HEAT SHOCK PROTEINS EXPRESSION DURING THERMAL RISK EXPOSURE IN THE
XEROTHERMOPHILIC ANT FORMICA CINEREA MAYR (HYMENOPTERA: FORMICIDAE)
P ŚLIPIŃSKI; JJ POMORSKI; K KOWALEWSKA
1. MATERIALS AND METHODS
Formica cinerea is a thermophilic and xerophilic oligotope of dry open habitats. It
occurs in sunny, sandy sites, bare or overgrown with sparse herbaceous vegetation.
Aggressive ants that live largely by predation though also intensively tend aphids.
1.1. COLONIES LOCALIZATION
All individuals used in both field and laboratory experiments were collected from two
closely located nests in a protected area of the Mazovian Landscape Park near the city of
Otwock (central eastern Poland N:52°10'26" E:18°51'9") from 29.06.2012 to 03.07.2012.
This location excluded the influence of chemical pollution on the Hsps expression level.
1.2. FIELD EXPERIMENT
20 individuals of F. cinerea (control group) were collected from the nest entrance in
the morning at about 8:30 a.m., when the nest temperature was low and the soil temperature
around the nest was 23.5ºC. Another group of 30 workers were collected directly from the
surface of the hot sand in the middle of the day (between 12:30 to 14:40 p.m.), when the soil
surface temperature was high and reached 47ºC to 54ºC. After collecting in the field, all
individuals were immediately killed with ethanol and moved into RNALater® (Ambion Inc.,
Austin, TX) to prevent RNA disintegration, cooled down and taken to the laboratory where
measurement of the Hsp expression was performed. Sand temperature was measured with an
infrared thermometer (DT8750).
1.3. LABORATORY EXPERIMENT
Workers of F. cinerea were collected from the same colonies as the individuals used
in the field experiment. Collection was conducted in the morning (8:14-8:36 a.m.) of three
days between 29.06.2012-03.07.2012, when the nest temperature was relatively low (19.3-
22.8ºC). Ants were transported to the laboratory inside large glass test tubes with a wad
closing the end of the tube (average time of transport was 2:00h, temperature of the transport
was 24.3-28.7ºC). Immediately after transport, the ants were placed in thermal chambers for
two hours at 40ºC and 45ºC (20 individuals per group) and put in RNAlater after thermal
stress. Such a long period of temperature application was necessary to estimate the threshold
of thermal resistance (loss of muscular control) for workers. After the exposition in 40ºC four
individuals out of 20 died (always all individuals were used for further analysis). After 1:15h
at 45ºC, seven ants died and the rest of individuals rapidly started to lose muscular control;
heat shock application for this group was stopped because apparently the threshold of thermal
resistance was reached. During the thermal exposure ants were kept in the same large glass
tubes as during the transport, closed with wad, with no food or water provided.
1.4. GENETIC METHODOLOGY
RNA AND CDNA SYNTHESIS
Total RNA was extracted using GenElute™ Mammalian Total RNA Miniprep Kit
(Sigma Aldrich). We used whole bodies of ants. After RNA isolation, concentrations for each
sample were measured with a NanoDrop 2000c (Thermo Scientific). Immediately after that,
the same amount of total RNA was used for cDNA synthesis using Enhanced Avian First
Strand Synthesis Kit (Sigma Aldrich). Samples of cDNA intended for the RT-PCR and
sequencing were stored at -20ºC.
SELECTION OF GENES AND PCR PRIMERS
There were no sequences of the Hsp genes available for the Formica genus in
GenBank database. Proper design of the specific primers for RT-PCR is essential, therefore
long sequences (of at least a few hundred base pairs) of Hsp and control gene were needed.
Based on all available sequences of Hsp genes of Formicidae species in GenBank, primers
for longer fragments were created. Elongation factor 1 (EF1) was chosen as a reference gene
because of sequence availability and because it was successfully used in previous animal
Real-Time PCR studies. Available data allowed for a design of the primers in Primer3
Software for Hsp60, Hsp75, Hsp90 and EF1 genes.
For the PCR reaction, conducted on Veriti® 96-Well Fast Thermal Cycler (Applied
Biosystems) REDTaq® ReadyMix™ PCR Reaction Mix (Sigma Aldrich) was used. PCR
product was purified and used for sequencing PCR accordingly to manufacturer’s protocols
for the 3500xL Genetic Analyzer (Applied Biosystems). Forward and reverse sequences were
obtained for each of the four genes from at least two specimens. Sequences were aligned and
checked by eye in Bioedit software and deposited in GenBank (KF931630-KF931633).
Primers for Real-Time PCR were designed in Beacon Designer™ Software based on
the longer sequences obtained earlier. Primers were as followed: for Hsp60: 5’-
TGGTTTCGGTGACAATAGA-3’ and 3’-CGTGATGATGACTTCTCCTA-5’; for Hsp75:
5’-TACCTATCAGCACCAAGTC-3’ and 3’- GCATCAATACTAATTCATCATACG-5’;
for Hsp90 5’-CTTGATTCCTGAGTATCTGAAC-3’ and 3’-TTATCCTCCGCCAATTCTT-
5’; for EF1 5’-GGTGTCAAGCAACTGATC-3’ and 3’-TACGACGATACTTCCTTCTT-5’.
Real-Time PCR was conducted on the Corbett Rotor-Gene 6000 Thermocycler
according to the manufacturer’s instructions. Real-Time 2x PCR Master Mix SYBR B (AA
Biotechnology) reagents with SYBR®Green dye was used. Each sample was run in doublet
of PCR reaction for target gene and doublet for reference gene. Reaction conditions for all
genes were: 3 min at 94ºC, 50 cycles of 30 s at 94ºC, 30 s at 58ºC and 30 s at 72ºC, and a
final extension of 5 min at 72ºC.
Melting curve was generated for each run to check for a possible contamination. For
each gene amplification, a serial dilution analysis was also performed to calculate reaction
efficiency. Reactions producing multiple or inconsistent products were not included in
analysis. For few samples even after repeating the first strand cDNA synthesis and the RT-
PCR step results were still inconsistent, which was probably caused by inhibitors still present
in the samples. Therefore, from an initial 20 to 30 individuals per group, only some (the exact
sample size is visible in Fig 1 and Fig 2) were used for further statistical analysis.
Corbett Rotor Gene 6000 software was used to calculate the critical threshold (CT)
for each reaction and tabulate results. Relative changes in expression level between target
genes and reference gene were calculated with the use of delta-delta CT method in the same
software. This method involves normalizing CT values within the sample of interest and the
control sample to an appropriate endogenous housekeeping gen (in this case EF1) and then
comparing these CT values between them.
1.5. STATISTICAL ANALYSIS
The Generalized Linear Model (in SPSS v. 20) with a linear scale response and
identity link function was applied for both parts of the experiment. In each case, data before
the GLM were log transformed.
In the field experiment, we measured the influence of heat stress during foraging in
risk prone conditions on the expression of three genes (Hsp60, Hsp75, Hsp90) and compare
this to a control group. In this case, we used gene expression as a dependent variable and
gene family and group (control vs. hot sand) as two explanatory variables. Both factors were
used as the main effect and the interaction between factors of the model was also applied
In the laboratory experiment, we assessed the influence of temperature (23ºC, 40ºC
and 45ºC) on the gene expression level. Gene expression value was a dependent variable and
temperature and gene family were explanatory variables. Both factors and interaction were
used at the same time as the main effect of the model (Table 2).
Table 1. Field experiment – parameter estimates, their 95% confidence intervals and
their significance of the generalized linear model explaining the influence of gene family
(Hsp60, Hsp75, Hsp90) and group (Control, Hot sand) factors (interaction between factors
marked with an asterisk) on gene expression level.
-1.538 ; -0.534
0.391 ; 1.811
0.787 ; 2.032
-1.169 ; 0.387
Hot sand (47-54ºC)
Table 2. Laboratory experiment – parameter estimates, their 95% confidence intervals
and significance of the generalized linear model, explaining the influence of gene family
(Hsp60, Hsp75, Hsp90) and temperature (23ºC, 40ºC and 45ºC) factors (interaction between
factors marked with an asterisk) on gene expression level.
-1.005 ; 0.411
-0.656 ; 0.612
0.425 ; 1.578
0.402 ; 2.11
-0.202 ; 1.484