Well, it looks like it works on USB 1.1 on Windows 2000 SP4 too without any extra drivers! Even though I say so myself, the choice to use HID and a standard USB audio.

ABSTRACT The secreted proteolytic activity of Aspergillus fumigatus is of potential importance as a virulence factor and in the industrial hydrolysis of protein sources. Fumigatus genome contains sequences that could encode a five-member gene family that produces proteases in the sedolisin family (MEROPS S53). Four putative secreted sedolisins with a predicted 17- to 20-amino-acid signal sequence were identified and termed SedA to SedD. SedA produced heterologously in Pichia pastoris was an acidic endoprotease.

Heterologously produced SedB, SedC, and SedD were tripeptidyl-peptidases (TPP) with a common specificity for tripeptide- p-nitroanilide substrates at acidic pHs. Purified SedB hydrolyzed the peptide Ala-Pro-Gly-Asp-Arg-Ile-Tyr-Val-His-Pro-Phe to Arg-Pro-Gly, Asp-Arg-Ile, and Tyr-Val-His-Pro-Phe, thereby confirming TPP activity of the enzyme. SedB, SedC, and SedD were detected by Western blotting in culture supernatants of A. Fumigatus grown in a medium containing hemoglobin as the sole nitrogen source.

A degradation product of SedA also was observed. A search for genes encoding sedolisin homologues in other fungal genomes indicates that sedolisin gene families are widespread among filamentous ascomycetes. Aspergillus fumigatus, the main causative agent of invasive aspergillosis in neutropenic patients, is a saprophyte that grows and sporulates in a humid environment on decaying organic materials.

Secreted proteolytic activity could be an important contributor to this virulence capability. Like many other ascomycetous fungi, A. Fumigatus grows well in a medium containing protein as the sole nitrogen and carbon source and secretes endoproteases and exopeptidases (for a review of Aspergillus endo- and exoproteases, see reference ). According to the MEROPS list () (, ), A. Fumigatus secreted endopeptidases may be classified into aspartic proteases of the pepsin family (MEROPS A1) (,, ), serine proteases of the subtilisin subfamily (MEROPS S8A) (,, ), and metalloproteases of two different families (MEROPS M35 and M36) (,,,, ). Aspergillus secreted exoproteases include aminopeptidases (MEROPS M28), carboxypeptidases (MEROPS S10), and dipeptidyl peptidases (MEROPS S9B and S9C).

Recently, a new endoprotease of the sedolisin family (MEROPS S53) was isolated from a commercial protease extract of Aspergillus oryzae and termed aorsin (). This enzyme was homologous to a tripeptidyl-peptidase (TPP) secreted by the same fungal species (GenBank accession no. ) and another TPP from Aspergillus niger (). The term sedolisin was recently introduced () for a group of similar acidic serine proteases of various origins. In particular, this family contains the human lysosomal TPP involved in hydrolysis of hydrophobic proteins. A hereditary deficiency of human lysosomal TPP results in infantile neuronal ceroid lipofuscinosis (, ).

A BLAST analysis based on the A. Fumigatus genome () revealed the presence of an apparently five member gene family encoding sedolisins.

Four of these enzymes have a predicted 17- to 20-amino-acid signal sequence for putative secretion and were named SedA, SedB, SedC, and SedD. The objective of this study was to provide a detailed description of these four putative sedolisin proteases secreted by A. Fumigatus and to validate predictions from the A. Fumigatus genome. This is the first time outside the patent literature () that TPPs secreted by fungi have been identified and characterized.

SedA, SedB, SedC, and SedD, together with the aspartic protease Pep1 (), constitute a set of proteases that could be used by A. Fumigatus to degrade proteins at acidic pHs. Genomic and cDNA libraries.An A. Fumigatus λgt11 cDNA library was previously constructed using mRNA from the CHUV192-88 strain grown at 30°C for 40 h in a liquid medium containing collagen as the sole nitrogen and carbon source (). Total RNA was extracted as described previously () and the mRNA purified by using oligo(dT) cellulose (Sigma, Taufkirchen, Germany) according to standard protocols (). The cDNA library was prepared from this mRNA using phage λgt11 (Promega, Madison, WI) and protocols supplied by the manufacturer.

Alternatively, another λ-phage cDNA library made from an A. Fumigatus D141 culture growing in the logarithmic phase (minimal medium, 37°C, 16 h) was used ().

DNA fragments for production of heterologous proteins.The cDNA coding for putative SedB and SedD was amplified by PCR. Target DNA was prepared from 10 6 clones of the cDNA libraries.

PCR was performed with homologous primers derived from genomic DNA sequences (P17 to P24 [Tables and ]). Two hundred nanograms of target DNA, 10 μl each of the sense and antisense oligonucleotides at a concentration of 42 mM, and 8 μl of deoxynucleotide mix (containing 10 mM of each deoxynucleoside triphosphate) were dissolved in 100 μl PCR buffer (10 mM Tris-HCl pH 8.3, 50 mM KCl, and 1.5 mM MgCl 2). To each reaction mixture, 2.5 U of AmpliTAQ DNA polymerase (Perkin Elmer, Boston, MA) was added. The reaction mixtures were incubated for 5 min at 94°C, then subjected to 25 cycles of 30 s at 94°C, 30 s at 55°C, and 60 s at 72°C, and finally incubated for 10 min at 72°C. Amplified DNA segments encoding putative SedA and SedC proteins were constructed and cloned into pUC18 to generate plasmids pSedA and pSedC.

For pSedA cloning, the following pairs of sense and antisense primers were used to amplify four fragments of A. Fumigatus genomic DNA: P1-P2, P3-P4, P5-P6, and P7-P8 (Table ). The PCR products were digested with EcoRI/BamHI, BamHI/SphI, SphI/SspI, and SspI/Asp718, respectively, and were inserted end-to-end into pUC18 digested with EcoRI/Asp718 to generate plasmid pSedAbp47-826. In a second step, the following pairs of sense and antisense primers were used to amplify pSedA bp47-826 and A. Fumigatus genomic DNA, respectively: P9-P10 and P11-P12 (Table ). The PCR products were digested with EcoRI/SacI and SacI /HindIII and were inserted end-to-end in pUC18 digested with EcoRI/HindIII to generate plasmid pSedA.

For pSedC construction, the following pairs of sense and antisense primers were used to amplify two contiguous fragments of A. Fumigatus genomic DNA: P13-P14 and P15-P16 (Table ). Subsequently, the two PCR products were digested with HindIII/Asp718 and Asp718/BglII and were inserted end-to-end into pUC18 digested with HindIII/BglII.

Proteolytic activities.Endoproteolytic activity was measured with 50 μl of a P. Pastoris culture supernatant and 50 μl of 0.2% resorufin-labeled casein at different pHs in sodium citrate buffer (50 mM final concentration; pH 2.0 to 7.0) in a total volume of 0.5 ml. After incubation at 37°C, the undigested substrate of the enzyme-substrate mixture was precipitated by trichloroacetic acid (4% final concentration) and separated from the supernatant by centrifugation. Five hundred microliters of Tris-HCl buffer (500 mM; pH 9.4) was added to the collected supernatant (neutralization step), and the A 574 of the mixture (1 ml) was measured. A blank was performed with 50 μl of a P.

Pastoris GS115 culture supernatant. For practical purposes, one milliunit of activity was defined as producing an increase in absorbance of 0.001 per min in a proteolytic assay (1 ml) at the optimal pH for activity. The assays were performed in triplicate. Exoproteolytic activity was tested with synthetic substrates supplied by Bachem (Bubendorf, Switzerland). Stock solutions were prepared at 100 mM concentration and stored at −20°C. Ala-pNA, Gly-Pro-pNA, Ala-Ala-pNA, Phe-Pro-Ala-pNA, Ala-Ala-Pro-pNA, and Ala-Ala-Pro-Leu-pNA were dissolved in ethanol. Ala-Ala-Phe-pNA was dissolved in dimethylformamide.

The reaction mixture contained a concentration of 5 mM substrate and the enzyme preparation (between 0.1 to 1.0 μg per assay) in 100 μl of 50 mM citrate buffer at different pH values for each Sed (between pH 2.0 and pH 7.0). After incubation at 37°C for 10 min, the reaction was terminated by addition of 5 μl of glacial acetic acid, followed by 0.9 ml of water, to the mixture. The pNA released was measured by spectrometry as a change in A 405. A control with a blank substrate and blank culture broth was carried out in parallel. The Sed TPP activities were expressed in milliunits (nanomoles of pNA released per minute) using Phe-Pro-Ala-pNA as the substrate.

For TPP kinetic analysis, purified heterologously produced SedB (see above) was tested with Phe-Pro-Ala-pNA as a substrate at various concentrations between 10 −3 and 10 −7 M at 20°C in 0.1 M sodium citrate buffer, pH 6.0. The final concentration of SedB in the reaction mixture was 2.5 × 10 −9 M. The absorption of liberated p-nitroaniline was monitored photometrically at 405 nm (Ultrospec 1000 photometer; Amersham Pharmacia). The Michaelis constant ( K m) and the turnover number ( k cat) were calculated on the basis of a standard Lineweaver-Burk plot. Peptide digestion by SedB was investigated with the synthetic peptide Ala-Pro-Gly-Asp-Arg-Ile-Tyr-Val-His-Pro-Phe (Sigma catalog no. The peptide was dissolved in 50 mM sodium citrate buffer, pH 6.0, at a final concentration of 0.5 mM, and purified heterologously produced SedB was added to a final concentration of 5 μg/ml. A control without SedB was carried out in parallel.

The reaction was stopped after a 2-h incubation at 37°C by addition of 1/20 volume of 100% formic acid, and samples were then analyzed with a Q-TOF Ultima Global mass spectrometer (Micromass) in MS scan mode. For this purpose, samples were diluted to 1 pmol/μl of the peptide starting material with acetonitrile-H 2O (50:50) and subsequently injected into the mass spectrometer using a HAMMEL syringe at an approximate flow rate of 300 nl/min. Acquisition of data was started after a constant spray was obtained and was carried out for at least 10 min for each digested and undigested sample. Data were interpreted with MassLynx 4.0 SP4 software (Micromass). Gene disruption.Gene disruption vectors were constructed using pAN7.1 () and 0.7- to 1.1-kb internal fragments of the respective sed gene. In detail, sed fragments were obtained by PCR using appropriate primers (Table, primers s1-1 to s4-2) and genomic A.

Fumigatus DNA as the template. The PCR products were first cloned into pCR-Script Amp SK(+) (Stratagene, La Jolla, CA). In a second step, the sed fragments were excised from the plasmid constructs with XbaI and HindIII or XbaI and NgoMIV, for which a site was introduced into the primers, and were ligated to the larger fragment of pAN7.1 digested with the same restriction enzymes. The plasmids generated were termed pΔsedA to pΔsedD. Undigested plasmids were used for subsequent gene-targeted disruption experiments.

For generation of a sedB disruption mutant only, the pΔsedB plasmid was digested with BsaAI prior to a second transformation experiment. A single BsaAI site was present in pΔsedB, approximately in the middle of the sedB gene fragment. Fumigatus NCPF 7367 was transformed according to a protocol that has been used for Aspergillus nidulans and A. Fumigatus (, ).

Transformation of 10 7 protoplasts either with 5 μg of undigested pΔsedA, pΔsedB, pΔsedC, or pΔsedD or with BsaAI-digested pΔsedB plasmid DNA typically yielded 100 to 200 hygromycin-resistant colonies. After overnight expression of the hygromycin B phosphotransferase gene ( HPH), the transformants were incubated on agar based on GYE medium (1% glucose, 0.5% yeast extract) containing 200 μg/ml hygromycin (Sigma) and were selected after 5 days of incubation at 20°C followed by overnight incubation at 42°C.

Transformants initially identified as hygromycin resistant were picked and subcultured again on agar containing hygromycin. The sed disruptants were identified by PCR of genomic DNA from various numbers of hygromycin-resistant colonies as a template and two pairs of specific primers (Table, primers s1-A to s4-D).

Each primer pair yields a product of the predicted size when the respective plasmid is integrated at a homologous site. In each primer pair, one primer hybridized with the transformation plasmid and the other primer hybridized with genomic DNA near the desired homologous integration locus, as shown for sedA disruption (Fig.

At least one disruption mutant for each sed gene was identified. Plasmid construct and predicted outcome of pΔsedA integration event.

A 1,027-bp internal PCR fragment of sedA was cloned into the pAN7.1 plasmid (). PAN7.1 carries the hygromycin resistance gene ( HPH) from Escherichia coli as a dominant selectable marker under the control of the GPD promoter (p GPD) and the TRPC terminator (t TRPC) from Aspergillus nidulans. Homologous recombination of the plasmid construct with sedA by a single-crossover event results in the generation of two incomplete copies of the A. Fumigatus gene [labeled sedA (N-term) and sedA (C-term)] separated by the linearized sequence of pAN7.1. The primers used to screen for the incomplete sedA genes are s1-A and s1-B (3′ fragment) and s1-C and s1-D (5′ fragment). For further information on primers, see Table.

Phylogenetic analyses and genomic data for Aspergillus.Amino acid sequences of Aspergillus fumigatus sedolisins (SedA to SedE) were first analyzed using the BLAST algorithm (blastp) () with a BLOSUM62 substitution matrix to determine the existence of homologous proteins in other fungal species. All fungal protein sequences displaying a BLAST score higher than 150 were included in further phylogenetic analyses. The sequences were then aligned using Clustal W () as implemented in BioEdit Sequence Alignment Editor software (). Phylogenetic analyses of A. Fumigatus sedolisins and homologous sequences from other fungi were performed in PAUP* v4.0b10 ().

Amino acid sequences were analyzed using maximum-parsimony (MP) and neighbor-joining (NJ) phylogenetic methods. The Dayhoff PAM model of protein evolution was used to compute the distances between the amino acid sequences. Analyses were performed using a heuristic search with the TBR branch swapping algorithm. The reliability of internal branches was assessed using the bootstrap method (), with 1,000 replicates. Phylogenetic trees were edited using TreeView (). In general, genomic data for A.

Fumigatus were provided by The Institute for Genomic Research () and The Wellcome Trust Sanger Institute (), while genomic data for A. Nidulans, Chaetomium globosum, and Stagonospora nodorum were provided by The Broad Institute ().

Coordination of the analyses of these data was enabled by an international collaboration involving more than 50 institutions from 10 countries and coordinated from Manchester, United Kingdom (; ). Production and structure of A. Fumigatus sedolisins.The nature of sedolisins secreted by A. Fumigatus was investigated by a reverse genetics approach using enzymes produced heterologously in Pichia pastoris. The cDNAs encoding putative SedB and SedD could be specifically amplified using 5′ sense and 3′ antisense primers (Tables and ) and DNA extracted from a pool of 10 6 clones of the A. Fumigatus cDNA libraries as the template. The intron-exon structures of the sedB and sedD genes were determined by comparing the cDNA sequences with the A.

Fumigatus genome sequence. No sedA or sedC cDNA was obtained by PCR. However, the possible intron-exon structures of the sedA and sedC genes (Table ) could be deduced following alignment with sedB and sedD cDNA sequences. DNA coding for putative SedA and SedC proteins was synthesized by adding end-to-end PCR-generated DNA fragments for the deduced exons and then cloning into pUC18. Driver Conexant Fusion 878a Windows 7 64 Bits there. The resulting plasmids were termed pSedA and pSedC.

Fumigatus cDNAs obtained by PCR and the sedA and sedC DNAs were cloned into P. Pastoris expression vectors (Table ) and expressed in P. Pastoris grown in a methanol-inducing medium. Thirty micrograms of heterologously produced protein per milliliter was obtained for SedA, SedB, and SedD, while the yield for SedC was 10 μg ml −1 of culture supernatant. The identities of the heterologously produced proteins were confirmed as described in Materials and Methods by ESI-liquid chromatography-MS/MS de novo sequencing of trypsin-digested bands of SedA to SedD. When sequences were subjected to a database search, the relevant heterologous Sed protein was identified as the only candidate in each inquiry.

The amino acid sequence coverage confirmed by de novo sequencing for mature SedA, SedB, and SedD was 44%, 40%, and 19%, respectively. The N terminus of SedC was not sequenced (see below), so no coverage value for this enzyme was determined.

Heterologously expressed SedA, SedB, SedC, and SedD polypeptides were glycoproteins, as determined from reductions in their molecular weights following treatment with N-glycosidase F (Fig., lanes 3 and 4). The apparent molecular mass of each deglycosylated heterologously expressed protein was lower than the calculated molecular mass of the complete polypeptide chain deduced from the nucleotide sequence of the encoding gene (Table ). Based on the N-terminal amino acid sequence obtained by Edman degradation, the mature SedA, SedB, and SedD proteins start at Arg-Ser-Pro-Leu-Pro, Thr-Ser-Thr-Cys-Asp, and Ala-Ala-Thr-Asn-Ser, respectively, and thus prove the existence of a prosequence for each of these enzymes. After the signal and propeptide portions of the polypeptide chain were subtracted, the calculated molecular masses for the mature SedA, SedB, and SedD proteins were consistent with the molecular masses of deglycosylated proteins estimated by SDS-PAGE.

Dell 745 Sound Driver Free Download on this page. The N terminus of SedC could not be determined by Edman degradation of the protein. Western blot of A.

Fumigatus culture supernatant and heterologously expressed enzymes. Fumigatus wild-type and mutated strains were grown in liquid hemoglobin medium at 37°C for 2 days on a rotatory shaker at 200 rpm.

The proteins of Aspergillus culture supernatants (lanes 1 and 2, mutant strain; lanes 5 and 6, wild-type strain) were concentrated 150-fold, and 30 μl of this solution was loaded onto SDS-PAGE (12.5% polyacrylamide) gels. As a control, 1.0 μg of matching heterologous enzymes from P. Pastoris culture supernatants was loaded in parallel (lanes 3 and 4). In all blots, supernatant preparations were used with (lanes 1, 3, and 5) and without (lanes 2, 4, and 6) prior treatment with N-glucosidase F. Western blots were revealed using antisera raised against parts of the respective heterologously expressed sedolisin (see Materials and Methods).

Enzymatic activities of heterologously produced A. Fumigatus sedolisins.Heterologously produced proteins were tested for endo- and exoproteolytic activity.

Only SedA had endoproteolytic activity as described in Materials and Methods. This enzyme was active between pH 3.0 and 6.5, with optimal activity at pH 5.5. At this pH, SedA activity was 15 mU ml −1 of P. Pastoris culture supernatant.

SedA had no exoprotease activity, as evidenced by the fact that it did not release pNA from any of the mono-, di-, tri-, or tetrapeptide substrates tested in this study (see Materials and Methods). The heterologously expressed SedB, SedC, and SedD proteins had no endoproteolytic activity, but they released pNA very efficiently when Phe-Pro-Ala-pNA or Ala-Ala-Phe-pNA was used as the substrate. Supernatants of P. Pastoris GS115 and KM71 had no activity on these tripeptide pNA substrates. SedB and SedC were active between pH 3.0 and 7.0, with an optimum at pH 6.0, while SedD was active between pH 1.5 and 6.0, with an optimum peak at pH 5.0. At the optimal pH, SedB, SedC, and SedD activities were 1,560, 400, and 480 mU ml of P. Pastoris culture supernatant −1, respectively.

None of these three enzymes had activity toward either Ala-Ala-Pro-pNA or mono-, di-, or tetrapeptide pNA substrates. Since SedB, SedC, and SedD share a common specificity for tripeptide-pNA substrates, they were considered TPPs. SedB was purified as described in Materials and Methods by ion-exchange chromatography and gel filtration for further characterization. At room temperature (20°C), with Phe-Pro-Ala-pNA as a substrate at the optimum pH, SedB had a k cat of 35 s −1 and a K m of 6.25 × 10 −5 M, leading to a k cat/ K m value of 5.6 × 10 5 M −1 s −1. The specific activity of SedB was 52 mU μg of protein −1. When purified SedB was incubated with the peptide Ala-Pro-Gly-Asp-Arg-Ile-Tyr-Val-His-Pro-Phe, mass peaks in accordance with the degradation products Ala-Pro-Gly, Asp-Arg-Ile, and Tyr-Val-His-Pro-Phe were detected by MS, thus confirming a TPP activity of the enzyme (Fig. MS scan analysis of protease activity of SedB.

(A) MS spectrum representing the undigested 11-amino-acid peptide A-P-G-D-R-I-Y-V-H-P-F in the absence of SedB. The sample (1 pmol/μl of peptide in acetonitrile-water [50:50]) was analyzed at a flow rate of 0.3 μl/min.

The chromatogram (not shown) and spectrum were recorded for 10 min and combined. Peaks at m/z 636.3 and m/z 424.6 represent double- and triple-charged whole peptides (insets). No digested di-, tri-, or pentapeptide fragments were observed. (B) MS spectrum of the SedB-digested 11-amino-acid peptide. Two peaks at m/z 244.1 and 403.2 were detected in accordance with the single-charged tripeptides A-P-G and D-R-I, respectively.

Two other peaks at m/z 662.3 and 331.6 represent the single- and double-charged pentapeptide Y-V-H-P-F. Detection of A. Fumigatus sedolisins in the culture supernatant. Sed-negative mutants were constructed as described in Materials and Methods. The TPP activity of the A. Fumigatus wild-type strain in comparison to that of sed-negative mutants was measured in hemoglobin liquid medium culture supernatants with Phe-Pro-Ala-pNA as a substrate at pH 5.0. At this pH, secreted A.

Fumigatus leucine aminopeptidase (Lap) and X-prolyl-peptidase (DppIV) () activities are not detectable, and thus interference by these enzymes was excluded. SedB-, sedC-, and sedD-negative mutants had 20, 90, and 80% of the TPP activity of the A. Fumigatus wild-type strain NCPF 7367, respectively. Therefore, TPP activity in the A.

Fumigatus culture supernatant was due mainly to SedB. None of the sed-negative mutants had a visible phenotype on Sabouraud agar or in hemoglobin liquid medium. Immunoblot analyses using polyclonal antibodies were performed to detect Seds in A.

Fumigatus culture supernatants. Antisera against SedA, SedB, and SedC were highly specific, while the anti-SedD antiserum cross-reacted slightly with SedA (data not shown). Western blot analysis unambiguously detected the presence of SedB, SedC, and SedD in A.

Fumigatus culture supernatants. Deglycosylated native enzymes and heterologously expressed enzymes from P. Pastoris had the same electrophoretic mobility (Fig. With the exception of glycosylated native SedD, which did not appear as a distinct band, the corresponding glycosylated heterologous and native Seds had approximately the same electrophoretic mobility. In contrast, only a 25-kDa protein was identified by anti-SedA or anti-SedD antibodies. This protein was absent in the sedA-negative mutant culture supernatant, leading to the conclusion that it is a degradation product of SedA that cross-reacts with the anti-SedD antibodies (Fig.

Phylogenetic analysis of A. Fumigatus sedolisins.BLAST searches performed with SedA to SedD displayed hypothetical proteins from complete genome sequences of various fungi. Matching sequences were obtained from the genomes of the ascomycetes Magnaporthe grisea (five sequences), Gibberella zeae (three sequences), Neurospora crassa (three sequences), Aspergillus nidulans (three sequences), Chaetomium globosum (seven sequences), and Stagonospora nodosum (six sequences) and from the genome of the basidiomycete Ustilago maydis (one sequence) (Fig. Additionally, A. Oryzae TPPa and aorsin sequences are displayed.

The best BLAST scores for nonfungal sequences were those for TPPs from vertebrates. The SedE sequence appeared to be highly divergent from SedA to SedD (18 to 21% sequence identity), and BLAST searches showed homology with the kumamolisin (sedolisin) precursors from a Bacillus sp., Alicyclobacillus sendaiensis, and Burkholderia pseudomallei. The sequences closest to SedA to SedD were Anidulans3 (55% identity), A. Oryzae TPPa (69% identity), Anidulans2 (63% identity), and Gzeae1 (47% identity), respectively (Fig. NJ phylogenetic tree inferred from sedolisin amino acid sequences. Aspergillus fumigatus sedolisins are boxed. Bootstrap values (1,000 replicates) higher than 50% are given at nodes.

Bar, number of substitutions per site. The aligned data set of SedA to SedE and putative orthologues from fungi and bacteria cited comprised 912 sites, among which 975 were variable (74%). Phylogenetic analyses performed using MP and NJ methods produced very similar tree topologies. The MP analysis produced a single most-parsimonious tree, which differed from the NJ tree mainly by the basal position of the human TPP1 sequence. The trees were rooted with the most divergent Sed sequence (SedE).

Amino acid sequence accession numbers at NCBI are as follows: SedA, CAE51075; SedB, CAE17674; SedC, CAE46473; SedD, CAE17675; SedE, EAL86850; A. Oryzae TPPa, BAC56232; A. Oryzae aorsin, BAB97387; Anidulans1, XP411296; Anidulans2, XP407157; Anidulans3, XP411338; Mgrisea1, EAA53127; Mgrisea2, EAA49282; Mgrisea3, EAA51305; Mgrisea4, EAA53282; Mgrisea5, EAA47813; Ncrassa1, XP329464; Ncrassa2, XP324260; Ncrassa3, XP329780; Gzeae1, XP383248; Gzeae2, XP390519; Gzeae3, XP382642; Umaydis, XP403733; TPP1 human, O14773; Bacillus, BAB85637; Alicyclobacillus, BAC41257; Burkholderia, YP_111568.

Amino acid sequence accession numbers at Broad Institute are as follows: Cglobosum1, CHG02181; Cglobosum2, CHG05416; Cglobosum3, CHG07661; Cglobosum4, CHG08553; Cglobosum5, CHG09918; Cglobosum6, CHG05162; Cglobosum7, CHG09915; Snodorum1, SNU10163; Snodorum2, SNU01084; Snodorum3, SNU12941; Snodorum4, SNU12886; Snodorum5, SNU09464; Snodorum6, SNU14758. DISCUSSION We identified and characterized four novel proteases secreted by A. Fumigatus that belong to the sedolisin family (MEROPS S53). The catalytic triad of sedolisins (Glu, Asp, Ser) was conserved in the four enzymes described here (see Fig. S1 in the supplemental material). SedA is an endoprotease similar to the previously described aorsin from A. Oryzae () (MEROPS S53.007).

No 46-kDa polypeptide chain comparable to that of the heterologously expressed enzyme was detected in A. Fumigatus culture supernatants by Western blot analyses. However, comparison of results obtained using a wild-type strain and sed mutants suggests that the 25-kDa protein identified by Western blotting was a SedA product following degradation of the mature enzyme by other endo- and exoproteases secreted by A. Fumigatus into media that contained proteins as their sole nitrogen source (). The other three A.

Fumigatus secreted sedolisins, SedB, SedC, and SedD, are TPPs like the human lysosomal enzyme of the sedolisin family. SedD activity clearly was more acidic than those of SedB and SedC.

The N-terminal amino acid sequences of heterologously expressed SedA, SedB, SedD (this work), and aorsin () indicate that the mature enzymes are made as preproproteins. N-terminal Edman degradation of SedC did not yield usable results. However, the presence of a prosequence was strongly supported by the molecular mass of deglycosylated SedC, which was comparable to those of the other Seds. Consistently, no peptide before alignment position 200 was detected by mass spectrometric de novo sequence analysis of any of the four secreted Seds (see Fig.

S1 in the supplemental material). Single TPPs from A.

Oryzae and A. Niger were previously described in a patent (). However, this is the first report of a fungal species secreting several TPPs of the sedolisin family.

The finding of putative homologous proteins in all of the fungal genomes available for the ascomycete subphylum Pezizomycotina ( A. Globosum, and S.

Nodorum) and in one basidiomycete genome ( U. Maydis) indicates that proteases of the S53 family are widespread among fungi. The sedolisin phylogenetic tree has a complex branching pattern of orthologues (versions of the same gene in different genomes that have been created by the splitting of taxonomic lineages) and paralogues (genes in the same genome that have been created by gene duplication events). The SedA sequence belongs to a basal, well-supported group that includes aorsin. This group is clearly distinguished from other groups, including all of the other sequences we analyzed, which emphasizes the unusual endoprotease activity of SedA and aorsin.

Most of the ascomycete sedolisin paralogues are distributed in distinct clades, suggesting a common origin from ancient gene duplications. In contrast, only one putative sedolisin gene was found in the genome of the basidiomycete Ustilago maydis, and no putative sedolisin genes were detected in the yeast genomes (ascomycete subphylum Saccharomycotina) screened. This would suggest that the duplications of sedolisins occurred after the ascomycete-basidiomycete split, followed by loss of all genes in the Saccharomycotina. The current disparity in protease gene numbers among filamentous ascomycetes (subphylum Pezizomycotina) has been observed previously for other fungal proteases () and likely results from further duplication and/or loss of genes under selective pressure. The interaction of secreted proteases with the environment is believed to confer considerable selective functions and possibly to preadapt fungi for decomposition of organic matter in saprophytism or in pathogenesis. Digestion of protein into amino acids and short peptides by Aspergillus spp.

Has been investigated extensively by the food fermentation industry. These fungi secrete various endo- and exoproteases that cooperate very efficiently. In Aspergillus, endoprotease cleavage of proteins at neutral pHs with the subtilisin Alp (, ) and metalloproteases (, ), Laps (), and DppIV () synergistically digests large peptides into amino acids and X-Pro dipeptides (, ).

Laps can degrade peptides from their N termini; however, X-Pro acts as a stop sequence. In a complementary manner, these X-Pro sequences can be removed by DppIV, thus allowing Laps access to the next residues.

Synergistic action of A. Oryzae Lap and DppIV at pH 7.5 leads to complete digestion of the Ala-Pro-Gly-Asp-Arg-Ile-Tyr-Val-His-Pro-Phe peptide into amino acids and X-Pro dipeptides (). Using the same peptide, we have shown that SedB can bypass Pro residues by degrading large peptides from their N termini when these Pro residues are in the second position. Although the substrate specificity of A. Fumigatus TPPs needs further investigation, these enzymes appeared to be active when the amino acid in the P1 or P′1 position (amino acids in positions 3 and 4 from the N terminus of the substrate peptide) is not a proline. Indeed, the remaining pentapeptide Tyr-Val-His-Pro-Phe was not digested by SedB, just as the synthetic substrate Ala-Ala-Pro-pNA was not digested by any of the TPPs tested. Filamentous fungi, like bacteria, yeasts, and specialized cells of plants and animals, express membrane proteins for uptake of amino acids, dipeptides, and tripeptides (,,,, ).

Large peptides cannot be used as nutrients. Therefore, SedA and the aspartic protease Pep1 () as endoproteases, together with SedB, SedC, and SedD as exoproteases, constitute a set of proteases that can be used by A. Fumigatus to degrade proteins at acidic pHs and to generate assimilable nitrogen sources in decomposing organic matter and composts. In addition, as A. Fumigatus regularly acidifies its culture supernatant in vitro (), it is likely to acidify its microenvironment in the living host. Thus, the Seds may also play a role in proper nutrition of the fungus during infection. Growth studies on protein medium and in the living host comparing the wild type and a triple-knockout sedB sedC sedD mutant should therefore follow our investigations as an intriguing future project.

Response to breast cancer chemoprevention can depend upon host genetic makeup and initiating events leading up to preneoplasia. Increased expression of aromatase and estrogen receptor (ER) is found in conjunction with breast cancer. To investigate response or resistance to endocrine therapy, mice with targeted overexpression of Esr1 or CYP19A1 to mammary epithelial cells were employed, representing two direct pathophysiological interventions in estrogen pathway signaling. Both Esr1 and CYP19A1 overexpressing mice responded to letrozole with reduced hyperplastic alveolar nodule prevalence and decreased mammary epithelial cell proliferation. CYP19A1 overexpressing mice were tamoxifen sensitive but Esr1 overexpressing mice were tamoxifen resistant. Increased ER expression occurred with tamoxifen resistance but no consistent changes in progesterone receptor, pSTAT3, pSTAT5, cyclin D1 or cyclin E levels in association with response or resistance were found. RNA-sequencing (RNA-seq) was employed to seek a transcriptome predictive of tamoxifen resistance using these models and a second tamoxifen-resistant model, BRCA1 deficient/Trp53 haploinsufficient mice.

Sixty-eight genes associated with immune system processing were upregulated in tamoxifen-resistant Esr1- and Brca1-deficient mice, whereas genes related to aromatic compound metabolic process were upregulated in tamoxifen-sensitive CYP19A1 mice. Interferon regulatory factor 7 was identified as a key transcription factor regulating these 68 immune processing genes. Two loci encoding novel transcripts with high homology to human immunoglobulin lambda-like polypeptide 1 were uniquely upregulated in the tamoxifen-resistant models. Letrozole proved to be a successful alternative to tamoxifen. Further study of transcriptional changes associated with tamoxifen resistance including immune-related genes could expand our mechanistic understanding and lead to biomarkers predictive of escape or response to endocrine therapies. Response to breast cancer chemoprevention can depend upon host genetic makeup and initiating events leading up to preneoplasia. Increased expression of aromatase and estrogen receptor (ER) is found in conjunction with breast cancer.

To investigate response or resistance to endocrine therapy, mice with targeted overexpression of Esr1 or CYP19A1 to mammary epithelial cells were employed, representing two direct pathophysiological interventions in estrogen pathway signaling. Both Esr1 and CYP19A1 overexpressing mice responded to letrozole with reduced hyperplastic alveolar nodule prevalence and decreased mammary epithelial cell proliferation.

CYP19A1 overexpressing mice were tamoxifen sensitive but Esr1 overexpressing mice were tamoxifen resistant. Increased ER expression occurred with tamoxifen resistance but no consistent changes in progesterone receptor, pSTAT3, pSTAT5, cyclin D1 or cyclin E levels in association with response or resistance were found.

RNA-sequencing (RNA-seq) was employed to seek a transcriptome predictive of tamoxifen resistance using these models and a second tamoxifen-resistant model, BRCA1 deficient/ Trp53 haploinsufficient mice. Sixty-eight genes associated with immune system processing were upregulated in tamoxifen-resistant Esr1- and Brca1-deficient mice, whereas genes related to aromatic compound metabolic process were upregulated in tamoxifen-sensitive CYP19A1 mice. Interferon regulatory factor 7 was identified as a key transcription factor regulating these 68 immune processing genes. Two loci encoding novel transcripts with high homology to human immunoglobulin lambda-like polypeptide 1 were uniquely upregulated in the tamoxifen-resistant models. Letrozole proved to be a successful alternative to tamoxifen. Further study of transcriptional changes associated with tamoxifen resistance including immune-related genes could expand our mechanistic understanding and lead to biomarkers predictive of escape or response to endocrine therapies. Introduction Targeted overexpression of estrogen receptor 1 ( Esr1) and cytochrome P450, family 19, subfamily A, polypeptide 1 ( CYP19A1) to mammary epithelial cells of genetically engineered mouse (GEM) model results in overexpression of their respective proteins, estrogen receptor (ER) alpha and aromatase, followed by development of mammary hyperplasia and invasive cancer ().

These GEM models represent two direct pathophysiological interventions in estrogen pathway signaling, the first at the receptor level and the second involving ligand. Mammary-targeted CYP19A1 expression increases mammary-localized aromatase activity but does not increase circulating estrogen levels and is sufficient to promote development of hyperplastic alveolar nodules (HANs) and cancer (). Both overexpression of Esr1 and CYP19A1 in mammary epithelial cells increase expression levels of progesterone receptor (PGR) and phosphorylated insulin growth factor receptor, components of a high ‘ER activity profile’ suggested to be a criterion for selection of an aromatase inhibitor over tamoxifen for treatment of breast cancer (). Esr1 overexpressing mice show intrinsic-type resistance to tamoxifen with development of mammary cancers on first exposure (). GEM models lacking expression of full-length breast cancer 1, early onset ( Brca1) conditionally targeted to mammary cells coupled with germ-line Trp53 haploinsufficiency (BRCA1 KO) also demonstrate tamoxifen resistance ().

BRCA1, more widely known for its role in DNA repair, has the capacity to repress ER function () and loss of BRCA1 function results in a mammary environment with increased sensitivity to estrogen in vivo (,). GEM overexpressing CYP19A1 treated with letrozole or mifepristone show significantly lower levels of hyperplasia but this is not found following ICI182,780 exposure (). Tamoxifen is a Food and Drug Administration-approved drug for breast cancer chemoprevention in both pre- and postmenopausal women; however, some treated women still develop ER+ or ER− breast cancer ().

Development of tamoxifen resistance in ER+ breast cancers is a well-recognized clinical challenge. Overexpressed molecules linked to tamoxifen resistance include cyclin D1 (), cyclin E (), signal transducer and activator of transcription (STAT) 3 () and STAT5 (). Aromatase inhibitors, such as exemestane and letrozole, with use limited to postmenopausal women, are effective in reducing mammographic density and invasive breast cancer (,) and are being investigated as chemopreventives for women carrying BRCA1/2 mutations (). There is a clinical need to identify women who are less or more likely to respond to available anti-hormonal agents () as diverse initiating events in breast cancer may respond differently to these agents. Tissue transcriptomes are defined using high-throughput RNA-sequencing (RNA-seq) (). Polyadenylated RNA, isolated from tissue and converted to complementary DNA for deep sequencing resulting in millions of short reads for reference genome mapping, yield an unbiased approach for evaluating differences in tissue transcriptomes.

Here, genes that were significantly up- or downregulated in each of the three GEM models compared with wild-type (WT) mice were identified and gene expression patterns between tamoxifen-sensitive and tamoxifen-resistant models compared to detect differentially expressed genes (DEGs) and pathways. The immune system is implicated in playing both facilitative and antitumor roles in carcinogenesis.

A therapy-resistant cancer microenvironment includes alterations in immune cell behavior (,). Immune cells within the tumor microenvironment can induce breast cancer stem cells by initiating epithelial to mesenchymal transition (). Immune system genes were found among the 95 genes correlated with breast cancer relapse on tamoxifen (). Members of the interferon (IFN) regulatory factor (IRF) family of proteins have been reported to impact both progression and regression of malignant disease.

In vitro continuous low exposure to IFN beta and activation of downstream genes including IRF9 leads to resistance to DNA damage and increased cell survival (). IRF4 mediates a pathway suppressing cisplatin-induced apoptosis in vitro (). High IRF7 expression in breast cancer correlates with longer metastasis-free survival (), whereas loss of IRF5 is linked to invasiveness ().

IRF1 can restore sensitivity to ICI182,780 in vitro (). Immune signatures in different types of breast cancer are being studied for their predictive value in prognosis and therapeutic response ().

GEM models are preclinical tools used to test how specific genetic interventions influence disease pathophysiology and treatment response. Well-defined genetic interventions can be introduced into GEM, whereas the genetics of disease pathophysiology are more challenging to define in human populations. HANs and ductal hyperplasia (DH) are mammary gland preneoplastic lesions correlating with increased risk of invasive mammary cancer development in mice (,,,,) similar to histological abnormalities in women at elevated risk for breast cancer (,). In Esr1 and CYP19A1 overexpressing mice, DH can be detected as early at 4 months of age, whereas HANs appear later, by 8 months of age (,). Here, the aromatase inhibitor letrozole was more effective than tamoxifen at reducing rates of mammary epithelial cell proliferation and preneoplasia in mouse models with mammary-targeted overexpression of either ER or aromatase.

Letrozole was an effective alternative to tamoxifen for reduction of preneoplasia in tamoxifen-resistant Esr1 overexpressing mice. Tamoxifen resistance in Esr1 overexpressing and BRCA1 KO mice was correlated with the presence of a gene signature indicating immune activation that included upregulation of Irf7 and downstream genes.

Histological and immunohistochemical analyses HANs were identified and counted on whole mounts of inguinal mammary glands fixed in Carnoy’s solution and stained in carmine alum (). DH, defined as mammary epithelium consisting of at least four epithelial cell layers, was identified on 5 μm sections of hematoxylin and eosin-stained sections of inguinal mammary gland fixed in 10% phosphate-buffered formalin and embedded in paraffin (,). Statistical analyses for histology and IHC Statistical analyses were performed (GraphPad Prism version 4.03 for Windows, La Jolla, CA) using Fisher’s exact for comparisons of HAN and DH prevalence, t-test, two-tailed, unpaired for HAN numbers and two-tailed unpaired Mann–Whitney for percentages of mammary epithelial cells demonstrating nuclear-localized Ki67, ER, PGR, cyclin D1, cyclin E, pSTAT3 and pSTAT5. Statistical significance was assigned at P ≤ 0.05.

Means and standard errors of the mean presented. RNA-seq procedures and analyses Total RNA was extracted from homogenized mammary glands using Trizol (Invitrogen) and 300 μl chloroform (Thermo Fisher Scientific, Waltham, MA) for phase separation, precipitated with 650 μl isopropanol (Thermo Fisher Scientific) and purified using an RNeasy Plus Mini Kit (Qiagen, Valencia, CA). Nanodrop (Thermo Scientific) was used to analyze each sample for concentration and integrity assessed by an Agilent Bioanalyzer 2100 (Agilent Technologies, Santa Clara, CA).

Polyadenylated (polyA) RNA was isolated from 1 µg total RNA and converted to complementary DNA using SuperScript II (Invitrogen). A TruSeq RNA Sample Preparation Kit (Illumina, San Diego, CA) was used to generate sequencing libraries according to the manufacturer’s instructions. Libraries were independently sequenced by both single- and paired-end methods using HiSeq 2000 (Illumina). FastQC was used to evaluate quality of sequenced reads (), and contaminated adaptor portions of sequenced reads were trimmed using the Trim Galore program (). RNA-seq samples showing low quality were discarded.

Single- and paired-end reads [See Gene Expression Omnibus (GEO) database number GSE63857] of each biological replicate (single-end: two replicates of WT, Esr1, CYP19A1 and Brca1 KO; paired-end: one WT, two replicates of Brca1 KO and three replicates of Esr1 and CYP19A1) were aligned to the mouse reference genome (mm9 assembly) using TopHat (). Cufflinks () was used to assemble transcriptomes, estimate the abundance of transcripts (fragments per kilobase of transcript per million mapped reads) and identify DEGs. Utilizing the two different RNA-seq sets (single- and paired-end), DEGs were defined as genes showing reproducible 2-fold up- or 0.5 downregulated expression in both single- and paired-end data. Totals of 291, 359 and 837 DEGs were identified in Esr1, CYP19A1 and Brca1 KO compared with WT mice.

Gene ontology, network analyses, motif analysis and coding potential assessment Identified DEGs were analyzed using BiNGO with default parameters () to infer characteristics of the mouse transcriptomes (for full list of terms associated with DEGs, see, available at Carcinogenesis Online). For the gene ontology and pathway annotation network analysis, 68 upregulated genes in both Esr1 and Brca1 KO but not CYP19A1 were analyzed with ClueGO with default parameters (, available at Carcinogenesis Online) (). GeneMANIA and MCODE defined core genes (highly interconnected genes) among the 68 genes according to available databases (,).

All the analyses were performed via the Cytoscape network analysis platform (). For motif analysis, the software tool Pscan was used to identify overrepresented transcription factor binding sites (TFBSs) on the promoter regions (−1 kbp upstream to transcription start site) of the 68 upregulated genes using the JASPAR motif database (,). The top TFBSs were defined by relative P value. Two highly relevant loci encoding several novel transcripts were identified using Cufflinks and manual examination. To distinguish coding and non-coding RNAs, coding potential of the transcripts was assessed using the Coding-Potential Assessment Tool (). Non-coding RNAs were defined with the optimized cutoff value for mouse (.

Cell culture, RNA interference, reverse transcriptase–PCR and western blot analyses Michigan Cancer Foundation (MCF)-7 cells (from Marvin Rich, Michigan Cancer Foundation, Detroit, MI), maintained in the Lombardi Cancer Center Tissue Culture Shared Resource, were last authenticated 22 September 2010 (Cell ID™ System, Promega, Madison, WI). Proliferating, subconfluent monolayers of MCF-7 cells were seeded at a density of 300 000 cells per well into 6-well tissue culture plates.

Letrozole was more effective than tamoxifen in reducing HANs and mammary epithelial cell proliferation in Esr1 and CYP19A1 overexpressing mice Mammary preneoplasia includes both larger HANs demonstrable on mammary gland whole mounts at low magnification and DH detectable on hematoxylin and eosin-stained sections at higher magnification. At 12 months of age, both lesions were present in untreated Esr1 and CYP19A1 overexpressing mice with 67% of Esr1 overexpressing mice demonstrating HANs and 40% showing DH, whereas prevalence in CYP19A1 overexpressing mice were 60 and 31%, respectively (–C). HAN prevalence ( P. Esr1 and CYP19A1 transgenic mice differentially activate ER expression, pSTAT3 and pSTAT5 upon treatment with tamoxifen or letrozole Percentages of mammary epithelial cells demonstrating nuclear-localized ER were measured before and after treatment as increased ER following treatment has been associated with tamoxifen resistance (). The percentage of ER expressing mammary epithelial cells was significantly higher with tamoxifen treatment in Esr1 overexpressing mice (control: 6.8 ± 1.3 versus tamoxifen: 15.0 ± 3.4, P. IRF7 and other immune associated genes are linked to tamoxifen resistance Transcriptome analysis by RNA-seq was conducted to determine if significant differences in mammary gland gene expression could be identified in tamoxifen-resistant Esr1 overexpressing and Brca1 KO mice compared with tamoxifen-sensitive CYP19A1 overexpressing mice prior to treatment. Totals of 837, 291 and 359 genes were defined as DEGs in the GEM ( Brca1 KO: 575 upregulated, 262 downregulated; Esr1: 221 upregulated, 70 downregulated; Cyp19A1: 114 upregulated, 245 downregulated) compared with the corresponding WT transcriptome ().

Gene ontology analyses of the identified DEGs revealed unique characteristics of the transcriptomes of the two tamoxifen-resistant GEM (,, available at Carcinogenesis Online). Genes associated with ‘immune system process’ and ‘response to stimulus’ were those most notably upregulated in tamoxifen-resistant Esr1 overexpressing and Brca1 KO mice, whereas genes related to ‘cellular aromatic compound metabolic process’ were specifically upregulated in CYP19A1 overexpressing mice. Non-overlapping genes related to immune response were upregulated in each tamoxifen-resistant model compared with WT mice, whereas cell adhesion genes were upregulated specific to CYP19A1 overexpression mice compared with WT mice (, available at Carcinogenesis Online). Sixty-eight genes were identified as being uniquely upregulated in the two tamoxifen-resistant mouse models compared with the tamoxifen-sensitive model ().

The 28 genes with the most significant differences in expression patterns are illustrated in. For further insight into functions of the 68 genes, gene ontology and network analyses were performed via Cytoscape. The majority of the 68 uniquely upregulated genes were definitively associated with immune response including IFN-alpha/beta/gamma pathways ( Ifit1, Ifit3, Gbp9, I830012O16Rik and H2-Ab1), antigen processing ( Ctse, H2-Ab1, H2-Q2, H2-Q7 and H2-Q8) and inflammatory response ( Alox5 and Serpine1) (,, available at Carcinogenesis Online). To work toward identifying a key transcription factor regulating these genes, a molecular network using GeneMANIA was recognized and a core network extracted from it using MCODE (). Significantly, Irf7, a key transcriptional regulator of IFN-dependent immune responses, was identified.

This transcription factor activates target genes by binding to a consensus DNA motif (AAAnnGAAA) in their promoters. To test whether the 68 genes were controlled by IRF7 or transcription factors, the significantly overrepresented TFBSs on the promoters of these 68 genes were examined ().

Intriguingly, the IRF-binding site was the most significant motif among the known 130 TFBSs. Two loci encoding novel transcripts showing upregulation specific to the tamoxifen-resistant mouse models were identified ().

At least four transcripts with high homology to human immunoglobulin lambda-like polypeptide (IGLL) 1, IGLL3P, IGLL5, rabbit IGLL1 and cow IGLL1 were expressed from the XLOC_010544 locus. Transcripts a, c and e are highly likely to be protein-coding genes () and transcripts b, c and e were specifically upregulated in the tamoxifen-resistant Esr1 overexpressing and Brca1 KO mice (). Promoter regions of these transcripts contain IRF-binding motifs (AAAnnGAAA). IRF7 IHC demonstrated nuclear-localized IRF7 in mammary epithelial cells in Esr1 overexpressing and Brca1 KO mice (). INTERFEROME, a database of IFN-regulated genes (), was interrogated for overlap between the 68 genes upregulated in the Esr1 overexpressing and Brca1 KO mice and known IFN-regulated genes and used to test for known IRF7-binding sites. Fifty-four percent (37 non-redundant genes) were identified as IFN types I and/or II regulated genes (, available at Carcinogenesis Online). Nine genes contained known IRF7 binding sites, seven additional genes contained unspecified IRF sites (including IRF7) and a further seven contained either an IRF1 or IRF2 site.

Fifteen genes were identified as having STAT sites, nine of these in concert with an IRF site. STAT1 is a known mediator of cellular response to IFNs. IHC demonstrated STAT1 nuclear localization in mammary epithelial cells of Esr1 overexpressing and Brca1 KO mice (insets, ).

Type I IFN alpha is produced by leukocytes and type II IFN by T cells and macrophages. CD45, CD3 and F4 80 are markers for white blood cell lineages, T cells and macrophages, respectively. Higher numbers of CD3 per 10 high power (×40) fields compared with CD45 labeled cells were found adjacent to mammary epithelial structures in all genotypes. These numbers ranged higher in Esr1 overexpressing (CD45: 22 ± 19, CD3: 60 ± 32) and Brca1 KO (CD45: 8 ± 7, CD3: 36 ± 20) mice than CYP19A1 overexpressing (CD45: 2 ± 1, CD3: 19 ± 8) and WT mice (CD45: 2 ± 1, CD3: 14 ± 3) ( and ). F4 80 immunoreactivity in mammary epithelial structures was present in all genotypes (). Cells demonstrating expression of CD45, CD3 and/or F4 80 were documented in mammary adenocarcinoma tissue from a Brca1 KO mouse tissue (), mammary lymph nodes and blood vessels (data not shown).

MCF-7 cells were used as a model system to determine the impact of reduced IRF7 expression on PARP14, IRGM1 and GBP7 (), predicted by the bioinformatics analysis to exhibit possible regulation by IRF7 (). A time course study was performed to characterize IRF7 expression at basal conditions and following transfection with scrambled negative control siRNA and IRF7-specific siRNA. IRF7 expression levels were increased at the protein ( and B) and RNA () levels by transfection with the scrambled negative control siRNA.

This increase was attenuated by transfection with IRF7-specific siRNA (−C). In comparison with the scrambled negative control siRNA samples, expression levels of PARP14, IRGM1 and GBP7 were all reduced at the day 3 (−F) but not the day 1 time point (data not shown) following transfection with the IRF7-specific siRNA. Discussion The greater efficacy of letrozole compared with tamoxifen for resolution of mammary preneoplasia shown here parallels results reported in women where aromatase inhibitors have shown a higher response rate (,,).

In the tamoxifen-resistant Esr1 mice, tamoxifen was unable to resolve the larger HANs that represent a later stage in disease progression although it was effective in reducing prevalence of the smaller DHs that occur at an earlier stage in disease progression, compatible with previous studies in this model that showed that tamoxifen could reduce ductal abnormalities if administered at 4 months of age (,). Tamoxifen resistance in this model is correlated with disease progression suggesting that the timing of a preventive intervention may be a key factor in its effectiveness. Timing of preventive interventions in relationship to disease progression is an issue to consider for breast cancer prevention in women as well. Significantly, the aromatase inhibitor letrozole was effective at reducing both early and late lesions.

If aromatase inhibitors are more uniformly effective at both early and later stages of disease progression, this could be a factor favoring their selection. At the same time, this is complicated by the fact that only tamoxifen has an acceptable safety profile in premenopausal women, the same population that may be most likely to exhibit early disease. The molecular effects of the anti-hormonal agents in the GEM models studied here showed significant parallels with reports from human populations. Decreased proliferation is linked to an effective response to anti-hormonal agents in women (). Letrozole, compared with tamoxifen, mediated a more dramatic reduction of levels of mammary epithelial cell proliferation and may have contributed to the more profound impact on preneoplasia. Esr1 overexpressing mice showed a statistically significant increase in ER levels, a finding associated with acquisition of tamoxifen resistance in vitro ().The higher, but variable and not statistically significant, apoptotic index found with tamoxifen treatment in the CYP19A1 overexpressing mice compared with Esr1 overexpressing mice suggests but does not prove that higher levels of apoptosis with tamoxifen in the CYP19A1 overexpressing mice contributed toward its greater effectiveness. Induction of apoptosis is linked to tamoxifen response (.