The level of expression and silencing of the Imnp, Spi and Lpc genes were investigated by quantitative real-time RT-PCR (qRT-PCR) in tick gut and ovary samples collected at day 5 of feeding (Figure 1). As a first step for standardization of the qRT-PCR, actin, tubulin, glucose 6-phosphate dehydrogenase (G6PDH), and phospholipid-hydroperoxide glutathione peroxidase (PHGPx) were evaluated as tick reference gene candidates for data normalization. The geNorm 14 analyses showed that none of reference gene candidates were stably expressed and therefore these candidates were considered inadequate for qRT-PCR normalization (data not shown). Consequently, the qRT-PCR was normalized to the total amount of RNA used to generate the cDNA and the transcription level was calculated as a relative expression using the formula: Relative expression_(sample) = 2^{[Ct(control)-Ct(sample)]}, where the control is the lowest Ct value for a given gene of interest (GOI). Efficiency of amplification of the qRT-PCR for the GOI ranged from 102 to 109% and melt curve analyses showed the absence of primer dimers and nonspecific amplification.

To assess gene silencing, female ticks were injected with dsRNA identical to the Imnp, Spi or Lpc genes and fed on a calf acutely infected with B. bovis, and at day 5 of feeding, 6 biological replicates of individual ovaries and guts were examined by qRT-PCR (Figure 1 and 2). The Imnp gene was silenced 92.9% (± 3.1%) and 93.6% (± 2.8%) in ovaries and guts of dsRNA-injected ticks, respectively, and its relative expression decreased more than 14 times (P = 0.0150) in ovaries and more than 32 times (P = 0.0406) in guts. The Spi gene was silenced 93.4% (± 7.0%) and 85.3% (± 26.4%) in ovaries and guts of dsRNA-injected ticks, respectively, and the relative expression of this gene decreased more than 15 times (P = 0.0234) in ovaries and more than 6 times (P = 0.0213) in guts. The Lpc gene was silenced 81.5% (± 9.6%) and 86.8% (± 12.5%) in ovaries and guts of dsRNA-injected ticks, respectively, and its relative expression decreased more than 8 times in ovaries (P = 0.0021) and guts (P = 0.0284).

To investigate the biological effect of gene silencing, dsRNA-injected female ticks were fed on a calf during acute B. bovis infection and tick fitness was evaluated (Table 1). Fifty two out of 180 female ticks (28.8%) fed to repletion in the Spi silenced group whereas 83 out of 180 females (46%) fed to repletion in the control group demonstrating that the silencing of the Spi gene significantly decreased (P = 0.0011) the number of engorged females. In contrast, the silencing of the imnp and Lpc genes had no effect on the number of engorged females. The average weight of engorged females in the Lpc silenced group was 364.3 mg and significantly higher (P = 0.0167) than the average weight of engorged females in the control group (342.4 mg). However, the silencing of imnp and Spi did not affect the weight of engorged females. Overall, the rate of oviposition was not affected by gene silencing. The average weight of egg masses in the Spi silenced group was 117.8 mg and significantly lower (P = 0.0329) than the average of weight of egg masses in the control group (141.1 mg). The silencing of Imnp and Lpc had no effect on the weight of egg masses. Fifty seven out of 82 egg masses (69.5%) hatched in the Imnp silenced group whereas 62 out of 93 egg masses (66.6%) hatched in the Lpc silenced group. In comparison, 74 out of 82 egg masses (90.2%) hatched in the control group showing that the silencing of Imnp or Lpc decreased significantly the percentage of hatching (P = 0.0016 and P = 0.0002, respectively). The silencing of the Spi gene did not affect the percentage of hatching. The silencing of Imnp decreased the percentage of larvae survival significantly (P < 0.0001) and 11 out of 57 larval progeny (19.3%) died 45 days after hatching whereas 100% (n = 74) of the larval progeny of the control group survived in the same period. In contrast, the silencing of Spi or Lpc showed no effect on the percentage larval survival. Additionally, the cumulative percentage of engorgement during the feeding period was significantly lower in the Spi silenced females than the control females at days 7 and 8 of tick feeding (P < 0.0001 and P = 0.0004, respectively) (Figure 3). The cumulative percentage of engorgement in the Imnp and Lpc silenced groups was not significantly different from the control group.

The larval stage of R. microplus is the only tick stage implicated in the natural transmission of B. bovis 1. Thus, in this study the Imnp, Spi or Lpc genes were silenced in R. microplus females fed on a B. bovis-infected calf and the larval progeny examined for the presence of the protozoan (Figure 4). The infection rate of B. bovis was significantly higher (P = 0.0198) in larval progeny from the Imnp silenced females than in larval progeny from the control group. As a result, 90% of the samples of larval progeny from the Imnp silenced females were infected with B. bovis whereas 30% of the samples of larval progeny from the control group were infected with the protozoan. Although not significant, there was a tendency of higher infection ratio of B. bovis in larval progeny from the Spi (P = 0.0698) and Lpc (P = 0.1789) silenced females than in the larval progeny from the control females, and consequently 80% and 70% of larval progeny from the Spi and Lpc silenced groups, respectively were infected with the protozoan.

The effect of B. bovis infection on the pattern of expression of the Imnp, Spi and Lpc genes was investigated in ovaries and gut of R. microplus females (Figure 5). Imnp was significantly up-regulated (P = 0.0142) 5 times in ovaries of ticks fed on the infected calf compared to ticks fed on the uninfected calf. Although not significant, there was a tendency of up-regulation of Imnp in guts of females fed on the B. bovis-infected calf. B. bovis infection did not significantly affect the level of expression of the Spi gene, despite the up-regulation in ovaries and guts of females fed on the infected calf compared to ticks fed on the uninfected calf. The Lpc gene was significantly up-regulated (P = 0.0049) more than 6 times in guts of female ticks fed on the B. bovis-infected calf compared to females fed on the uninfected calf. Even though Lpc was up-regulated more than 4 times in ovaries of ticks fed on the infected calf, the difference was not statistically significant.

Two Holstein calves 3-4 months of age that tested negative for B. bovis by nested PCR 27 were used in this study. The animals were maintained according to protocols approved by the University of Idaho Institutional Animal Care and Use Committee. One calf remained uninfected throughout the experiment and the other calf was inoculated intravenously with approximately 1.4 × 10⁸ B. bovis-infected erythrocytes (T2Bo strain) 32. Parasitemia in peripheral blood was examined by quantitative real time PCR to amplify the single copy B. bovis msa-1 gene. For the msa-1 quantitative real time PCR, specific primers (5' gatgcgtttgcacatgctaag 3' and 5' cgggtacttcggtgctctca 3') and probe (FAM 5'-cacgctcaagtaggaaattttgttaaacctgga-3' TAMRA) were used following the assay conditions described elsewhere 27. At day 6 after the B. bovis inoculation, the infected calf started showing clinical signs of acute babesioses marked by a decrease of 20% in packed cell volume and temperature above 39°C. The La Minita strain of R. microplus used in this study has been maintained at the University of Idaho Holm Research Center since 1999 as previously described 33. To obtain uninfected and unfed adult ticks, approximately 40,000 larvae (from 2 g of eggs) were placed under a cloth patch on the uninfected calf. After 13-14 days, engorged nymphs were manually removed with forceps and incubated in vitro at 26°C to induce molting to adult stage. Unfed adult ticks were sexed and the females used for gene expression and silencing experiments. The biological effect of gene silencing was investigated in female ticks fed until repletion on the B. bovis-infected calf whereas the pattern of gene expression was examined in females fed until repletion on either the B. bovis-infected calf or the uninfected calf.

Specific primers were designed based on the cDNA sequences of the Imnp (TC12157), Spi (TC9311) and Lpc (TC8123) genes described in the R. microplus Gene Index Project http://compbio.dfci.harvard.edu/tgi/tgipage.html (Table 2). PCR products from each one of the GOI were individually cloned into pCR^® 4-TOPO^® (Invitrogen, Carlsbad, CA, USA), sequenced and analyzed in silico for the presence of putative small interfering RNA (siRNA) by the algorithm siRNA Target Finder (Ambion, Austin, TX, USA). A fragment with approximately 400 bp from each GOI containing the highest number of putative siRNA was amplified by PCR, cloned into pCR^®II-TOPO^® (Invitrogen) and used as template for the dsRNA synthesis (Table 2). The MEGAscript^® Transcription Kit (Ambion) was used for the dsRNA synthesis following the manufacturer's protocol. The dsRNA molecules were checked by electrophoresis on agarose gel, quantified by spectrophotometry and kept at -20°C until used for tick injection.

Freshly molted unfed females were used for the dsRNA injection. Three groups with 200 female ticks each received a single injection with dsRNA identical to the target region of the Imnp, Spi or Lpc genes. Another group also with 200 female ticks was injected with buffer (0.1 mM EDTA) and used as control. For injection, individual females were placed on double-sided tape with the ventral side upwards and injected with either 1 μl of dsRNA (approximately 1 × 10¹¹ molecules dissolved in buffer 0.1 mM EDTA) or 1 μl of EDTA buffer through the coxal membrane at the base of the 4th leg on the right ventral side. Injections were accomplished using a 10 μl syringe with a 33 gauge needle (Hamilton, Bonaduz, Switzerland) and the microprocessor controlled UMP3 injection pump apparatus (World Precision Instruments, Berlin, Germany). After the injection, the female ticks of each group plus an equal number of males were placed under individual stockinet sleeves glued to the side of the B. bovis-infected calf (2 days after the experimental B. bovis infection). At day 5 post-dsRNA injection, 6 partially engorged females from each group were dissected and individual ovaries and guts collected in RNAlater (Ambion) to evaluate gene silencing.

After the dsRNA injection, individual stockinet sleeves containing each experimental group were monitored daily for the presence of engorged females. The effect of gene silencing on engorgement and survival of female ticks was investigated by assessing the cumulative percentage of engorged females during the feeding period. Individual engorged females were weighed and put in 24-well plates at 26°C to assess oviposition. At day 14 after the beginning of oviposition, eggs laid by individual females were weighed, put in individual vials and incubated at 26°C to evaluate hatching. Hatching was evaluated at 30 days after the egg masses were collected and was defined as the presence of any larvae in eggs from each individual female. The larval progeny was maintained in individual vials at 26°C for 45 days and the larvae survival was determined as the presence of any live larvae in larval progeny from individual females.

A qRT-PCR was standardized for each GOI to assess gene expression and silencing. Total RNA was extracted from individual ovary or gut samples using the RNAqueous^® Kit (Ambion) according to the manufacturer's protocol and quantified by Qubit^® Fluorometer (Invitrogen). The RNA samples were treated with DNase I (Invitrogen) following the manufacturer's protocol and 200 ng of total RNA of each sample were used for cDNA synthesis using the Superscript^® Vilo™ cDNA Synthesis Kit (Invitrogen) following the manufacturer's protocol. For the qRT-PCR, specific primers for each GOI were designed to avoid the gene region used to synthesize the dsRNA (Table 2). For the standardization of the qRT-PCR, the R. microplus actin, tubulin, G6PDH and PHGPx genes were evaluated as tick reference gene candidates (Table 2). The geNorm applet 14 was used to examine the stability of expression of the reference gene candidates. The qRT-PCR for the GOI and reference gene candidates were performed in a CFX96™ Real-Time PCR Detection System using the Express SYBR^® GreenER™ Supermix Kit (Invitrogen). The cycling conditions consisted of a Uracil-DNA Glycosylase inactivation step of 50°C for 30 sec, initial denaturation of 95°C for 2 min followed by 40 cycles of 95°C denaturation for 15 sec and annealing/extension of 60°C for 45 sec. Reactions were performed in duplicate in 20 μl using 200 nM of each primer and 2 μl of a 1/20 dilution of cDNA as template. An inter-run calibrator was included to assess inter-run variations. The CFX Manager™ Software (Bio-Rad, Hercules, CA, USA) was used to analyze the qRT-PCR data. Efficiency of amplification and melt curve analyses were performed to evaluate analytical sensitivity and specificity of the qRT-PCR for each GOI.

The presence of B. bovis in larval progeny of the silenced and control R. microplus females was examined by nested PCR with primers to the B. bovis msa1 gene as previously described 27. Briefly, a total of 10 larval progeny samples (containing approximately 100 larvae per sample) of individual females from each experimental group were tested. Genomic DNA (gDNA) was extracted from each larval sample using the Puregene^® Genomic DNA Purification Kit (Gentra, Minneapolis, MN, USA) following the manufacturer's protocol and quantified by spectrophotometry. The nested PCR were performed using the HotStarTaq^® Plus Master Mix Kit (QIAGEN, Valencia, CA, USA) in 20 μl containing 0.2 μM of each primer. For the first run of the nested PCR, 100 ng of larvae gDNA were used as template with 55°C of annealing temperature. For the second run, 2 μl of a 1/100 dilution of the first PCR run was used as template with 60°C of annealing temperature. The presence of amplifiable DNA was examined by PCR in all negative samples using primers specific for the R. microplus actin gene described in Table 2.

The Instat^® software (GraphPad Software, Inc., USA), version 3.06, was used to perform the statistical analyses. The relative gene expression and weights of engorged females and egg masses were compared by t-test. The percentages of engorged females, oviposition, hatching, larvae survival and infection rate of B. bovis were compared by Chi-squared test.