Among environmental conditions, temperature is a major influence on insects, affecting development, sex ratio, and longevity (Harrison et al. 1985; Bleicher & Parra 1990; Davis et al. 2006; Keena 2006). Temperature also influences behavioral aspects of insects, such as mating, spatial orientation, and walking speed (Langer et al. 2004; Colinet & Hance 2009). Morphological and physiological plasticity due to temperature changes often are observed through alterations in development time, fecundity, and size. Generally, at lower temperatures, development often is extended, while at higher temperatures, the adult insect is smaller and has a faster development (Atkinson 1994; Angilletta & Dunham 2003; Angilletta et al. 2004; Pigliucci 2005; Sibly et al. 2007).

Alterations in development time and size frequently are related to changes in metabolic rates due to temperature (Kingsolver & Huey 2008; Angilletta 2009). Effects on oxygen consumption, carbon excretion, and respiration rates also can be altered (Neven 1998). Besides metabolic alterations, temperature changes also can affect the nervous and endocrine systems of insects (Neven 2000). Furthermore, temperature affects the plant host, altering food quality for herbivores, which may modify the insect's response to temperature (Acreman & Dixon 1989).

Among hemipterans, aphids are set apart by their distinct life cycle of alternating asexual and sexual reproduction. When compared to other insect groups, aphids exhibit a much greater phenotypic plasticity due to environmental effects. Temperature plays an important role in aphid life history, influencing development rate, size, fecundity, polymorphism, mating, and migration (Lees 1963; Dixon & Glen 1971; Dixon 1972; Leather & Dixon 1982; Liu 1994; Dixon 2000; Collins & Leather 2001; Müller et al. 2001). Aphids exhibit a high level of phenotypic plasticity (changes in morphology or physiology in response to an environmental condition). Consequently, analysis of growth, development rates, and fecundity of individual aphids have been reliable tools to predict aphid population dynamics under different conditions (Leather & Dixon 1984; Acreman & Dixon 1989; Dixon 1990, 2000).

The sugarcane aphid, Melanaphis sacchari Zehntner (Hemiptera: Aphididae), historically has been an important sugar cane (Saccharum officinarum L.; Poaceae) pest in the US, vectoring sugar cane yellow leaf virus (Schenck & Lehrer 2000; Singh et al. 2004). However, in 2013, outbreaks of M. sacchari on sorghum were reported in Texas, Louisiana, Oklahoma, and Mississippi (Villanueva et al. 2014; Bowling et al. 2016). In 2014, M. sacchari tripled its range, reaching 12 sorghum-producing states with infestation occurring early in the crop season. In 2015, all of the 17 states producing sorghum in the US reported major infestations of M. sacchari (Villanueva et al. 2014; Kerns et al. 2015; Bowling et al. 2016). In Louisiana and Texas, aphid populations can reach over 900 aphids per leaf with yield declines of 60 to 100%, requiring up to 4 insecticide applications to control M. sacchari infestations (Brewer et al. 2017).

By genotyping with 52 microsatellite makers, Harris-Shultz et al. (2017) showed that this outbreak is associated with 1 clonal lineage. Subsequently, Nibouche et al (2018) confirmed that the outbreak is associated mainly with 1 M. sacchari genotype, and that this genotype was not present in the US prior to the outbreaks. During the sexual phase, aphids produce several clones that will thrive in different environments (Douglas & van Emden 2007). A superclone arises when 1 of these clones becomes widespread in different environments in a high frequency and over time (Vorburger et al. 2003; Chen et al. 2013; Harrison & Mondor 2011). Because of this sorghum-associated M. sacchari clone's extensive distribution and high frequency in the US, it can be classified as a superclone, and it is expected to show differential fitness responses across distinct temperatures ranges (Vorburger et al. 2003; Harrison & Mondor 2011; Chen et al. 2013). Studies on the M. sacchari response to temperature changes are inconclusive. Previous work in Asia with M. sacchari indicated that temperatures over 20 °C decreased longevity and fecundity while increasing mortality (van Rensburg 1973a; Chang et al. 1982; Kawada 1995; Abe et al. 2011). However, the M. sacchari intrinsic rate of increase (r_m) is known to increase with rising temperatures (Abe et al. 2011), and M. sacchari populations build up faster in summer than in winter, when an individual female can produce up to 96 nymphs (Chang et al. 1982), resulting in infestations of up to 30,000 aphids per plant (van Rensburg 1973b). In a study after the sorghum outbreaks in the USA, Michaud et al. (2017) observed that M. sacchari on sorghum plants took 5 to 6 d to develop at 23 °C.

Although it is known that temperature affects M. sacchari population growth, specific studies detailing the effect of temperature on a colony collected after the current outbreaks in the US has not been conducted. In addition, before economic thresholds can be developed, information on population growth under simulated field conditions is needed. Therefore, the objective of this study was to study development of M. sacchari population dynamics on sorghum under 6 constant temperatures to estimate upper and lower developmental thresholds and optimum developmental temperature.

Materials and Methods

STATISTICAL ANALYSIS

The biological variables (longevity, development time, nymphs per female, reproductive period, and nymphs per female per d) and the population estimation variables (mean generation time, net reproductive rate, doubling time, and finite rate of increase) were analyzed as randomized block design experiment. PROC MIXED procedures in SAS (SAS Version 9.4, SAS Institute, Cary, North Carolina, USA) were used for all datasets. Analysis of variance tests were used to detect presence of differences among treatments for each variable, and Tukey-Kramer analysis at 0.05% of significance allowed us to compare the least square means and determine whether paired temperatures treatments were different for each variable. Age-specific survival graphs were plotted using SigmaPlot (SigmaPlot Version 14.0, Systat Software Inc., San Jose, California, USA).

Temperature-dependent thresholds under constant temperature regimes were estimated using a Logan model (Logan et al. 1976) as modified by Lactin et al. (1995), as seen below:

Where r(T) is the mean developmental rate at temperature T (°C). Fitted parameters ρ (rate of increase at optimal temperature), T_max (upper developmental threshold), Δ (difference between optimal and upper developmental threshold), and λ (which allows the curve to intercept the x-axis), were estimated using Marquardt's method on PROC NLIN (SAS Version 9.4, SAS Institute, Cary, North Carolina, USA).

The average temperatures per mo from 2002 to 2017 for each continental location in the US were obtained upon special request from the National Oceanic and Atmospheric Administration (NOAA 2017). The average temperature per mo for 15 years was then calculated for each state.

Results

LIFE TABLE ANALYSIS

Temperature affected M. sacchari development time (F = 510.12; P < 0.0001). Melanaphis sacchari reached adulthood faster when it was reared at both 25 and 30 °C (Table 1). Development time decreased by 4.4 d when the temperature increased from 15 °C (12.2 d) to 20 °C (7.8 d) (F = 19.48; df = 5, 149; P > 0.0001), and decreased by 3.1 d when the temperature increased again from 20 °C (7.8 d) to 25 °C (4.8 d) (F = 28.61; P > 0.0001). Differences were not detected in development time when the temperature increased from 25 to 30 °C (4.8 d); however, at 32 °C (5.9 d) the development time increased by approximately 1 d (F = 10.15; P < 0.0001). The aphid was not able to complete development at a constant 35 °C.

Table 1.

Mean (± SD) development time (d), reproductive period (d), nymphs per female and longevity (d) of Melanaphis sacchari collected on sorghum under constant 15, 20, 25, 30, and 32 °C on grain sorghum.

The amount of time in which the female remained reproductively active (reproductive period) was affected by temperature (F = 152.15; P < 0.0001) (Table 1). The longest reproductive period, 18.8 and 15.3 d, occurred at 15 and 20 °C, respectively. Raising the temperature to 25 °C decreased reproductive activity to 8.9 d (F = 6.63; P < 0.0001). Females reduced their reproductive period to 1.5 d at 30 °C (F = 12.29; P < 0.0001); however, differences were not detected in the reproductive period when temperature increased from 30 to 32 °C. Likewise, temperature treatments affected fertility of M. sacchari females (F = 172.15; P < 0.0001) (Table 1). The greatest production of nymphs per female was at 20 °C; females produced an average of 49.8 nymphs, ranging from 4.0 to 111.0 nymphs per female. High fecundity rates also occurred at 15 and 25 °C, with 36.4 and 40.0 nymphs per female, respectively. Increasing the temperature to 30 °C caused a decrease (F = 13.94; P < 0.0001) in nymph production per female. At 30 °C and 32 °C, females had the lowest nymph production of 4.1 and 5.1 nymphs produced per female, respectively, and at these temperatures, many females reached adulthood but did not produce any nymphs.

The average lifespan from d 1 until death (longevity) was affected by temperature, and decreased with increasing temperature (F = 250.02; P < 0.0001) (Table 1). The longest M. sacchari longevity was achieved at 15 °C with insects living for 32.3 d on average, and the shortest longevity was observed at 35 °C with insects living only for an average of 2.8 d.

Age-specific survivorship (lx) decreased linearly with the increase in temperature (Fig. 1). At 15 °C, the greatest age-specific survivorship was observed at 75 d. Observations up to 53 d at 20 °C, and up to 31 d at 25 °C were recorded. Even at the highest constant temperatures of 30 and 32 °C, aphids could survive for approximately 20 d, but at 35 °C, only a few individuals survived until d 10. Survival started to diminish after d 7 at a temperature of 20 °C, at d 4 at 25 °C, and at d 2 at the other temperatures (Fig. 1).

The highest net reproduction rate of 50.4 occurred at 20 °C (Table 2). At 15 and 25 °C, the R₀ values were 31.1 and 34.0, respectively, and the lowest R₀ values of 3.6 and 4.5 were found at 30 and 32 °C, respectively. The intrinsic rate of increase was highest at 25 °C (0.405 ± 0.030), indicating that population increases fastest at this temperature, while aphids kept at the colder temperatures had lower r_m (Table 2). The finite rate of increase (λ_F) was highest at 25 °C at 1.5 nymphs per female per d. The maximum and minimum population doubling times (DT) were 5.6 at 30 °C and 1.7 d at 25 °C (Table 2).

Fig. 1.

Age-specific survivorship (proportion of alive individuals per day) of Melanaphis sacchari under 6 constant temperatures on grain sorghum.

THERMAL REQUIREMENTS

When we fitted the observed data in the Logan Lactin-modified model (Lactin et al. 1995), the lower developmental threshold for M. sacchari development was 8.6 °C, and the maximum developmental threshold was 37.8 °C, with an optimum development rate at 28.3 °C (Table 3, Fig. 2).

Based on M. sacchari lower thermal requirements, shaded areas in Table 4 represent the mo in which M. sacchari would be actively developing for each state. In Florida and Louisiana, M. sacchari populations are constantly developing over the yr, meaning that regardless of the rate, nymphs are developing to adults and reproducing. In Mississippi, Georgia, and Texas, M. sacchari development is ceased only in the mo of Jan when temperatures are below the 8.6 °C development threshold. Contrarily, in Idaho, Maine, Minnesota, Montana, New Hampshire, North Dakota, Wisconsin, and Wyoming, M. sacchari have only 5 mo, from May through Sep, in which the average temperature allows active development.

Table 2.

Intrinsic rate of increase (r_m), net reproductive rate (R₀) (nymphs per female), mean generation time (d), doubling time (d), and finite rate of increase (nymphs per female per d) of Melanaphis sacchari under constant temperatures on grain sorghum.

Discussion

The lower developmental threshold of 8.6 °C for M. sacchari and the upper developmental threshold of 37.8 °C observed in the present study are higher than the threshold for most aphid species. Even though lower developmental thresholds of different aphid species can vary from 3.6 °C (Carter et al. 1982; Zhou et al. 1989) to 11.8 °C (Bayhan et al. 2005), lower developmental thresholds above 7.0 °C are not common in the literature. High lower development thresholds like our results are unusual among aphids that colonize sorghum. Rhopalosiphum padi (L.) (Hemiptera: Aphididae) exhibits above average thermal thresholds (Asin & Pons 2001; Ma & Ma 2007); however its lower development threshold does not reach 7.0 °C (Elliot & Kieckhefer 1989; Auad et al. 2009; Park et al. 2017) with an optimum temperature for development between 25 °C and 28 °C (Dean 1974; Asin & Pons 2001), and upper development threshold < 30 °C (Dean 1974; Elliot and Kieckhefer 1989; Asin & Pons 2001; Auad et al. 2009). Under 7 constant temperatures ranging from 10 to 33 °C, Sipha flava Forbes (Hemiptera: Aphididae) and Schizaphis graminum Rondani (Hemiptera: Aphididae) (aphids which can infest sorghum) showed lower thermal thresholds of 2.1 °C and 5.7 °C, respectively, with an optimum temperature for development between 20 °C and 26 °C (Oliveira et al. 2009; Tofangsazi et al. 2010).

Table 3.

Estimated parameters of Lactin model for constant temperature regimes: ρ = rate of increase at optimal temperature, T_max = upper developmental threshold, Δ = difference between optimal and upper developmental threshold, and λ = value that allows the curve to intercept x-axis.

In the literature, we could not find developmental thresholds for M. sacchari on any of its host plants to make related comparisons. Our developmental threshold data indicate a great tolerance of M. sacchari to elevated temperatures; however, the metabolic and morphological features that allow this species to have optimum development above 25 °C are not clear, but heat shock proteins may be an important component of this condition.

Heat shock proteins act as chaperones and prevent protein denaturation under heat stress (Okada et al. 2014; King & MacRae 2015), and species-specific production of heat shock proteins have been reported (Sharma et al. 2007; Wang et al. 2013; Li et al. 2017). For R. padi, heat shock proteins were induced when the aphid was exposed to temperatures ranging from 36 to 38 °C, indicating that for this species, heat shock proteins are an important component in heat tolerance (Li et al. 2017).

While most aphid species start to suffer in temperatures above 25 °C, M. sacchari had faster development at 25 and 30 °C and higher upper developmental thresholds. Symbionts, such as Serratia symbiotica Sabri et al. (Enterobacteriaceae), are thought to increase aphid heat tolerance (Montllor et al. 2002; Russell & Moran 2006; Heyworth & Ferrari 2015) through nutritional compensation when heat impairs Buchnera aphidicola Munson et al. (Enterobacteriaceae) (Koga et al. 2003; Russell & Moran 2006). Secondary endosymbionts conferring heat tolerance are in aphids more frequently found in tropical areas (Henry et al. 2013). Currently, there are no reports of the secondary endosymbionts harbored by M. sacchari.

Fig. 2.

Melanaphis sacchari constant temperature dependent development rate Lactin model estimation. Solid line representing the predicted values by Logan-Lactin model and dots represent the observed values.

Development, survivorship, growth, and fecundity of individual aphids can predict population trends (Leather & Dixon 1984; Acreman & Dixon 1989; Dixon 2000), and this has been used to predict population growth in diverse environmental conditions (Zuniga et al. 1985; Sumner et al. 1986; Warrington et al. 1987; Fereres et al. 1989). The intrinsic rate of increase observed in M. sacchari on sorghum corroborates previous reports of a greater M. sacchari population increment during hotter periods (van Resburg 1973a, b; Chang et al. 1982; Kawada 1995). Lopes-da-Silva et al. (2014) found an intrinsic rate of increase of M. sacchari on sorghum at 24 °C of 0.30, a much smaller value than the present study found at 25 °C (0.405). We found that r_m doubled when the temperature increased from 15 to 25 °C, and the same rate was observed for M. sacchari on sorghum by Abe et al. (2011). However, Abe et al. (2011) observed that the r_m increased from 0.390 at 25 °C to 0.450 under 30 °C, whereas in the present study the r_m decreased from 0.405 to 0.124 when the temperature increased from 25 to 30 °C. At 24 °C, M. sacchari R₀ was 27.70 (Lopesda-Silva et al. 2014), while in the present study, M. sacchari population growth from 1 generation to the next was nearly 6 times higher at 25 °C. These markedly different biological responses under similar temperature conditions may be part of the explanation for the recent M. sacchari sorghum outbreaks in the US.

The findings of this study reveal key aspects of M. sacchari under different temperatures that have not been investigated before, and may shed some light on the recent outbreaks in the US. Even though this study was not designed to understand the mechanisms of M. sacchari responses to different temperatures, the hypotheses raised here give perspective for future studies. Although laboratory conditions of low densities and constant temperature are not consistent with field conditions, a laboratory study can provide valuable information about life history and population dynamics, because comparisons of r_m are the most reliable way to predict population performance under different conditions. In addition, to estimate economic thresholds, the observations made in the present study are essential.

Therefore, these findings will assist in planning control measures. For instance, using the mo average for the past 15 yr (NOAA 2017) and M. sacchari lower developmental threshold, we suggest that monitoring of remnant sorghum (patches of sorghum that survived winter) in Florida, Georgia, Louisiana, Mississippi, and Texas has to start before the crop season, because M. sacchari population did not stop developing during winter, or stopped only for a brief period on far northern areas of states with an extensive latitude, such as Georgia and Texas. In addition, because populations have been in constant development, colonization pressure of sorghum fields is expected to be higher in these states. In the states where M. sacchari has a narrower window of development, monitoring frequency can be reduced, which in turn decreases production costs from scouting. Thus, the use of M. sacchari thermal thresholds to adjust the monitoring frequency of sorghum fields according to actual climate conditions reduces production costs and prevents unexpected outbreaks.

Table 4.

Fifteen years average temperature (°C) in the US continental states throughout the year. Shaded area corresponds to Melanaphis sacchari active development.