Effects of the protease inhibitor protein on primary symbiotic bacteria of pea aphid
By real-time fluorescence quantitative PCR analysis, the effects of the protease inhibitor protein on the number of B. aphidicola in pea aphids were investigated. groEL gene was chosen to reflect the number of B. aphidicola as the previous report [24]. Relative number of B. Aphidicola in aphids fed on the diet containing different concentration of the protease inhibitor protein were calculated. The results showed that the pea aphid fed artificial diet containing 100 and 500 µg/ml the protease inhibitor protein for 4 days had only 73% and 54% of B. aphidicola respectively, compared with the control. The quantity of B. aphidicola in the pea aphid fed with 500 µg/ml protease inhibitor protein reduced significantly compared with the control (Fig. 1) (F2,8=5.33, df = 2,8, P < 0.05).
Effects of the protease inhibitor protein on B. aphidicola membrance protease gene expression
The effects of protease inhibitor protein Xbpi-1 on the gene expression of three B. aphidicola membrance proteases were investigated. The results showed that the three membrance protease expressions were increased significantly at 100 µg/ml by 4.09, 5.10, 3.53 fold respectively. The expressions of protease hflK and lepB were also increased at 500 µg/ml (Table 1). However, the expression of protease ftsH was decreased in aphids at 500 µg/ml (-2.49 fold compared with that of the control).
Table 1
The effects of Xbpi-1, protease inhibitor protein, on Buchnera aphidicola protease gene expression in pea aphids by real- time fluorescence quantitative RT-PCR.
Treatment
|
Expression fold difference (log2)
|
hflK
|
lepB
|
ftsH
|
Ck
|
0.000001 ± 0.050569a
|
0.000001 ± 0.094321a
|
0.000001 ± 0.001624a
|
Xbpi-1 (100 µg/ml)
|
4.099772 ± 0.094323c
|
5.100794 ± 0.145534c
|
3.538084 ± 0.278181b
|
Xbpi-1 (500 µg/ml)
|
1.885456 ± 0.291693b
|
4.125319 ± 0.280705b
|
-2.491182 ± 0.118481c
|
Data in the table are shown as mean ± SE. Means followed by the same letter are not significantly different.
Effects of the protease inhibitor protein on cultivable symbiotic bacteria in pea aphid
The protease inhibitor protein inhibited the aphid-cultivable symbiotic bacteria growth. There were colonies growing in the control group and no visible colonies in the treatment plates containing the protease inhibitor protein (100 µg/ml) (Fig. 2). The bacteria colonies on the control group can be divided into 2 categories according to their morphology. One type of colony was yellow, round and larger (colony number from 1–2 for different replications). The other one appeared white and smaller (colony number from 6–9 for different replications). By molecular techniques, 16S rDNA sequences of bacteria colonies were obtained and blasted by the NCBI database (Table 2). We found that the 16S rDNA sequence of yellow strain (YELLOW-1) was very close to the 16S rDNA sequence of Staphylococcus sciuri gb|HQ154580.1|, and only one nucleotide difference. This bacterium was reported in the pea aphid honeydew [25]. It may secret chemical pheromones to attract hoverflies–one kind of aphid predator. The 16S rDNA sequences of the other white strains (WHITE-1-4) shared 98% similarity. They had the highest similarity with the 16S rDNA sequence gb|JN644522.1|, belonged to an epidermal bacterium Staphylococcus epidermidis. This bacterium was reported in the gut of wild culex quinquefasciatus mosquito [26], but not reported in aphids before.
Table 2
Information of aphid cultivable bacteria by Blast against GenBank database
Bacterial strain
|
Length of sequence
(bp)
|
Accession number
|
Identities
(%)
|
Species of identification sequence comes from
|
YELLOW-1
|
1458
|
gb|HQ154580.1|
|
99
|
Staphylococcus sciuri strain R10-5A
|
WHITE-1
|
1466
|
gb|JN644522.1|
|
99
|
Staphylococcus epidermidis strain JDM2_4A
|
WHITE-2
|
1466
|
gb|JN644522.1|
|
99
|
Staphylococcus epidermidis strain JDM2_4A
|
WHITE-3
|
1464
|
gb|JN644522.1|
|
99
|
Staphylococcus epidermidis strain JDM2_4A
|
WHITE-4
|
1473
|
gb|JN644522.1|
|
99
|
Staphylococcus epidermidis strain JDM2_4A
|
WHITE-5
|
1463
|
gb|JN644522.1|
|
99
|
Staphylococcus epidermidis strain JDM2_4A
|
Effects of protease inhibitor protein on pea aphid mortality
The effects of Xbpi-1 protein on pea aphid mortality were detected. Results showed that the aphid mortality was 13.8, 19.8% and 47.8% respectively, after aphids fed on diets containing Xbpi-1 protein at the concentrations of 0 (the control), 100 and 500 µg/ml for 4 days (Fig. 3). The mortality of pea aphid of treatment group (diet containing 500 µg/ml protease inhibitor protein) increased significantly compared with the control (F2,15=26.13, df = 2, 15, P < 0.001).
Sequencing and sequence alignment
After removal of adaptor sequences, ambiguous sequencing reads and low-quality reads, 29,993,490 and 36,617,886 high-quality clean reads were obtained from the control and treatment groups, respectively. All high-quality reads were mapped to the genome sequences with 16,562,318 control (CK) and 21,729,520 (PIP) using the top hat program [27]. There were 15,076,489 and 19,362,763 reads uniquely mapped to the genome, respectively. Of these reads, 98.22% and 98.19% were mapped to the exon regions, respectively (Fig. 4).
Differentially expressed gene (DEG) statistics
Among these genes, we focused on the differentially expressed genes. There were 213 genes those were significantly differentially expressed between the control and treatment groups in our study (remove the gene records of which the read counts < 20) (p < 0.05 and fold change > 1.5 or fold change < 0.67). Among those genes, 187 were up-regulated, and 26 were down-regulated in PIP treatment samples comparing with those of the control samples (Fig. 5). Of these genes, 81 had defined functions containing heat shock protein, cuticle protein, energy mechanism, protease, binding protein, transcription factors, peroxidase, G-protein-coupled receptor, immune related proteins and so on (Table 3).
Table 3. Differential Gene Expression
|
Transcript_id
|
gene_id
|
locus
|
identification sequence
|
log2-fold change
|
p value
|
q value
|
ACYPI34023-RA
|
acyp2eg0032998
|
GL353789:1948-6100
|
uncharacterized protein
|
-3.32315
|
5.48E-08
|
3.40E-05
|
ACYPI086488-RA
|
LOC100575796
|
GL350115:165328-166841
|
DNA repair helicase rad3/xp-D
|
-2.85341
|
5.75E-13
|
1.07E-09
|
ACYPI35543-RA
|
LOC100573825
|
GL349698:1112300-1114895
|
uncharacterized protein
|
-2.13864
|
7.88E-09
|
7.34E-06
|
ACYPI007459-RA
|
ACYPI007459
|
GL349649:65637-67551
|
uncharacterized protein
|
-2.10767
|
9.33E-12
|
1.49E-08
|
ACYPI009117-RA
|
acyp2eg0020355
|
GL350004:775205-777838
|
heat shock protein 70 B2-like
|
2.07107
|
3.11E-13
|
6.95E-10
|
ACYPI007961-RA
|
acyp2eg0000583
|
GL349623:1728254-1730648
|
Heat shock cognate 71
|
2.28951
|
2.22E-15
|
1.08E-11
|
ACYPI069455-RA
|
acyp2eg0029488
|
GL350856:40987-43713
|
heat shock protein 70 A1-like
|
2.39213
|
1.33E-15
|
1.08E-11
|
ACYPI004698-RA
|
acyp2eg0000333
|
GL349622:1460225-1468910
|
Acyrthosiphon pisum heat shock protein 70 A1-like
|
2.44622
|
2.89E-15
|
1.08E-11
|
ACYPI000903-RA
|
acyp2eg0004663
|
GL349652:265322-267312
|
heat shock protein 68-like
|
2.4631
|
8.54E-10
|
1.19E-06
|
ACYPI083434-RA
|
acyp2eg0013336
|
GL349786:388704-391066
|
heat shock 68 KD protein cognate
|
2.54302
|
7.57E-14
|
2.12E-10
|
ACYPI062698-RA
|
acyp2eg0029611
|
GL350880:10489-12795
|
uncharacterized protein
|
2.608
|
0.000213868
|
0.0318806
|
ACYPI008994-RA
|
acyp2eg0020797
|
GL350028:164676-167549
|
uncharacterized protein
|
2.63433
|
7.78E-09
|
7.34E-06
|
ACYPI061343-RA
|
acyp2eg0030512
|
GL351140:7285-10308
|
uncharacterized protein
|
3.40795
|
2.18E-05
|
0.00566211
|
ACYPI065480-RA
|
acyp2eg0032727
|
GL353380:0-1015
|
uncharacterized protein
|
4.77093
|
1.01E-09
|
1.25E-06
|
ACYPI007593-RA
|
LOC100166743
|
GL349635:1317279-1328235
|
uncharacterized protein
|
1.79769e+308
|
1.07E-08
|
9.23E-06
|
GO assignments of DEGs
By GO analysis, there were 213 DEG sequences assigned to GO terms based on BLAST matches with sequences of known function (Fig. 6). The transcripts were assigned to biological processes, cellular components and molecular functions. Among the biological process terms, a high percentage of genes were assigned to response to stress processes (32.14%), followed by proteolysis (7.14%) and G-protein-coupled receptor signaling pathway (7.14%). The cellular component terms showed a significant percentage of genes assigned to plasma membrane, intracellular organelles and integral to membrane, whereas molecular function assignments were predominantly associated with ATP binding (25%), nucleic binding (8.3%) and hydrolase activity (8.3%).
Quantitative real-time PCR (qRT-PCR) validation
To validate the differentially expressed gene (DEGs) determined by the transcriptome results, we selected six interested genes and compared the gene expression profiles of CK and PIP using qRT-PCR. Heat shock 70 − 1 protein (Hsp70-1) and heat shock 68 like protein were up-regulated in PIP treatment samples by 2.22 and 2.58 folds, respectively. Three other genes, peptidase, gamma-glutamyl transpeptidase 1-like, and GroE L (came from Buchnera aphidicola), were down-regulated in PIP treatment samples with lower values for the qRT-PCR method (Fig. 7). Using Cufflinks for gene assembly and comparing these results with the known transcription gene model, we found new transcription regions (coverage > 2, and no overlap with the known genes within 200 bp). The expression profiles of the new transcript from two samples are shown in Fig. 8. They were predicted to have 201 and 228 new transcripts in the CK and PIP samples, respectively. The alternative splicing events were predicted by Astalavista software. The difference in the CK and PIP samples may suggest that PIP has effects on aphid gene transcription models.