We have initiated a pilot study on large-scale whole mount in situ hybridization of Macrostomum lignano. We established an efficient strategy for screening a large number of genes for their expression pattern by whole-mount in situ hybridization in adult M. lignano (Fig. 1). Our goal was to identify stem cell and germ cell marker genes to analyze their spatial and temporal expression patterns and to study their function. For the current pilot study we have selected 96 genes based on annotations of the M. lignano EST project 43 [see Additional file 1]. Genes were selected from different categories 44 like e.g. RNA-binding, cell cycle, transcription factors, chromosome/nuclear structure, growth factors. We have upscaled processing of probe synthesis and converted necessary steps of the in situ hybridization protocol onto the BioLane HTI robot. A number of genes showed cell- and tissue-specific expression patterns (Fig. 2). For example, expression in the gonads was observed for vasa-, MCM-, insulin-, piwi-, smad4-, hsp70-, boule- and tnf-like genes (2A, B, D, E, I, J, L). Most of these genes were expressed in both gonads, the testes and the ovaries, except for two genes. Boule was expressed mainly in the male gonads (2J) and a complementary pattern could be observed with hsp70, which expression was restricted to the female gonads (2I). Piwi was expressed at a basic level in the stem cells in addition to the expression in both gonads (2E). Specific cell types were stained by sir2- and notch-like in situ hybridization (2G, K). The distribution of sir2-like transcripts was in the gut region, as well as in the pharyngeal glands (2G) and notch-like transcripts were found in the pharynx-region (2K). Tol1 showed an expression pattern biased to the anterior region of the animal, with strong expression in the head (2H). A subset of cells was labeled by DNA supercoiling factor-, smad4-, notch- and tnf-like riboprobes in the tail plate (2C, F, K, L).

Hydroxyurea (HU) inhibits DNA-Synthesis and arrests cells in early S-Phase. From literature it is known that HU primarily affects spermatogenesis 45. Our results corroborate these findings for Macrostomum. We observed a decrease of macvasa expression in the testes of HU-treated animals. Currently, it is not clear whether spermatogonia, spermatocytes or spermatides are affected specifically. However, by means of histological sections of macvasa in situ hybridizations it will be possible to further address this question.

Here we demonstrate that the vasa-like gene macvasa was expressed in testes and ovaries of adult M. lignano in control animals (Fig. 3C). In order to look at macvasa expression after HU-treatment of M. lignano, in situ hybridizations were performed. HU-treated animals do not show macvasa expression in the male gonads (Fig. 3D). In ovaries, however, the level of expression appeared only slightly reduced or unchanged compared to control animals (Fig. 3D).

In addition to in situ hybridization experiments the effects of HU-treatment on proliferating cells were examined by labeling S-phase cells with BrdU. The distribution of S-phase cells in control animals showed two bands of proliferating cells along the lateral sides and clusters of S-phase cells within the testes (Fig. 3A, see also 8). After 48 hours of 2.5 mM HU incubation no significant differences between the control- and the treated group were detected. After 3.5 days about 50% of the S-phase cells disappeared and after 5.5 days BrdU-labeled cells could only be seen in the gonads. After 8.5 days and 10 days only about 10–40 S-phase cells remained stained in the reproductive organs (Figure 3B).

We have irradiated M. lignano with doses of 10, 20, 40, and 80 Gray, respectively. Effects on cell proliferation were examined by BrdU/phos-H3 stainings and macpiwi gene [see Additional file 2] expression after 3 hours, 1 day, 7 days and 21 days (Fig. 4, 5).

The number of mitoses was decreased significantly 3 hours after irradiation and remained at a low level at 1 day postirradiation (Fig. 4D). Cell proliferation has partially recovered one week after irradiation but reflected a correlation to the irradiation levels. Low levels of irradiation (10 and 20 Gray) picked up to about 70% of the number of mitoses compared to control animals while higher doses (40 and 80 Gray) got back to less than 30%, respectively. After 3 weeks postirradiation the number of mitoses at all irradiation doses was even higher than the number of mitotic cells of control animals (Fig. 4D). We could not collect data on the number of S-phase cells 3 h and 24 h postirradiation. However, earlier experiments revealed that cell proliferation of somatic stem cells was strongly reduced after 2 h and 24 h at irradiation levels of 25, 50, 100, 150, and 200 Gray (Ladurner unpublished). In the current experiment, we observed differences in the number of S-phase cells between lower doses and higher doses of irradiation after 1 week. At lower doses (10, 20 Gy) the number of S-phase cells was comparable with control animals. In contrary, at higher doses (40, 80 Gy) BrdU incorporation was only detected in gonadal cells (Fig. 4B). Three weeks postirradiation animals recovered completely and the pattern of BrdU incorporation corresponded to the distribution of BrdU labeled cells of control animals (compare Fig. 4A,C).

The piwi homologue macpiwi was expressed in both testis and ovary, as well as in somatic stem cells (Fig. 5A). Macpiwi expression was still present in somatic neoblasts and gonadal cells 3 h post irradiation (Fig. 5B). After 1 day recovering, macpiwi expression was observed in the gonads, but lacking in somatic stem cells (Fig. 5C). Comparable to cell proliferation results, the influence of the irradiation dose (10 or 20 Gy versus 40 or 80 Gy) became apparent 1 week postirradiation. Animals irradiated with 10 or 20 Gy do not show an alteration in macpiwi expression compared to control animals (data not shown). However, worms treated with 40 or 80 Gy, have lost macpiwi expression in testis and somatic stem cells (Fig. 5D). After 3 weeks postirradiation, macpiwi was expressed comparable to control animals in testes, ovaries and in somatic stem cells (Fig. 5E).

In order to identify differentially expressed genes specific for stem cells and germ line cells, a cDNA subtraction library was produced. cDNA of HU treated animals was subtracted from control animals to enrich for respective genes. The differential screening yielded 148 clones. Of these 147 were sequenced and used for BLAST analysis. BLAST hits were assigned to 10 gene categories in terms of their associated biological processes, cellular components and molecular functions similar to the M. lignano EST project 43 and are depicted in a pie chart according to their percentage (Figure 6). A list of the top ten BLAST results for all clones with E values ≤ 1 × 10^-2 is shown in a table [see Additional file 3]. In the category "RNA binding" e.g. SNRPF protein (similar to Mus musculus SNRPF) could be identified (E value of 6e-26). The gene codes for small nuclear ribonucleoprotein polypeptide F that belongs to the snrnp sm proteins family. The molecular function is RNA binding, RNA splicing factor activity and a transesterification mechanism. It is involved in mRNA processing, nuclear mRNA splicing via spliceosome and located in the nucleus. In the category "Intracellular signalling" an Arginine kinase (E-value of 2e-06) was identified, as well as a protein phosphatase Y (Serine/threonine specific protein phosphatase, Drosophila, E value of 4e-09), a sperm cAMP-dependent protein kinase catalytic subunit (E value of 2e-35), casein kinase I (E value of 2e-17) and several phosphatases. In the "transduction" category a FMRF-amide-gated Na-channel (E value of 1e-04) and a CocoaCrisp Protein (Rattus norvegicus, E value of 5e-04), which belongs to a family of extracellular domains (Human glioma pathogenesis-related protein GliPR). The category "other" includes a serine protease that is newborn-larvae specific (Trichinella spiralis, E value of 2e-04). In the "receptor" category a kremen protein 1 precursor (E value 5e-4) and Kremen2 (Xenopus laevis, E value 2e-4) are included. In "Metabolism" a putative tubulin-tyrosine ligase (E value 2e-4) and a cysteine sulfinic acid decarboxylase (E value of 4e-19) was found.

HU treatment of M. lignano lead to a substantial reduction in the number of S-phase cells after 3.5 days of HU incubation. At 8.5 days and 10 days HU exposure, somatic neoblasts did not show BrdU incorporation. In the gonads, however, a small number of BrdU labeled cells were present. Moreover, expression of the germ line marker macvasa was restricted to the ovaries and developing eggs. Testes were completely devoid of macvasa expression. A similar phenomenon was observed in mice. There, spermatogonia and spermatocytes were found to be affected selectively by HU 45. In vivo HU exposure induced testicular germ cell apoptosis and damaged cells were eliminated by phagocytosis by neighbouring cells. In Cnidarians, HU has an effect on nerve cells. The production of nerveless Hydra attenuata by hydroxyurea treatments was shown by Sacks and Davis 46.

For Macrostomum lignano, two groups of animals – HU treated and controls – were used to generate a subtraction library enriched for differentially expressed genes specific for stem cells and germ line cells. Suppressive Subtractive hybridization was also used to e.g. isolate differentially expressed genes involved in interactions between the snail host and the parasitic flatworm Schistosoma mansoni, for investigating mechanisms of wound healing 47 or the identification of progenitor-cell-specific genes 48.

We have demonstrated that M. lignano can be used for high throughput large scale whole mount in situ hybridization. With the development of high-throughput robot systems large-scale in situ hybridization, screens have been performed in several organisms such as e.g. Drosophila, Ciona, Xenopus, Schmidtea, Danio, Medaka and Mus 495051525354555657. In these screens, the high potential of large scale in situ hybridization for the discovery of new genes has been shown. A large number of cell-, tissue- and organ specific markers were identified. A variety of valuable molecular markers of developmental studies became available from these screens. In M. lignano, we could identify gonad-specific genes such as insulin-, smad4-, tnf-, MCM- and hsp70-like, which were predominantly expressed in the ovaries. On the contrary, a boule-like gene was expressed only in the testes and with vasa and piwi, we found markers that are present in both gonads of the simultaneous hermaphrodite M. lignano. These genes will prove to be useful markers for experiments dealing with sex allocation or mating experiments.

The expression patterns of sir2 and insulin-like qualify these genes as candidates for studying aging. Riboprobes of four other genes, DNA supercoiling factor-, smad4-, notch-, tnf-like probably stain adhesive organ cell bodies. These findings may be used for differentiation and regeneration studies. With high throughput large scale whole mount in situ hybridization, genes with expression patterns interesting for different research groups become readily available and can be cloned and sequenced from the respective cDNA library.

Irradiation of M. lignano with different doses of gamma rays has shown that an irradiation sensitive population of proliferative cells exist in this animal. Few hours postirradiation, the somatic compartment of S-phase and mitotic cells has been strongly reduced while cell proliferation was still present in the gonads. After one week, BrdU incorporation was comparable to control animals irradiated with 10 or 20 Gray while mitoses were reduced to 70%. Higher doses (40 and 80 Gy) resulted in a lack of somatic stem cell proliferation while S-phase cells were visible in the gonads. At three weeks postirradiation animals of all irradiation levels showed normal S-phase cell distribution pattern. In contrast, irradiation of different planarian species with 30 Gy always resulted in a complete absence of proliferating stem cells 2428305859. In M. lignano, after 3 h postirradiation macpiwi was still expressed in mesenchymal and gonadal cells. In cancer cells, gene expression profiling revealed an increase of expression of genes involved in e.g. in DNA repair 60. Piwi-like genes are not considered as repair genes but play a role in regulation of stem cells and germ cells. This molecular function of piwi-like genes could result in the presence or absence of expression after irradiation. At one week postirradiation, macpiwi expression completely disappeared in the testes and somatic stem cells of worms irradiated with 40 or 80 Gy. Interestingly, in mice, the expression of e.g. growth factor genes was decreased specifically in testicular germ cells while somatic cells (Sertoli) maintained expression.

The recovery of cell proliferation in Macrostomum could have different explanations: First, somatic stem cells entered cell cycle arrest upon irradiation but were able to re-enter the cell cycle after 2–3 weeks of recovery. Second, non cycling neoblasts became activated and re-established the neoblast pool. A number of non cycling neoblasts have been observed in M. lignano 61. Third, gonadal stem cells recover the somatic stem cell population. An entry of cells from the ovary has been observed to contribute to the somatic stem cell population during regeneration in planarians 62. Fourth, methodological considerations have to be taken into account. It might be possible that differences in the application of the respective irradiation levels in M. lignano and in planarians are not equal despite the fact that the same doses, i.e. the number of Gray, were given. The actual "effective dose" applied could account for observed differences in the sensitivity to irradiation of macrostomid and planarian taxa. A complete elimination of S-phase cells in M. lignano was not achieved while planarians required transplantation enriched neoblasts of donor to be rescued from 30 Gy irradiation dose 59. Moreover, Macrostomum irradiated with 40 or 80 Gy regenerated and remained alive while planarians disintegrated.

BLAST analyses of the 147 obtained differentially expressed clones revealed a majority of sequences that did not show any match with BLAST databases (54.3%, "no match"). This group may contain some novel stem cell and germ line specific genes in M. lignano, in particular genes that are connected with testis-specific expression pattern. Surprisingly, we have not found genes homologous to known stem cells or proliferation genes despite the dramatic reduction of cells in S-phase after HU treatment (Figs. 3A,B). Most likely the majority of mRNA was not reduced substantially after HU incubation resulting in a low representation of such genes in the SSH library.

The Macrostomum EST library was cloned directedly into Sport 6.1 vectors (Invitrogen) with T7 for antisense transcription 43. PCRs with M13 standard primers were performed to generate templates for RNA probe synthesis. PCR conditions were 5 minutes at 95°C, 35 cycles (95°C for 45 s, 55°C for 45 s, 72°C for 1 min 20 s), 72°C for 2 minutes. Riboprobes were synthesized with the RNA in vitro transcription master mix (final volume of 10 μl per well) using the T7 transcription polymerase from Promega. Transcription was performed after manufacturer's instructions except for the use of DIG-labeling mixture from Roche. 1 μl of each transcription was checked by gel electrophoresis and 9 μl transcription reaction were diluted in 500 μl HybMix and used directly for the in situ hybridization screen.

For the pilot screening, batches of ~1000 animals were pretreated for the prehybridization similar to the whole mount in situ protocol with the following exceptions: both de- and re-hydration series were always performed in methanol. For hybridization, about 20 animals each were distributed into the mesh vessels of the robot system (HTI Biolane, Hölle and Hütter AG). Mesh vessels were placed into plastic vials. Pre-hybridization was perfomed in 100% HybMix for 2 h at 55°C (identical to the standard Macrostomum in situ hybridization protocol below). For the screen, the whole transcription reaction was diluted in 500 μl HybMix and used. For hybridization, the samples were re-incubated at 55°C for 1.5 days with the riboprobes. The stringent washes with SSC mixtures were performed according to Macrostomum standard protocol. Vials were placed first into the robot rack, which then was placed into the robot plastic container. Washes were then performed in a 55°C oven outside the robot and media were exchanged manually. Starting from the MAB washes the in situ post washes were performed within the robot. Color development was performed manually with NBT/BCIP in 24-well plates at 37°C within 15 minutes. This was followed by ethanol and PBS washes, as well as embedding and photographing of the samples according to the standard protocol.

Animals were relaxed in 1:1 7,14% MgCl2:artificial sea water (ASW) for 5 minutes and 10 minutes in pure 7,14% MgCl2. Fixation was done in 4% PFA in 0.1 M PBS for one hour at room temperature. Fixative was removed with three 5-minutes 1× PBS-Tween (0.1%) washes, and animals were dehydrated by ascending Methanol series and stored at -20°C. Whole mount in situ hybridization was performed using a newly developed protocol for M. lignano 63 based on a modified in situ protocol for Hydra 6465. Animals were rehydrated by an Ethanol series (100 % Ethanol for 5 × 5 minutes, 75%, 50% and 25% Ethanol with DEPC-treated H₂O for 5 minutes each) followed by three washes with 1× PBS-Tween. Proteinase-K treatment (20 μg/ml) was done at room temperature for 17 minutes and stopped with Glycine (4 mg/ml in 1× PBS-Tween). Animals were incubated 3 × 5 minutes in 1× PBS-Tween, 2 × 5 minutes in 0.1 M TEA, 2 × 5 minutes in 0.1 M TEA with acetic anhydride (400:1), 2 × 5 minutes in 0.1 M TEA with acetic anhydride (200:1) and 2 × 5 minutes in 1× PBS-Tween. Animals were refixed in 4% PFA in 0.1 M PBS for 20 minutes at room temperature followed by 5 × 5 minutes 1× PBS-Tween washes. Animals were heat-fixed at 80°C for 20 minutes, shaking. Subsequently an incubation in 1× PBS-Tween/HybMix (1:1) for 10 minutes at room temperature and in 100% HybMix (50% formamide, 5× SSC, 100 μg/ml heparin, 0.1% Tween, 0.1% CHAPS, 200 μg/ml yeast tRNA, 1× Denhardt's) for 10 minutes at 55°C was performed. Animals were prehybridized in fresh HybMix at 55°C for 2 hours. Riboprobe was added (working concentration 0.075 ng/μl for macvasa and 0.1 ng/ml for macpiwi) after denaturation (7 minutes at 95°C and snap chilled on ice). Hybridization time was 2.5 days at 55°C for macvasa and 1.5 days at 55°C for macpiwi, shaking continuously at 300 rpm. Stringent washes at 62°C replaced HybMix subsequently with 2× SSC, followed by two washes with 2× SSC/0.1% CHAPS at 62°C for 30 minutes each and two MAB washes (10 minutes each at room temperature). Animals were blocked in 1% blocking solution (Roche) in MAB at 4°C for two hours. DIG-AP-antibody (Roche) incubation followed for four hours at 4°C (1:2000 in blocking solution). Animals were washed in MAB at room temperature overnight and washed with MAB for 70 minutes changing the solution frequently. 2 × 5 minutes incubation steps followed in NTMT (0.1 M NaCl, 0.05 M MgCl2), 0.1 M Tris (pH 9.5), 0.1% Tween-20). Colour development was performed with a NBT/BCIP system (Roth) in the dark at 37°C for 55 minutes for macpiwi. For macvasa developing time was 10 minutes at 37°C and 8 minutes at room temperature. Frequent Ethanol washes were done followed by rehydration (3 × 5 minutes PBS). Animals were embedded in self-hardening Glycerin/Mowiol 4–88 solution. In situ hybridizations were analyzed and pictures taken using a Leica DM5000 microscope and a Pixera Penguin 600CL digital camera. Assembly and editing was performed with Adobe^® Photoshop^® 7.0 Software. Template DNA for producing DIG-labeled riboprobe was produced by standard PCR techniques. For macpiwi, primer couples were TGCTCAAGCTGGTGTTGCTGGTC and GTCTTGTTGTTGTGCCGCGTGAG and for macvasa, primer couples were CCCGGTCGCTCTTATACTGACTCC and GGTTGGCACGGAACATCCTCTC. Probes were generated using a DIG RNA labeling KIT SP6/T7 (Roche) according to the manufacturer's protocol, revealing a macpiwi probe of 864 bp and a macvasa probe of 1067 bp. Partial sequences of vasa-(ANGU3256) and piwi-like (ANGU7606) genes were obtained from the Macrostomum lignano EST database 44. Detailed information on both genes will be published separately.

Test animals were kept under normal culture conditions 66 on glass petri dishes in a climate chamber on a diatom layer of Nitzschia curvilineata in Guillard's nutrient-enriched artificial seawater f/2 medium 67. Antibiotics were added (Kanamycin 1:1000). Hydroxyurea (HU) concentration in the medium was 4.0 mM.

In order to identify differentially expressed genes specific for stem cells and germ line cells, a subtraction library was created by culturing two groups of animals on algae for 15 days. Group A consisted of HU-treated M. lignano (4 batches at 250 animals each) by soaking the animals in HU-solution. Group B were non-treated M. lignano (4 batches at 250 animals each). After 10 days of gradually increasing HU incubation (2.8 mM to 4.0 mM) control stainings with BrdU were performed and still showed 30–60 S-phase cells per animal in the gonads. The incubation time was continued until 15 days and total RNA was isolated. Total RNA of both groups of animals (HU-treated and non-treated) was checked on 1.2% Agarose Gel and sent to EVROGEN^® (Moscow, Russia). Evrogen made cDNA of both RNA samples, cloned 40 ng of purified cDNA into the pAL9 vector and used E. coli for transformation. Subtractive hybridization was performed using the SSH method (Suppression Subtractive Hybridization as described in 6869. The subtraction revealed 147 differentially expressed clones.

Worms were soaked for 30 minutes in 5 mM BrdU (Bromodeoxyuridine, Sigma). After BrdU incorporation, animals were washed twice with f/2 and relaxed in 1:1 7.14% MgCl₂:artificial sea water (ASW) for 5 minutes, followed by 15 minutes in pure 7.14% MgCl₂. Animals are fixed for 1 hour at RT in 4% PFA in 0.1 M PBS, followed by multiple washing with PBS-T (0.1%). Protease treatment (0.22 μg/ml Protease XIV (30 min, 37°C) was followed by incubation for 10 minutes in 0.1 N HCl on ice and further denaturation in 2 N HCl (1 h, 37°C). A 3 × 5 minutes wash in PBS-T to remove acid completely was followed by 30 minutes blocking in BSA -T. Primary antibody (mouse-anti-BrdU, Roche), 1:600 in PBS-BSA-T was incubated at 4°C overnight. After washing with PBS-T (3 × 5 minutes), specimens were incubated with a secondary antibody (FITC-goat-anti-mouse (DakoCytomation), 1:150 at RT in darkness for 1 h and washed again with PBS-T for 3 × 5 minutes. Specimens were mounted in Vectashield (Vector Laboratories) and animals observed with a Leica DM5000 fluorescent microscope. For BrdU/phosH3 double labeling, the anti-phosphorylated H3 antibody (Upstate Biotechnology) was added to the primary antibody incubation in a concentration of 1:150 in PBS-BSA-T. For detection, a TRITC-swine-anti-rabbit antibody (DakoCytomation) was used at a final dilution of 1:150. The number of mitoses was counted using a Leica DM5000 fluorescent microscope and a mechanical tally counter. Because of the low number of mitotic cells, the use of a software-based quantification according to 70 was not necessary.

Worms in f/2 culture medium of 5 mm depth were irradiated with 10, 20, 40 or 80 Gray (Gy) using gamma rays of an IBL-437C-instrument with a Cs-137 source (660 KeV; 700 cGy/min; 189 TBq nominal Activity). Animals were killed 3 hours, 24 hours, 7 days and 21 days after initial irradiation and examined for cell division (BrdU/H3-P) as well as macpiwi expression.