-gal histochemistryunderestimates the extent of gene expression while immunohisto-chemistry has the superior sensitivity (Pereira et al., 2006; Coue-gnas et al., 2007). In addition, the lower intensity of the X-galstaining could be due to possible instability of the mutant protein.Still identifying the regions with high levels of Stam2 expressionby X-gal staining could be not considered a disadvantage, as it re-veals restricted positive areas most relevant for the STAM2 func-tion research.

Fig. 2. Visualization of b-galactosidase activity in the representative heterozygous mouse embryos by histochemical staining with X-gal. (A and B) Lateral views of E9.5 andE11.5 embryos, respectively, (C) the developing heart of E11.5 embryo, (D) cross section of the neural tube of E11.5 embryo, (E) sagital section of E14.5 embryo head, (F)sagital section of E17.5 embryo telencephalon, (G) transversal section of E18.5 brain, (H) the testis, and (I) the ovary at E18.5, (J) cryo section of the skin at E16.5, (K) thetongue and oral cavity epithelia at E18.5, (L) the pituitary, and (M) the adrenal gland at E18.5. AH – adenohypophysis, C – cortex, CP – cortical plate, FP – floor plate, H – heart,HC – hippocampus, HG – hindgut, IZ – intermediate zone, MA – medulla of adrenal gland, MO – medulla oblongata, NT – neural tube, N – notochord, NH – neurohypophysis,OCE – oral cavity epithelium, P – pons, SC – spinal cord, TE – tongue epithelium, TS – seminiferous tubules. White bars – 1 mm; black bars – 0.1 mm.

72 M. Curlin et al. / Gene Expression Patterns 12 (2012) 68–76

The endocytic pathway can be important for the embryonicdevelopment in several ways, but the most intriguing is the estab-lishment of the gradient of morphogens via planar transcytosis,and their subsequent selective uptake and degradation. The candi-date regions for these processes correspond to the regions of Stam2expression and include the notochord, floor plate and roof plate ofthe neural tube, telencephalic cortex, hippocampus and the heart.STAM2 was as well suggested to participate in exocytic extrusionof synaptic and other secretory granules (Murai and Kitamura,2000; Komada and Kitamura, 2001). One of the regions of Stam2expression was the nervous tissue and the endocrine organs likepituitary and adrenal glands near the end of gestation, all of themat this stage already highly secretory active. Other regions ofexpression are polarized epithelial tissues in the skin, tongue andoral cavity, where sorting and correct delivery of the proteins todifferent membrane domains depends on accurate vesicle traffick-ing and sorting (Nelson and Yeaman, 2001; Altschuler et al., 2003).

In addition, Stam2 showed an intriguing strong and specificexpression pattern in the cortex of the telencephalon (neopallialcortex). An appearance of strong Stam2 expression in the interme-diate zone at the time of the cortex differentiation and zone strat-ification and further spreading of the expression into the corticalplate during the intensive neural migration and cortical layers for-mation indicate a possible Stam2 function in the processes of thecortex differentiation comprising complex cell specifications.

When compared to other members of the ESCRT-0 complex, Hrsand STAM1, the observed Stam2 expression was more restrictedthan Hrs expression in mouse and rat embryos (Avantaggiato et al.,1995; Komada and Soriano, 1999; Tsujimoto et al., 1999; Yamadaet al., 2001; Tamai et al., 2008). Furthermore, a comparison of ourstudy on Stam2 and published data on Stam1 expression (Yamadaet al., 2001) reveals that Stam2 and Stam1 mRNA distributions over-lap. On the other hand our b-galactosidase activity study showedthat Stam2 is expressed in the organs where Stam1 expression was

Fig. 3. Visualization of b-galactosidase activity in the representative cryosections of homozygous mouse embryos by histochemical staining with X-gal. Bar (the same for A, Band C) – 1 mm. (A) b-Galactosidase activity on the sagital cryosection of E14.5 embryo. (B) b-Galactosidase activity on the sagital cryosection of E16.5 embryo. (C) b-Galactosidase activity on the transversal cryosections of E16.5 embryo. The locations of the sections are indicated by horizontal arrows in the B. IZ – intermediate zone of theneopallial cortex, P – pituitary, TG – grigeminal ganglion, MO – medulla oblongata, SC – spinal cord, DRG – dorsal root ganglion, T – tongue, M – skeletal muscles, BA – brownadipose tissue, Ht A – heart atrium, Ht V – heart ventriculum, N – nasal cavity mucosa, Lu – lungs, L – liver, G – gut, K – kidney, S – spleen, Pa – pancreas, US – urogenital sinus,Ph – phalus.

M. Curlin et al. / Gene Expression Patterns 12 (2012) 68–76 73

not reported: skin, oral and nasal cavity mucosa, urogenital sinus,pancreas, muscles, testis, ovary, adrenal glands and pituitary.Expression in the nasal cavity mucosa and olfactory bulbs is in accor-dance to the described Stam2 expression in the olfactory system inadult mice (Furic Cunko et al., 2008).

The insertion of the pKC199bgeo vector into the second intronof Stam2 gene was expected to impair its function. However, asmentioned above, no overt phenotype was detectable in the homo-zygotes. Similar result was also reported in the case of the com-plete loss of STAM2 function in homozygous mouse mutantsgenerated by gene targeting mutagenesis (Yamada et al., 2002).The embryonic lethality of Stam1 and Stam2 double knockout in-deed confirms that STAMs are important for the embryonic devel-opment. One of the reasons that the loss of STAM2 duringdevelopment was not accompanied with phenotypic changes couldbe the compensation of its loss by STAM1. Although current datashow differences in their expression pattern, they do posses sev-eral overlapping expression domains. Our data actually suggestfor Stam2 two different expression levels, one low which couldbe detected as a low signal by in situ RNA hybridization and immu-nohistochemistry. This differs to stronger level of expression whichwas clearly detected by all three methods. If the same is valid forStam1, we could speculate that these low levels of expression couldhelp to compensate loss of either STAM1 or STAM2 in case of theirindividual loss of function, but in case of double loss of function,their importance for development is revealed and results with em-bryo lethality.

3. Conclusions

In conclusion, we provide new data revealing that the main re-gions of Stam2 expression during development are confined to thedeveloping nervous system, heart, lungs, skin, tongue, oral cavityepithelia, intestine, pancreas, kidney, urogenital sinus, phalus, tes-tis, ovary, pituitary and adrenal glands, muscles, and brown adi-pose tissue. These data provide additional information for further

analyzes of the functional roles of the STAM proteins duringdevelopment.

4. Experimental procedures

4.1. Animals

This study was performed on heterozygous E9.5–E18.5 embryosfrom the gene trap mouse line Stam2Gt1Gaj and wild type inbredstrain C57Bl/6NCrl.

Gene trap vector pKC199bgeo, containing splice acceptor ofHoxc9 gene in frame with bgeo (lacZ/neomycin phosphotransferasefusion gene) upstream to the SV40 polyadenylation signal (Thomaset al., 2000), was used for transformation of ES cells by electropor-ation. After selection for 10 days by 250 mg/ml G418 (GIBCO BRL,Gaithersburg, MD), resistant clones were isolated and stained withX-gal for the presence of b-galactosidase. The positive clones wereexpanded; the chimeras were generated by morula aggregationand germ line transmission achieved by mating them withC57Bl/6 mice. The mouse line Stam2Gt1Gaj was selected for furtherinvestigation and kept as heterozygous through subsequent breed-ing with C57Bl/6NCrl mice. Homozygous animals of Stam2Gt1Gaj

line were obtained by intercrossing of heterozygotes.All the experiments were approved by institutional Ethical

Committee and were in agreement with the Croatian Society forLaboratory Animal Science and the International Council for Labo-ratory Animal Science.

4.2. Sequence determination of the trapped gene

The sequence of the trapped gene was obtained by 50 and30RACE (Frohman et al., 1988; Curlin et al., 2002). The obtainedsequence was compared to the known cDNA sequences by BLASTsearch at the National Center for Biotechnology Information(http://www.ncbi.nlm.nih.gov).

Fig. 4. Stam2 expression in E11.5 and E16.5 embryos detected by in situ RNA hybridization and immunohistochemistry confirmed the expression pattern obtained by X-galstaining. (A) b-Galactosidase activity on the cryosection of E16.5 embryo. (B) Stam2 in situ RNA hybridization on E16.5 embryos. Left – antisense (AS) Stam2 specific probe,right – sense (S) Stam2 specific probe, as negative control. Bars in (A) and (B) – 1 mm. (C) b-Galactosidase activity on the sagital section of the telencephalic cortex of E16.5embryo, (D) STAM2 and b-galactosidase immunohistochemistry of the sagital section of the telecenphalic cortex of E16.5 embryo. Bars in (C) and (D) – 0.1 mm. (E) STAM2 andb-galactosidase immunohistochemistry colocalisation in the ventral part of the neural tube of E11.5 embryo, bar – 10 lm. Ct – neopallial cortex, P – pituitary, SC – spinal cord,DRG – dorsal root ganglion, NA – cartilage primordia of neural arches, T – tongue, M – skeletal muscles, Ht V – heart ventriculum, Ht A – heart atrium, Lu – lungs, L – liver, G –gut, K – kidney, AR – adrenal gland, MZ – marginal zone, CP – cortical plate, IZ – intermediate zone, VZ – ventricular zone.

74 M. Curlin et al. / Gene Expression Patterns 12 (2012) 68–76

4.3. Southern blotting and PCR

DNA for Southern blot analysis was isolated from the adult livermouse tissue and cut by restriction enzymes BglI or SacI (Promega).The digoxygenin (DIG-dUTP, Boehringer Mannheim) labeled DNAprobes were produced by PCR using bgeo specific primers(50TTGGCGTAAGTGAAGCGAC30 and 50AGCGGCTGATGTTGAACTG30)and Stam2 gene specific primers (50TGACAGGGTTGGAAGCACTC30

and 50ATAAAGCTCTGACTCTCCG30). Both, bgeo and Stam2 DNAprobes were approximately 500 bp long. The detection of the tem-plate – probe hybrids was performed using a chemiluminescencesubstrate ‘‘CDP Star’’ (Boehringer Mannheim). For PCR genomicDNA was isolated from the tail tips and three primers used for

genotyping were LOB31 (50GCTTTACAGTGGGGATACAT30), LOB5A(50TTATGGCTTTTAGGCAATCT30) and lac2 (50CTGCAAGGCGATTAAGTTGG30).

4.4. Detection of b-galactosidase activity

4.4.1. Whole mount stainingThe heterozygous and wild type (negative control) embryos

were obtained by mating heterozygous mutant males with femalesof C57Bl/6 strain. The morning of the vaginal plug was taken asstage E0.5. The isolated embryos were fixed in a mixture of 2% form-aldehyde and 0.2% glutaraldehyde (Sigma) in phosphate bufferedsaline (PBS) for 1 h either as whole embryos (E8.5–E12.5) or cut

Table 2Stam2 expressing regions identified by X-gal staining were confirmed by in situ RNAhybridization and immunohistochemistry. b – Background staining; ‘‘�’’ – no signal;‘‘�/+’’, ‘‘+’’, ‘‘++’’ and ‘‘+++’’ – arbitrary estimated intensities of signal strength.

Embryonic day E11.5 E16.5

X-gal IHC X-gal ISH IHCNotochord + +

Nervous systemSpinal cord (roof plate) � + � + �/+Spinal cord (basal plates) + ++ � + �/+Spinal cord (floor plate) + +++ + + +Spinal cord (raphe) + + +Spinal cord (gray matter) + + +Oblongata, pons (floor plate) + + +Oblongata, pons (raphe) + + +Oblongata, pons (gray matter) + + +Cortex (ventricular zone) �/+ + +Cortex (intermediate zone) ++ ++ +++Cortex (cortical plate) �/+ + �/+Cortex (marginal zone) + + +Hippocampus �/+ + +Olfactory bulbs + + +Trigeminal ganglia + + +

HeartAtrium + + ++ + ++Ventriculum + + + + +Lungs + + +Nasal cavity mucosa + + +

Digestive systemIntestine and stomach b + +Oral cavity mucosa � � �/+Tongue ++ �/+ +Kidneys + + +Urogenital sinus + + +

Endocrine systemPituitary + + +Testes +++ + +Skin ++ ++ ++Muscles + + +

M. Curlin et al. / Gene Expression Patterns 12 (2012) 68–76 75

sagitally or transversally (E13.5 and later). After fixation they wererinsed three times in PBS, and incubated overnight at 37 �C in the X-gal solution (1 mg/ml X-gal, 10 mM K3Fe(CN)6, 10 mM K4Fe(CN)6,and 2 mM MgCl2 in PBS), washed in PBS for 24 h and cleared inascending concentrations of glycerol in PBS up to 70%.

4.4.2. Cryosection stainingThe homozygous and wild type embryos were obtained by mat-

ing heterozygous female with heterozygous male mouse. The geno-type of the embryos was determined by PCR with DNA isolatedfrom the embryonic tail. The isolated embryos were fixed for 1 hin a mixture of 2% formaldehyde and 0.2% glutaraldehyde (Sigma)in phosphate buffered saline (PBS), washed two times in the PBSand frozen to �80 �C by 1 min immersion in the prefreezed isopen-tane. The selected embryos were cut on a cryostat (Leica CM3000,Leica Instruments GmbH, Germany) into 40–50 lm thick sections,mounted on the glass slides, air-dried for 1–3 h, washed two timesin PBS, incubated in the X-gal solution overnight at 37 �C. Thestained sections were washed in PBS, dehydrated in ascending con-centration of ethanol, cleared in Histo-Clear (National Diagnostics,USA) and covered with coverslips using mounting agent Histo-mount (National Diagnostics, USA). The micrographs of the sectionswere taken by computer scanning (EPSON EFFECTION 4870PHOTO).

4.5. In situ hybridization

In situ RNA hybridization for Stam2 gene was performed by amodified method described previously (Sperk et al., 1992). DNA

oligonucleotides complementary to mouse Stam2 gene (50-GGATGACAGTGACGCCAACTGGTGGCAAGGAGAAAATCACAGAGG-30) were labeled with [35S] thio-dATP (Perkin–Elmer) using termi-nal deoxynucleotidyltransferase (Roche). Unfixed frozen mouseembryos at E16.5 were cut on a cryostat (Leica CM3000, LeicaInstruments GmbH, Germany) into 20 lm thick sections andthaw-mounted on poly-L-lysine coated slides (Menzel–Glaser).After fixation by immersion in 2% paraformaldehyde, sections wererinsed in 0.1 M PBS (pH 7.4) and were acetylated with 0.25% aceticanhydride in triethanolamine (pH 8.2). The sections were furtherrehydrated in ascending series of alcohol. Hybridization was per-formed overnight at 53 �C in the hybridization buffer supple-mented with dextran sulfate, dithiothreitol (Sigma), and the 35Slabeled oligonucleotide probe. Next day the sections were washedwith sodium citrate buffer at 60 �C and dipped in 70% ethanol. Thesections were further exposed to Kodak BioMax MR Film for2 weeks at room temperature.

4.6. Immunohistochemistry

Two different stages of embryos were used for immunohisto-chemical analysis (E11.5 and E16.5). Whole embryos were fixedby immersion in 4% paraformaldehyde (Sigma) at 4 �C overnight.Embryos were rinsed in PBS and then transferred to 10% sucrosefollowed by 30% sucrose in PBS at 4 �C. Sagital sections were cuton a cryostat at �20 �C. Thickness of the sections was 30 lm for16.5 day-old embryos and 40 lm for 11.5 day-old embryos. Sec-tions were mounted on SuperFrost microscope slides (Menzel–Glaser).

Section were further incubated with primary antibodies againstSTAM2 (Rabbit polyclonal, diluted 1:50, Abcam, ab63372) and b-galactosidase (Chicken polyclonal, diluted 1:50, Abcam, ab9361)at 4 �C overnight. Primary antibodies were diluted in PBS contain-ing 0.2% Triton X-100 and 2% goat serum. Fluorescent secondaryantibody incubation was carried out for 2.5 h at room temperature.The secondary antibodies were: Alexa Fluor 488 goat anti-rabbit(Invitrogen, A11008) and Alexa Fluor 546 goat anti-chicken (Invit-rogen, A11040). All secondary antibodies were diluted in a PBS at aconcentration 1:200. Sections were rinsed in PBS and coverslippedusing Fluoromount (Sigma–Aldrich) mounting media. Fluorescentlabeling was viewed using a confocal microscope Zeiss LSM 510Meta.

Acknowledgments

This work was supported by Grants of Ministry of Science andTechnology, Republic of Croatia (108-1081870-1902), and Unityfor Knowledge Fund, Republic of Croatia (UKF 35/08). The produc-tion of the gene trap mutant mouse line was supported by the MaxPlanck Society, Germany.

References

Altschuler, Y., Hodson, C., Milgram, S.L., 2003. The apical compartment: traffickingpathways, regulators and scaffolding proteins. Curr. Opin. Cell Biol. 15, 423–429.

Avantaggiato, V., Torino, A., Wong, W.T., Di Fiore, P.P., Simeone, A., 1995. Expressionof the receptor tyrosine kinase substrate genes eps8 and eps15 during mousedevelopment. Oncogene 11, 1191–1198.

Bache, K.G., Raiborg, C., Mehlum, A., Stenmark, H., 2003. STAM and Hrs are subunitsof a multivalent ubiquitin-binding complex on early endosomes. J. Biol. Chem.278, 12513–12521.

Conibear, E., 2002. An ESCRT into the endosome. Mol. Cell. 10, 215–216.Couegnas, A., Schweitzer, A., Andrieux, A., Ghandour, M.S., Boehm, N., 2007.

Expression pattern of STOP lacZ reporter gene in adult and developing mousebrain. J. Neurosci. Res. 85, 1515–1527.

Curlin, M., Kostovic-Knezevic, L., Gajovic, S., 2002. Gene trap mutagenesis of threegenes expressed during mouse embryo development. Periodicum. Biologorum.104, 47–54.

76 M. Curlin et al. / Gene Expression Patterns 12 (2012) 68–76

Endo, K., Takeshita, T., Kasai, H., Sasaki, Y., Tanaka, N., Asao, H., Kikuchi, K., Yamada,M., Chenb, M., O’Shea, J.J., Sugamura, K., 2000. STAM2, a new member of theSTAM family, binding to the Janus kinases. FEBS Lett. 477, 55–61.

Eyer, J., Peterson, A., 1994. Neurofilament-deficient axons and perikaryal aggregatesin viable transgenic mice expressing a neurofilament-beta-galactosidase fusionprotein. Neuron 12, 389–405.

Frohman, M.A., Dush, M.K., Martin, G.R., 1988. Rapid production of full-lengthcDNAs from rare transcripts: amplification using a single gene specificoligonucleotide primer. Proc. Natl. Acad. Sci. USA 85, 8998–9002.

Furic Cunko, V., Mitrecic, D., Mavric, S., Gajovic, S., 2008. Expression pattern andfunctional analysis of mouse Stam2 in the olfactory system. Coll. Antropol. 1(Suppl. 32), 59–63.

Gajovic, S., Chowdhury, K., Gruss, P., 1998. Genes expressed after retinoic acid-mediated differentiation of embryoid bodies are likely to be expressed duringembryo development. Exp. Cell. Res. 242, 138–143.

Hu, X., Chen, J., Wang, L., Ivashkiv, L.B., 2007. Crosstalk among Jak-STAT, Toll-likereceptor, and ITAM-dependent pathways in macrophage activation. J. LeukocBiol. 82, 237–243.

Hurley, J.H., Emr, S.D., 2006. The ESCRT complexes: structure and mechanism of amembrane-trafficking network. Annu. Rev. Biophys. Biomol. Struct. 35, 277–298.

Katzmann, D.J., Odorizzi, G., Emr, S.D., 2002. Receptor downregulation andmultivesicular-body sorting. Nat. Rev. Mol. Cell. Biol. 3, 893–905.

Kirkin, V., Dikic, I., 2007. Role of ubiquitin- and Ubl-binding proteins in cellsignaling. Curr. Opin. Cell Biol. 19, 199–205.

Komada, M., Kitamura, N., 2001. Hrs and hbp: possible regulators of endocytosisand exocytosis. Biochem. Biophys. Res. Commun. 281, 1065–1069.

Komada, M., Soriano, P., 1999. Hrs, a FYVE finger protein localized to earlyendosomes, is implicated in vesicular traffic and required for ventral foldingmorphogenesis. Genes Dev. 13, 1475–1485.

Letournel, F., Bocquet, A., Perrot, R., Dechaume, A., Guinut, F., Eyer, J., Barthelaix, A.,2006. Neurofilament high molecular weight-green fluorescent protein fusion isnormally expressed in neurons and transported in axons: a neuronal marker toinvestigate the biology of neurofilaments. Neuroscience 137, 103–111.

Luzio, J.P., Parkinson, M.D., Gray, S.R., Bright, N.A., 2009. The delivery of endocytosedcargo to lysosomes. Biochem. Soc. Trans. 37, 1019–1021.

Mizuno, E., Kawahata, K., Kato, M., Kitamura, N., Komada, M., 2003. STAM proteinsbind ubiquitinated proteins on the early endosome via the VHS domain andubiquitin-interacting motif. Mol. Biol. Cell. 14, 3675–3689.

Murai, S., Kitamura, N., 2000. Involvement of hrs binding protein in IgE receptor-triggered exocytosis in RBL-2H3 mast cells. Biochem. Biophys. Res. Commun.277, 752–756.

Nelson, W.J., Yeaman, C., 2001. Protein trafficking in the exocytic pathway ofpolarized epithelial cells. Trends. Cell. Biol. 11, 483–486.

Pandey, A., Fernandez, M.M., Steen, H., Blagoev, B., Nielsen, M.M., Roche, S., Mann,M., Lodish, H.F., 2000. Identification of a novel immunoreceptor tyrosine-basedactivation motif-containing molecule, STAM2, by mass spectrometry and itsinvolvement in growth factor and cytokine receptor signaling pathways. J. Biol.Chem. 275, 38633–38639.

Pereira, C., Maamar-Tayeb, M., Burke, A., Perez-Polo, R., Herndon, D.N., Jeschke,M.G., 2006. Immunohistochemical staining of transgenic beta-galactosidase inburned skin is a better indicator of transfection efficiency than histochemicaltechniques. J. Immunol. Methods 315, 75–79.

Piper, R.C., Luzio, J.P., 2007. Ubiquitin-dependent sorting of integral membraneproteins for degradation in lysosomes. Curr. Opin. Cell Biol. 19, 459–465.

Raiborg, C., Rusten, T.E., Stenmark, H., 2003. Protein sorting into multivesicularendosomes. Curr. Opin. Cell Biol. 15, 446–455.

Raiborg, C., Stenmark, H., 2009. The ESCRT machinery in endosomal sorting ofubiquitylated membrane proteins. Nature 458, 445–452.

Ren, X., Kloer, D.P., Kim, Y.C., Ghirlando, R., Saidi, L.F., Hummer, G., Hurley, J.H., 2009.Hybrid structural model of the complete human ESCRT-0 complex. Structure 17,406–416.

Sperk, G., Marksteiner, J., Gruber, B., Bellmann, R., Mahata, M., Ortler, M., 1992.Functional changes in neuropeptide Y- and somatostatin-containing neuronsinduced by limbic seizures in the rat. Neuroscience 50, 831–846.

Takata, H., Kato, M., Denda, K., Kitamura, N., 2000. A hrs binding protein having a Srchomology 3 domain is involved in intracellular degradation of growth factorsand their receptors. Genes Cells 5, 57–69.

Tamai, K., Toyoshima, M., Tanaka, N., Yamamoto, N., Owada, Y., Kiyonari, H., Murata,K., Ueno, Y., Ono, M., Shimosegawa, T., Yaegashi, N., Watanabe, M., Sugamura, K.,2008. Loss of hrs in the central nervous system causes accumulation ofubiquitinated proteins and neurodegeneration. Am. J. Pathol. 173, 1806–1817.

Thomas, T., Voss, A.K., Chowdhury, K., Gruss, P., 2000. A new gene trap constructenriching for insertion events near the 50 end of genes. Transgenic Res. 9, 395–404.

Tsujimoto, S., Pelto-Huikko, M., Aitola, M., Meister, B., Vik-Mo, E.O., Davanger, S.,Scheller, R.H., Bean, A.J., 1999. The cellular and developmental expression ofhrs-2 in rat. Eur. J. Neurosci. 11, 3047–3063.

Williams, R.L., Urbé, S., 2007. The emerging shape of the ESCRT machinery. Nat. Rev.Mol. Cell. Biol. 8, 355–368.

Yamada, M., Takeshita, T., Miura, S., Murata, K., Kimura, Y., Ishii, N., Nose, M.,Sakagami, H., Kondo, H., Tashiro, F., Miyazaki, J., Sasaki, H., Sugamura, K., 2001.Loss of hippocampal CA3 pyramidal neurons in mice lacking STAM1. Mol. Cell.Biol. 21, 3807–3819.

Yamada, M., Ishii, N., Asao, H., Murata, K., Kanazawa, C., Sasaki, H., Sugamura, K.,2002. Signal-transducing adaptor molecules STAM1 and STAM2 are required forT-cell development and survival. Mol. Cell Biol. 22, 8648–8658.