Genomic analysis of the secretion stress response in the enzyme-producing cell factory Aspergillus niger

Abstract

Background: Filamentous fungi such as Aspergillus niger have a high capacity secretory system and are therefore widely exploited for the industrial production of native and heterologous proteins. However, in most cases the yields of non-fungal proteins are significantly lower than those obtained for fungal proteins. One well-studied bottleneck appears to be the result of mis-folding of heterologous proteins in the ER during early stages of secretion, with related stress responses in the host, including the unfolded protein response (UPR). This study aims at uncovering transcriptional and translational responses occurring in A. niger exposed to secretion stress.

Results: A genome-wide transcriptional analysis of protein secretion-related stress responses was determined using Affymetrix DNA GeneChips and independent verification for selected genes. Endoplasmic reticulum (ER)-associated stress was induced either by chemical treatment of the wild-type cells with dithiothreitol (DTT) or tunicamycin, or by expressing a human protein, tissue plasminogen activator (t-PA). All of these treatments triggered the UPR, as shown by the expression levels of several well-known UPR target genes. The predicted proteins encoded by most of the up-regulated genes function as part of the secretory system including chaperones, foldases, glycosylation enzymes, vesicle transport proteins, and ER-associated degradation proteins. Several genes were down-regulated under stress conditions and these included several genes that encode secreted enzymes. Moreover, translational regulation under ER stress was investigated by polysomal fractionation. This analysis confirmed the post-transcriptional control of hacA expression and highlighted that differential translation also occurs during ER stress, in particular for some genes encoding secreted proteins or proteins involved in ribosomal biogenesis and assembly.

Conclusion: This is first genome-wide analysis of both transcriptional and translational events following protein secretion stress. Insight has been gained into the molecular basis of protein secretion and secretion-related stress in an effective protein-secreting fungus, and provides an opportunity to identify target genes for manipulation in strain improvement strategies.

Figure 1

GeneChip results were confirmed for some genes using Northern blotting and RT-PCR. Examples are provided for both Northern blots (A) and RT-PCR (B). Note that the RT-PCR for the hacA mRNA was designed to indicate enhanced splicing of the hacA mRNA intron under stress conditions (DTT, tunicamycin and production of t-PA). This is shown as a relative increase in the amount of the processed (lower band) hacAⁱ form of the mRNA compared to the unprocessed higher band (hacA^u). Probing or PCR for an actin gene was used as a non-stress-responsive control transcript.

Figure 2

Hierarchical clustering of records in a dendrogram (tree graph) based on the similarity of the signal log ratios obtained in each of the duplicate stress studies. Records (188) were selected based on differential expression in the tPA comparisons. This tree is representative for multiple clusterings performed using signals or signal log ratios. The genes have been rearranged into their cluster order and are represented on the vertical axis. The experiments are represented on the horizontal axis. The significance of the colour scale is indicated.

Figure 3

Venn diagrams of the numbers of overlapping and non-overlapping induced (≥ 1.5 fold) or repressed (≤ 1.5 fold) genes on the A. niger array after exposure to DTT or tunicamycin (Tun) and in the t-PA producing strain (t-PA).

Figure 4

(A) Representative absorbance profile for RNA separated by velocity sedimentation through a 15–60% sucrose gradient. Fractions are numbered from the top to the bottom of the gradient. (B) RNA was extracted from each fraction and subjected to electrophoresis through a formaldehyde gel. The ribosomal RNA distribution profile (25S, 18S and 5S rRNA; indicated by arrowheads) enables the the assignment of OD₂₅₄ peaks, corresponding to the 40S and 60S ribosomal subunits and to intact ribosomes (80S). (C) RT-PCR analysis with hacA primers were performed from each fraction of collected gradients from treated and non-treated cells. The full length hacA^u mRNA (arrow) as well as low-molecular-weight version of hacAⁱ (arrowhead) can be visualized.

Figure 5

Model of the secretory pathway under different ER stress conditions (t-PA secretion, tunicamycin and DTT) together with examples of genes that are transcriptionally induced or repressed. The A. niger gene designation is provided where previously known or, otherwise, the S. cerevisiae gene name is provided. Red, genes up-regulated by 3 conditions; orange, genes up-regulated by 2 conditions; yellow, genes up-regulated by 1 condition; light blue, genes down-regulated by 1 condition; blue, genes down-regulated by 2 conditions. N, nucleus; ER, endoplasmic reticulum; E, endosome; V, vacuole; G, Golgi. ERAD is ER-associated degradation.