Differential centrifugation is one of the most common initial steps taken in protein purification from cells or tissues according to (Lodish, H., et al., 2016.); to study the function and structure of a protein. The aim of this study is to use differential centrifugation to produce a fraction enriched in nuclei and mitochondria from a rat liver homogenate, obtained using a motor driven pestle. To assess if a good separation has occurred, the recovery of the mitochondria in the fractions will be measured using the activity of succinate dehydrogenase (SDH) as a marker. SDH is an enzyme found only in mitochondria. Thus, if any activity is detected in a cell fraction, this would mean some mitochondria are present in that fraction. Indicating that the centrifugation has not occurred well and the fractions are impure. Overall, the specific SDH activity of the Mitochondria Fraction was high as expected and the Homogenate Fraction Total SDH Activity. However the % recovery of SDH was the highest in the SF this was unexpected and could have been due to inaccuracies in carrying out the protocol. Overall majority of the results conclude what was intended, slight changes to laboratory equipment and a change in some of the methods used would mean less contamination and purer fraction could be obtained avoiding results such as finding SDH activity in the supernatant.
The results in table 1 were obtained by using differential centrifugation. The speed of the centrifuge is an important factor in determining if separation will occur. A low g force is used to separate molecules with a high density in this case nuclei, whereas a high g force is used to separate soluble organelles/proteins such as the mitochondria.
The samples were stored at -20oC to prevent SDH losing its catalytic activity, and the different fractions were suspended into an isotonic buffer to prevent damage to organelles due to osmotic imbalance.
To obtain a series of pure fractions, centrifugation of the homogenate was done twice at 1500 x g for 10 min to sediment the nuclei from the other organelles such as whole cells, endomembrane organelles, and the mitochondria producing supernatant I and II. The same principle was used to obtain the mitochondrial fraction by combining Supernatant I and II, but at 20,000 x g force producing supernatant III / IV, which were combined together to produce supernatant fraction (SF). To obtain the final total volume both NF (10mL) and MF (8mL) were suspended in isotonic buffer.
TABLE 1: Total volume for each cellular fraction
Total Volume (mL)
Exact volume of the homogenate remaining
AFTER taking 3 X 1 mL H aliquots.
Nuclei fraction (NF) before taking the aliquots
Mitochondrial Fraction (MF) before taking the aliquots
Supernatant fraction (SF)
FROM Combined supernatants lll and IV
Table 1. Total volume for each of the fraction. Homogenate volume is lower as 3 x 1mL of aliquots were taken to be later used to measure the protein content of each subcellular fraction and to measure the succinate dehydrogenase activity (SDH) by using red formazan essay. Conversely the SF fraction has a high volume as Supernatant III and IV were combined.
Measuring protein content in each subcellular fraction using a biuret assay
To measure the protein content in each of the subcellular concentrations, a Biuret assay was used. In this the proteins form a blue complex with alkaline Cu2 , this colour has an absorbance peak at 550nm. Although the reaction is specific for peptide bonds as stated by… the reaction is not sensitive and thus can only be used with crude extracts, which contain large amounts of protein. The biuret reaction also requires alkaline condition to achieve this 0.1M of NaOH was used. Known amounts of bovine serum albumin (BSA), were reacted with biuret reagent, the absorbance values were measured to create a standard calibration graph to calculate the protein concentration in each of the subcellular concentrations as can be seen below in table 3.
Table 2. Averaged absorbance values for each BSA protein amount
Values for BSA standard curve
Protein amount (mg)
Absorbance at 550nm
TABLE 3: Protein amount in homogenate and subcellular fractions
Average absorbance (550nm)
0.1017-0.089 = 0.0127
0.173- 0.089= 0.084
Protein amount in sample’s aliquot (mg)
Protein concentration in fraction (mg/mL)
Protein amount in fraction’s Total Volume (mg)
OVERALL PERCANTAGE RECOVERY
Calculations used to obtain the figures in table 3:
To find out the protein amount in sample’s aliquot (mg) the equation y = mx c was used from the graph created from table 2 as can be seen above.
Y = average absorbance of each of the fractions taken at 550nm
Y = (0.117 0.0013) / 0.0064
X = 18.48mg
To work out the protein concentration in fraction, divide the protein amount in sample aliquot by the volume taken to produce the reaction tubes to be used for the Biuret assay.
For the homogenate 0.1mL was used
Protein amount in homogenate aliquot 18.48mg divided by 0.1mL volume taken
= 184.062 mg/mL
To calculate protein amount in fraction’s total volume, multiply it by total volume of fraction found in table 1
= 184.062 * 11.5ml=2116.719 mg
To calculate the percentage of protein recovery in comparison to the initial homogenate (HF)
HF = 100%
% of MF = (299.880 mg / 2116.719 mg) x 100% = 14.1672%
Rounded to 3dp = 14.167%
To obtain reasonable values of SDH activity, the fractions were diluted with phosphate extraction buffer to the approximate protein concentration given in the first column of Table 4. To get the suggested concentration given in table 4 for each of the fractions in a volume of 2000 ?L, the fold dilution was calculated using the equation C1 *V1 = C2*V2 as shown below.
TABLE 4. Dilution of Fractions in PBS
Volume of homogenate or fraction required (?L)
Volume of diluent required (?L) to make up to 2000 µL total
Actual concentration used (mg/mL)*
to approx. 5 mg/mL
Rounded up: 54 ?L
V1 = 54 microliter
to approx. 2 mg/mL
5.468/2 = 2.734
2000/2.734 = 731.528
Rounded up: 732 ?L
2000 – 731.528 = 1268.472
V2= 2ml(2000 ?L)
to approx. 1 mg/mL
33.320/1 = 33.32
2000/33.32 = 60.024
Rounded up: 60 ?L
2000 – 60.024 = 1939.976
Rounded up to:
Value from Table 3, row C
Fold dilution was calculated using the equation C1 *V1 = C2*V2 as shown below.
Example: Nuclear fraction (NF)
The protein concentration in fraction (mg/mL) from table 3 column C is divided by the approximate protein concentration given in the first column of Table 4 mg/mL. In this case
5.468 (mg/mL) / 2 (mg/mL) = 2.734 concentration fold
To work out the suggested concentration in a volume of 2000 ?L:
2.734 / 2000 (?L) = 731.528 ?L
Rounded up to 732 ?L
To work out the volume of diluent required (?L) to make up to 2000 µL total
2000 (?L) – 731.528 (?L) = 1268.472
Rounded up to 1268
Actual concentration used (mg/mL)*
V2= 2ml (2000?L)
C2 = (C1*V1) / V2
= (5.458 (mg/mL) * 732 (?L) / 2000 (?L)
= 2.001 mg/mL
For TABLE 5: Determining the amount of Formazan content at an absorbance of 490nm
In this experiment Succinate dehydrogenase is being used as a marker enzyme, it is found in the citric acid cycle in the inner membrane of mitochondria as stated by … it catalyses the conversion of succinate oxidation into fumarate. This reaction happens simultaneously however as both reagents are colourless, we cannot see the reaction taking place that’s why a coupled reaction is used as shown below.
Succinate FAD ? fumarate FADH2
The overall reaction is known as an oxidation reaction. This is because when SDH breaks down the succinate into fumarate, 2 electrons are donated/lost to the enzyme bound cofactor flavin adenine dinucleotide (FAD 2 electrons ? FADH2) to produce FADH2.
FADH2 INT ? FAD formazan
Since we cannot measure the SDH directly, the formation of formazan is measured instead which forms a deep red compound. This happens when FADH2 reduces tetrazolium salt (INT). This way the Succinate acts as a donor of electrons and INT the final electron acceptor.
TABLE 5: Determining the amount of Formazan content at an absorbance of 490nm
Average test absorbance values (obtained over a 10 min incubation)
Corrected value 0.000
0.227 -0.004 = 0.223
0.255 – 0.004 = 0.251
Corrected value 0.000
Corrected value 0.000
Corrected value 0.000
The control values were subtracted from the test results, and test ½ were added and average was taken.
Example HF = 0.004 subtracted from test 1 (0.227 -0.004 = 0.223) and test 2 (0.255 – 0.004 = 0.251)
(0.223 0.251) / 2 = 0.237 average absorbance value
As little formation of formazan in the control tubes can be seen even though no sodium succinate was added, this is because INT (tetrazolium salt) is a non-specific electron acceptor, this means it can be reduced by different redox enzymes transferring electrons from many compounds in the cell such as ascorbic acid, hydroquinones, and menadione. However, since SDH is highly specific for succinate the control sample (containing no succinate) will be used to subtract any INT reduction which may produce red formazan due to other enzymes than SDH.
TABLE 6. Calculating Total and Specific activity and % recovery of activity in the fractions in relation to the homogenate
Obtained over 10 min incubation period
Total Succinate Dehydrogenase (SDH) Activity (nmoles/min)
Amount of substrate formed per minute
% Recovery of Activity (rel.Homogenate)
Specific SDH Activity
(rounded to 3dp)
4.716 X 10-3
3.2512 X 10-3
4.477 X 10-3
Example of how to calculate the total SDH activity of HF
Average Absorbance value of homogenate: 0.117 (Obtained over 10min incubation)
Apply Beer lambert law A? = ?? L [c]
0.237= 20100 M-1 cm-1 x 1 cm x [c ]
0.237 / 20100 = [c]
[c] =1.1791 X 10-5M = (moles/L)
Convert to µmoles/L by multiplying by 106
(1.1791 X 10-5 moles/L) X (106 µmoles/L) = 11.7910 µmoles/L of formazan in 10min
However we want to find out the total activity of SDH in the whole fraction to calculate that:
Firstly need to calculate per minute:
Note: This would be the concentration of Formazan (11.7910 mM) or the amount of formazan (11.7910 mM) produced and dissolved in 1 litre of ethyl acetate. However, we only dissolved in 4mL
11.7910µmoles/L of formazan in 10min
(11.7910 ?moles/1000mL) X 4mL = 0.04716µmoles in 10 min (in reaction)
0.04716 (?moles) x 1000 nmoles/ ?mole = 47.16 nmoles in 10min (in reaction)
47.16 nmoles in 10min (in reaction)
47.16 nmoles / 10min = 4.7164nmoles per min (in reaction)
Formazan product was derived from the SDH activity of 0.2mL diluted fraction
4.7164nmoles per min in 0.2ml
To find out in 1mL multiply by 5 (mg/mL)
4.7164nmoles x 5 (mg/mL) = 23.58nmoles per min in 1mL of diluted homogenate
From table 3, we can find out the protein concentration in fraction mg/mL in this case it is 184.062mg/mL
184.062 (mg/mL) / 5 (mg/mL) = 36.812 (3.dp) this is the dilution factor used to dilute the sample with extraction buffer to obtain the approximate protein concentration given in first column of table 4.
5mg/mL (Homogenate protein concentration/volume)
23.58 (nmoles/permin/1mL) X 36.812 = 868.10nmoles/per min in 1mL (undiluted homogenate)
868.10 (nmoles/permin/mL) X 11.5 (mL) = 9983 (nmoles/min)
(11.5 total fraction volume obtained from table 1)
Total SDH Activity of homogenate (HF) = 9983 (nmoles/min)
Example of how to calculate % recovery of activity in contrast to homogenate
% of recovery 100%
% of MF = (Total SDH X MF mitochondria fraction) / (Total SDH Homogenate fraction)
2998.8 (nmoles/min) / 9983 (nmoles/min) X 100% = 30.01%
Example of how to calculate specific SHD activity for the homogenate fraction
Specific SDH activity refers to the activity of an enzyme per milligram of total protein as stated by ..
(Total SDH Activity nmoles/per min) / (Protein amount in fraction’s Total Volume mg)
For HF = 9983(nmoles/min) / 2116.719 (mg) = 4.716 nmoles / min / mg (Specific SDH Activity)
Convert to µmoles = To do this divide the amount of substance by 1000
= 4.716 X 10-3 µmoles / min / mg protein
Overall, the results obtained showed a relatively good separation had occurred. The evidence for this can be seen in table 3 column D, in the differing amounts of protein in each fractions total volume, as well as the differing values in specific SDH activity and total succinate dehydrogenase activity from table 6, both show that the organelles have been efficiently isolated from each other. Out of the three fractions obtained from the homogenate, the supernatant had the highest, Total SDH Activity this was unexpected, since SDH is only found in the mitochondria, the MF fraction should have had the highest value, which indicates some fraction contamination, as there is mitochondria activity detected in the supernatant fraction.
The reason for this could have been due to, the inaccuracies caused in the initial experiment (table 1) when decanting the supernatant fraction between each centrifugation, some of the pellet could have been poured in with the supernatant affecting all of the following results.
From looking at table 3, the percentage of protein recovery in relation to the initial homogenate was not 100%, instead there was a 35.266% loss. The reason for this could be as proteins are held by relatively weak forces such as hydrophobic, van der Waal and salt bridges according to … to create a biologically active structure. The bonds can be easily disrupted by temperature, as the tubes were constantly being taken out of ice and returned this could have also had a significant effect on the catalytic activity of SDH, as the room temperature was not cold. During homogenisation, cells like lysosomes can also burst releasing enzymes such as protease which are known to degrade proteins within the source material, to minimize the activity of protease the temperature needs to be maintained at 25oC according to …, therefore there is a high probability of protease being one of the causes for low recovery. However to reduce this effect BSA (bovine serum albumin) was added, which prevents proteolysis of the target protein according to…
To improve the separation of the organelles, an alternative would be rate zonal centrifugation which should provide purer fractions. This method works on the basis of different masses of water soluble proteins, the proteins are separated by centrifugation through a solution of increasing density, usually sucrose is used. The heavier particles sediment the fastest followed by smaller and less dense ones, this creates discrete bands which can be collected as fractions and assays can be performed. This is especially useful in separation of other organelles from the mitochondria. A small amount of detergent is added, and absorbed by the cell through endocytosis, making the lysosomes less dense which gives a clean uncontaminated separation from the mitochondria.
In, the experiment a hand-held homogenizer was used; this perhaps was not the most accurate way of releasing cell content as insufficient force was used so whole bits remained in the solution. To improve this, a rotor based or an ultrasonic homogeniser could have been used which would provide more efficiency in organelle separation.
The results obtained in table 3 column A (average absorbance for the fractions), for the homogenate and nuclear fraction were much higher than the absorbance values, taken for the BSA standard curve. This could have been due to inaccuracies in pipetting, and collecting the samples such as not shaking the tubes properly, or due to an error in centrifugation of the samples in the initial experiment.
The supernatant had a high protein concentration, reference table 3 row D, this is because a high concentration of soluble proteins, were left in the solution such as peroxisomes, lysosomes and other large macromolecules, fragments of Endoplasmic reticulum, and microsomes. Organelles specific marker molecules could be used to extract these. For example cytochrome c protein is present only in mitochondria, so the presence of this protein in a fraction of lysosomes would indicate contamination. Also, catalase is present only in peroxisomes; ribosomes, only in the rough ER and acid phosphatase, only in lysosomes.
In conclusion the aim of this particular subcellular fractionation was achieved, a fraction enriched in nuclei and a fraction enriched in mitochondria were obtained. The specific SDH activity was the highest in the mitochondrial fraction as was predicted and the total SDH activity in the homogenate. However the percentage activity recovered was the highest in the supernatant fraction which indicates contamination of mitochondria in the fraction. Overall there has been some contamination, however majority of the results conclude what was intended, slight changes to laboratory equipment and a change in some of the methods used would mean less contamination and purer fraction could be obtained avoiding results such as finding SDH activity in the supernatant.
? Lodish, H.F., 2016. Molecular cell biology Eighth edition..; Global., New York: W.H. Freeman.
? Hames, B.D., ebrary, Inc
The Cognitive and Decision-making Process of M. Sabuleti Foraging During Food Shortages
ABSTRACT OF RESEARCH PLAN:
Myrmica sabuleti is a species of eusocial ant which exhibits a reproductive division of labor and age-dependent task specialization. The mushroom bodies are a part of the insect brain responsible for olfactory integration and experience-based learning. This portion of the brain varies in size depending on the task specialization of the individual ant. 10 colonies of ants will be given different combinations of nutrient-rich and -poor food to force specific decision-making during foraging. The morphology of the brains of the ants in each colony will be compared to see if there is a link between decision-making and the mushroom bodies.
A. Specific Aims:
This study aims to explore the relationship between decision-making and ant brain morphology during times of environmental stress.
Our hypothesis is that decision-making while foraging will affect the mushroom bodies. Mushroom bodies are a section of the insect brain associated with olfactory integration and experience-based learning (Gronenberg, et al, 1996).
This study will utilize 10 colonies of Myrmica sabuleti to test the hypothesis. A portion of the colonies will be given nutrient-poor food or limited quantities of nutrient-rich food without any other choices. Another portion of the colonies (decision-making colonies) will be given the choice between varying quantities of nutrient-rich food and nutrient-poor food. A small sample of ants from each colony will be taken weekly and dissected to measure the size of their mushroom bodies. We will compare the average size of the mushroom bodies and the change in the size of the mushroom bodies over time between each colony. We will also track the number of ants visiting each food source every day to track their foraging behavior.
Our prediction is that colonies forced to make foraging decisions will have larger mushroom bodies than non-decision-making colonies. By observing these differences, we can begin to make conclusions regarding how environmental stressors affect the neurology of social insects and how brain morphology mediates adaptive behavioral changes.
Eusocial insects are species of insects characterized by a highly organized social structure of interrelated individuals whose fitness depend on the fitness of the collective nest and reproductive capabilities of the queen (Nowak, et al., 2010). For such a large number of individual organisms to function as a single superorganism requires levels of nuanced communication that makes their neurology inherently interesting (Perry, et al. 2013). Recent studies done on eusocial insects, most notably on honeybees, have shown that their social behaviors, communication methods, and cognitive abilities are far more advanced than previously anticipated (Menzel, 2012). By understanding these complex networks of communication and adaptive behavior, we can better understand how insect foraging functions within an ecosystem.
Myrmica sabuleti is a widely studied species of eusocial ant native to Europe. They exhibit age-dependent task determination (Gronenberg, et al, 1996). Communication between individuals relies primarily on olfactory chemical signaling (Cammaerts, M., et al, 2014). Foraging ants rely mainly on olfactory cues to locate food sources (Detrain, C., et al, 1999).
The decision-making capabilities of ants, and how this directly involves their neurology, have been studied but are still relatively unexplored (Detrain, C., et al, 1999) or have not established causal links (Robinson, et al., 2009). Past studies done on ant neurology have lead us to focus on the mushroom bodies of foraging ants (Gronenberg, et al, 1996). Mushroom bodies, or corpora pedunculata, are the parts of the insect brain responsible for olfactory learning and memory (Perry, et al, 2013). The morphology of mushroom bodies are also highly variable depending on the role an individual plays within the colony. Young, nonforaging ants and queen ants tend to have smaller mushroom bodies than foraging ants (Gronenberg, et al, 1996). Because ant foraging is based heavily in integrating olfactory information, it is essential for foraging ants to be able to distinguish between scents in order to make decisions that maintain the wellbeing of the whole colony.
While some of the traits described inherent to most eusocial insects and ant species, Myrmica sabuleti poses specifically relevant questions because it is a host for parasitic butterflies (Witek, et al., 2014). Maculinea butterflies, such as Phengaris alcon and Phengaris arion leave larvae inside of a M. sabuleti nest to either prey on ant larvae or eat food provided by nurse ants (Witek, et al., 2014). Maculinea butterflies are threatened species and the stability of a specific population can be accurately predicted by analyzing the related population of Mymica ants (Thomas, et al., 2009). However, hosting parasitic butterflies causes a significant strain on a colony’s resources and the wellbeing of its brood (Thomas, et al., 1992; Witek, et al., 2016). While the food shortages simulated in this experiment will not perfectly represent the affect of a parasitic butterfly in a colony, the results may provide insights into how a colony can adapt its behavior when under environmental stress, which will benefit conservation efforts.
C. Experimental Design and Methodology:
10 colonies of M. sabuleti will be used for this experiment. Each colony will have approximately 200 workers and 1 – 2 queens. They will be kept indoors in separate nest boxes in an appropriate substrate, with a regular 12 hour light/dark cycle and appropriate temperature.
Before any data collection takes place, we will rear single cohort colonies without brood. Previous studies have shown that young, non-foraging ants still experience olfactory-based learning within the nest due to tasks such as brood care and nest maintenance (Gronenberg, et al, 1996). By rearing the ants without brood, their mushroom bodies should stay relatively undeveloped until they specialize in foraging, and therefore should display a more clear trend over time.
Because ants are invertebrates, they are not protected under any animal testing legislation.
Preliminary data collection:
A period of one month will be reserved to collect preliminary data necessary for the controls during the next step in this experiment. The nests will be fed an excess of mealworms (protein source) combined with a sucrose solution. These food sources will be added at the beginning of the day and removed at the end. The weight of each food source will be taken to calculate the amount eaten in one day. The average amount of food eaten every day will be used as a guide to determine the “sufficient” amount of food.
As will be explained below, during the experimental data collection period, a citrus scent will be added to nutrient-poor food in order to mediate olfactory learning in the decision-making colonies. During the preliminary data collection period, we will run T-maze tests on a random sample of ants to make sure that the ants are neither attracted to or adverse to the scent. These ants will not be returned to the source colonies. If we find that the citrus scent biases the decision-making of these ants, we will test back-up aromatic chemicals such as vanilla or carvone oil (this is the last resort because it is the most difficult to obtain).
Each nest box will also be set up with a Broodminder tool. This tool is typically used for monitoring bee hives but can be adjusted for our purposes. It is placed underneath a nest box and provides real-time updates on the nest’s weight and temperature. The weight of a nest box is directly correlated to the number of individuals within the box. This information will be used to monitor the wellbeing of each colony throughout the experiment.
This period of data collection will also be used to fine tune the camera/video setup and procedures necessary for the next step. We will use the video data to determine the 3 hours of the day in which the nests are most actively foraging.
Foraging behavior treatments:
Once preliminary data has been collected, the nests will be treated with different food sources so we can observe the changes in foraging behavior. Complete food contains the mealworm and sucrose-solution in the proportions observed during preliminary data collection. Protein-poor food contains half the necessary proportion of mealworms. Sucrose-poor food contains half the necessary proportion of sucrose solution. Additionally, both protein- and sucrose-poor food will include a citrus scent (nontoxic) in order to mediate olfactory-based learning.
Sufficient quantity means the food is present in the necessary quantity (by weight) observed in the preliminary data collection. Limited quantity is 2/3 the quantity of food determined necessary.
Table 1 – Treatments for each nest
Nutrient-rich food in sufficient quantities.
Nutrient-rich food in limited quantities.
Protein-poor food in sufficient quantities.
Protein-poor food in limited quantities.
Sucrose-poor food in sufficient quantities.
Sucrose-poor food in limited quantities.
Choice between nutrient-rich food and protein-poor food in sufficient quantities.
Choice between nutrient-rich food and protein-poor food in limited quantities.
Choice between nutrient-rich food and sucrose-poor food in sufficient quantities.
Choice between nutrient-rich food and sucrose-poor food in limited quantities.
Nest 1 serves as the negative control because it will be given the same amounts of food given during the preliminary data collection and is not expected exhibit any changes in behavior.
Nests 2 – 6 will either be given a nutrient-poor diet, or a shortage of food. These serve as non-decision-making experimental groups. Food-shortage colonies are expected to increase their amount of foraging behavior in search of more food, even once the limited food source is depleted. Nutrient-poor colonies are expected to increase the amount of visits to the food sources because they will need a higher quantity of food to meet their nutritional needs. The observed behavioral changes in these groups will allow us to control for the effects of food shortages and malnutrition in the decision-making colonies.
Nests 7 – 10 will be referred to as the decision-making colonies, as they are the only colonies given two different food sources to choose from. In the case of colonies given two different food sources, 50% of the food by weight will be nutrient-rich and 50% will be nutrient-poor. Here, we expect to see a higher proportion of foraging ants to visit the nutrient-rich food over the nutrient-poor food. As mentioned earlier, the unique scent of the nutrient-poor food sources is expected to help foragers differentiate between the two food sources.
Every nest will be recorded from above for one hour a day. This hour will be determined randomly from the pool of three hour periods where the nests are most active, as determined during the preliminary data collection. During this 1 hour period, the following data will be recorded: the number of exits from the underground portion of the nest and the number of visits to each food source.
Because of the sheer number of ants and their small size, it would be unrealistic to tag ants in order to track individual behavior. Due to this, the number of exits, and number of forager visits to the food sources may include individual foragers who leave the nest multiple times or who visit a food source multiple times over the course of a single hour. This also means its possible for the number of visits to a food source to be higher than the number of exits from the nest.
During this period of data collection, we will continue to monitor the health of the colonies using Broodminder data. We will continue to collect data for 5 months or until the colonies show signs of hibernation.
During both the preliminary and experimental periods, we remove a random sample of 5 workers from each nest once a week for dissection. This sample size may decrease if the population size of the nests decreases to a point where the colony’s wellbeing is at risk. We will cut open the head capsules and dissect the brains in isotonic saline. The brains will then be rinsed in a buffer solution, rinsed in water and dried in ethanol, then finally sectioned in 10 µm slices. The slices will be dyed with Methylene blue. These slices can then be observed under a microscope. These procedures have been used in previous studies to observe brain morphology in ants (Gronenberg, et al, 1996). Because head size is highly variable between individuals, we will measure the total brain volume, then calculate the percentage of brain volume taken by the mushroom bodies.
We expect to see a correlation between decision-making and the morphology of the mushroom bodies. We will compare the morphology of the decision-making ants to the nutrient-poor ants and food-shortage ants to make sure that the food treatments alone were not responsible for any of the observed changes.
For each colony, we will have collected the following data: average % brain volume of the mushroom bodies, average visits to each food source, average exits from the nest. We can pool this data by week and plot each nest’s datapoints by time. In the control and non-decision-making colonies, we expect to see no difference in their foraging behaviors over time.
The decision-making colonies have two sets of data for food source visits, because each had two food sources. We will run paired T tests on these data to test for a difference between the number of visits to each food source. We expect to see a bias in food source visits towards the nutrient-rich food once the ants have established the relationship between the scent and nutritional value of each food source. A significant increase in visits to the nutrition-rich food source confirms that the ants are capable of making decisions while foraging and have been exhibiting this behavior over the course of the study.
We will run a second set of ANOVA tests on the weekly mushroom body data for each nest. A pairwise analysis will then be used to determine which weeks showed the significant increases in size (if the ANOVA test shows significant difference in the data sets). Decision-making colonies are expected to show a significant increase in the size of the mushroom bodies. The control and non-decision-making colonies are expected to not show a significant increase in their mushroom bodies. The lack of a significant trend in the non-decision-making experimental colonies will confirm that malnutrition or food shortages alone are not responsible for the changes in brain morphology of the ants. The lack of significant trend in the control colony will confirm that the change brain morphology is not an inherent physiological trait.
Table 2 – Proposed Budget
Tar Heel Ants
Pumice (substrate, 10lb bag)
BroodMinder-W weight scale
Frozen crickets (1lb bag)
Granulated sugar (1lb bag)
Citric acid (food grade, 5oz)
Carvone oil (5 oz)
Vanilla extract (1oz)
Methylene blue dye (4 oz)
Note: I will be able to access the equipment needed to slice and mount the ant brains in the mouse lab.
Initial M. sabuleti queens and colonies will be collected from the Philadelphia zoo and Central Park zoo, then additional workers will be reared for a period of three months before the study begins to ensure enough individuals for each colony.
D. Literature cited: