Expression of Duplicate msa Genes in the Salmonid Pathogen Renibacteriumsalmoninarum

Renibacterium salmoninarum is a gram-positive bacterium responsiblefor bacterial kidney disease of salmon and trout. R. salmoninarumhas two identical copies of the gene encoding major solubleantigen (MSA), an immunodominant, extracellular protein. Todetermine whether one or both copies of msa are expressed, reporterplasmids encoding a fusion of MSA and green fluorescent proteincontrolled by 0.6 kb of promoter region from msa1 or msa2 wereconstructed and introduced into R. salmoninarum. Single copiesof the reporter plasmids integrated into the chromosome by homologousrecombination. Expression of mRNA and protein from the integratedplasmids was detected, and transformed cells were fluorescent,demonstrating that both msa1 and msa2 are expressed under invitro conditions. This is the first report of successful transformationand homologous recombination in R. salmoninarum.

INTRODUCTION

Top

Introduction

Bacterial kidney disease (BKD) in salmonid fish is a debilitating,systemic condition that is characterized by granulomatous lesions,primarily of the kidney and other internal organs. The diseaseis widely reported throughout North America, the British Isles,northern continental Europe, and Japan (14), with infectionprevalences among some hatchery populations in the western UnitedStates approaching 100% (8). BKD may be endemic among feralsalmon, where the prevalence of infection can exceed 30% (20,32). The listing of multiple stocks of wild salmon in the UnitedStates under the Endangered Species Act has resulted in substantialrecovery and restoration efforts, including the rearing of captivebroodstock. BKD poses a significant threat to those efforts.

The causative agent of BKD is Renibacterium salmoninarum, afastidious and slowly growing gram-positive diplococcobacillus(6, 9, 44). The bacterium is transmitted horizontally, probablyby a fecal-oral route (1), and vertically in the egg (11, 24).It produces an abundant, 57-kDa protein associated with theextracellular surface (13, 16, 46). This heat-stable protein,called major soluble antigen (MSA), is released by bacteriain situ (45), and it has been associated elsewhere with agglutinatingactivity (5, 7), immunomodulation (48), and virulence (2). Cloningof the msa gene was reported in 1992 (4), and subsequently itwas recognized that two identical copies existed (30), a rareoccurrence among prokaryotes. Because of the abundance of theMSA protein, we hypothesized that both copies of the msa geneare expressed.

To test this hypothesis, the ability to transform R. salmoninarumwas required. Until now, there has been no report of geneticmanipulation of R. salmoninarum. Here, we report the first successfultransformation of this bacterium and the chromosomal integrationof two reporter plasmids by single-crossover (SCO) homologousrecombination. Based on expression of the integrated plasmids,both msa1 and msa2 are expressed in R. salmoninarum under invitro culture conditions.

MATERIALS AND METHODS

Top

Materials and Methods

Bacterial strains, media, and culture conditions.
The type strain R. salmoninarum ATCC 33209 (American Type CultureCollection), which expresses abundant levels of MSA (5, 16,47), was used as the parental strain for transformation. R.salmoninarum isolate MT239 (2, 5, 41) was used in the fluorescenceanalysis. Liquid cultures were grown in modified KDM2 broth(1% Bacto Peptone, 0.05% yeast extract, 0.05% L-cysteine; adjustedto pH 6.5) at 15°C. Plate cultures were grown on modifiedKDM2 agar (1% Bacto Peptone, 0.05% yeast extract, 0.05% L-cysteine,10% newborn calf serum, 5% R. salmoninarum-conditioned medium[10]; adjusted to pH 7.5). Culture purity was determined byGram staining, and identity was confirmed by direct fluorescentantibody staining with an anti-R. salmoninarum polyclonal antibody(Kirkegaard and Perry Laboratories). Kanamycin selection wasat 50 µg/ml for both liquid and plate cultures. Recombinantstrains 42-8.5, 42-8.6, 43-6.1, and 43-6.2 were transformedwith pLDR109, and recombinant strains 43-7.1, 43-7.2, 43-8.1,and 43-8.2 were transformed with pLDR111.

Plasmid constructions were performed in Escherichia coli Top10F' {F' [lacI^q Tn10(Tet^r)] mcrA ${Delta}$ (mrr-hsdRMS-mcrBC) F80lacZ ${Delta}$ M15 ${Delta}$ lacX74 recA1 araD139 ${Delta}$ (ara-leu)7697 galU galK rpsL (Str^r) endA1nupG}.

Plasmids.
Plasmids used in this study are listed in Table 1. A 722-bpXbaI fragment from pEGFP (Clontech) containing the open readingframe (ORF) of enhanced green fluorescent protein (GFP) wasligated into the XbaI site of pZErO2.1 (Invitrogen), creatingpLDR104. The presumptive promoter regions of msa1 and msa2 wereobtained by PCR amplification of pCO2 and pCO4, respectively,which each contain the entire msa ORF and more than 1 kb offlanking sequence (30). The upstream primer for the msa1 PCRwas 5'-ATCATCGGATCCACGGGCTAACCCCGCCCT-3', and the upstream primerfor the msa2 PCR was 5'-ATCATCGGATCCAGTACTCGTATCGGCCTC-3'. Thedownstream primer for both reactions was 5'-AAAGGTACCATGGAAGAATTCCCTTCACCT-3'.The primers included BamHI or NcoI restriction sites near the3' end to facilitate cloning. PCR products were ligated intopCRII (Invitrogen), and BamHI-NcoI fragments were excised andcloned into pLDR104 to create the reporter plasmids. PlasmidpLDR109 contains 558 bp of 5' sequence flanking the msa1 ORF,102 bp of the msa ORF, and the entire gfp ORF. Plasmid pLDR111is identical to pLDR109 except that it contains 580 bp of 5'sequence flanking the msa2 ORF in place of the msa1 sequences.Plasmids were cycle sequenced with BigDye Terminator reactions(PE Biosystems) to confirm that the msa reporter plasmids encodeda chimeric protein bearing the N-terminal portion of MSA (includinga putative signal peptide [4]) fused to GFP.

View this table:
[in this window]
[in a new window]
TABLE 1. Plasmids used

Transformation of R. salmoninarum.
Plasmid DNA was prepared with column filters for plasmid purification(Qiagen) and was resuspended in water to a concentration of100 ng/µl. R. salmoninarum liquid cultures grown for 5to 7 days to an optical density at 525 nm (OD₅₂₅) of 1.0 to1.5 were harvested by centrifugation for 20 min at 10,000 xg. Pellets were resuspended in an equal volume of glycine medium(modified KDM2 broth supplemented with 0.5 M sucrose, 20 mMMgCl₂, 2 mM CaCl₂, and 1% glycine), and the cells were incubatedwith agitation at 15°C for 45 to 48 h. On the day of electroporation,1.5-ml aliquots of the glycine-treated cells were centrifugedand washed once with electroporation solution (0.5 M sucrosewith 10% glycerol). Washed cell pellets were resuspended in50 µl of electroporation solution and equilibrated at15°C for at least 15 min. Plasmid DNA (500 ng in 5 µl)was added, and the mixture was transferred to a 1-mm cuvette(Invitrogen) which had been equilibrated to 15°C. A singlepulse (12,000 V/cm, 50 mA, 50 W, 50 µF, 200 ${Omega}$ ) was deliveredwith a commercial electroporator (Invitrogen), and the mixturewas immediately transferred to 1 ml of ice-cold expression medium(modified KDM2 broth supplemented with 0.5 M sucrose, 20 mMMgCl₂, and 2 mM CaCl₂). Electroporated cells were allowed torecover at 15°C with agitation for 22 to 24 h. At the endof the recovery period, electroporated cells were selected onkanamycin-containing KDM2 plates at 15°C. Plates were examinedweekly.

Chromosomal DNA analysis.
Chromosomal DNA was prepared as previously described (33) andanalyzed by standard Southern blotting (38). Probes were randomprime labeled with digoxigenin-11-dUTP, and hybridized probeswere detected by chemiluminescence according to the manufacturer'sinstructions (Boehringer Mannheim). The msa probe was synthesizedfrom a 1.06-kb BglII fragment within the msa ORF of pCO2, andthe gfp probe was synthesized from an 0.76-kb XbaI fragmentcontaining the gfp ORF from pLDR104. PCR analysis of the integrationjunction was performed with copy-specific upstream primers (msa1primer, 5'-CATCCTGCGGCGAAGCACCCCTAAAC-3'; msa2 primer, 5'-CCACCACCGTCGAAGCCGAGAGTTG-3')and a common downstream primer (5'-AGGGTCAGCTTGCCGTAGGTGGCATC-3').

mRNA analysis.
Total RNA was isolated by a modification of the method of Yimand Rubens (49). Briefly, 25 ml of cells was grown to an OD₅₂₅of 0.8 and harvested by centrifugation for 20 min at 10,000x g. Cell pellets were resuspended in water and lysed in thepresence of acidified phenol, sodium acetate, and sodium dodecylsulfate (SDS) by vigorous vortexing with glass beads (150- to212-µm diameter; Sigma). After chloroform was added, thelysate was harvested by centrifugation, extracted twice withphenol and once with chloroform, and precipitated with sodiumacetate and isopropyl alcohol. The resulting pellet was resuspendedin water, and the nucleic acid concentration was measured byspectrophotometry. Northern blot analysis was performed by standardmethods with formaldehyde agarose gels (38). Probes were randomprime labeled with digoxigenin-11-dUTP, and hybridized probeswere detected by chemiluminescence according to the manufacturer'sinstructions (Boehringer Mannheim).

Protein analysis.
Equivalent numbers of R. salmoninarum cells in late logarithmicgrowth (OD₅₂₅ between 1.0 and 1.2) were harvested by centrifugationfor 10 min at 10,000 x g, and the cell pellets were resuspendedand lysed at 99°C in 1x sample buffer (0.31 M Tris, pH 6.8;2.5% glycerol; 0.5% SDS; 1.25% ß-mercaptoethanol).Heat-denatured samples were analyzed by SDS-polyacrylamide gelelectrophoresis and the standard immunoblot method with alkalinephosphatase visualization (18). GFP was detected with monoclonalantibody JL-1 (Clontech), and MSA was detected with monoclonalantibody 3H1 (47).

Analysis of fluorescence in transformants.
Liquid cultures were grown to an OD₅₂₅ between 0.5 and 0.8,and 0.5 ml of cells was harvested by centrifugation for 10 minat 10,000 x g, washed with 10 mM Tris (pH 7.5), and resuspendedin 200 µl of 10 mM Tris (pH 7.5). Serial halving dilutionswith 10 mM Tris were made in 96-well immunoassay plates (Costar).Cell density was read at 595 nm with an ELx808IU microplatereader (Bio-Tek), and fluorescence was read (excitation at 485nm; excitation at 530 nm) with an FLx800 fluorescence microplatereader (Bio-Tek). For each clone, fluorescence units were regressedon optical densities by a simple linear model, and regressionslopes were compared by Student's t test.

RESULTS

Top

Results

Transformation of R. salmoninarum.
Our previous efforts to transform R. salmoninarum with a plasmidcontaining the ColE1 origin of replication and a kanamycin resistancegene suggested that transient transformation was possible (datanot shown). The regions immediately upstream of the msa1 andmsa2 genes (558 and 580 bp, respectively) and 102 bp of theMSA ORF were amplified by PCR and cloned directly adjacent tothe GFP ORF, which had been inserted into the commercially availablevector pZErO2.1. The fusion protein from this construct wouldcontain the first 34 amino acids of MSA preceding the entireGFP. Plasmid pLDR109 contains the putative promoter sequencefrom msa1, and pLDR111 contains the promoter sequence from msa2.A control plasmid, pLDR104, contains the GFP reporter withoutR. salmoninarum sequences.

Plasmid DNA was introduced into R. salmoninarum ATCC 33209 byelectroporation. In addition to pLDR109 and pLDR111, parallelaliquots of bacteria were electroporated with no DNA, pZErO2.1,or pLDR104 as controls. All electroporated aliquots were incubatedon KDM2 plates containing kanamycin. By 16 weeks of incubation,no colonies developed from the aliquots that received no DNA.Four weeks after electroporation, colonies were visible at a20x magnification for all remaining aliquots. However, microscopiccolonies from aliquots receiving pZErO2.1 and pLDR104 failedto increase in size with continued incubation up to 16 weeks.After 11 weeks of incubation, larger fluorescent colonies appearedonly on plates inoculated with aliquots receiving pLDR109 andpLDR111. The cells from these fluorescent colonies were gram-positivediplococcobacilli and were labeled by the anti-MSA antibody3H1. This morphology and phenotype are consistent with thoseof R. salmoninarum. Integration frequencies varied between 5.8x 10^-8 and 8.8 x 10^-6/µg of DNA for six independent transformations.

Genetic analysis of transformed R. salmoninarum.
From the larger fluorescent colonies, eight clones (four fromtransformation by pLDR109 and four from transformation by pLDR111)were subjected to extraction procedures for chromosomal andplasmid DNA. The plasmid preparations were analyzed by gel electrophoresisand were used to transform E. coli for selection by kanamycinresistance. No plasmids were observed by gel electrophoresis,and no kanamycin-resistant E. coli colonies were isolated. Theseresults suggested that the plasmids may have integrated, inwhole or in part, into the R. salmoninarum chromosome. To testthis possibility, chromosomal DNA from the eight clones wasanalyzed by Southern blotting. Among the eight clones, the hybridizationpattern of BamHI-digested DNA was identical with that of eitherthe msa (Fig. 1A) or the gfp (Fig. 1B) probe. The pattern observedwas consistent with integration into the corresponding msa promoterregion by homologous recombination (Fig. 1C).

View larger version (54K):
[in this window]
[in a new window]
FIG. 1. Demonstration of homologous recombination in R. salmoninarum. Blot images were captured with an ImageStation 440CF with 1D Image Analysis software, version 3.5.3 (Kodak). (A) Chromosomal DNA from four integrant transformants with the msa1 reporter plasmid pLDR109, four integrant transformants with the msa2 reporter plasmid pLDR111, and the parental strain ATCC 33209 was digested with BamHI and probed with a BglII fragment from the msa ORF. (B) The blot from panel A was stripped and reprobed with an XbaI fragment containing the gfp ORF. In both panel A and panel B, molecular size markers are in the lane labeled "m.w.," and the sizes are shown in bases. (C) Schematic representation of the integrative plasmids and the proposed organization of the integration sites. The sizes of BamHI-digested chromosomal fragments predicted to hybridize to the msa or gfp probe are shown. The msa ORF is an open box, and the gfp ORF is a stippled box. Chromosomal DNA flanking the msa ORF is shown as a thin line. The cloned msa promoter region is a thick gray line, and the corresponding endogenous region is a thick black line. The remaining plasmid sequence is a thick hatched line, and the chromosomal sequences flanking the endogenous msa ORF and promoter are thin black lines. BamHI (B) sites are indicated. Small arrows indicate the positions of primers described in Results. The diagram is not drawn to scale.

If the integration occurred by precise homologous integration,the junction between the cloned promoter and the chromosomalpromoter sequences could be assessed. One method of assessmentthat we performed was by PCR with one primer positioned withinthe gfp ORF and the second primer positioned in the chromosomalregion flanking the cloned msa promoter (see primer positionsin Fig. 1C). Only an integration into the homologous regionwould result in a PCR product, and the sizes of the expectedPCR products would be 813 bp (for msa1) and 874 bp (for msa2).Integrations that varied 50 or more bp from the homologous sitewould result in a detectably larger PCR product or no product.PCR of chromosomal DNA from all eight clones generated productsof the predicted sizes, whereas the parental strain and thetwo reporter plasmids failed to produce specific PCR products(Fig. 2). From these results, we conclude that the regions fromthe msa genes that are cloned into the gfp reporter plasmidsare sufficient to mediate homologous recombination by SCO inR. salmoninarum.

View larger version (40K):
[in this window]
[in a new window]
FIG. 2. PCR assessment of the integration junction. PCR of chromosomal DNA from integrant transformants, the parental strain ATCC 33209, and each of the reporter plasmids was performed with a copy-specific primer and a primer positioned within the gfp ORF (see Fig. 1C for positions). Molecular size markers are in the lane labeled "100 bp," and the sizes are shown at right. The predicted sizes of precise SCO integrations by pLDR109 and pLDR111 are 880 and 874 bp, respectively. The gel image was captured with an ImageStation 440CF with 1D Image Analysis software, version 3.5.3.

The stability of the integrated plasmids was tested by successivepassage in broth culture without kanamycin selection. After2 weeks in the absence of selection (approximately 14 doublings),chromosomal DNA was analyzed by Southern blotting as previouslydescribed. The hybridization patterns with the msa and gfp probeswere identical to those in Fig. 1A and B (data not shown), suggestingthat the integrated plasmids do not require selection for maintenance.

Expression of GFP in transformed R. salmoninarum.
The purpose of the reporter plasmids was to determine whetherone or both of the msa genes were expressed. Northern blot analysisof total RNA with a gfp probe showed steady-state transcriptsfrom all eight of the transformed strains (Fig. 3). The sizeof these transcripts was approximately 800 bp, which is consistentwith the transcription start site occurring within the putativepromoter regions of the msa genes.

View larger version (65K):
[in this window]
[in a new window]
FIG. 3. Analysis of gfp transcript expression by Northern blotting. Total RNA was prepared from integrant transformants and the parental strain ATCC 33209, blotted, and probed with an XbaI fragment containing the gfp ORF. The blot image was captured with an ImageStation 440CF with 1D Image Analysis software, version 3.5.3.

To determine whether protein from the reporter plasmid was expressedin R. salmoninarum, mid-logarithmic-phase broth cultures ofeach of the eight clones were harvested, and equivalent numbersof cells were lysed and examined for GFP and MSA expressionby immunoblot analysis. All eight of the clones displayed acell-associated, 27- to 29-kDa protein that reacted with anti-GFPmonoclonal antibody JL-1, while the parental strain 33209 lackedthis protein (Fig. 4A). In contrast, all eight clones and theparental strain expressed abundant quantities of MSA protein(Fig. 4B). There was no evidence that the cell-associated levelsof MSA protein were significantly affected by the integrationof either reporter plasmid.

View larger version (82K):
[in this window]
[in a new window]
FIG. 4. Analysis of GFP and MSA by immunoblotting. Samples were prepared from the integrant transformants and the parental strain ATCC 33209. Protein molecular mass standards are in the lane labeled "m.w.," and the sizes are shown to the right. Cell lysate from E. coli expressing GFP from pEGFP is in the farthest right lane. Blot images were captured with an ImageStation 440CF with 1D Image Analysis software, version 3.5.3. (A) Whole-cell lysates of equivalent numbers of cells were probed with anti-GFP monoclonal antibody JL-1. (B) Whole-cell lysates of equivalent numbers of cells were probed with anti-MSA monoclonal antibody 3H1.

To determine whether the transformed R. salmoninarum clonesdisplayed increased fluorescence as a result of GFP expression,serial dilutions of broth cultures of the transformed clones,the parental strain 33209, and another untransformed isolateof R. salmoninarum (MT239) were measured for fluorescence (excitationat 485 nm, emission at 530 nm) and for cell density (OD₅₉₅).For each strain, a regression of fluorescence on cell densitywas determined, and the regression slopes and y intercepts werecompared. Although untransformed R. salmoninarum exhibits aslight autofluorescence, the transformed clones displayed significantlygreater fluorescence (P < 0.01) than did the untransformedisolates of R. salmoninarum (Fig. 5). There were no significantdifferences in regression slopes or y intercepts among the eighttransformed clones (P > 0.05). Epifluorescence microscopyof the transformed clones revealed that all eight clones werevisibly fluorescent either as colonies on plates or in wet mountsof resuspended cells (data not shown). Based on these results,we reach several conclusions. First, GFP is expressed from eachof the integrated reporter constructs. Second, the expressedGFP is properly folding because increases in fluorescence canbe observed. Finally, the levels of expression of GFP from thetwo msa promoters are similar.

View larger version (20K):
[in this window]
[in a new window]
FIG. 5. Fluorescence of integrant transformants, the parental strain ATCC 33209, and a second, untransformed isolate of R. salmoninarum, MT239. Serial halving dilutions of broth cultures were measured for OD₅₉₅ (x axis) and fluorescence at 485-nm excitation and 535-nm emission (y axis). (A) Comparison of four integrant transformants with the msa1 reporter plasmid pLDR109 with untransformed R. salmoninarum. (B) Comparison of four integrant transformants with the msa2 reporter plasmid pLDR111 with untransformed R. salmoninarum.

DISCUSSION

Top

Discussion

Until now, there have been no reports of genetic manipulationof R. salmoninarum. The ability to introduce DNA into this problematic,gram-positive pathogen would open a new arena of research. Thisis the first report of both successful transformation and site-specificintegration of exogenous DNA into the R. salmoninarum chromosome.Chromosomal analyses show that the reporter plasmids integratedby an SCO event, perhaps through a Campbell insertion mechanism(3).

Based on the integrated GFP reporter plasmids, both msa1 andmsa2 are expressed under in vitro culture conditions. This findingwas not unexpected, because the ORFs of msa1 and msa2 are identical(30), suggesting that there may be selective pressure on bothcopies. While gene duplication is widely observed among bacterialgenomes (23, 42), the occurrence of identical copies of genesis unusual except for ribosomal DNA and tDNA (34). Among geneswith high (>95%) nucleotide identity, all copies are typicallyexpressed. Duplicate methane monooxygenase gene clusters inMethylococcus capsulatus Bath are nearly identical (99.996%),and disruption of any of the genes within either cluster reducedthe growth rate, indicating that each copy was expressed (40,43). Similarly, insertional inactivation of nearly identical(99.879%) amoA genes within duplicated ammonia monooxygenasegene clusters in Nitrosomonas europaea showed that both copiesare expressed (19, 28). Duplicate genes tufA and tufB both appearto contribute to the synthesis of the encoded gene product,elongation factor EF-Tu, a highly abundant cytoplasmic proteinin E. coli (12, 15, 39, 50). Gene duplication is a way to amplifyproduction of a gene product or to ensure the synthesis of anessential, multifunctional protein. Expression of both msa1and msa2 may be required for high levels of MSA. In vivo, MSAoccurs in such high abundance that it is the principal antigenrecognized by a polyclonal immunosorbent diagnostic assay forR. salmoninarum in salmon kidney tissue (31). Gene duplicationalso provides a way to differentially regulate expression ofthe same protein. For example, lysyl-tRNA synthetase genes inE. coli are differentially expressed, with lysS displaying constitutivebehavior and lysU exhibiting inducible behavior (21, 37). Whileboth copies of msa have identical sequences up to 40 bp upstreamof the ORF, the sequences beyond are quite divergent. If promotersor other regulatory sequences occur in the divergent region,msa1 and msa2 may be differentially expressed under in vivoconditions.

The differences in the amounts of GFP from the reporter genesand MSA from the endogenous msa genes are striking. BecauseMSA is a high-abundance protein both in culture and in infectedfish (35, 45), we expected GFP expression to be substantial.Although the GFP encoded by the integrated genes has a reportedin vivo half-life of >24 h (27), it is possible that GFPstability was reduced by the presence of the MSA sequences atthe amino terminus of the fusion protein. Alternatively, thefusion protein may lack sequences that confer stability on theMSA protein. Although MSA can undergo proteolysis at elevatedtemperatures, it exhibits no degradation after 10 h at 15°C(17, 36), the optimal culture temperature for R. salmoninarum.The high levels of MSA observed in tissues of infected fish(up to 200 µg/g) and its stability at ambient environmentaltemperatures suggest that MSA is persistent under a range ofconditions.

Each of the reporter plasmids integrated into the targeted msagene copy. As a result, the endogenous 5'-flanking sequencewas preserved, and the msa-gfp ORF fusion was subject to thefull promoter context for each msa gene. In this respect, anintegrated reporter is superior to an extrachromosomal reportergene because no assumptions about the size or context of thepromoter need to be made. While SCO integration is more frequentlyused for gene inactivation or heterologous gene expression,this and other studies (29) demonstrate its utility for monitoringpromoter activity.

The vector plasmid pZErO2.1 and the promoterless GFP reporterconstruct pLDR104 failed to produce integrants, suggesting thatthe plasmids lacked sequences that ensured sustained extrachromosomalreplication. However, we found evidence that the ColE1-basedplasmid could persist in R. salmoninarum for a transient period.The appearance of microscopic kanamycin-resistant colonies ofR. salmoninarum after 4 weeks of incubation in all electroporationgroups, except the group receiving no DNA, indicated that theplasmids had not been rapidly destroyed by an endogenous restrictionsystem and that the kanamycin gene was expressed. These microscopiccolonies failed to enlarge with prolonged (>16 weeks) incubation,leading us to speculate that the initial transformants may havereceived multiple plasmids which were subsequently segregatedto daughter cells. As a result, the ColE1-based plasmid behavedas a suicide vector, favoring integrations under selective pressure.

The frequency of recombinants obtained was extremely low. Thismay be a limitation of the length of the homologous regionsin the reporter plasmids, a site-dependent effect on recombinationfrequency, or a result of suboptimal electrotransformation conditions.The transformation efficiency by plasmid insertion among gram-positivebacteria demonstrates a strong dependence on the length of thehomologous site (25). Because the 600- to 700-bp homologousregions used in our GFP reporter plasmids are well above theshortest lengths of homology capable of recombination observedfor Lactococcus sake (300 bp [26]), Streptococcus pneumoniae(96 bp [25]), and Bacillus subtilis (70 bp [22]), we concludethat low recombination efficiency was not due to this factor.In order to determine whether there is a site-dependent effecton recombination by the msa promoter regions, comparisons withsimilarly sized inserts from other parts of the R. salmoninarumchromosome would be needed. When this comparison was made forS. pneumoniae, there was a very small site-dependent effecton recombination frequency among nondisrupted or nonessential,disrupted gene targets (25). Because only one of the msa genecopies was targeted by each reporter plasmid and MSA proteinexpression was not reduced, it is not likely that the integratedreporter plasmids disrupted an essential gene function. Thissuggests that any site-dependent effect on recombination frequencywas not strong. On the other hand, suboptimal electrotransformationconditions would exert a strong influence on the frequency ofrecombinants recovered. While insertional mutagenesis of targetedgenes is now possible by the method described here, generatingrandom mutagenic libraries with a high degree of saturationwill require substantial improvements in transformation efficiency.

ACKNOWLEDGMENTS

We thank Cynthia Masada for helpful discussions and review ofthe manuscript. The monoclonal antibody 3H1 was a gift fromJames Winton and Ronald Pascho.

This work was supported by CSREES grant 00-35204-9225 from theU.S. Department of Agriculture, and A.M.C. is an undergraduateintern supported by this grant. Additional support was providedby the National Marine Fisheries Service, National Oceanic andAtmospheric Administration, U.S. Department of Commerce.

FOOTNOTES

* Corresponding author. Mailing address: Northwest Fisheries Science Center, REUT Division, 2725 Montlake Blvd. East, Seattle, WA 98112. Phone: (206) 860-3279. Fax: (206) 860-3467. E-mail: Linda.Rhodes{at}noaa.gov.

REFERENCES

Top