Honey bee PTEN--description, developmental knockdown, and tissue-specific expression of splice-variants correlated with alternative social phenotypes

Abstract

Background: Phosphatase and TENsin (PTEN) homolog is a negative regulator that takes part in IIS (insulin/insulin-like signaling) and Egfr (epidermal growth factor receptor) activation in Drosophila melanogaster. IIS and Egfr signaling events are also involved in the developmental process of queen and worker differentiation in honey bees (Apis mellifera). Here, we characterized the bee PTEN gene homologue for the first time and begin to explore its potential function during bee development and adult life.

Results: Honey bee PTEN is alternatively spliced, resulting in three splice variants. Next, we show that the expression of PTEN can be down-regulated by RNA interference (RNAi) in the larval stage, when female caste fate is determined. Relative to controls, we observed that RNAi efficacy is dependent on the amount of PTEN dsRNA that is delivered to larvae. For larvae fed queen or worker diets containing a high amount of PTEN dsRNA, PTEN knockdown was significant at a whole-body level but lethal. A lower dosage did not result in a significant gene down-regulation. Finally, we compared same-aged adult workers with different behavior: nursing vs. foraging. We show that between nurses and foragers, PTEN isoforms were differentially expressed within brain, ovary and fat body tissues. All isoforms were expressed at higher levels in the brain and ovaries of the foragers. In fat body, isoform B was expressed at higher level in the nurse bees.

Conclusion: Our results suggest that PTEN plays a central role during growth and development in queen- and worker-destined honey bees. In adult workers, moreover, tissue-specific patterns of PTEN isoform expression are correlated with differences in complex division of labor between same-aged individuals. Therefore, we propose that knowledge on the roles of IIS and Egfr activity in developmental and behavioral control may increase through studies of how PTEN functions can impact bee social phenotypes.

Figure 1. Schematic diagram of the organization of the honey bee PTEN gene.

(A) cDNA sequences of three honey bee PTEN isoforms (A, B and C) were compared to the genomic sequences gleaned from the Honey Bee Genome Resources database (http://www.ncbi.nlm.nih.gov/genome/seq/BlastGen/BlastGen.cgi?taxid=7460) to define the exons (including alternative exons) and introns. (B) Three alternate splice forms were cloned and their structure is shown.

Figure 2. Alignment of the honey bee PTEN proteins toward other known PTEN amino acid sequences.

The honey bee isoforms were aligned with the sequences of D. melanogaster (AF161258), Nasonia vitripennis (NP_001128398), Tribolium castaneum (XP_974994), Aedes aegypti (ABM21568) and Homo sapiens (NP_000305) using ClustalW version 1.82. The PTEN signature motif (HCXXGXXR) is highlighted (grey), the phosphatase domain underlined with a solid black line, the C2 domain underlined with a dashed line, and PDZ binding motif is double underlined. The Genbank accession numbers for three honey bee PTEN isoforms A, B and C are FJ_969918, FJ_969919 and FJ_969920, respectively.

Figure 3. PTEN RNAi during larval development.

Test of gene knockdown in honey bee larvae fed worker (A) vs. queen (B) diet in each of two separate experiments (n = 24). The larvae were fed with a higher dosage (450 µg/ml) of dsRNA which elicited significant PTEN knockdown (A,B). Bars represent mean ± s.e, different letters (a, b) denote significantly different groups, main effect ANOVA, p<0.05).

Figure 4. PTEN isoform- and tissue-specific expression in adult forager and nurse bees of the same chronological age.

RT-qPCR was used to determine isoform-specific PTEN transcript levels in (A–C) Brain; (D–F) Ovary; and (G–I) Fat body. Age matched (20-day old bees) nurses and foragers from single-cohort colonies were used for the expression analysis. All RT-qPCR samples were run in triplicate. Brain, ovaries and fat body from 3 individual bees were pooled by tissue to make up one biological sample (n = 6). Bars represent mean ± s.e, different letters (a, b) indicate significant differences as determined with a Fisher LSD post-hoc test, p<0.05.

Figure 5. PTEN isoform-specific correlation plots by tissue.

(A) Isoforms A and B, (B) Isoforms A and C (C) Isoforms B and C. Relative levels in brain (red circles), ovary (blue squares) and fat body (green triangles) are shown. Open circles, squares and triangles represent the forager data points while the closed symbols represent the nurse bee data points. As a general pattern, expression of PTEN isoforms is positively correlated in brain and ovary (Pearson's correlation, p<0.05), but not in the fat body of the workers (p>0.05).

Figure 6. Possible tissue-specific PTEN functions in a regulatory network of honey bee behavior.

In brain and ovary, PTEN (red circles) isoforms A, B and C, which could potentially down-regulate insulin/insulin-like signaling (IIS), are more abundant in foragers than in nurses. However, foraging behavior is positively associated with IIS via the release of insulin-like peptide 1 (ilp-1, orange ellipse) in the brain (pink) . Ilp-1 may cause juvenile hormone (JH, green ellipse) levels to increase ; JH is also positively correlated with foraging behavior and may enhance IIS by feedback suppression of vitellogenin (Vg, violet ellipse), a proposed negative regulator of IIS ,,,. These relationships contradict the repression of IIS by elevated PTEN in forager brain tissue. In contrast, suppressed IIS by PTEN in ovary tissue is consistent with the reduced reproductive propensity of foragers . In fat body (yellow), PTEN isoform B, ilp-1 and insulin-like peptide 2 (ilp-2) are elevated in nurses compared to foragers (K. Ihle, unpublished data; and results in this paper), but effects on metabolic biology are currently unclear. The ilp gene products from fat body or brain may also take part in remote signaling to other organs . In this illustration, larger-size circles/ellipses, and thicker arrows (positive)/blocked arrows (negative) denote higher levels of expression, enhancement and suppression, respectively. Dotted arrows indicate the yet unresolved effects on worker phenotypes.