Charles Dueber
Capstone
Fish 495
The Effects of β-Cyclodextrin on Immunity in the Pacific Oyster
(Crassostrea gigas)
Global aquaculture is a major source of food and revenue for many countries around the world. However, disease outbreaks have placed major constraints on aquaculture production and proven to be very costly. Immunostimulants such as β-cyclodextrin are an emerging preventative treatment option for the aquaculture industry. The primary objective of this study was to determine if β-cyclodextrin could be used to increase the expression of the defense genes interleukin 17 and tissue inhibitor of metalloproteinase in Pacific oysters. A second objective was to examine whether exposure to β-cyclodextrin or Vibrio tubiashii altered the expression of the epigenetic gene DNA methyltransferase 1. These objectives were tested by comparing gene expression between three treatment groups which consisted of control, β-cyclodextrin exposed, and V. tubiashii exposed oysters. It was found that exposure to β-cyclodextrin resulted in significant upregulation of interleukin 17 but not tissue inhibitor of metalloproteinase. The expression of DNA methyltransferase 1 was not significantly different between the three treatment groups. These findings indicate that β-cyclodextrin could be a suitable immunostimulant for invertebrates.
Aquaculture is the fastest growing food producing sector in the world and its growth is expected to continue to meet the demands of a rapidly increasing human population (Bostock et al. 2010). In 2008, global aquaculture produced 52.5 million tonnes of product worth $98.5 billion (Bostock et al. 2010). However, if aquaculture is to continue to grow quickly and successfully, the problem of disease outbreaks needs to addressed. Disease poses one of the greatest challenges to the aquaculture industry because outbreaks are very costly and place major constraints on production. The intensive culture practices of many facilities contribute to disease outbreak and transmission through high animal densities and high nutrient loads. Disease outbreaks in global aquaculture results in an estimated loss of at least $3 billion dollars per year (Subasinghe et al. 2001). Unfortunately, the problem of disease outbreak is compounded by the fact that there are very few treatment options available due to strict regulations in many countries.
The treatment of disease outbreaks in countries such as Canada, U.S., Europe, and Japan is limited to a handful of drugs whose use is strictly regulated (Heurer et al. 2009). The use of these drug based treatment options are further limited as they can only be used for certain diseases in specific culture species. In many of these countries, there is also a maximum tissue residue level for aquaculture products which requires that antibiotic levels in tissues decline below a certain percentage prior to harvesting (FAO 2002). This protects consumers but can also harm culturists as emergency harvesting cannot be conducted if antibiotic treatment fails and tissue levels are too high. However, a large majority of global aquaculture takes place in countries with few or no regulations on the use of antibiotic (Heurer et al. 2009). This has lead to a lack of prevention and treatment options in strictly regulated countries while the overuse of antibiotics has become prevalent in unregulated countries. The overuse of antibiotics in a many areas around the world is especially concerning due to the fact that it has already led to antibiotic resistant pathogens (Cabello 2006). Antibiotic resistant pathogens pose a risk not only to aquaculture but to human and environmental health as well (Cabello 2006). The overuse of antibiotics and lack of treatment options could possibly be mitigated through the use of immunostimulants.
Immunostimulants are one of the most promising solutions to dealing with disease in aquaculture. An immunostimulant is a chemical, drug, or stressor that boosts defense mechanisms associated with the innate immune system (Sakai 1999). Immunostimulants work by enhancing the humoral and cellular response of the innate immune system which prime the immune response by upregulating host defense mechanisms prior to pathogen exposure (Maqsood et al. 2011). However, stimulation of the innate immune system has varied depending on the immunostimulant and the target species. This has led to the cultivation of multiple immunostimulants or even formulations comprised of several different types. Immunostimulants that are currently being used in aquaculture include glucans, peptidoglycan, chitin, chitosan, yeast, and vitamins (Maqsood et al. 2011). An emerging immunostimulant in the field of aquaculture is β-cyclodextrin.
β-cyclodextrin is a novel immunostimulant that has been commonly used in the food and pharmaceutical industry (Del Valle 2004). Recent research has shown that β-cyclodextrin can be used to stimulate increased expression of defense genes associated with the innate immune system of zebrafish larvae (Goetz et al. 2012). However, the mechanism through which it stimulates the innate immune system is not well understood. The purpose of this study was to determine whether β-cyclodextrin could be used to enhance the immune response of the pacific oyster (Crassostrea gigas). It was hypothesized that exposure to β-cyclodextrin would result in the upregulation of the defense genes interleukin 17 (IL-17) and tissue inhibitor of metalloproteinase (TIMP) comparable to the response elicited by exposure to Vibrio tubiashii in Pacific oysters. It was also hypothesized that exposure to β-cyclodextrin would result in a change in the expression of DNA Methyltransferase 1 (DNMT1) between treated and control oysters.
An agar plate was used to culture Vibrio tubiashii and consisted of 20 ml of 5X lysogeny broth (LB), 1 g of 1% NaCl, and 1.5 g of agar. The plate was inoculated with V. tubiashii and incubated at 37 °C for 24 hours. Three individual bacteria colonies were then transferred from the culture plate to 50 ml conicals containing 10 ml of LB. The three conicals each contained one bacterial colony and were incubated with a fourth control conical containing only 10 ml of LB at 37 °C for 24 hours. After incubation, the three conicals were evenly divided between two flasks containing 500 ml of LB each. The flasks were then incubated at 37 °C for 24 hours. After 24 hours, the flasks were centrifuged at 4,300 RPM for 30 minutes. The supernatant was discarded and the remaining bacteria pellets were dried, weighed, and stored at -20 °C. The bacteria pellets had a net weight of 2.026 g and were resuspended in 25 mL of seawater prior to oyster exposure.
Pacific oysters (n=35) were allowed to acclimate to experimental seawater (12°C) for 72 hours and then randomly distributed into one of three treatments: control, cyclodextrin (300 μg/ml), and V. tubiashii (505 μg/ml). For each treatment, there were two test chambers that consisted of clear, square plastic Sterilite (Townsend, MA, USA) containers (34.6 cm L x 21.0 cm W x 12.4 cm H) that were filled with 4 L of water and contained 6 oysters each. There was one exception to this, control group A which only had five oysters due to the fact that one died during acclimation. Exposures lasted a total of 24 hours after which oysters were measured, weighed, and tissue samples were taken. Sampling was done by removing gill and mantle tissue which was then placed on dry ice and stored at -80 °C until further use.
RNA (~60 mg) was isolated from all gill tissue samples using TRI Reagent and following the manufacturer’s protocol from Molecular Research Center, Inc. (Cinncinnati, OH, USA). RNA samples were then DNase-treated using the Turbo DNA-free kit to remove possible genomic DNA carry over from RNA isolation following the manufacturer’s “rigorous” protocol from Ambion Inc. (Foster City, CA, USA). The removal of genomic DNA from DNased samples was verified using real-time PCR. RNA (0.88 μg) was then reversed transcribed using M-MLV reverse transcriptase and oligo dT primers following the manufacturer’s protocol from Promega Corporation (Madison, WI, USA).
Quantitative PCR was conducted using a DNA Engine Opticon 2 Real-time PCR thermocycler from Bio-Rad Laboratories, Inc. (Hercules, CA, USA). All samples were ran in triplicate for each gene in 20 μL reactions that consisted of 1.0 μL of cDNA, 0.5 μL of forward primer (10 μM), 0.5 μL of reverse primer (10 μM), 10 μL of 2X Sso Fast EvaGreen Supermix (Bio-Rad), and 8.0 μL of nuclease free water. Negative controls contained 1.0 μL of water instead of cDNA and were run for each primer set. The genes of interest that were tested were IL-17, TIMP, and DNMT1 (Table 1). The cycling parameters were as follows: 95°C for 30 seconds, 40 cycles of 95°C for 5 seconds, and 60°C for 5 seconds. After quantification, a melt curve analysis was implemented by increasing the temperature from 65 to 95°C at a rate of 0.2°C/s with fluorescence readings every 0.5°C. The average gene efficiency and cycle threshold (CT) was calculated from qPCR amplification data using real-time PCR Miner Software for IL-17 and TIMP (Zhao and Fernald 2005). The average gene efficiency and CT was calculated from qPCR amplification data manually for DNMT1 using the equation shown below due to exponential fit failure with PCR Miner. The relative concentrations of mRNA (R0) were calculated using the equation R0= 1/(1+ average gene efficiency)^CT. The R0 for each gene was normalized using elongation factor 1 alpha (EF1-α) (Table 1). The normalized R0 for each gene of interest was then used to calculate the fold increase over the minimum R0 for all individuals. All data were analyzed using two-tailed t-tests for pairwise comparisons between the three experimental groups. Data were considered to be statistically different if the p-value was less than or equal to 0.05.
Table 1. Primer sequences used for expression analysis of defense and epigenetic genes in gill tissue of Pacific oysters.
Gene Name | Forward (F) and Reverse (R) Primers (5’3’) | Database and Accession Number | Paper Reference |
EF1-α | F: AACAAGGAGCAACGATGGGT R: GGTGATACCACGTTCACGCT | Genbank: AB122066 | N/A |
IL-17 | F: ACTGAGGCTCGATGCAAGTG R: AGCCTTCTTGCTTCATGTGG | Genbank: EF190193 | Roberts et al. 2008 |
TIMP | F: AACATCCGGTTTTGTTTCCA R: CTTCTGGTACACAGAGGTCAAT | Genbank: EW777598 | Roberts et al. 2009 |
DNMT1 | F: ATTGAGGTCCGTCTCTCCGA R: CTCACACGCCTTACCCTCAG | GigasDatabase: CU994437 | N/A |
The 35 Pacific oysters used in this study had an average (± SD) weight of 93.8 ± 15.2 g, length of 87.2 ± 7.3 cm, and width of 58.4 ± 6.0 cm. Statistical pairwise comparisons between the three treatments groups showed no significant difference between the three groups in terms of weight, length, or width.
For expression of IL-17 in gill tissue, the control group showed an average (± 2*SE) expression of 8.1 ± 4.3 (Fig. 1). The cyclodextrin and V. tubiashii groups had expression values of 46.4 ± 22.7 and 70.8 ± 37.4, respectively (Fig. 1). A statistically significant difference (p< 0.05) was observed between the control and cyclodextrin groups as well as the control and V. tubiashii groups. Both cylodextrin and V. tubiashii were significantly higher in terms of IL-17 expression than the control. However, there was not a significant difference in expression of IL-17 between the cyclodextrin and V. tubiashii exposed oysters.
Figure 1. The relative mRNA expression levels of IL-17 in gill tissue of Pacific oysters 24 hours after exposure to cyclodextrin and V. tubiashii. Error bars show standard error multiplied by two.
The average (± 2*SE) expression of TIMP for the control group was 8.0 ± 4.3 while the average expression for cyclodextrin exposed oysters was 14.9 ± 6.7 (Fig. 2). The V. tubiashii exposed oysters showed an average expression level of 25.1 ± 14.5 (Fig. 2). The average expression level of the V. tubiashii group was significantly higher than that of the control group. It was found that there was no significant difference in terms of TIMP expression between the cyclodextrin and control group as well as the cyclodextrin and V. tubiashii group.
Figure 2. The relative mRNA expression levels of TIMP in gill tissue of Pacific oysters 24 hours after exposure to cyclodextrin and V. tubiashii. Error bars show standard error multiplied by two.
For expression of DNMT1, the control group showed an average (± 2*SE) expression of 105 ± 65.6 (Fig. 3). The cyclodextrin and V. tubiashii groups had expression values of 64.7 ± 51.4 and 104.0 ± 72.9, respectively (Fig. 3). There was not a significant difference found in terms of DNMT1 expression in any pairwise comparisons between the three experimental groups. The expression of DNMT1 had the greatest error and variability associated with it out of the three genes that were tested.
Figure 3. The relative mRNA expression levels of DNMT1 in gill tissue of Pacific oysters 24 hours after exposure to cylodextrin and V. tubiashii. Error bars show standard error multiplied by two.
To our knowledge, this is the first study to demonstrate the upregulation of a defense gene in an invertebrate upon exposure to β-cyclodextrin. A significant increase in the expression IL-17 was observed in Pacific oysters exposed to β-cyclodextrin that was comparable to the response elicited by exposure to V. tubiashii (Fig. 1). Upregulation of TIMP was not significantly different from the control group which implies that β-cyclodextrin may not stimulate expression of this particular defense gene (Fig. 2). Unpublished research has indicated that β-cyclodextrin can be used to stimulate the innate immune system of fish resulting in the upregulation of IL-1B, IL-6, and tissue necrosis factor (F.Goetz, Unpublished Data, 2012). Expression of defense genes was sufficiently elevated by β-cyclodextrin that no mortality was observed when zebrafish larvae were exposed to a lethal concentration of bacterial LPS (F.Goetz, Unpublished Data, 2012). However, the vertebrate immune system differs substantially from that of invertebrates. The immune response of vertebrates is regulated by the innate and adaptive immune systems while the invertebrate immune response is comprised solely of the innate immune system. Recent research suggests that the two systems of the vertebrate immune response are intimately linked and work as more of a combinational system than two separate systems (Magnado´ttir et al. 2006). In mammals, IL-17 production is activated by cells associated with the adaptive immune system which illustrates one example of the complex, cooperative relationship between the two immune systems (Yao et al. 1995). Due to these differences, it was previously unknown whether β-cyclodextrin could be used to stimulate an invertebrate's immune system.
The use of β-cyclodextrin to increase the expression of IL-17 in oysters is significant due to the vital role that it plays in the innate immune system. IL-17 is a defense gene that was recently described and belongs to the inflammatory cytokine family. The exact functions of IL-17 has yet to be entirely defined but it is known to play a major role in the immune system (Aggarwal and Gurney et al. 2002). One of the major functions of IL-17 in the immune system is to amplify the immune response by initiating the production of chemokines, cytokines, and cell surface markers (Pappu et al. 2010). In mammals, IL-17 is capable of stimulating humoral factors of the innate immune system such as prostaglandins, matrix metalloproteinases, and chemokines (Gaffen et al. 2006). At this time there has not yet been any research conducted on the role that IL-17 may play in regulating the expression of other genes in the invertebrate immune system. However, given the role of IL-17 in the mammalian immune system, it is probable that IL-17 is regulating the expression of other genes in the invertebrate immune system upon exposure to a pathogen. Assuming that the role of IL-17 is just as substantial in amplifying the expression of other defense genes in the invertebrate immune system, upregulation due to cyclodextrin exposure could prime the immune system for a substantial response when presented with a pathogen.
TIMPs are multifunctional proteins that are excreted into the extracellular matrix where they play an important role in regulation (Gomez et al. 1997). In oysters, it has been found that TIMP is only expressed in hemoycytes which are phagocytic cells that play an important role in internal defense (Montagnani et al. 2001). It has also been demonstrated that TIMP expression is strongly increased when oysters are subjected to shell damage or bacterial challenge (Montagnani et al. 2001). TIMP expression for cyclodextrin exposed oysters in this study was not significantly different from the control group but it was also not significantly different from the V. tubiashii exposed group. The lack of significant difference between cyclodextrin and V. tubiashii exposed groups suggests that it is possible that upregulation of TIMP may have occurred and had declined as previous studies have shown that increased expression 12 hours after exposure to bacteria (Montagnani et al. 2001). The lack of significant expression of TIMP could also be due to the fact that non-living V. tubiashii bacteria were used in this study while it has previously been demonstrated that living bacteria are twice as effective at stimulating TIMP expression (Montagnani et al. 2007). Due to a lack of significant difference in TIMP expression between the control and β-cyclodextrin groups, it does not appear that cyclodextrin elicits a strong response for this particular gene. However, upregulation of IL-17 may be representative of a much larger immune response so commercial applications of β-cyclodextrin should be considered.
Immunostimulants have been used primarily for the culture of fish but the numerous diseases plaguing the culture of invertebrates suggests a need for more treatment options. The increased expression of IL-17 demonstrated in this study suggest that β-cyclodextrin could be used to as an immunostimulant for the culture of invertebrates (Fig. 1). Aquaculture facilities culturing adult crustaceans could effectively use cyclodextrin as an immunostimulant given certain culture methods such as ponds but molluscs are often outplanted in large bays or estuaries where immunostimulants could not be effectively utilized. However, cyclodextrin could be effectively employed as an immunostimulant for larviculture of molluscs and crustaceans in hatcheries. Hatcheries often have to deal with diseases that cause larval mortality given the high density, temperature, and nutrient load associated with rearing conditions (Prado et al. 2005). Bacterial pathogens are most common in hatcheries especially Vibrio species to which larval and juvenile molluscs and crustaceans are the most susceptible (Bachère 2003). The need for a prophylactic immunostimulant was recently demonstrated in 2006 and 2007 as there were several outbreaks of V. tubiashii were reported in shellfish hatcheries along the West Coast of the U.S. (Elston et al. 2008). The outbreaks of V. tubiashii in these hatcheries were associated with an increase in water temperature associated with an El Nino event (Elston et al. 2008). Vibrio outbreaks at hatcheries are not a spatially isolated occurrence either as outbreaks have also taken place at shellfish hatcheries along the Galicia coast in Spain and were attributed to three previously unknown Vibrio species (Prado et al. 2005). As temperature slowly increases due to global climate change, the frequency of Vibrio outbreaks in hatcheries around the world would also be expected to increase especially as the effects from climate change become more pronounced. This scenario appears likely as previous research suggests that increases in marine temperatures are currently resulting in an increase in marine disease outbreaks (Harvell et al. 2009). Unfortunately, the invertebrate culturing portion of the aquaculture industry is poorly equipped to handle such a scenario as there are few reasonable, preventative measures that could be employed. In such a scenario, β-cyclodextrin could be used prophylactically to prevent disease outbreaks by stimulating the immune response during periods of elevated temperature.
However, further research needs to be done to determine the extent of protection that β-cyclodextrin stimulation provides. Future research should focus on conducting a bacterial challenge using live V. tubiashii and larval oysters. It has been demonstrated that beta- β-cyclodextrin can be used to stimulate the immune system of a Pacific oyster, but a bacterial challenge would determine whether stimulation provides sufficient protection against mortality. If it is found that β-cyclodextrin provides protection against V. tubiashii then further research should also focus on additional bacterial challenges using a variety of pathogens to determine whether β-cyclodextrin can be used for the prevention of other diseases. Lastly, significant increases in expression due to β-cyclodextrin exposure were observed only for IL-17 so it would also be beneficial to evaluate increases in expression for other defense genes such as tissue necrosis factor, defensin, etc. This would aid increase our understanding of the immune response elicited by cyclodextrin and possibly suggest what pathogens β-cyclodextrin is best suited to prevent.
The effects of exposure to β-cyclodextrin and V. tubiashii on oyster epigenetics was also explored by looking alterations in DNMT1 expression. Epigenetic mechanisms are capable of inducing changes in gene expression without changing the underlying genetic code (Roberts and Gavery 2012). Epigenetic mechanisms that alter gene expression include histone modification and DNA methylation. In humans, DNMT1 is responsible for cytosine methylation and also plays a role in down regulating gene expression (Fuks et al. 200). In insects, DNMT1 is involved in the post replication maintenance of methylated DNA (Lyko and Maleszka 2011). In relation to disease, it has been observed that viruses and psychiatric diseases in humans can cause the upregulation of DNMTs (Paschos and Allday 2010; Ptak and Petronis 2008). Alterations in the expression of DNMT is significant because it indicates changes in the expression of other genes. Changes in DNA methylation have also been implicated in phenotypic plasticity which may allow invertebrates to respond to environmental changes that occur quickly (Roberts and Gavery 2012). There was no observed response in DNMT1 expression after short term exposure to β-cyclodextrin or V. tubiashii. The lack of a significant difference in DNMT1 expression between the three treatment groups is most likely because the exposure time was too short. It would be expected that expression of DNMT1 in either treatment group would change with long term exposure.
Aggarwal, S. and A. L. Gurney. 2002. IL-17: prototype member of an emerging cytokine family. Journal of Leukocyte Biology 71:1-8.
Bachère, E. 2003. Anti-infectious immune effectors in marine invertebrates: potential tools for disease control in larviculture. Aquaculture 227:427-438.
Bostock, J., B. McAndrew, R. Richards, K. Jauncey, T. Telfer, K. Lorenzen, D. Little, L. Ross, N. Handisyde, and I. Gatward. 2010. Aquaculture: global status and trends. Philosophical Transactions of the Royal Society B: Biological Sciences 365:2897-2912.Cabello, F. C. 2006. Heavy use of prophylactic antibiotics in aquaculture: a growing problem for human and animal health and for the environment. Environmental Microbiology 8:1137-1144.
Defoirdt, T., P. Sorgeloos, and P. Bossier. 2011. Alternatives to antibiotics for the control of bacterial disease in aquaculture. Current Opinion in Microbiology 14:251-258.
Elston, R. A., H. Hasegawa, K. L. Humphrey, I. K. Polyak, and C. C. Hase. 2008. Re-emergence of Vibrio tubiashii in bivalve shellfish aquaculture: severity, environmental drivers, geographic extent and management. Diseases of Aquatic Organisms 82:119.
FAO. 2002. The state of world fisheries and aquaculture: antibiotic residues in aquaculture products. Food and Agriculture Organization of the United Nations. 2002:74-83.
Fuks, F., W. A. Burgers, A. Brehm, L. Hughes-Davies, and T. Kouzarides. 2000. DNA methyltransferase Dnmt1 associates with histone deacetylase activity. Nature Genetics 24:88-91.
Gomez, D., D. Alonso, H. Yoshiji, and U. Thorgeirsson. 1997. Tissue inhibitors of metalloproteinases: structure, regulation and biological functions. European Journal of Cell Biology 74:111-122.
Harvell, D., S. Altizer, I. M. Cattadori, L. Harrington, and E. Weil. 2009. Climate change and wildlife diseases: when does the host matter the most? Ecology 90:912-920.
Heuer, O. E., H. Kruse, K. Grave, P. Collignon, I. Karunasagar, and F. J. Angulo. 2009. Human Health Consequences of Use of Antimicrobial Agents in Aquaculture. Clinical Infectious Diseases 49:1248-1253.
Lyko, F. and R. Maleszka. 2011. Insects as innovative models for functional studies of DNA methylation. Trends in Genetics 27:127-131.
Magnadottir, B. 2006. Innate immunity of fish. Fish and Shellfish Immunology 20: 137-151.
Montagnani, C., F. Le Roux, F. Berthe, and J.-M. Escoubas. 2001. Cg-TIMP, an inducible tissue inhibitor of metalloproteinase from the Pacific oyster Crassostrea gigas with a potential role in wound healing and defense mechanisms. FEBS Letters 500:64-70.
Montagnani, C., J. Avarre, J. De Lorgeril, M. Quiquand, V. Boulo, and J.-M. Escoubas. 2007. First evidence of the activation of Cg-timp, an immune response component of Pacific oysters, through a damage-associated molecular pattern pathway. Developmental and Comparative Immunology 31:1-11.
Pappu, R., V. Ramirez-Carrozzi, N. Ota, W. Ouyang, and Y. Hu. 2010. The IL-17 Family Cytokines in Immunity and Disease. Journal of Clinical Immunology 30:185-195.
Paschos, K. and M. J. Allday. 2010. Epigenetic reprogramming of host genes in viral and microbial pathogenesis. Trends in Microbiology 18:439-447.
Prado, S., J. L. Romalde, J. Montes, and J. L. Barja. 2005. Pathogenic bacteria isolated from disease outbreaks in shellfish hatcheries. First description of Vibrio neptunius as an oyster pathogen. Diseases of Aquatic Organisms 67:209-215.
Ptak, C. and A. Petronis. 2008. Epigenetics and complex disease: from etiology to new therapeutics. Annual Review of Pharmacology and Toxicology 48:257-276.
Roberts, S., G. Goetz, S. White, and F. Goetz. 2009. Analysis of genes isolated from plated hemocytes of the Pacific oyster, Crassostreas gigas. Marine Biotechnology 11:24-44.
Roberts, S. and M. Gavery. 2012. Is there a relationship between DNA methylation and phenotypic plasticity in invertebrates? Frontiers in physiology 2:116.
Roberts, S., Y. Gueguen, J. De Lorgeril, and F. Goetz. 2008. Rapid accumulation of an interleukin 17 homolog transcript in (Crassostrea gigas) hemocytes following bacterial exposure. Developmental and Comparative Immunology 32:1099-1104.
Sakai, M. 1999. Current research status of fish immunostimulants. Aquaculture 172:63-92.
Subasinghe, R.P., M.G. Bondad-Reantaso, and S.E. McGladdery. 2001. Aquaculture development, health, and wealth. Technical Proceedings of the Conference on Aquaculture in the Third Millennium Bangkok: NACA and FAO. 2001:167-191.
Zhao, S., and Fernald, R. D. (2005). Comprehensive algorithm for quantitative real-time polymerase chain reaction. Journal of Computational Biology. 12: 1045–1062.