Assessment of Various Standard Fish Diets on Growth and Fecundity of Platyfish (Xiphophorus maculatus) and Medaka (Oryzias latipes)

Abstract

Several small freshwater fish species are utilized as models for human conditions and disease in biomedical research. Research animal diets are generally tailored to optimize growth, fecundity, and produce healthy research animals. However, a lack of reference diets presents a barrier in comparative studies between aquatic animal models and even among laboratories using the same species. Therefore, the objective of this study was to determine feeding regime and dietary effects on growth and fecundity in two commonly used freshwater fish, platyfish and medaka. From 1 through 6 months of age, platyfish and medaka were fed one of three feeding regime/diets: (1) our custom feeding regime consists of commercial flake food, beef liver paste, and live brine shrimp (CON); (2) a commercially available zebrafish diet, Gemma (GEM); and (3) a laboratory defined reference feeding regime (WAT). Weight, size, brood numbers, and survival rates for both species were measured monthly. Numbers of platyfish fry and hatch rate of medaka embryos were also determined. We observed that custom feeding regime (CON) fed platyfish and medaka grew larger, exhibited a higher survival rate, and had higher fecundity than WAT or GEM fed fish. These observations suggest that diets and regimes designed for zebrafish are not optimal to maintain platyfish or medaka. Thus, base diets, with clearly defined components and regimes, need to be developed with compositions that can be adjusted in a species-specific manner.

Introduction

It has been demonstrated that dietary variations may lead to phenotypic variations in laboratory animals.¹ Previous data indicate that zebrafish reared on different diets vary significantly in growth, reproduction, and survival^2–5 demonstrating the importance of diet as a controllable environmental factor. Standardization of diets for mouse and rat model organisms in the 1970s contributed to the minimization of nutritional effects on experimental results and thereby enhanced data reproducibility.^6,7 Research to determine optimal nutrient requirements and develop standardized diets has been carried out for other vertebrate laboratory animals and zebrafish.⁸ However, such defined diets have not yet been established for other common laboratory fish models (e.g., medaka, platyfish, swordtail, mosquitofish, guppies, stickleback, and spotted gar).

These nonzebrafish fish models are often reared on custom feeding regime/diets developed specifically by the host facility and may or may not be supplemented with commercial feed having unknown or poorly defined components or nutrients. Commercial diets vary in composition and nutritional value resulting in a lack of nutritional control in laboratory fish husbandry.⁵ Furthermore, many research laboratories utilize diets intended for aquaculture production that are formulated to develop fish that grow rapidly, but may neglect long-term health needs of fish used in research studies.² Therefore, diet is an “extrinsic factor” that should be precisely controlled to produce uniformly healthy research animals and to enhance transparency and reproducibility among biomedical research fish models.

To assess the need of the research community for developing a reference diet, the Office of Research Infrastructure Programs (ORIP/NIH) held a workshop (July 30, 2018) entitled: Defined Diets for Zebrafish and Other Aquatic Biomedical Research Models: Needs and Challenges (https://orip.nih.gov/about-orip/workshop-reports). Key outcomes were that reference diets are prerequisites to study nutritional requirements of fish and that such reference diets should provide husbandry characteristics comparable to (not exceeding) other currently available commercial diets. Thus, reference diets need to support rearing and growth of juveniles, as well as adult performance such as overall health, fertility, and fecundity. The use of reference diets and testing of specific components will advance our knowledge of the nutritional requirements of fishes in biomedical research. To assess the utility of a recently proposed laboratory-defined reference diet for zebrafish for the husbandry of several aquatic species, and across several facilities, ORIP supported the current study with an administrative supplement to the grants of the Kent laboratory (Oregon State University, Corvallis), the Zebrafish International Resource Center (ZIRC, University of Oregon, Eugene), and the Xiphophorus Genetic Stock Center (XGSC, Texas State University, San Marcos).

As the most commonly utilized fish model, a laboratory-defined diet has been developed and proposed for zebrafish (Danio rerio) by the Watts laboratory at the University of Alabama at Birmingham (WAT).^2,9,10 Another diet, Gemma (GEM; Skretting), is commercially available and has been utilized in numerous zebrafish facilities. To assess the proposed zebrafish reference diet, we compared the effectiveness of GEM, WAT, and our current fish diet at the XGSC to maintain two nonzebrafish model organisms, Xiphophorus and Oryzias. Xiphophorus is a genus including 26 known species that were collected in Central and South America and are available in the XGSC.¹¹ Xiphophorus is a classic genetic model that is best known for cancer genetics studies leveraging negative epistasis-induced spontaneous tumorigenesis within Xiphophorus interspecies hybrids (i.e., X. maculatus-X. hellerii and X. birchmanni-X. malinche hybrids).^12–15 The Xiphophorus model is also utilized to investigate a broad range of studies (e.g., behavior, evolution, viviparity^16–22). Oryzias, commonly known as Japanese rice fish, resides in rice patties in eastern Asian countries. Oryzias is commonly used for toxicological, neurodegenerative disease, cancer research, and developmental studies.^12,23–26 The XGSC feeding regime was adopted from a feeding protocol developed decades ago. It contains three portions, Zeigler flakes (Zeigler Bros, Inc., Gardners, PA), live Artemia nauplii (BIO-MARINE^®, Hawthorne, CA), and Gordon's beef liver paste. Although the live feed and beef liver paste portions are known variables between batches, this custom feeding regime has been effective in maintaining the Xiphophorus species and Oryzias lines. Herein, our main research goal for this study was to investigate if the two zebrafish diets, GEM and WAT, could be extrapolated for Xiphophorus and Oryzias maintenance. We compared growth and fecundity of Xiphophorus and Oryzias on the XGSC custom feeding regime, GEM, or WAT for 6 months. Increases in both hepatic glycogen and lipid are reported effects of nutritional status, developmental status, sex, dietary fat and caloric contents, and contaminant exposure in several fish species.^27–29 Therefore, we also performed histology in evaluating the hepatic glycogen and lipids of fish subjected to different feeding regime/diets. The data obtained from this study will determine if the zebrafish diets are appropriate for feeding Xiphophorus and Oryzias and can help to frame future directions in the development of a standardized reference diet for these two species and additional various fish models.

Materials and Methods

Animals

This study was conducted in accordance with the ethical guidelines for animal research approved by the Institutional Animal Care and Use Committee at Texas State University (protocol #7234). Platyfish (Xiphophorus maculatus strain JPWild) and medaka (Oryzias latipes strain Carbio) were obtained from XGSC at Texas State University, San Marcos, TX. Three 37.8 L tanks per dietary treatments were used for each species. Environmental temperature was a constant 25°C, and fish were maintained on a 13-h light/11-h dark cycle. At 1 month of age, platyfish were weighed and measured before separation into three tanks per diet (nine tanks total) at a density of n = 10 fish/tank. At 1 month of age, medaka were weighed and measured before placing into three tanks per diet (nine tanks total) at a density of n = 25 fish/tank. The average weight for Xiphophorus was 54.6 and 43.6 mg for medaka (Supplementary Tables S2 and S3).

Dietary treatments

Per dietary group, fish were fed twice daily Monday to Saturday (8:00 am and 4:00 pm) and once on Sunday (8:00 am) at 3% body weight (BW) per day. The three dietary treatments were as follows: an in-house feeding regime by XGSC consisting of Zeigler flakes (Zeigler Bros, Inc.), live A. nauplii (BIO-MARINE), and Gordon's beef liver paste (CON, Table 1)³⁰; Skretting Gemma Micro150 for juvenile and 300 for adult fish (Skretting Zebrafish, Westbrook, ME; GEM; Table 1); and a zebrafish reference diet developed by the University of Alabama, Birmingham (WAT, Table 1). Zeigler flakes, GEM, and WAT pellets were ground to a fine powder before feeding. Proximate and elemental analysis of diets was performed by Eurofins, Inc., commercial laboratory for GEM, WAT, and Zeigler flakes (Table 1). Nutritional contents were obtained from the manufacturer for A. nauplii and estimated from the USDA food database for beef liver paste.

Table 1.

Proximate and Elemental Analysis of Diets

	Treatmenta
	CON			GEM	WAT
	BLPb	Artemia (cysts)	Aquatox		Juvenile	Adult
Crude protein, %	13.6	62.8	51.8	60.1	43.1	44.2
Crude fat, %	8.3	6.2	11.8	18.9	22	9.5
Moisture, %	—	6	3.3	6.4	8.5	7.9
Nitrogen, %	—	—	8.3	9.6	6.9	7.1
Ash, %	—	4.9	11.2	12.4	5.2	14.9
Crude fiber, %	2.2	0.6	0.6	0.4	2.8	2.8
Metabolizable energy, kcal/lb	—	—	1,550	1,592	1,760	1,366
Digestible energy, kcal/lb	—	—	1,810	1,896	2,014	1,567
Nitrogen free extract, calculated	—	—	21.34	1.9	18.4	20.7
Calcium, %	0.25	—	2.41	1.72	0.48	0.49
Iron, %	0.02	—	0.02	0.03	0.01	0.05
Magnesium, %	0.02	—	0.17	0.25	0.08	0.1
Phosphorous, %	0.11	—	1.45	1.96	0.08	0.83
Potassium, %	0.18	—	0.79	1.02	0.09	0.8
Sodium, %	0.61	—	0.71	1.04	0.61	0.63
Sulfur, %	—	—	0.61	1.48	0.43	0.43
Copper, ppm	0.9	—	11	19	6.1	8.4
Manganese, ppm	—	—	41	60	9.8	12
Zinc, ppm	27	—	126	202	46	47

^a CON = beef liver paste (BLP), Artemia, and Zeigler Aquatox flake; GEM = Skretting Gemma Micro150; WAT = laboratory-defined reference diet developed by Steven Watts at University of Alabama Birmingham.

^b Proximate and elemental analysis conducted by Eurofins, Inc., for all ingredients except for Artemia, for which values were provided by the manufacturer, and BLP, for which values were estimated using USDA Food database.

Fish were fed at 3% BW per day. Flake food portions were controlled using custom, 3D printed, feeding spoons at the Zebrafish International Resource Center (ZIRC; www.thingiverse.com/thing:2855202). Spoon volumes were adapted to the specific densities of flake diets and number of fish per tank. The juvenile phase was defined as 1–3 months of age, and the adult phase was 4–6 months of age. For CON group juvenile platyfish, fish received 50% of dietary allowance as Zeigler Aquatox flake and 50% as live A. nauplii. Adult CON platyfish received 25% of dietary allowance as Zeigler Aquatox flake, 50% as live A. nauplii, and 25% as beef liver paste. Medaka of all age groups received 50% Zeigler Aquatox flake and 50% live A. nauplii. For GEM, both juvenile and adults of the two fish species were fed 3% BW of diet per day. The WAT diet has two different formulations for juvenile and adult fish (Table 1) and was fed to different growth phases of the two species accordingly.

Growth measurements

Monthly, fish were placed into 4 mL of anesthetic solution (0.01% MS-222) then weighed and placed over grid paper, and photographed. Body length and width were determined using ImageJ software.³¹

Reproductive measurements

At 3 months of age, platyfish tanks were inspected for fry daily, and medaka eggs were collected in the morning before feeding (7:30 am). Platyfish fry, if present, were separated into an additional tank and monitored for survival for 14 days. Medaka eggs were inspected for chorionic threads and fertilization status. Fertilized eggs per dietary group were maintained in 40 mL of E2 Embryo Media (https://wahoo.cns.umass.edu/sites/default/files/2019-01/E2_solution.pdf) with Methylene Blue that was changed daily. Hatchlings were documented to calculate hatch rate and were monitored for survival for 14 days.

Calculations and data analyses

Weight, length, and width changes were expressed as percent change compared to baseline measurements: [(final value/initial value)*100]. Body condition factor (BCF)³² was calculated as: [(weight (g)/(length (cm)³)*100] as a simple and rapid assessment tool to evaluate animal health and welfare.

Statistical analysis was conducted using R interface RStudio Version 1.3.1093 to compare differences in growth, reproduction, and survival between three dietary treatments in two fish species over 6 months. Data were analyzed using mixed model ANOVA with diet as a fixed factor and tanks nested as random factors within diet. Before analysis, a Shapiro–Wilk normality test was conducted with the “stats” package on the overall data with alpha set at 0.01. If data were normally distributed (p > 0.01), a parametric ANOVA test was performed using “stats” package. Post hoc and pair-wise analyses were conducted using t-tests. Non-normally distributed data (p < 0.05) were evaluated using the Kruskal–Wallis rank sum test with the “stats” package, and for significant observations, post hoc tests were conducted using Nemenyi tests for multiple comparisons of rank sums with the “PMCMR” package. Results were organized into bar plots using the “sciplot” package.

Histology

Fish were euthanized by hypothermia and processed for histology using our routine method for zebrafish,³³ in which slides that are mid-sagitall sections are prepared. This provides sections from two halves of the fish in which essentially all organs are visible for evaluation. Histologic sections prepared from both fish species were blindly evaluated separately. All organs were examined, and the occurrence of vacuolation of hepatocytes in liver that are indicative of either glycogen or lipid was scored.^28,34 Specifically, scoring was as follows: For glycogen, 1 = slight cell swelling, ground glass to microvascular cytoplasm, may be patchy distribution; 2 = almost all hepatocytes with lacy angular vacuolation, cytoplasm intact, and nucleus generally central; 3 = multiple vacuoles consistent with glycogen, with apparent disruption of cell membrane and eccentric nucleus. For lipid, 1 = up to about one quarter of cells have vacuoles consistent with lipid, characterized by spherical inclusions with distinct margins; 2 = approximately one quarter to two thirds of cells consist of lipid vacuoles.

Results

WAT and GEM negatively affected platyfish and medaka growth

For both platyfish and medaka, we measured weight, length, and width monthly and calculated BCF. For platyfish, CON significantly increased weight, length, and width at all time points versus WAT (p-value <0.05; Fig. 1A–C) and significantly increased width versus GEM from month 2 to 5 (p-value <0.05; Fig. 1C; Supplementary Fig. S1).

FIG. 1.

Dietary impacts on growth of platyfish. Bar graphs showing (A) increase in weight, (B) increase in length, (C) increase in width, and (D) body condition factor of platyfish treated with different diets. *Indicates the two dietary groups in comparison that are statistically significant (p-value <0.05). ^#Means there is no statistically significant difference.

As with platyfish, the medaka CON group had significantly increased weight increase versus the WAT dietary group but had the same width and length gain (except month 3) as WAT with no difference in BCF (Fig. 2). The GEM diet only exhibited a delay in weight gain until month 3 (Fig. 2A), with no significant difference in length or width gain from the CON group (Fig. 2B, C).

FIG. 2.

Dietary impacts on growth of medaka. Bar graphs showing (A) increase in weight, (B) increase in length, (C) increase in width, and (D) body condition factor of medaka treated with different diets. *Indicates the two dietary groups in comparison that are statistically significant (p-value <0.05). ^#Means there is no statistically significant difference.

WAT and GEM decreased platyfish and medaka survival rate

By the end of the regime/dietary treatment (i.e., 6 months), platyfish in the CON group had a 95% ± 9% survival rate, which was greater (p-value <0.05) than the GEM group (67% ± 13%). WAT group had a similar survival rate to CON group (p-value = 0.15; 76% ± 16%; Table 2).

Table 2.

Survival of Platyfish and Medaka Over a 6-Month Period by Diet

	Treatmenta			p-value
	CONb	GEM	WAT	CON vs. GEM	CON vs. WAT	GEM vs. WAT
Platyfish, %c	95 ± 9	67 ± 13	76 ± 16	0.03	0.15	0.63
Medaka, %	100 ± 0	91 ± 3	91 ± 2	<0.01	<0.01	0.42

^a CON = beef liver paste (BLP) (excluded for medaka), Artemia, and Zeigler Aquatox flake; GEM = Skretting Gemma Micro150/300; WAT = laboratory-defined reference diet.

^b Values are presented as Mean (SEM).

^c Percent survival calculated as: [(final number of fish in tank/initial number of fish in tank – fish sacrificed)^*100].

Both GEM and WAT diets decreased medaka survival over the course of the experiment. We observed a 100% survival rate for CON group, while the GEM group survival rate was 91% ± 3% (p-value <0.05) and the WAT group survival rate was 91% ± 2% (p-value <0.05; Table 2).

WAT and GEM reduced fecundity of platyfish and GEM reduced fecundity of medaka

To assess fecundity of platyfish and medaka, we documented brood sizes of viviparous platyfish and clutch sizes of oviparous medaka. Newborns were only observed in CON group for platyfish. At month 6, a total of 21 fry were collected from CON group. All fry survived after 14 days (data not shown).

For medaka, there were more eggs (p-value <0.05) collected from CON (76.6 ± 70.5) versus GEM (16 ± 13.4; Table 3) and WAT but no difference (p-value = 0.06) between CON and WAT groups (18.4 ± 16.2). Diet did not affect hatch rate (Table 3), but hatchlings from CON group exhibited a higher survival rate than those from GEM group (p-value <0.05; Table 3).

Table 3.

Medaka Reproductive Outcomes by Diet

	Treatmenta			p-value
	CONb	GEM	WAT	CON vs. GEM	CON vs. WAT	GEM vs. WAT
No. of eggs	76.6 ± 70.5	16 ± 13.4	18.4 ± 16.2	0.04	0.06	0.4
Hatched, %c	53 ± 11	43 ± 40	38 ± 25	0.5	0.23	0.23
Survived, %d	87 ± 9	47 ± 43	60 ± 42	0.02	0.14	0.17

^a CON = Artemia and Zeigler Aquatox flake; GEM = Skretting Gemma Micro150/300; WAT = laboratory-defined reference diet developed by Steven Watts at University of Alabama Birmingham.

^b Values are presented as Mean ± SEM.

^c Percent eggs hatched calculated as: [(number of eggs hatched/number of eggs collected)^*100].

^d Percent survival of hatchlings calculated as: [(number of hatchlings survived/number of eggs hatched)^*100].

All feeding regime/diet exhibited the same level of liver glycogen and lipid in platyfish and medaka

For platyfish, glycogen accumulation was variable, with glycogen scores ranging from 1 to 3, and exhibited no statistically significant difference between CON and WAT or between CON and GEM (Supplementary Table S4 and Fig. 3A–D). The liver lipid scores were also not statistically significant between groups.

FIG. 3.

H&E stained histologic sections of livers of medaka and platyfish fed with different diets. Bar = 30 μm. (A) Medaka CON diet, glycogen score = 1, no appreciable lipid or and minimal glycogen. (B) Medaka WAT diet, glycogen score = 1, lipid score = 2. Hepatocytes with spherical vacuoles with smooth edges (arrows) and eccentric nuclei are consistent with lipid storage. (C) Medaka GEM diet, glycogen score = 3. Note vacuoles with irregular edges and lacy appearance. (D) Medaka GEM diet. Multiple granulomas (G) indicative of mycobacteriosis. (E) Platyfish CON diet, glycogen score = 1. (F) Platyfish GEM diet, glycogen score = 3.

Glycogen scores ranged from mild to moderate among all medaka samples, regardless of regime/dietary group (Supplementary Table S4 and Fig. 3E, F). Two females in the WAT group showed moderate lipid accumulation (Score 2), whereas mild lipid accumulation was noted in one female in the CON group. However, there was not a statistically significant difference between groups for either glycogen scores or lipid scores.

Discussion

The WAT regime mainly stunted growth-related parameters (Figs. 1 and 2), while the GEM diet mainly decreased survival and fecundity (Tables 2 and 3). Both findings are potentially rooted in nutritional programming as WAT and GEM groups were subjected to suboptimal nutrition as juveniles. This could have predisposed both test species to adverse outcomes as adults.³⁵ More specifically, suboptimal nutrition during the early developmental stages could negatively impact development, causing long-term consequences.^29,36,37 Thus, the delayed growth in the early juvenile phase for Xiphophorus and medaka fed GEM and WAT could have resulted in unfavorable reproductive and mortality outcomes that were observed in the adult phase. There were no differences among regime/diet groups in BCF in either fish species suggesting that, even though there was more size and weight gain in CON fed fish, growth in all groups was proportional (Figs. 1D, 2D and Supplementary Table S1).

This study is the first investigation of the effects of zebrafish diets on Xiphophorus and medaka. We conclude that zebrafish feeding regimes and diets are not the most appropriate for nonzebrafish species.

There are two aspects to the underperformance of the two zebrafish diets. First is the dietary composition (Table 1). The three feeding regimes/diets are substantially different in composition (e.g., crude protein and lipids). These differences may impact performance. Scrutinizing the composition of the three diets tested in this study showed that WAT diet contains the lowest Zn. Zn deficiency was reported to affect gene expression related to development, cell growth, and proliferation, and Zn has been shown to affect early development in zebrafish.^38–41 In other aquatic species (e.g., crab), dietary Zn level influenced weight gain.⁴² These support the idea that low Zn composition in WAT may lead to the stunted growth of medaka and platyfish. Therefore, it is important, for the future development of Xiphophorus and medaka standardized diet to consider Zn concentration. The second aspect for underperformance of the WAT and GEM is the evolutionary distance and different evolutionary niches between the fish species used in this study and zebrafish. Xiphophorus and Oryzias diverged ∼18 millions years ago. In contrast, there are ∼300 million years of evolutionary distance between zebrafish and both Xiphophorus and Oryzias. Their native habitats are also diverged (i.e., medaka: east Asia, Xiphophorus: South America, zebrafish: south Asia). Their adaptions to respective native niches may explain why their specific nutritional needs are varied and indicates that a diet developed for one species may not be optimal for maintaining other species. The CON feeding regime led to larger and more prolific platyfish and medaka. However, this does not mean the CON is the most appropriate for platyfish or medaka feeding. Comparison between our CON fed medaka growth to that from medaka that were fed differently (i.e., paramecium for larvae, A. nauplii and commercial powder feeds for juvenile and adult ad libitum) showed that the CON fed medaka grew slower from ∼3-month old (Supplementary Fig. S2;⁴³). By 4-month old, the medaka fed ad libitum with Artemia and powdered feeds were twice the size of CON fed medaka (Supplementary Fig. S2;⁴³). Given this evidence, it is important to consider the species-specific nutritional requirement, composition of diet, quality of dietary composition, and quantity of feeding in further feeding regime development.

Increased vacuolation of hepatocyte cytoplasm may indicate storage of either lipid or glycogen in a well-fed fish or it may represent a toxicant related degenerative change caused by lipid distension of organelles such as the endoplasmic reticulum and Golgi apparatus or an accumulation of free fluid in the cytoplasm.⁴⁴ Unspecified vacuolation, therefore, can be an indication of either good or poor general health. The liver histology of both species for all dietary groups showed that there were minimal changes in the liver. Vacuolation of the hepatocyte cytoplasm is consistent with accumulation of glycogen, and a few fish showed cytoplasmic changes that are consistent with lipid accumulation. Although there were differences in growth and fecundity between tested diets, the liver changes in the present study in both fish species were considered mild, within the range of what would be expected for healthy fish. Histology allowed for detection of other histopathologic changes, and one medaka showed lesions consistent with mycobacteriosis. This is occasionally seen in Oryzias in research [15] and, given that only one fish was affected, we conclude that no confounding underlying infections or diseases compromised our study.

In addition to identifying the effects of the WAT and GEM diet on nonzebrafish species, this study also highlights a need to better understand nutritional requirements and eventually standardize Xiphophorus and medaka diets. A portion of the current custom CON (i.e., BLP) was developed by fish hobbyists almost a century ago and was adapted by the XGSC in the 1930s. The beef liver used to prepare the paste originates from nontraceable sources resulting in unknown nutritional variables. Therefore, there is an imminent need to produce a controlled diet that replicates the CON nutrient composition for Xiphophorus fish. As discussed earlier, standardization of diets for fish models requires consideration of species-specific nutritional needs, as well as potential transgenerational impacts. Although WAT and GEM led to slower growth and decreased fecundity than the custom CON diet, with known ingredients, they form a solid foundation to further develop standardized diets for Xiphophorus, Oryzias, and potentially other fish species. It would also be the best practice to develop and implement a base diet for all fish biomedical models and provide traceable nutrient additions to tailor the nutritional needs essential for proper growth and development for each species.

Authors' Contribution

All persons who meet authorship criteria are listed as authors, and all authors certify that they have participated sufficiently in the work to take public responsibility for the content. Furthermore, all authors have read and approved of the final article.

Disclosure Statement

The authors do not declare conflict of interest.

Funding Information

This work is supported by National Institutes of Health (NIH), Office of Research Infrastructure Programs (ORIP) administrative supplement grant to R24 OD011120 to XGSC, and P40 OD011021 to ZIRC, and NIH, NCI grant R15-CA-223964. This work is also part of graduate thesis entitled “Standardizing and optimizing rearing methods of animal models: Influence of diet on growth and reproduction in Xiphophorus maculatus and Oryzias latipes” (https://digital.library.txstate.edu/handle/10877/13491?show=full). The thesis author is Ms. Crystal Russo. The degree she qualifies is Master of Science in Integrated Agricultural Sciences. The granting institution is Texas State University.

Supplementary Material

Supplementary Figure S1

Supplementary Figure S2

Supplementary Table S1

Supplementary Table S2

Supplementary Table S3

Supplementary Table S4