Limits...
Transcriptional response of honey bee larvae infected with the bacterial pathogen Paenibacillus larvae.

Cornman RS, Lopez D, Evans JD - PLoS ONE (2013)

Bottom Line: We identified 75 genes with significantly higher expression and six genes with significantly lower expression.However, analysis of Drosophila homologs of differentially expressed genes revealed spatial and temporal patterns consistent with developmental asynchrony as a likely confounder of our results.The consistently responsive genes in our test set included a hymenopteran-specific protein tyrosine kinase, a hymenopteran specific serine endopeptidase, a cytochrome P450 (CYP9Q1), and a homolog of trynity, a zona pellucida domain protein.

View Article: PubMed Central - PubMed

Affiliation: Bee Research Laboratory, Agricultural Research Service of the United States Department of Agriculture, Beltsville, Maryland, United States of America. scott.cornman@gmail.com

ABSTRACT
American foulbrood disease of honey bees is caused by the bacterium Paenibacillus larvae. Infection occurs per os in larvae and systemic infection requires a breaching of the host peritrophic matrix and midgut epithelium. Genetic variation exists for both bacterial virulence and host resistance, and a general immunity is achieved by larvae as they age, the basis of which has not been identified. To quickly identify a pool of candidate genes responsive to P. larvae infection, we sequenced transcripts from larvae inoculated with P. larvae at 12 hours post-emergence and incubated for 72 hours, and compared expression levels to a control cohort. We identified 75 genes with significantly higher expression and six genes with significantly lower expression. In addition to several antimicrobial peptides, two genes encoding peritrophic-matrix domains were also up-regulated. Extracellular matrix proteins, proteases/protease inhibitors, and members of the Osiris gene family were prevalent among differentially regulated genes. However, analysis of Drosophila homologs of differentially expressed genes revealed spatial and temporal patterns consistent with developmental asynchrony as a likely confounder of our results. We therefore used qPCR to measure the consistency of gene expression changes for a subset of differentially expressed genes. A replicate experiment sampled at both 48 and 72 hours post infection allowed further discrimination of genes likely to be involved in host response. The consistently responsive genes in our test set included a hymenopteran-specific protein tyrosine kinase, a hymenopteran specific serine endopeptidase, a cytochrome P450 (CYP9Q1), and a homolog of trynity, a zona pellucida domain protein. Of the known honey bee antimicrobial peptides, apidaecin was responsive at both time-points studied whereas hymenoptaecin was more consistent in its level of change between biological replicates and had the greatest increase in expression by RNA-seq analysis.

Show MeSH

Related in: MedlinePlus

Relative expression of honey bee genes, binned by Drosophila expression cluster.For each developmental gene-expression cluster defined by the modENCODE project [30], the closest honey bee homolog of each Drosophila gene in that cluster was identified by BLAST and binned accordingly. The left panel shows the log2 relative abundance of each honey bee gene binned into the given cluster (green squares), together with all other honey bee genes (grey squares). The values are plotted as a function of transcript length (X axis), which contributes to the variance in differential expression estimates and could potentially co-vary with expression cluster. The middle panel shows a histogram of log2 abundance for all honey bee genes mapped to the cluster. The right panel shows the median relative expression (scaled from 0 to 1) of Drosophila genes in each expression cluster during embryonic, larval, and pupal developmental stages, thereby illustrating the characteristic pattern of gene expression in each cluster.
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC3675105&req=5

pone-0065424-g004: Relative expression of honey bee genes, binned by Drosophila expression cluster.For each developmental gene-expression cluster defined by the modENCODE project [30], the closest honey bee homolog of each Drosophila gene in that cluster was identified by BLAST and binned accordingly. The left panel shows the log2 relative abundance of each honey bee gene binned into the given cluster (green squares), together with all other honey bee genes (grey squares). The values are plotted as a function of transcript length (X axis), which contributes to the variance in differential expression estimates and could potentially co-vary with expression cluster. The middle panel shows a histogram of log2 abundance for all honey bee genes mapped to the cluster. The right panel shows the median relative expression (scaled from 0 to 1) of Drosophila genes in each expression cluster during embryonic, larval, and pupal developmental stages, thereby illustrating the characteristic pattern of gene expression in each cluster.

Mentions: We then downloaded whole-organism microarray data for defined developmental stages of Drosophila, available from modENCODE [30] and viewable in Flybase [37]. Inspection of these expression data revealed a frequent pattern in which homologs of differentially expressed honey-bee genes have rapidly increasing expression early in larval development, achieve high maxima in mid-larval stages, and then rapidly decline in late larval development. Roy and colleagues [30] defined over 30 developmental gene-expression clusters from these data, which we used to bin differentially expressed genes (Table 1). Expression cluster 7 was highly represented among these genes (28 of the 69 genes with Drosophila homologs, or 40.6%), and has a single strong peak at the L1 stage of Drosophila development (Fig. 4 bottom-right panel). Moreover, we can reverse the direction of the comparison and identify all A. mellifera homologs (best TBLASTX match with an expectation threshold of 1E-10) of D. melanogaster genes in each expression cluster and investigate the pattern of change in each group, rather than just examining significantly differentially expressed genes. Plotting the expression differentials of honey bee homologs to Drosophila genes for each of the most-frequent expression clusters in Table 1 (clusters 4, 7, 12, 25, and 29), cluster 7 displays a clearly bimodal pattern of differential expression with the primary mode at a log2 difference of approximately 2 and a secondary mode near zero (Fig. 4). The mean log2 differential in infected bees for cluster 7 genes was 1.06.


Transcriptional response of honey bee larvae infected with the bacterial pathogen Paenibacillus larvae.

Cornman RS, Lopez D, Evans JD - PLoS ONE (2013)

Relative expression of honey bee genes, binned by Drosophila expression cluster.For each developmental gene-expression cluster defined by the modENCODE project [30], the closest honey bee homolog of each Drosophila gene in that cluster was identified by BLAST and binned accordingly. The left panel shows the log2 relative abundance of each honey bee gene binned into the given cluster (green squares), together with all other honey bee genes (grey squares). The values are plotted as a function of transcript length (X axis), which contributes to the variance in differential expression estimates and could potentially co-vary with expression cluster. The middle panel shows a histogram of log2 abundance for all honey bee genes mapped to the cluster. The right panel shows the median relative expression (scaled from 0 to 1) of Drosophila genes in each expression cluster during embryonic, larval, and pupal developmental stages, thereby illustrating the characteristic pattern of gene expression in each cluster.
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC3675105&req=5

pone-0065424-g004: Relative expression of honey bee genes, binned by Drosophila expression cluster.For each developmental gene-expression cluster defined by the modENCODE project [30], the closest honey bee homolog of each Drosophila gene in that cluster was identified by BLAST and binned accordingly. The left panel shows the log2 relative abundance of each honey bee gene binned into the given cluster (green squares), together with all other honey bee genes (grey squares). The values are plotted as a function of transcript length (X axis), which contributes to the variance in differential expression estimates and could potentially co-vary with expression cluster. The middle panel shows a histogram of log2 abundance for all honey bee genes mapped to the cluster. The right panel shows the median relative expression (scaled from 0 to 1) of Drosophila genes in each expression cluster during embryonic, larval, and pupal developmental stages, thereby illustrating the characteristic pattern of gene expression in each cluster.
Mentions: We then downloaded whole-organism microarray data for defined developmental stages of Drosophila, available from modENCODE [30] and viewable in Flybase [37]. Inspection of these expression data revealed a frequent pattern in which homologs of differentially expressed honey-bee genes have rapidly increasing expression early in larval development, achieve high maxima in mid-larval stages, and then rapidly decline in late larval development. Roy and colleagues [30] defined over 30 developmental gene-expression clusters from these data, which we used to bin differentially expressed genes (Table 1). Expression cluster 7 was highly represented among these genes (28 of the 69 genes with Drosophila homologs, or 40.6%), and has a single strong peak at the L1 stage of Drosophila development (Fig. 4 bottom-right panel). Moreover, we can reverse the direction of the comparison and identify all A. mellifera homologs (best TBLASTX match with an expectation threshold of 1E-10) of D. melanogaster genes in each expression cluster and investigate the pattern of change in each group, rather than just examining significantly differentially expressed genes. Plotting the expression differentials of honey bee homologs to Drosophila genes for each of the most-frequent expression clusters in Table 1 (clusters 4, 7, 12, 25, and 29), cluster 7 displays a clearly bimodal pattern of differential expression with the primary mode at a log2 difference of approximately 2 and a secondary mode near zero (Fig. 4). The mean log2 differential in infected bees for cluster 7 genes was 1.06.

Bottom Line: We identified 75 genes with significantly higher expression and six genes with significantly lower expression.However, analysis of Drosophila homologs of differentially expressed genes revealed spatial and temporal patterns consistent with developmental asynchrony as a likely confounder of our results.The consistently responsive genes in our test set included a hymenopteran-specific protein tyrosine kinase, a hymenopteran specific serine endopeptidase, a cytochrome P450 (CYP9Q1), and a homolog of trynity, a zona pellucida domain protein.

View Article: PubMed Central - PubMed

Affiliation: Bee Research Laboratory, Agricultural Research Service of the United States Department of Agriculture, Beltsville, Maryland, United States of America. scott.cornman@gmail.com

ABSTRACT
American foulbrood disease of honey bees is caused by the bacterium Paenibacillus larvae. Infection occurs per os in larvae and systemic infection requires a breaching of the host peritrophic matrix and midgut epithelium. Genetic variation exists for both bacterial virulence and host resistance, and a general immunity is achieved by larvae as they age, the basis of which has not been identified. To quickly identify a pool of candidate genes responsive to P. larvae infection, we sequenced transcripts from larvae inoculated with P. larvae at 12 hours post-emergence and incubated for 72 hours, and compared expression levels to a control cohort. We identified 75 genes with significantly higher expression and six genes with significantly lower expression. In addition to several antimicrobial peptides, two genes encoding peritrophic-matrix domains were also up-regulated. Extracellular matrix proteins, proteases/protease inhibitors, and members of the Osiris gene family were prevalent among differentially regulated genes. However, analysis of Drosophila homologs of differentially expressed genes revealed spatial and temporal patterns consistent with developmental asynchrony as a likely confounder of our results. We therefore used qPCR to measure the consistency of gene expression changes for a subset of differentially expressed genes. A replicate experiment sampled at both 48 and 72 hours post infection allowed further discrimination of genes likely to be involved in host response. The consistently responsive genes in our test set included a hymenopteran-specific protein tyrosine kinase, a hymenopteran specific serine endopeptidase, a cytochrome P450 (CYP9Q1), and a homolog of trynity, a zona pellucida domain protein. Of the known honey bee antimicrobial peptides, apidaecin was responsive at both time-points studied whereas hymenoptaecin was more consistent in its level of change between biological replicates and had the greatest increase in expression by RNA-seq analysis.

Show MeSH
Related in: MedlinePlus