P458 Comparison of PCR protocols for detecting Histoplasma capsulatum and Coccidioides spp. DNA through a multicenter study

Poster session 3, September 23, 2022, 12:30 PM - 1:30 PM

 

Introduction

In-house real-time PCR (qPCR) is increasingly used for the diagnosis of endemic mycoses and diverse assays are in use in specialized laboratories. External quality control is currently lacking.

Objective

To compare the performance of different molecular detection protocols for the detection of Histoplasma capsulatum and Coccidioides spp. in a multicenter study involving five European laboratories.

Methods

Two test sample panels were sent to each laboratory which performed the analysis with their in-house assays. Recipients were blinded to sample content. The Histoplasma-panel included 14 samples representing a range of concentrations of Histoplasma DNA (n = 7), as well as a negative control and DNA from other fungi to test for specificity (Paracoccidioides lutzii n = 1; Blastomyces dermatitidis n = 1; Aspergillus fumigatus n = 1; Emergomyces spp. n = 2, and Candida albicans n = 1). The Coccidioides-panel included 10 samples representing a range of DNA concentrations of Coccidioides posadasii (n = 6), as well as a negative control and DNA from other fungi to test specificity (Uncinocarpus reesii n = 1; Trichophyton violaceum n = 1; and Candida albicans n = 1). Regarding techniques used, four laboratories used Histoplasma qPCRs, and one laboratory a conventional PCR and a broad-range PCR (brPCR) for fungal DNA. Four laboratories used different Coccidioides qPCRs and one laboratory a brPCR to detect Coccidioides DNA.

Results

Concerning the Histoplasma panel, qPCR assays were the most sensitives and agreement in the lowest detected amount of Histoplasma DNA was very suitable, ranging from 1 pg to 4 pg [<1 genomic equivalent (mean sensitivity: 96.4%)]. The lowest detected amount of Histoplasma DNA by cPCR (sensitivity 71.4%) and the brPCR (sensitivity 42.9%) was 0.1 and 10 pg, respectively. Overall, sensitivity ranged from 42.9-100% (mean 83.3%). Overall specificity ranged from 78.6%-100%, with false positive results occurring with high DNA concentrations (200 pg/μl) of Blastomyces spp. in two laboratories that used qPCR, Emergomyces spp. by qPCR in one laboratory and Aspergillus in one laboratory that used cPCR. Concerning the Coccidioides panel, sensitivity ranged from 33.3-100% (mean 76.6%), and agreement of the lowest detected amount of Coccidioides DNA by qPCR ranged from 1-16 pg (<1 genomic equivalent) (mean sensitivity: 87.5%) and in the brPCR 10 pg (sensitivity 33.3%). Specificity was between 87.5%-100%, with one false positive result occurring with high DNA concentrations (20 pg/μ) of Uncinocarpus in one laboratory using qPCR.

Conclusion

Specific protocols based on qPCR showed better sensitivity than conventional and brPCR. These methods are useful for the rapid and sensitive detection Histoplasma and Coccidioides. Application of these tests on clinical samples may speed up diagnosis and potentially limit laboratory exposure to these fungi. Comparisons of in-house tests are essential to assess the performance and detect potential cross-reactivities and achieve a consensus.

Roles of the multiplex real-time PCR assay and β-D-glucan in a high-risk population for intra-abdominal candidiasis (IAC)

Multiplex quantitative real-time PCR (MRT-PCR) using blood can improve the diagnosis of intra-abdominal candidiasis (IAC). We prospectively studied 39 patients with suspected IAC in the absence of previous antifungal therapy. Blood cultures, MRT-PCR, and β-D-glucan (BDG) in serum were performed in all patients. IAC was defined according to the 2013 European Consensus criteria. For MRT-PCR, the probes targeted the ITS1 or ITS2 regions of ribosomal DNA. Candidaemia was confirmed only in four patients (10%), and IAC criteria were present in 17 patients (43.6%). The sensitivity of MRT-PCR was 25% but increased to 63.6% (P = .06) in plasma obtained prior to volume overload and transfusion; specificity was above 85% in all cases. BDG performance was improved using a cutoff > 260 pg/ml, and improvement was not observed in samples obtained before transfusion. In this cohort of high risk of IAC and low rate of bloodstream infection, the performance of non-culture-based methods (MRT-PCR or BDG) was moderate but may be a complementary tool given the limitations of diagnostic methods available in clinical practice. Volume overload requirements, in combination with other factors, decrease the accuracy of MRT-PCR in patients with IAC.

Differential Diagnosis of Fungal Pneumonias vs. Tuberculosis in AIDS Patients by Using Two New Molecular Methods

Opportunistic fungal pneumonias (OFP) are the main cause of death in AIDS patients worldwide. Diagnosis of these infections is often late as tuberculosis (TB) is frequently the first suspicion. In addition, diagnostic tools have limitations and are unavailable in disadvantaged regions. To perform the differential diagnosis of the main fungi causing OFP in AIDS patients (Histoplasma capsulatum, Cryptococcus neoformans/C. gattii and Pneumocystis jirovecii) vs. the Mycobacterium tuberculosis complex (MTBC), two new assays were developed: (i) a multiplex real-time PCR (MRT-PCR) and (ii) a simple and cost-effective method based on real-time PCR and the analysis of melting curves after amplification (MC-PCR). Both of the techniques were optimized and standardized “in vitro”, showing a suitable reproducibility (CV ranged between 1.84 and 3.81% and 1.41 and 4.83%, respectively), a 100% specificity and detection limits between 20 and 2 fg of genomic DNA per 20 µL of reaction. A validation study was performed by retrospectively using 42 clinical samples from 37 patients with proven fungal infection or TB, and 33 controls. The overall sensitivity for the MRT-PCR assay and the MC-PCR assay was 88% and 90.4%, respectively. Both techniques were fast, sensitive and reproducible, allowing for the detection of these pathogens and the performance of a differential diagnosis.

Low sensitivity of conventional fungal agars in fungemia by Rhodotorula mucilaginosa: description of two cases

Background

Although most bloodstream yeast infections are caused by Candida spp., infections by rare or less common species have increased in recent years. Diagnosis of infections caused by these species is difficult due to the lack of specific symptoms and adequate diagnostic tools.

Cases presentation

We describe two cases of fungemia by Rhodotorula mucilaginosa within a few months of each other, in a secondary Spanish hospital. In both cases, diagnosis was challenging. Blood subcultures in conventional fungal media were persistently negatives and the use of non-conventional fungal media was essential for isolating the yeasts and achieving a correct diagnosis. 1–3 beta-d-glucan detection and a panfungal PCR assay were helpful techniques to confirm the diagnosis

Conclusion

It is highly important to establish an early diagnosis for fungemia. The process is challenging because often non-specific symptoms are presents. When yeasts grow in blood cultures other genera than Candida spp. could be the cause of infection. Patient risk factors should be assessed to incorporate alternative culture media and the available rapid diagnostic test, in order to provide an early recognition of the pathogen.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12941-021-00427-w.

Multiplex PCR Based Strategy for Detection of Fungal Pathogen DNA in Patients with Suspected Invasive Fungal Infections

A new and easy polymerase chain reaction (PCR) multiplex strategy, for the identification of the most common fungal species involved in invasive fungal infections (IFI) was developed in this work. Two panels with species-specific markers were designed, the Candida Panel for the identification of Candida species, and the Filamentous Fungi Panel for the identification of Aspergillus species and Rhizopusarrhizus. The method allowed the correct identification of all targeted pathogens using extracted DNA or by colony PCR, showed no cross-reactivity with nontargeted species and allowed identification of different species in mixed infections. Sensitivity reached 10 to 1 pg of DNA and was suitable for clinical samples from sterile sites, with a sensitivity of 89% and specificity of 100%. Overall, the study showed that the new method is suitable for the identification of the ten most important fungal species involved in IFI, not only from positive blood cultures but also from clinical samples from sterile sites. The method provides a unique characteristic, of seeing the peak in the specific region of the panel with the correct fluorescence dye, that aids the ruling out of unspecific amplifications. Furthermore, the panels can be further customized, selecting markers for different species and/or resistance genes.

Timely Diagnosis of Histoplasmosis in Non-endemic Countries: A Laboratory Challenge

Human histoplasmosis is a fungal infection caused by the inhalation of microconidia of the thermally dimorphic fungi Histoplasma capsulatum. Autochthonous cases of histoplasmosis have been diagnosed in almost every country, but it is considered an endemic infection in specific areas of the world. Many of them are popular travel destinations or the source of migratory movements. Thus, the vast majority of the registered cases in non-endemic countries are imported. They correspond to people having been exposed to the fungus in endemic locations as immigrants, expatriates, transient workers or tourists, with reported cases also associated to organ donation. Misdiagnosis and delays in initiation of treatment are not uncommon in cases of imported histoplasmosis. They are associated to high fatality-rates specially in patients with compromised cellular immunity in which progressive disseminated forms develop. The diagnosis of this infection in non-endemic countries is hampered by the lack of clinical suspicion and a dearth of available diagnostic tools adequate to offer rapid and accurate results. Non-culture-based assays such as nucleic-acid amplification tests present as a suitable alternative in this situation, offering improved sensitivity and specificity, shortened turnaround time, and increased biosafety by avoiding culture manipulation. In non-endemic regions, molecular techniques are being used mainly in laboratories from countries that have registered an increase in the incidence of imported cases. However, the number of published techniques is limited and lack consensus. Efforts are currently under way to standardize nucleic acid amplification-based techniques for its implementation in areas registering a rising number of imported cases.

African histoplasmosis: new clinical and microbiological insights

African histoplasmosis is defined as the fungal infection caused by Histoplasma capsulatum var. duboisii (Hcd). Studies focused on distinguishing Hcd and H. capsulatum var. capsulatum (Hcc), which coexist in Africa, are scarce or outdated, and African strains are continuously underrepresented. In this work, 13 cases of African patients with histoplasmosis diagnosed in the Spanish Mycology Reference Laboratory have been reviewed showing that 77% had disseminated disease and AIDS as underlying disease although Hcd infection has been classically considered a rare presentation in AIDS patients. Strains isolated from these patients and other clinical and reference strains were studied by assessing classical identification methods and performing a three-loci multi-locus sequence analysis (MLSA). Classical identification methods based on biochemical tests and measurement of yeast size proved to be useless in distinguishing both varieties. The MLSA defined an African cluster, with a strong statistical support, that included all strains with African origin. Finally, mating type was also determined by using molecular methods revealing an unequal mating type distribution in African strains. In conclusion, historical statements and classical identification methods were useless to distinguish between varieties, whereas molecular analyses revealed that all strains with African origin grouped together suggesting that traditional classification should be revised. Further investigation is required in order to unravel traditional concepts about Hcd infection and support results obtained in this work.

Copy Number Variation of Mitochondrial DNA Genes in Pneumocystis jirovecii According to the Fungal Load in BAL Specimens

Pneumocystis jirovecii is an unculturable fungus and the causative agent of Pneumocystis pneumonia, a life-threatening opportunistic infection. Although molecular diagnosis is often based on the detection of mtLSU rRNA mitochondrial gene, the number of copies of mitochondrial genes had not been investigated. We developed and optimized six real-time PCR assays in order to determine the copy number of four mitochondrial genes (mtSSU rRNA, mtLSU rRNA, NAD1, and CYTB) in comparison to nuclear genome (DHPS and HSP70) and tested 84 bronchoalveolar fluids of patients at different stages of the infection. Unexpectedly, we found that copy number of mitochondrial genes varied from gene to gene with mtSSU rRNA gene being more represented (37 copies) than NAD1 (23 copies), mtLSU rRNA (15 copies) and CYTB (6 copies) genes compared to nuclear genome. Hierarchical clustering analysis (HCA) allowed us to define five major clusters, significantly associated with fungal load (p = 0.029), in which copy number of mitochondrial genes was significantly different among them. More importantly, copy number of mtLSU rRNA, NAD1, and CYTB but not mtSSU rRNA differed according to P. jirovecii physiological state with a decreased number of copies when the fungal load is low. This suggests the existence of a mixture of various subspecies of mtDNA that can harbor different amplification rates. Overall, we revealed here an unexpected variability of P. jirovecii mtDNA copy number that fluctuates according to P. jirovecii’s physiological state, except for mtSSU that is the most stable and the most present mitochondrial gene.

Ribosomic DNA intergenic spacer 1 region is useful when identifying Candida parapsilosis spp. complex based on high-resolution melting analysis

The epidemiology of Candida parapsilosis and the closely related species C. orthopsilosis and C. metapsilosis has changed in recent years, justify the need to identify this complex at the species level. In this study we investigate the intergenic spacer 1 (IGS1) of the ribosomal DNA (rDNA) to evaluate the utility of this gene region as a phylogenetic molecular marker and the suitability of a high-resolution melting (HRM) strategy based on this region for identification of members of the C. parapsilosis spp. complex. We sequenced the IGS1 and the internal transcribed spacer (ITS) regions of the rDNA from 33 C. parapsilosis sensu lato strains. Although both regions are useful in identifying species, comparative sequence analysis showed that the diversity in the IGS1 region was higher than in the ITS sequences. We also developed an HRM analysis that reliably identifies C. parapsilosis spp. complex based on the amplification of 70 bp in the IGS1 region. All isolates were correctly identified with a confidence interval >98%. Our results demonstrate that HRM analysis based on the IGS1 region is a powerful tool for distinguishing C. parapsilosis from cryptic species.

Development of a single tube multiplex real-time PCR to detect the most clinically relevant Mucormycetes species

Mucormycetes infections are very difficult to treat and a delay in diagnosis could be fatal for the outcome of the patient. A molecular diagnostic technique based on Real Time PCR was developed for the simultaneous detection of Rhizopus oryzae, Rhizopus microsporus and the genus Mucor spp. in both culture and clinical samples. The methodology used was Molecular beacon species-specific probes with an internal control. This multiplex real-time PCR (MRT-PCR) was tested in 22 cultured strains and 12 clinical samples from patients suffering from a proven mucormycosis. Results showed 100% specificity and a detection limit of 1 fg of DNA per microlitre of sample. The sensitivity was 100% for clinical cultured strains and for clinical samples containing species detected by the PCR assay. Other mucormycetes species were not detected in clinical samples. This technique can be useful for clinical diagnosis and further studies are warranted.

Phenotypic characteristics of isolates of Aspergillus section Fumigati from different geographic origins and their relationships with genotypic characteristics

BACKGROUND

Epidemiological studies worldwide have shown that A. fumigatus exhibits important phenotypic and genotypic diversity, and these findings have been of great importance in improving the diagnosis and treatment of diseases caused by this fungus. However, few studies have been carried out related to the epidemiology of this fungus in Latin America. This study's aim is to report on the epidemiology of the fungus by analyzing the phenotypic variability of Aspergillus section Fumigati isolates from different Latin American countries and the relationship between this variability, the geographical origin and genotypic characteristics.

METHODS

We analyzed the phenotypic characteristics (macro- and micromorphology, conidial size, vesicles size, antifungal susceptibility and thermotolerance at 28, 37 and 48°C) of A. section Fumigati isolates from Mexico (MX), Argentina (AR), Peru (PE) and France (FR). The results were analyzed using analysis of variance (ANOVA) and Tukey's multiple comparison test to detect significant differences. Two dendrograms among isolates were obtained with UPGMA using the Euclidean distance index. One was drawn for phenotypic data, and the other for phenotypic and genotypic data. A PCoA was done for shown isolates in a space of reduced dimensionality. In order to determine the degree of association between the phenotypic and genotypic characteristics AFLP, we calculated the correlation between parwise Euclidean distance matrices of both data sets with the nonparametric Mantel test.

RESULTS

No variability was found in the macromorphology of the studied isolates; however, the micromorphology and growth rate showed that the PE isolates grew at a faster rate and exhibited the widest vesicles in comparison to the isolates from MX, AR and FR. The dendrogram constructed with phenotypic data showed three distinct groups. The group I and II were formed with isolates from PE and FR, respectively, while group III was formed with isolates from MX and AR. The dendrogram with phenotypic and genotypic data showed the same cluster, except for an isolate from FR that formed a separate cluster. This cluster was confirmed using PCoA. The correlation between the phenotypic and genotypic data of the isolates revealed a statistically significant association between these characteristics.

CONCLUSIONS

The PE isolates showed specific phenotypic characteristics that clearly differentiate them from the rest of the isolates, which matches the genotypic data. The correlation between the phenotypic and genotypic characteristics showed a statistically significant association. In conclusion, phenotypic and genotypic methods together increase the power of correlation between isolates.

Utility of real-time PCR for the detection of Paracoccidioides brasiliensis DNA in the diagnosis of imported paracoccidioidomycosis

An increase in immigration from endemic regions has resulted in a number of cases of paracoccidioidomycosis (PCM) being imported into Spain. A molecular diagnostic technique based on real-time PCR was developed for the detection of Paracoccidioides brasiliensis DNA in both culture and patients' clinical samples. A Molecular Beacon probe was used, labelled with FAM and directed at the ITS1 region of ribosomic DNA. The detection limit of the technique developed was 1 fg of fungal DNA per microl of sample. This procedure proved to be very reproducible and specific. The technique was tested with cultures of 12 clinical strains and on samples from two patients with proven PCM. Real-time PCR was positive for all the culture strains, as well as those from both patients. By samples, the technique was positive in sputum and tissue biopsies but less useful on blood samples. Samples were analyzed several months after patient treatment, detecting a small amount of fungal DNA in one respiratory sample. This technique of real-time PCR is a sensitive method for rapid diagnosis of paracoccidioidomycosis and could serve to monitor patients after treatment has begun.

[Assessment of a quantitative PCR method for clinical diagnosis of imported histoplasmosis]

OBJECTIVE

Evaluation of the usefulness of a quantitative real-time polymerase chain reaction-based (RT-PCR) technique for clinical diagnosis of histoplasmosis.

METHODS

Primers and probes were designed on the basis of sequences from the ITS regions of ribosomal DNA of 20 clinical strains of Histoplasma capsulatum. LightCycler procedures (Roche Applied Science) were used with probes marked by fluorescence resonance energy transfer (FRET). Reproducibility, sensitivity, and specificity were analyzed. In addition, an internal control was designed to identify false negative results by PCR inhibition. The RT-PCR assay was tested in 22 clinical samples from 14 patients with proven histoplasmosis. In addition, 30 samples from patients with febrile neutropenia or mycoses other than histoplasmosis, and from healthy volunteers were analyzed as controls.

RESULTS

The limit of detection of the assay was 1 fg of genomic DNA per microl of sample. The PCR-based technique was reproducible and highly specific. Positive results were obtained in 11/14 (78.6%) patients and in 17/22 (77.3%) clinical samples. RT-PCR was positive in 100% of respiratory secretions and bone marrow samples, but only 70% of sera (p < 0.01). Mean fungal DNA value was 23.1 fg/microl in serum and 4.85 x 10(3) fg/microl in respiratory and bone marrow samples. RT-PCR results were positive in serum from three HIV patients for which antibody detection by immunodiffusion was negative. Specificity was 100%, since PCR results were negative for all the control samples.

CONCLUSION

Thes RT-PCR technique is a sensitive, specific method for early diagnosis of histoplasmosis, particularly when respiratory secretions or bone marrow samples are analyzed. The reliability is lower in serum, but it can be used as an additional, complementary technique to culture and serology in HIV patients.

Detection of imported histoplasmosis in serum of HIV-infected patients using a real-time PCR-based assay

A new real-time PCR-based assay was used for detecting DNA of Histoplasma capsulatum in serum samples collected from four HIV-infected patients with proven histoplasmosis. The assay targeted the ITS1 region of rDNA and its in vitro sensitivity, specificity and reproducibility were evaluated. The technique detected DNA of H. capsulatum in all of the HIV-infected patients with proven histoplasmosis (4/4, 100%). The PCR result was positive for seven of the ten (70%) samples studied. The assay's specificity was determined to be 100%, since the method was negative for 25 other serum samples (10 from patients with proven aspergillosis and 15 from healthy controls). The PCR assay is a new and promising diagnostic alternative and further investigation is warranted.

Detección de Aspergillus spp. mediante PCR en tiempo real en un modelo murino de infección pulmonar

OBJECTIVES

Assessment of a real-time PCR technique for the detection and quantification of fungal DNA in a murine model of pulmonary aspergillosis.

METHODS

Male ICR specific pathogen-free mice were used in the studies. The animals were divided into groups: immunosuppressed and intranasally inoculated with various inoculum sizes (10(6), 10(5), 10(4), and 10(3) conidia/mL) of a clinical isolate of Aspergillus fumigatus. When symptoms of pulmonary aspergillosis were detected, the mice were killed and the lungs removed for culture and real-time PCR determination. The PCR reactions used primers that amplified a region of Aspergillus spp. ribosomal DNA. Survival time per experimental group was calculated and correlation coefficients with inoculum size, colony counts and PCR results were determined.

RESULTS

Average survival time was significantly associated with the size of the inoculum. Pulmonary colony count was positive for 90% of the infected mice, but there was no statistical relationship between count values and either survival time or inoculum size. Real-time PCR was positive in 100% of the animals and was significantly associated with survival time and inoculum size (p < 0.01).

CONCLUSION

Real-time PCR is a reliable procedure for the quantification and evaluation pulmonary infection due to A. fumigatus in animal models.

Targeted gene disruption of the 14-alpha sterol demethylase (cyp51A) in Aspergillus fumigatus and its role in azole drug susceptibility

The role of Aspergillus fumigatus 14alpha-sterol demethylase (Cyp51A) in azole drug susceptibility was assessed. Targeted disruption of cyp51A in azole-susceptible and -resistant strains decreased MICs from 2- to 40-fold. The cyp51A mutants were morphologically indistinguishable from the wild-type strain, retaining the ability to cause pulmonary disease in neutropenic mice.

Glucan synthase complex of Aspergillus fumigatus

The glucan synthase complex of the human pathogenic mold Aspergillus fumigatus has been investigated. The genes encoding the putative catalytic subunit Fks1p and four Rho proteins of A. fumigatus were cloned and sequenced. Sequence analysis showed that AfFks1p was a transmembrane protein very similar to other Fksp proteins in yeasts and in Aspergillus nidulans. Heterologous expression of the conserved internal hydrophilic domain of AfFks1p was achieved in Escherichia coli. Anti-Fks1p antibodies labeled the apex of the germ tube, as did aniline blue fluorochrome, which was specific for beta(1-3) glucans, showing that AfFks1p colocalized with the newly synthesized beta(1-3) glucans. AfRHO1, the most homologous gene to RHO1 of Saccharomyces cerevisiae, was studied for the first time in a filamentous fungus. AfRho proteins have GTP binding and hydrolysis consensus sequences identical to those of yeast Rho proteins and have a slightly modified geranylation site in AfRho1p and AfRho3p. Purification of the glucan synthase complex by product entrapment led to the enrichment of four proteins: Fks1p, Rho1p, a 100-kDa protein homologous to a membrane H(+)-ATPase, and a 160-kDa protein which was labeled by an anti-beta(1-3) glucan antibody and was homologous to ABC bacterial beta(1-2) glucan transporters.

The nucleotide sequence of Saccharomyces cerevisiae chromosome IV

The complete DNA sequence of the yeast Saccharomyces cerevisiae chromosome IV has been determined. Apart from chromosome XII, which contains the 1-2 Mb rDNA cluster, chromosome IV is the longest S. cerevisiae chromosome. It was split into three parts, which were sequenced by a consortium from the European Community, the Sanger Centre, and groups from St Louis and Stanford in the United States. The sequence of 1,531,974 base pairs contains 796 predicted or known genes, 318 (39.9%) of which have been previously identified. Of the 478 new genes, 225 (28.3%) are homologous to previously identified genes and 253 (32%) have unknown functions or correspond to spurious open reading frames (ORFs). On average there is one gene approximately every two kilobases. Superimposed on alternating regional variations in G+C composition, there is a large central domain with a lower G+C content that contains all the yeast transposon (Ty) elements and most of the tRNA genes. Chromosome IV shares with chromosomes II, V, XII, XIII and XV some long clustered duplications which partly explain its origin.

The nucleotide sequence of Saccharomyces cerevisiae chromosome XIV and its evolutionary implications

In 1992 we started assembling an ordered library of cosmid clones from chromosome XIV of the yeast Saccharomyces cerevisiae. At that time, only 49 genes were known to be located on this chromosome and we estimated that 80% to 90% of its genes were yet to be discovered. In 1993, a team of 20 European laboratories began the systematic sequence analysis of chromosome XIV. The completed and intensively checked final sequence of 784,328 base pairs was released in April, 1996. Substantial parts had been published before or had previously been made available on request. The sequence contained 419 known or presumptive protein-coding genes, including two pseudogenes and three retrotransposons, 14 tRNA genes, and three small nuclear RNA genes. For 116 (30%) protein-coding sequences, one or more structural homologues were identified elsewhere in the yeast genome. Half of them belong to duplicated groups of 6-14 loosely linked genes, in most cases with conserved gene order and orientation (relaxed interchromosomal synteny). We have considered the possible evolutionary origins of this unexpected feature of yeast genome organization.

The sequence of a 20.3 kb DNA fragment from the left arm of Saccharomyces cerevisiae chromosome IV contains the KIN28, MSS2, PHO2, POL3 and DUN1 genes, and six new open reading frames

We report the sequence of a 20 300 bp DNA fragment from the left arm of Saccharomyces cerevisiae chromosome IV. This segment contains 13 complete open reading frames (ORFs) and part of another ORF, altogether covering 84.2% of the entire sequence, five of which correspond to the previously characterized KIN28, MSS2, PHO2, POL3/CDC2 and DUN1 genes. One putative protein, D2358p, shares considerable homology with an O-sialoglycoprotein endopeptidase from Pasteurella haemolytica serotype A1. The putative product of D2325 contains the characteristic consensus motif of triacylglycerol lipases. D2320p and D2352p have a putative 'leucine-zipper' structure and a RNA-binding region Rnp-1 signature, respectively.

The sequence of a 21.3 kb DNA fragment from the left arm of yeast chromosome XIV reveals LEU4, MET4, POL1, RAS2, and six new open reading frames

The nucleotide sequence of a fragment of 21 308 bp from the left arm of Saccharomyces cerevisiae chromosome XIV has been determined. Analysis of the sequence revealed 13 open reading frames (ORFs) longer than 300 bp, four of which correspond to the previously identified genes LEU4, MET4, POL1 and RAS2. One putative protein, N2160, shares considerable homology (32% identity) with a hypothetical protein encoded by a gene located on chromosome XV as well as with human OCRL protein (36% identity), involved in Lowe's syndrome. N2185 contains ten predicted transmembrane segments and is similar to another putative protein (YKL146) from yeast.