University Makerere

MAKERERE UNIVERSITY

COLLEGE OF NATURAL SCIENCES DEPARTMENT OF ZOOLOGY, ENTOMOLOGY AND FISHERIES SCIENCES

MICROBIAL CONTAMINANTS IN WILD HARVESTED AND TRADED EDIBLE LONG-HORNED GRASSHOPPER, Ruspolia differens (ORTHOPTERA: TETTIGONIIDAE) IN UGANDA

SIMON LABU (BA. Educ., Dip. Educ. KYU)

2018/HD13/1871U

SUPERVISORS:

DR. PERPETRA AKITE DR. PATRICE KASANGAKI DR. JAMES P EGONYU

A DISSERTATION SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE AWARD OF MASTER OF SCIENCE DEGREE IN ZOOLOGY OF MAKERERE UNIVERSITY

APRIL, 2021

DECLARATION This dissertation is my original work and has not been submitted for an award to any other University before. Contributions from authors and/individuals involved at the different stages of developing this dissertation have been duly acknowledged with due reference.

Signed……… …………………… Date: 8th April 2021

SIMON LABU

APPROVAL We, the undersigned, agree to, and approve the final submission of the findings presented in this Dissertation:

DR. PERPETRA AKITE

Signed: …… ………………… Date: 8th April 2021

Makerere University, P.O. Box 7062, Kampala, Uganda.

DR. PATRICE KASANGAKI

Signed: Date: 8th April 2021

National Livestock Resources Research Institute (NaLIRRI), P. O. Box 5704, Kampala,

Uganda.

DR. JAMES P. EGONYU

Signed: Date: 8th April, 2021

International Centre of Insect Physiology and Ecology (icipe), P.O. Box 30772-00100,

Nairobi, Kenya.

DEDICATION To my father, Mr. Augustine Sokuton and my mother, Beatrice Chelangat who dearly took care and educated me right from primary up to University level.

ACKNOWLEDGEMENT This research was funded by the German Federal Ministry for Economic Cooperation and Development (BMZ) commissioned and administered through the Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ) Fund for International Agricultural Research (FIA), grant number: 012345678. I acknowledge icipe’s core funding provided by UK’s Foreign Common wealth & Development Office (FCDO); the Swedish International Development Cooperation Agency (SIDA); the Swiss Agency for Development and Cooperation (SDC); the Federal Democratic Republic of Ethiopia and the Government of Kenya. I express my heartfelt appreciations to my mentors and supervisors: Dr. Perpetra Akite, Dr. Patrice Kasangaki and Dr. James P. Egonyu for their encouragement, positive criticism and untiring guidance during this study and dissertation write-up. Thank you so much for all the technical guidance and moral support you offered to me throughout my study. Surely, I owe you a lot and may God bless you abundantly. Additional support and guidance provided by Dr. Moses Chemurot and Dr. Eric Sande and the entire staff of the Department of Zoology, Entomology and Fisheries Sciences (Makerere University) is greatly acknowledged. I also greatly appreciate the staff of icipe, Nairobi (Dr. Sevgan, Dr. Fathiya, Levi and Dr. James,) who supported and guided me while conducting my laboratory experiments. Thanks to all my colleagues, Francis, Fred, Sam, Violet and Charles for the advice and encouragement. Last but not least, my deepest gratitude to my beloved wife Lucy Cherop for her unconditional love, support and encouragement. Your words were a source of inspiration to me. And to my sons; Emmanuel and Elisha, and my daughter Esther, your unconditional love and inspiration were great.

iii

TABLE OF CONTENTS DECLARATION ...... i

DEDICATION ...... ii

ACKNOWLEDGEMENT ...... iii

LIST OF FIGURES ...... vi

LIST OF TABLES ...... vii

LIST OF ACRONYMS AND ABBRREVIATIONS...... viii Abstract ...... ix

CHAPTER ONE: INTRODUCTION ...... 1

1.1 Background ...... 1 1.2 Problem statement ...... 3 1.3 General Objective ...... 3 1.3.1 Specific objectives; ...... 4 1.4 Hypotheses ...... 4 CHAPTER TWO: LITERATURE REVIEW ...... 5

2.1 Introduction ...... 5 2.2 Biology of Ruspolia differens...... 5 2.3 Economic potential of edible insects ...... 6 2.4 Edible insects and policy regulations regarding their consumption ...... 6 2.5 Nutritional value of Ruspolia differens ...... 7 2.6 Microbial contamination of edible insects ...... 7 CHAPTER THREE: MATERIALS AND METHODS ...... 9

3.1 Study area ...... 9 3.2 Sampling design, sampling points and sample collection ...... 10 3.3 Species identity and characterisation of bacteria and fungi in/on Ruspolia differens ...... 12 3.3.1 Sample preparation and serial dilutions ...... 12 3.3.2 Isolation and analyses of bacteria and fungi in R. differens ...... 13 3.3.3 Microscopic identification and characterisation of bacterial and fungal isolates ...... 14 3.3.4 Molecular identification and characterisation ...... 15 3.3.4.1 DNA extraction ...... 15 3.3.4.2 PCR-amplification ...... 15 3.4 Evaluating bacterial and fungal loads in Ruspolia differens at the different points along the value chain in Uganda...... 18 3.5 Data analysis ...... 18

CHAPTER FOUR: RESULTS ...... 19

4.1 Identification and characterisation of bacteria and fungi in Ruspolia differens along the value chain in Uganda...... 19 4.2 Bacterial and fungal loads (safety levels) in Ruspolia differens from different sampling points along the value chain in Uganda...... 23 4.2.1. Microbial loads in R. differens samples from other districts sold at the distribution points in Kampala district ...... 25 CHAPTER FIVE: DISCUSSION ...... 27

5.1 Species of bacteria and fungi recorded in/on Ruspolia differens in Uganda ...... 27 5.2 Bacterial and fungal safety levels in/on Ruspolia differens in Uganda...... 29 5.2.1. Microbial loads in R. differens samples from other districts sold at the distribution points in Kampala district ...... 31 5.3 Health implications of the pathogenic bacteria and fungi detected in R. differens...... 31 CHAPTER SIX: CONCLUSIONS AND RECOMMENDATIONS ...... 32

6.1 Conclusions ...... 32 6.2 Recommendations ...... 32 REFERENCES...... 34

APPENDICES ...... 42

Appendix 1: Diversity of microbe species isolated from Ruspolia differens at different sampling points along the value chain in Uganda...... 42 Appendix 2: Species of bacteria and fungi isolated from R. differens samples from Kampala and Masaka districts...... 46 Appendix 3: Trapping of Ruspolia differens using local drums in Uganda...... 48 Appendix 4: Whole R. differens at trapping points being packaged for transportation ...... 48 Appendix 5: Handling processes involved in Ruspolia differens while in transit and at commercial distribution points in Kampala ...... 49 Appendix 6: Laboratory analyses involved during DNA extraction ...... 50 Appendix 7: Samples of extracted DNA of the two microbes ...... 50

LIST OF FIGURES Figure 1: Locations of Ruspolia differens sampling sites in the two districts ...... 9 Figure 2: Layout of Ruspolia differens sampling design ...... 10 Figure 3: Sampling points of Ruspolia differens along the value chain...... 11 Figure 4: Preparation of serial dilutions for microbial analyses...... 12 Figure 5: Pure cultures of bacteria and fungi isolated from Ruspolia differens ...... 14 Figure 6: Bands of bacteria and fungi in Ruspolia differens viewed under KETA GL imaging system trans illuminator ...... 17 Figure7: Mean bacterial (A) and fungal (B) counts in Ruspolia differens from different sampling points in Uganda in the first and second swarming seasons of 2019...... 24 Figure 8: Mean bacterial (A) and fungal (B) counts in Ruspolia differens samples from Masaka and Kampala in the first and second swarming seasons of 2019...... 25 Figure 9: Mean bacterial and fungal counts in Ruspolia differens from distribution points in Kampala originating from Fort Portal, Kabale and Kasese districts in the second swarming season of 2019...... 26

LIST OF TABLES Table 1: Gram stain tests of bacterial cultures from R. differens and their identities by molecular techniques...... 21 Table 2: Bacterial and fungal species isolated from Ruspolia differens in Uganda from different points along the value chain in 2019...... 22 Table 3: Proportional occurrence (%) of microbes classified in R. differens along the value chain .... 23

vii

LIST OF ACRONYMS AND ABBRREVIATIONS BLAST: Basic Local Alignment Search Tool cfu: Colony forming unit

DNA: Deoxyribonucleic acid

DRC: Democratic Republic of Congo

EFSA: European Food Safety Authority

EU: European Union

FAO: Food and Agricultural Organization

FASFC: Federal Agency for the Safety of the Food Chain

GLM: Generalized linear Model

GPS: Global Positioning System icipe: International Centre of Insect Physiology and Ecology

LPCB: Lactophenol cotton blue

PCR: Polymerase chain reaction pH: Power of Hydrogen rRNA: Ribosomal ribonucleic acid

SHC: Superior Health Council

USA: United States of America

WHO: World Health Organization

viii

Abstract Edible insects are now globally recognised as alternative sources of food and feed. Currently, over 470 spp of insects are consumed in Africa alone, with Ruspolia differens being the most common in central Uganda. Harvesting and trading of these insects is done informally and the post-harvest handling, processing and marketing involves several people. These coupled with other factors increase the risk of microbial contamination of R. differens. This study identified and characterised bacterial and fungal microbes found in R. differens along the value chain, and also evaluated the safety levels of R. differens at the different points, including their different districts of origin. Samples of whole R. differens were collected from wild vegetation, trapping sites and markets. Additionally, samples of plucked and deep-fried ready-to-eat R. differens were collected from the markets. The samples were cultured on standard media for microbial quantification, and pure cultures were identified and characterised using molecular techniques. There were seven species each of bacteria and fungi recorded in R. differens samples, with harvesting drums harbouring more microbes compared to the other points along the value chain. The key pathogenic bacteria detected in marketed R. differens were Staphylococcus sciuri, Acinetobacter baumannii and Serratia marcescens, all of which were absent in wild-caught whole R. differens. The bacterial and fungal counts in deep fried ready-to-eat R. differens were ~3- and 2-fold lower, respectively, than in raw samples, and the values in deep fried R. differens were within the recommended microbial limits for edible insects. In terms of bacterial loads, the most unsafe samples were the whole R. differens samples from the market followed by those from trapping points. The fungal counts in the raw R. differens were comparable across the sampling points, making their safety concerns comparable along the value chain. Districts of origin did not compromise the safety of R. differens sold in Kampala. These results demonstrate that R. differens obtained at the trapping sites and markets are contaminated with potentially harmful microbes, with varying levels of safety concerns. Processing through deep frying greatly minimises the health risks associated with consumption of R. differens through lowering the loads of microbial contaminants. However, the specific handling practices at distribution points, trapping points and markets (for ready-to-eat R. differens) which may be responsible for introducing microbes into R. differens still needs to be addressed.

Key words: markets, food safety, food processing, nsenene, trapping sites

CHAPTER ONE: INTRODUCTION 1.1 Background Edible insects have gained global recognition in recent years as alternative sources of food and feed with a comparable or superior nutritional content than conventional animal products (Van Huis et al., 2013; Kelemu et al., 2015; Dobermann et al., 2017). The rich nutritional value of edible insects coupled with high moisture content provide a conducive environment for microbial contamination, survival and growth (Klunder et al., 2012; Van Huis et al., 2013). Intrinsic factors (for example, pH, moisture content, osmotic pressure and water activity) and extrinsic factors (such as, temperature, light, relative humidity, number and type of microorganisms present) influence microbial survival and growth in foods (Charpe et al., 2019; Garofalo et al., 2019). Microorganisms can also be introduced along the food handling chain, causing food spoilage or disease outbreaks (Ng’ang’a et al., 2019; Ssepuuya et al., 2019).

The number and type of microorganisms in foods is highly influenced by the environment where the food originated and the sanitary conditions under which it is handled and processed (Van Huis et al., 2013). The symbiotic associations of many insect species with some fungi and bacteria (Engels and Moran, 2013) enables them to thrive and survive in the insects’ gut (Colman et al., 2012) and possibly causing spoilage once the insect dies. Several studies have highlighted the presence of different genera of bacteria and fungi in fresh and processed (ready-to-eat) insects for example, the migratory locust, Locusta migratoria (Orthoptera: Acrididae) and mealworm, Tenebrio molitor (Coleoptera: Tenebrionidae) (Stoops et al., 2016; Garofalo et al., 2017); crickets, Acheta domesticus and Gryllodes sigillatus (Walker) (Orthoptera: Gryllidae) (Osimani et al., 2017; Vandeweyer et al., 2017); and the long-horned grasshopper, Ruspolia differens (Serville) (Orthoptera: Tettigoniidae) (Ng’ang’a et al., 2019; Ssepuuya et al., 2019).

In East Africa, the long-horned grasshopper, R. differens commonly known as nsenene by various tribes forms a major part of the food culture, constituting 5–10% protein intake of some populations (Ayieko et al., 2012; Kinyuru et al., 2010). This insect exists predominantly in two colourmorphs, i.e. light brown and light green (Leonard et al., 2020). Swarms of R. differens occur twice a year (from March to May and November to December) during rainy seasons (Agea et al., 2008; Mmari et al., 2017; Ssepuya et al., 2017). These insects are harvested at night using high light intensity fluorescent bulbs to attract them onto

1 slanting silver iron sheets that slide the catch into collection drums (Okia et al., 2017). Materials such as moist cassava flour or cooking oil are applied on the drum walls and iron sheets to make them slippery to prevent the insects from escaping. These materials may however predispose the trapped insects to contamination by harmful microbes hence reducing their quality and increasing health risks to consumers.

Currently, there are attempts aimed at reducing the factors which predispose the nutritious insect (R. differens) to microbial contamination. Such attempts include mass rearing (Kelemu et al., 2015) and modifying the wild harvesting drums (Sengendo et al., 2021). Mass rearing prevents contaminations from the wild while modified drums prevents escape of the catch without using local materials (Okia et al., 2017) which compromises on the safety. As opposed to local harvesting drums, modified drums have three compartments separated by wire meshes of different sizes to filter out smaller by-catches, separating them from R. differens reducing further contamination.

After harvesting, trappers manually transfer R. differens from the drums into aerated polythene or sisal bags. The harvest is then sold to dealers who transport whole insects by public means to distribution points in other districts. In such points, the insects are sold to retailers who then process them before serving final consumers or other middlemen. The involvement of several people during the post-harvest handling processes is likely to compromise R. differens quality and safety (Woh et al., 2017; Ssepuuya et al., 2019). Processing of R. differens is usually carried out for value addition and to increase on shelf stability, but food processing methods may impact on safety and nutritional value (Bokulich et al., 2016; Nyangena et al., 2020). Processing R. differens begins with sorting out non- target insects and plucking (removal of wings, antennae and legs) followed by washing with clean water. To speed up the cleaning process, materials such as wood ash, maize flour or cassava flour are employed to increase friction (Ssepuuya et al., 2019). The plucked grasshoppers are usually deep fried, cooled and sold ready-to-eat or stored in cardboard boxes. Deep fried grasshoppers have an increased shelf life of up to twenty one weeks (Ssepuuya et al., 2016) which can enable R. differens business to continue throughout the year despite the seasonality of the swarms.

Several studies (for example, Sepuuya et al. 2019; Megido et al., 2017 & Stoops et al., 2016) recommended the application of microbial limits of minced meat for edible insects before

2 insects specific standards were established. Recently, the Uganda National Bureau of Standards (UNBS, 2019) and the Kenya Bureau of Standards (KEBS, 2020), published standards indicating microbial limits for edible insects. According to UNBS, the recommended maximum limits of bacterial and fungal counts are 5.0 log cfu/g and 3.0 log cfu/g respectively. On the other hand, KEBS recommends the maximum bacterial and fungal counts of 5.0 log cfu/g and 2.0 log cfu/g respectively. These limits are fairly comparable with the standards recommended for minced meat by the Uganda National Bureau of Standards (UNBS, 2019) at 6.0 log cfu/g for bacteria and 4.0 log cfu/g for fungus respectively. However, the Codex Alimentarius commission (FAO/WHO, 2002) recommends a higher total bacterial count limit of 7.0 log cfu/g.

1.2 Problem statement The consumption of, and trading in Ruspolia differens is on the rise in Uganda especially in the central and western parts of the country (Agea et al., 2008; Ssepuya et al., 2017). Despite the nutritive value, R. differens and other edible insects harvested from the wild may be contaminated with harmful bacteria and fungi (Schabel, 2008; Gallo, 2018). Also, the post- harvest handling of foods e.g. R. differens which is highly fragmented with several people and processes involved along the value chain increases it’s risk of contamination with harmful microbes (Woh et al., 2017). However, there is only piecemeal data on levels of microbial contamination of R. differens and from only a few selected points along the value chain (e.g. (Ng’ang’a et al., 2019; Ssepuuya et al., 2019). As such, it is difficult to design any insect food safety policy without first bridging this gap. This is because the quality of R. differens at any one point along the value chain has a direct bearing on the level of its safety at the subsequent point (s). This study therefore aims to provide the missing information given the prominence of these insects in food enterprises in Uganda.

1.3 General Objective To generate information on the microbial contaminants found in Ruspolia differens consumed in Uganda in order to guarantee the consumer safety, and provide data that would enable policy formulation for entomophagy.

1.3.1 Specific objectives; i. To identify and characterise bacterial and fungal microbes found in Ruspolia differens along the value chain in Uganda.

ii. To evaluate bacterial and fungal safety levels in Ruspolia differens at the different points along the value chain in Uganda.

1.4 Hypotheses Ho: There is no significant difference in species identity of bacteria and fungi in Ruspolia differens characterised along its value chain in Uganda.

Ho: Ruspolia differens obtained from any point of the value chain in Uganda is not safe for human consumption.

CHAPTER TWO: LITERATURE REVIEW 2.1 Introduction In recent years, edible insects are being promoted as a future alternative source of protein and most communities’ world over are beginning to embrace entomophagy. The most commonly consumed insect species belong to order Coleoptera (31%), followed by Lepidoptera (18%), Hymenoptera (14%) and Orthoptera (13%) (Jongema, 2015). Previous studies indicate that the protein and energy content of edible insects are quite comparable to those of other conventional sources of protein e.g. meat (Rumpold and Schlüter, 2013). In Uganda, the long-horned grasshopper is one of the most nutritious and economically important edible insect. Its consumption has been rising over the years especially in the central and western parts of the country with no regulations put in place (Agea et al., 2008; Ssepuya et al., 2017).

2.2 Biology of Ruspolia differens Ruspolia differens, also known as long-horned grasshoppers are arthropods belonging to class Insecta, order Orthoptera and family Tettigoniidae (Matojo and Hosea, 2013). Family Tettigoniidae has more than 6,000 species characterized by the long antennae (exceeds the body length) and powerful mandibles (Matojo and Njau, 2010). The hind limbs are long and strong (for jumping/taking off) with four tarsal segments; males have tegminal stridulatory organs and front tibial tympanum (Jang, 2011).

The distribution of R. differens is widespread from South West Africa through the Congo forests to South and East Africa (Matojo and Njau, 2010). Ruspolia differens usually occur in tropical grasslands and/farmlands where they are mainly active at night feeding on wild grass, flowers and seeds of cereals especially millet and maize cracked using their powerful mandibles (Matojo and Hosea, 2013).

Males usually produce very loud continuous hissing calls for up to five minutes to attract females. Fertilization is internal; males introduce sperm into the ovipositor through the aedeagus and females deposit fertilised eggs in soft plant tissues, tree branches or in the soil inform of egg pods. Nymphs hatch in 1-2 months and reach adult maturity in 2-3 months (Matojo and Njau, 2010). Nymphs are similar to adults but with less developed wings.

2.3 Economic potential of edible insects Commercialisation of edible insects has been an ongoing practice for decades at different levels. Some insects like mopane worms are usually exported from African countries to Europe, with Belgium and France importing the largest quantities of these insects (Chakravorty et al., 2011). The Democratic Republic of Congo (DRC) is the main country in Africa exporting mopane worms (Chakravorty et al., 2011).

Consumers buy processed and unprocessed insects at village/local markets, commercial distribution points, retail supermarkets and from street vendors (Ssepuuya et al., 2019). Insects are also consumed in restaurants depending on the extent to which anthropo- entomophagy is recognized in a given region (Megido et al., 2014). Fresh raw and cooked edible insects in Thailand are often sold at local markets, wholesale supermarkets and minimarts; available either as precooked or uncooked in frozen packages in supermarkets or found in markets when ready-to-eat and in microwaveable packages (Belluco et al., 2013).

In Uganda, traders of R. differens are mainly dominated by men and characterised by wholesalers who buy the insects from collectors and in turn sell to retailers, who subsequently sell to consumers along the roadside or highway vehicle-stopping points in Kampala (Agea et al., 2008). By trading in edible insects, most economically disadvantaged persons are able to sustain their families and educate their children (Siulapwa et al., 2014). In Tanzania, commercialisation of R. differens is prominent in Lake Victoria regions of Mwanza, Geita and Kagera. Fresh R. differens harvested in the region are sold by harvesters to retailers who may eventually sell the insects to consumers or roast and pack in small packages of 0.5 to 5kg (Mmari et al., 2017).

2.4 Edible insects and policy regulations regarding their consumption Insects, (both edible and none edible) are key life forms in many ecosystems, functioning as pollinators, decomposers and aerating soil through their burrowing amongst the many functions. They are also important food sources for birds, reptiles, amphibians and mammals, including humans (Durst et al., 2008). The World Health Organisation (WHO) records indicate that edible insects are a diet for more than two billion people worldwide (WHO, 2018). The nutritional benefits and low environmental stress of edible insects (Giovanni, 2005) is further expected to boost the consumption trends.

The fact that edible insects can make a good substitute for other alternative sources of protein for example, meat and fish has been attracting attention from several organisations (Giovanni, 2005). As a result, there have been several attempts carried out to scale up edible insects production and to develop standards that protect consumers (Mmari et al., 2017). Uganda and Kenya for example through their quality assurance regulatory bodies (UNBS and KEBS respectively) recently published standards for use in edible insects. Attempts of setting a regional standard are also underway to protect consumer health and ensure quality of edible insects and insect-enhanced products in the regional and international food trade (Van Huis et al., 2013). 2.5 Nutritional value of Ruspolia differens Most edible insects are reported to contain more proteins, fats and carbohydrates than the equivalent of beef or fish (Kinyuru et al., 2010). Edible insects also contain higher energy content than soybeans, maize, beef, fish and other beans (Belluco et al., 2013). The animal protein from R. differens is 44% crude protein (Rumpold and Schlüter, 2013) with good ratios of essential amino acids compared to those from plant protein (Kinyuru et al., 2010). Fats also form a big part of the insects’ biomass accounting for about 26.8 % on average of dried insect (Belluco et al., 2013). The crude fat content of mature R. differens is about 48% (Kinyuru et al., 2010) while pupal and larval stages of other insects have higher levels of fat (Rumpold and Schlüter, 2013). In addition to fats and proteins, high levels of vitamins (A = 2.12 mg/g, and E = 2.01 mg/g) among others with minerals (iron, zinc, potassium, magnesium, phosphorus, sodium magnesium) have also been recorded in R. differens (Kinyuru et al., 2010).

2.6 Microbial contamination of edible insects Edible insects may have associated micro-organisms that can influence their safety as food. They may be infected with pathogenic microorganisms like bacteria, viruses, fungi and protozoa during harvesting or some can occur intrinsically in insects (EFSA, 2015).

In most regions practicing entomophagy, edible insects are still obtained from the wild (Rumpold and Schlüter, 2013) although some like mopane warms are currently being produced commercially in other parts. The environments from which edible insects thrive and feed are getting contaminated with microorganisms from different sources including human wastes (Adwan and Salama, 2016). As such, there are increased chances with which edible insects harvested from the wild are contaminated with harmful microorganisms.

In addition, the methods of harvesting, processing, packaging and storage/marketing of edible insects in most developing countries are still informal, traditional, and carried out under unhygienic conditions (Ssepuuya et al., 2019). Commercialisation of R. differens in Uganda for example is characterised by a long chain of people and processing steps which increases its risk of microbial contamination (Ssepuuya et al., 2019; Woh et al., 2017). The Post- harvest handling and movement of foods such as R. differens can therefore be a complex process involving a number of intermediaries (Ssepuuya et al., 2019) at different locations/points. These makes the food susceptible to microbial contamination, especially if the conditions are not well managed (Woh et al., 2017; Ssepuuya et al., 2019).

CHAPTER THREE: MATERIALS AND METHODS 3.1 Study area This study was conducted in two purposively selected districts of Masaka and Kampala in central Uganda (Figure 1). Masaka district was selected because of the high quantities of R. differens usually harvested than other districts in the country, while Kampala is known to have the highest concentration of R. differens business activities (Agea et al., 2008).

Figure 1: Locations of Ruspolia differens sampling sites in the two districts

Masaka is bordered by Sembabule district in the Northwest, Mpigi in the North, Rakai to the west and south, and Kalangala district in the east. It is located between 30°3’–31°3’E and 0°15’–0°30’S with a total area of about 6413.3 km2 of which land area is 3214 km2. It has a bi-modal rainfall pattern with peaks experienced between March to May and September to December. The mean annual rainfall in Masaka ranges between 1,100 mm to 1,200 mm with temperatures ranging from 18°C to a maximum of about 28°C (Masaka District Local Government, 2020).

Kampala district on the other hand is the capital city of Uganda with a population of 1,507,000 by the year 2014. It is located at 0°19′N, 32°35′E with an altitude of 1,189 m above sea level. The mean annual rainfall ranges between 1,750 mm to 2,000 mm and the temperature ranges between a minimum of about 17℃ to a maximum of 28℃ (NEMA, 2014). Kampala district is bordered by Mukono district to the East and Wakiso to the south, north and west. It is divided into five boroughs that oversee local planning: Kampala Central Division, Kawempe Division, Makindye Division, Nakawa Division, and Rubaga Division (UBOS, 2014).

3.2 Sampling design, sampling points and sample collection Samples of R. differens weighing 200 g per batch were collected during the 14th - 15th of June (first swarming season) and 23rd - 24th of November (second swarming season) 2019 from purposively selected sampling points (Figures 2 and 3) in Masaka and Kampala districts. Coordinates of the sampling points were taken with a Global Positioning System (GPS) and plotted on Arc Map using the Arc GIS software version 10.3 (Esri Eastern Africa Ltd, Naiobi, Kenya) (Figure 1).

Figure 2: Layout of Ruspolia differens sampling design

Figure 3: Sampling points of Ruspolia differens along the value chain

In Masaka district, a total of 72 batches of R. differens (36 batches per swarming season) were obtained from the selected sampling points as follows; (i) Nine batches (3 replicates per site) of live whole R. differens collected from wild vegetation using a sweep net at Nyendo-vila road, Kayirikiti A and Kayirikiti B. (ii) Nine batches (3 replicates per site) of fresh whole R. differens collected from the harvesting drums at Kayirikiti A, Nyendo-vila road and Kitaka trapping points. (iii) Nine batches of fresh plucked and 9 batches of deep-fried R. differens (3 replicates of each type) at Nyendo, Masaka Municipal and Ssaza markets.

In Kampala district, a total of 45 batches of R. differens were also obtained from the purposively selected sampling points as follows; (i) Nine batches of fresh whole R. differens from the distribution points of Busega (3 replicates originating from Fort-portal district) and Katwe (3 replicates each of grasshoppers originating from Kasese and Kabale districts). These were only purchased in the second swarming season because sample collection in the first season commenced when supplies at the distribution points had become scarce. (ii) Eighteen batches in each swarming season (9 batches of fresh plucked and 9 batches of deep-fried R. differens) were purchased in triplicates from Nakasero, Kalerwe and Kasubi markets.

Ruspolia differens batches collected were each immediately placed in a sterile zip lock bag, kept in a cool box with flaked ice and transported to Makerere University, Department of Zoology, Entomology and Fisheries Sciences for preservation in Haier deep freezer HCF- 300HTQ at -20°C (Haier deep freezer, India) prior to transportation for microbial analysis at the International Centre of Insect Physiology and Ecology (icipe), Nairobi.

3.3 Species identity and characterisation of bacteria and fungi in/on Ruspolia differens Standard culturing protocols for both bacteria and fungi were carried out in laboratories at the international centre for insect physiology and ecology in Nairobi. Detailed step-by-step procedures are given below.

3.3.1 Sample preparation and serial dilutions From each batch of R. differens, 5 g were ground using a sterile pestle and mortar to make homogenates. These homogenates were suspended in 45 ml buffered saline in 50 ml falcon tubes to form stock solutions. Serial dilution was done by transferring 1 ml of the stock solution into another 9 ml phosphate-buffered saline in a 27 ml universal tube using a sterile pipette. The 10-fold dilution was repeated until the desired dilutions were obtained as shown by Figure 4.

Figure 4: Preparations of serial dilutions for microbial analyses (Adopted from Cummings, 2007).

3.3.2 Isolation and analyses of bacteria and fungi in R. differens Two media were used; the nutrient agar and MacConkey. The nutrient agar is a generalist media while the MacConkey is a highly specific media used for culturing only the gram- negative bacteria. Pre-prepared sterile triplicate plates of MacConkey agar and nutrient agar made according to manufacturer’s specifications were used to culture and enumerate bacterial colonies. Inoculation of the samples into plates was done aseptically by spreading 1 ml of the sample from the dilutions between 10-2 to 10-8 (Ssepuuya et al., 2019) followed by incubation at 37℃ for 24-72 hours to allow bacterial growth. Plates with less than 300 colonies were counted with the aid of a hand-held tally counter.

Sterile triplicate plates of Potato dextrose agar (PDA) and Sabouraud dextrose agar (SDA) each containing 0.05g/l of chloramphenicol made according to manufacturer’s specifications were used for enumerating fungal colonies. Inoculation of individual samples into plates was done aseptically by spreading 0.1ml of the sample from dilutions between 10-2 to 10-5 (Ssepuuya et al., 2019) followed by incubation at 30℃ for 5 days. After 5 days of growth establishment, fungal counts were obtained with the aid of a dissecting microscope.

The mean for each triplicate plates of bacteria and fungi given by the formulae Mean = (∑x/ n) were obtained and multiplied by the dilution factor (d) to get the total number of viable cells per unit weight (expressed as colony forming unit per gram (cfu/g)) of the sample (Omoya and Akinyosoye, 2013), i.e. cfu/g = (∑x/ n)d.

Where; ∑x = total sum of the triplicate colony counts n = 3 (number of plates) d = dilution factor

Plates with colonies beyond the required range were sterilised and discarded. Sub-culturing of bacteria and fungi was done for plates with mixed colonies to obtain pure colonies (Figure 5). This was done by using a sterile wire loop to streak each separate colony onto new plates; MacConkey agar/ nutrient agar for bacteria and PDA/ SDA for fungi. Pure bacterial colonies were then incubated at 37℃ for 24 hours while pure fungal isolates were incubated at 30℃ for 5 days (Omoya and Akinyosoye, 2013).

Figure 5: Pure cultures of bacteria (A = Nutrient agar, B = MacConkey agar) and fungi (C =PDA, D = SDA) isolated from R. differens

3.3.3 Microscopic identification and characterisation of bacterial and fungal isolates Pure bacterial isolates were classified by gram staining (Public Health England, 2019). For each isolate, a smear was prepared, fixed by heating gently and the slide flooded with 0.5% crystal violet dye. After 30 seconds, the slide was rinsed using water, flooded with 1% Lugol’s iodine and covered with fresh iodine. The smear was then decolorised using 95% ethanol, rinsed with water and flooded with 0.1% counterstain safranin. The slide was washed, blot dried and examined using oil immersion objective under a total magnification of 1000x (LEICA EZ4HD Stereo microscope, Leica microsystems Inc.). Pure cultures were later identified to species level using molecular techniques.

Fungal isolates were identified morphologically by direct microscopic mount (Public Health England, 2019), using Lactophenol cotton blue mount (LPCB). A drop of LPCB was placed on a microscope slide and a mounting needle was used to gently remove a portion of the pure colony and placed in the LPCB drop. The slide was then covered gently using a coverslip to make a thin mount while avoiding air bubbles. Excess LPCB stain was blotted off and the

14 slide examined under high power (x 400) with reduced light. Molecular techniques were later used for identification of fungal microbes to species level.

3.3.4 Molecular identification and characterisation 3.3.4.1 DNA extraction Pure bacterial isolates from individual colonies were aseptically inoculated into 7 ml luria broth in a sterile 50 ml falcon tube placed into an innova 44 incubator shaker (Eppendorf) set at 180 rpm and 37℃ for 18 hours. The cells were then harvested by centrifugation at 2,000 relative centrifugal force for 3 minutes into 1.5 ml Eppendorf tubes (Eppendorf, Hamburg, Germany). Bacterial DNA was then extracted using the isolate II Genomic DNA Kit (Bioline), following the manufactures protocol.

On the other hand, fungal colonies from pure cultures were harvested by scrapping mycelia/ spores using a sterile blade into 1.5 ml Eppendorf tubes each containing three sterile beads (Gatheru et al., 2019). Fungal DNA was then extracted following the plant DNA extraction kit (Bioline, London, UK). For individual batches, the DNA extracted (for bacteria and fungi) were separately quantified using a Nanodrop 2000/2000c spectrophotometer (Thermo Fischer Scientific, Wilmington, DE, USA) and the samples stored in a deep freezer at -20℃ until further downstream processing.

3.3.4.2 PCR-amplification Multiple copies of extracted DNA (for individual batches of bacteria and fungi) were produced through Polymerase Chain Reactions (PCR). For bacteria, the universal primers targeting bacterial small sub-unit rRNA genes from the V1 to the V2 hyper variable region (Rinttilä et al., 2004) i.e., 27F (5- ‘AGAGTTTGATCMTGGCTCAG-3’) and 1492R (5- ‘TACCTTGTTACGACGACTT-3’) respectively were used. The PCR reactions were carried out in a final 20 µl volume containing 5x MyTaq reaction buffer (Bioline) (5 mM dNTPs,

15mM MgCl2 stabilizers and enhancers), 0.5 pmol/µl of each primer, 0.25 mM MgCl2, 0.0625 U/µl MyTaq DNA polymerase (Bioline) and 15 ng/µl of DNA template. The PCR reactions were set up in a mastercycler Nexers thermal cycler (Eppendorf). The cycling conditions involved an initial denaturation step at 95℃ for 10 min, 35 cycles of denaturation step at 94℃ for 1 min, an annealing step at 54℃ for 45 sec and an extension step at 72℃ for 1 min, followed by a final extension at 72℃ for 10 min (Gusmão et al., 2007). The expected product size was approximately 1,500 base pairs; the amplified PCR products were resolved through

15 a 1.2% agarose gel. Bands of DNA on the gel (Figure 6A) were analyzed using KETA GL imaging system trans illuminator (Wealtec Corp, Meadowvale Way Sparks, NV, USA).

For fungi, PCR amplification was carried out using the universal fungal internal transcribed spacer primers: ITS5, (5’-TCCTCCGCTTGATATTGATATGC-3’)-forward primer and ITS4 (5’-GGAAGTAAAAGTCGTAACAAGG-3’)-reverse primer (White et al., 1990). All PCR reactions were carried out in a final 20 µl volume containing master mix (5x MyTaq reaction buffer (Bioline), 0.5 pmol/µl of each primer, 0.25 mM MgCl2, 0.0625 U/µl MyTaq DNA polymerase (Bioline)) and 15 ng/µl of DNA template. Polymerase chain reactions were set up in a mastercycler Nexers thermal cycler (Eppendorf). The cycling conditions were an initial denaturation step at 95℃ for 10 min, 35 cycles of denaturation step at 94℃ for 1 min, an annealing step at 54℃ for 1 min and an extension step at 72℃ for 1 min, followed by a final extension at 72℃ for 10 min (Gusmão et al., 2007). The PCR amplified products were resolved through a 1.2% agarose gel and DNA bands on the gel were analyzed using KETA GL imaging system trans illuminator (Wealtec Corp, Meadowvale Way Sparks, NV, USA) to confirm success (Figure 6B). The expected product size ranged between 450-600 base pairs. Successively amplified products of fungi and bacteria were sent to Macrogen Inc Europe Laboratory, Amsterdam, Netherlands, for bi-directional sequencing.

To identify and characterise bacterial and fungal contaminants in R. differens along the value chain, pure isolates were grouped according to their Gram reactions (for bacteria) and division (based on morphology for fungi). All sequences from pure isolates (19 for bacteria and 30 for fungi) were edited and submitted to the Genebank homology searches with Megablast searches (https://blast.ncbi.nlm.nih.gov/Blast.cgi) for identification. The highest hit percentage in the NCBI Megablast search result was considered the isolates most likely identity.

Figure 6: Bands of bacteria (A) and fungi (B) in Ruspolia differens viewed under KETA GL imaging system trans illuminator

3.4 Evaluating bacterial and fungal loads in Ruspolia differens at the different points along the value chain in Uganda The procedure described under sub-sections 3.3.1 and 3.3.2 were followed to determine loads (cfu/g) of bacteria and fungi in R. differens from Masaka, Kampala and at different points of the value chain. The standard plate counts of edible insects in Uganda ( maximum total bacterial and fungal counts) are 5.0 log cfu/g and 3.0 log cfu/g respectively (UNBS, 2019). These loads were used as a reference to determine the safety of R. differens at different points of the value chain. These standard plate counts are relatively comparable to those used for edible insects in Kenya (with the maximum total bacterial count = 5.0 log cfu/g and maximum fungal count = 2.0 log cfu/g) (KEBS, 2020).

3.5 Data analysis To test for any difference in the percentage occurrence of different microbes (bacterial/fungal species) at different points along the value chain, the non-parametric kruskal-Wallis multiple comparison test with Bonferroni P-values adjustment was performed. To evaluate bacterial and fungal safety levels in Ruspolia differens at the different points along the value chain in Uganda, data on bacterial and fungal loads expressed as log cfu/g were tested for normality by plotting normal distribution curves and determining the homogeneity of variance. Because the data were normally distributed, one-way analysis of variance with 95% confidence interval was performed to test for differences in microbial loads (cfu/g) at different sampling points and in whole R. differens samples (originating from Kabale, Fort portal and Kasese districts) marketed in Kampala. Means were separated using Bonferroni adjustment. A two- sample t-test was used to determine if bacterial and fungal counts were significantly different between samples from Kampala and Masaka. Line indicating standard maximum limit for the loads were added to the plots. All analyses were carried out in R-statistical computer software version 4.0.0 (R Development Core Team, 2019) at α = 0. 05.

CHAPTER FOUR: RESULTS 4.1 Identification and characterisation of bacteria and fungi in Ruspolia differens along the value chain in Uganda The highest percentage (68.4 %) of the bacteria isolated from R. differens was Gram positive and the least (31.6 %) was Gram-negative. From samples that were Gram-positive, 46.2 % were recorded in whole R. differens from the harvesting drums (at trapping points), 15.4 % were recorded each in whole insects from wild vegetation, markets and in plucked R. differens from markets. Deep fried ready-to-eat R. differens from markets recorded the least percentage (7.6 %) of Gram-positive bacteria (Table 1, 2 and 3). From samples that were Gram-negative, the highest percentage (50.0 %) was recorded in whole R. differens from markets followed by trapping points (33.3 %), plucked in markets (16.7 %) and none in wild collected and deep-fried insects from markets (Table 2). The fungal species isolated from R. differens samples at different sampling points along the value chain were only of two divisions, i.e., Basidiomycota (60%) and Ascomycota (40 %) (Table 1, 2 and 3).

All the 49 sequences of bacteria and fungi in/on R. differens samples matched for identity with sequences of the database had similarities of between 90% and 100% to the sequences in the GeneBank. Thirty of the sequences originated from pure fungal isolates while 19 came from pure bacterial isolates. Seven species each of bacteria and fungi were recorded at different sampling points along the value chain of R. differens in Uganda (Table 2).

There were variations recorded in the distribution of microbial species and strains at different sampling points along the value chain (Table 2; Appendix 1). The highest number of bacterial (05) and fungal (04) species were recorded in R. differens samples from collection drums (at trapping sites) and the least number (01 species each of bacteria and fungi) were recorded in deep fried ready-to-eat insects from the markets. Only one species of bacteria (Bacillus cereus with 6 strains) was recorded at all sampling points along the value chain. Staphylococcus sp (1 strain) and Bacillus thuringiensis (2 strains) were only recorded in R. differens samples from the collection drums, Acinetobacter baumannii from distribution points and Roseomonas sp (1 strain) in plucked R. differens samples sold at markets only.

The fungal species Trichoderma asperellum (with 9 strains) was recorded in R. differens samples from all sampling points, while Papiliotrema laurentii (1 strain) was recorded only in whole R. differens samples from the wild. The pathogenic bacterium B. cereus and the

19 fungus T. asperellum were recorded in deep fried ready-to-eat insects and in plucked samples sold at the markets. Most species of microbes recorded in R. differens samples from the trapping sites and distribution points or markets were not recorded in samples from the wild (Table 2).

Table 1: Gram stain tests of bacterial cultures from R. differens and their identities by molecular techniques. Sampling Point Season Bacterial classification by Identity by molecular Gram staining technique W-HdN1 1 + Staphylococcus sp. strain W-HdN2 1 + Staphylococcus sciuri W-HdB1 2 + Staphylococcus sciuri W-DpK1 2 + Staphylococcus sciuri P-MA1 2 + Staphylococcus sciuri W-HdB2 1 + Bacillus thuringiensis W-HdB3 1 + Bacillus thuringiensis W-SnA1 1 + Bacillus cereus W-SnA2 2 + Bacillus cereus W-DpA2 2 + Bacillus cereus W-HdC1 1 + Bacillus cereus P-MB1 1 + Bacillus cereus Df-MA1 2 + Bacillus cereus W-HdA2 2 - Serratia marcescens W-HdB2 2 - Serratia marcescens W-DpB1 2 - Serratia marcescens W-DpK3 2 - Serratia marcescens W-DpB2 2 - Acinetobacter baumannii P-MB2 1 - Roseomonas sp

(+) = Gram-positive; (-) = Gram-negative; (1 and 2) = First and Second swarming seasons respectively; W = Whole R. differens samples (HdN- from harvesting drums at Nyendo; HdB- from harvesting drums at Kayirikiti B; HdA- from harvesting drums at Kayirikiti A); W-Sn = Whole R. differens samples from the sweep net (A- from Kayirikiti); W-Dp = Whole R. differens samples from distribution points (K- Katwe, B- Busega) in Kampala; Df-A = Deep fried R. differens samples from markets (A- Masaka, B- Kampala) and P = plucked R. differens samples from (A- Masaka and B-Kampala) markets.

Table 2: Bacterial and fungal species isolated from Ruspolia differens in Uganda from different points along the value chain in 2019 ((✓) = Detected, () = Not detected). Contaminant microbe Whole insect Processed insect in the market

Wild Harvesting Markets Plucked Deep fried vegetation drums Bacteria A) Gram positive Bacillus cereus ✓ ✓ ✓ ✓ ✓ Bacillus thuringiensis  ✓    Staphylococcus sp. Strain  ✓    Staphylococcus sciuri  ✓    B) Gram negative Serratia marcescens  ✓ ✓   Acinetobacter baumannii   ✓   Roseomonas sp    ✓  Fungi A) Basidiomycota Rhodotorula ✓ ✓ ✓   muscilaginosa Rhodotorula dairenensis  ✓  ✓  Trichosporon asahii   ✓ ✓  Papiiliotrema laurentii ✓     B) Ascomycota Trichoderma asperellum ✓ ✓ ✓ ✓ ✓ Clavispora lusitaniae ✓ ✓ ✓ ✓  Pichia kudriavzevii   ✓ ✓ 

Table 3: Proportional occurrence (%) of microbes classified in R. differens along the value chain

Contaminant microbe Whole insect Processed insect in the market

Wild Harvesting Markets Plucked Deep fried vegetation drums Gram-positive bacteria 15.4 46.2 15.4 15.4 7.6 Gram-negative bacteria 0.0 33.3 50.0 16.7 0.0 Basidiomycota 25.0 25.0 25.0 25.0 0.0 Ascomycota 18.2 18.2 27.3 27.3 9.0

4.2 Bacterial and fungal loads (safety levels) in Ruspolia differens from different sampling points along the value chain in Uganda. The mean bacterial counts in R. differens samples varied significantly with sampling point in

both the first (F2,43 = 91027, p < 0.001) and second (F4,46 = 33954, p < 0.001) swarming seasons of 2019. The bacterial counts in deep fried R. differens were significantly lower than those in raw samples from all other sampling points, ranging from plucked samples in the first swarming season to whole samples at commercial distribution points during the second swarming season (Figure 7A). The bacterial counts in deep fried R. differens were also lower than the maximum standard limit recommended for edible insects in Uganda (Figure 7A). The bacterial counts in raw R. differens from all the sampling points along the value chain were however beyond the maximum standard limit for edible insects in Uganda (Figure 7A). Bacterial counts in whole R. differens samples from the market in the second swarming season were significantly higher than those in plucked samples in the market, and whole samples from trapping points (Figure 7A).

The mean fungal counts were also significantly different in R. differens from different

sampling points in the two swarming seasons of 2019 (F4,44 = 5973, p < 0.001 and F4,46 = 3509, p < 0.001 for first and second seasons respectively). The fungal counts in deep fried R. differens were significantly lower than those in raw samples from all sampling points, ranging from samples from collection drums in the first season to samples from the wild in the second season (Figure 7B). The fungal counts in deep fried R. differens were below the maximum standard limit for edible insects, while the fungal counts in the raw insects from all

23 the sampling points along the value chain were beyond the recommended maximum standard limit (Figure 7B). There were however no significant differences in fungal counts in raw R. differens samples from all sampling points in both swarming seasons of 2019 (Figure 7B).

Figure 7: Mean bacterial (A) and fungal (B) counts in Ruspolia differens from different sampling points in Uganda in the first and second swarming seasons of 2019. Error bars represent standard errors of the mean, different letters on columns within a season indicate significant differences between means and cfu indicates colony forming units. (—) and (…) indicate the maximum legally acceptable limits of bacterial (A) and fungal (B) counts.

Furthermore, the mean bacterial counts in R. differens samples from Masaka district in the first swarming season were significantly higher than those in samples from Kampala district (t = 7.3, p < 0.001: Figure 8A). However, there was no significant difference in bacterial counts in R. differens samples between the two districts in the second swarming season (t = 0.29, p > 0.05). On the other hand, the mean fungal counts in R. differens samples were not significantly different between Masaka and Kampala districts in both the first and second swarming seasons (t = 1.06, p = 0. 63 and t = 1.32, p > 0.05 respectively: Figure 8B).

Figure 8: Mean bacterial (A) and fungal (B) counts in Ruspolia differens samples from Masaka and Kampala in the first and second swarming seasons of 2019. Error bars represent standard errors of the mean, different letters on columns within a season indicate significant differences between means and cfu indicates colony forming units.

4.2.1. Microbial loads in R. differens samples from other districts sold at the distribution points in Kampala district Samples of whole R. differens from the distribution points (Busega and Katwe) in Kampala were found to originate from Fort-portal, Kasese and Kabale districts. There was no significant difference in bacterial counts in whole R. differens samples from the distribution points based on the district of origin (F2,6 = 1.81, p > 0.05: Figure 9A). Similarly, there was no significant difference in fungal counts in whole R. differens samples from the distribution points based on their district of origin (F2,6 = 0.63, p > 0.05: Figure 9B). Although the loads of bacteria and fungi did not vary by district of origin, the loads were all significantly above safe limits by prescribed standard.

Figure 9: Mean bacterial (A) and fungal (B) counts in Ruspolia differens from distribution points in Kampala originating from Fort Portal, Kabale and Kasese districts in the second swarming season of 2019. Error bars represent standard errors of the mean, similar letters on columns indicate no significant differences between means and cfu indicates colony forming units. (—) and (…) indicate the maximum legally acceptable limits of bacterial (A) and fungal (B) counts.

CHAPTER FIVE: DISCUSSION 5.1 Species of bacteria and fungi recorded in/on Ruspolia differens in Uganda Ruspolia differens from different sampling points along the value chain were found to harbour seven species of bacteria and seven species of fungi. The highest number of microbe species was recorded in whole R. differens samples from trapping points and the least number was in the deep-fried (ready-to-eat) R. differens from markets. These findings partly corroborates with the report by Ssepuuya et al. (2019) which indicated that raw R. differens harbours different genera of pathogenic microbes for example Bacillus, Acinetobacter and Staphylococcus among others. Previous studies of other edible insects for example, Tenebrio molitor larvae and grasshopper Locusta migratoria (fresh and processed - ready-to-eat) also reported the same genera of bacteria and fungi (Stoops et al., 2016; Garofalo et al., 2017).

The microbe species (bacteria and fungi) recorded in R. differens could be attributed to the high nutrient and moisture content (Kinyuru et al., 2010), which provides a conducive atmosphere for microbial contamination, growth and survival (Klunder et al., 2012). Also, some of the microbe species recorded for example Bacillus cereus are reportedly associated with many insect taxa, aiding their survival by synthesizing essential nutrients (Khanna et al., 2013). The high number of microbe species recorded in raw R. differens samples from the trapping points than other points along the value chain could be attributed to unhygienic practices during trapping. For example, smearing drum walls and iron sheets with moist cassava flour to prevent escape of the catch (Okia et al., 2017), and contamination through contact with trapped non-target species (Sengendo et al., 2021). Cassava flour processed under unhygienic conditions could be contaminated with microbes including those that produce mycotoxins (Kaaya and Eboku, 2010; Omara et al., 2020).

Furthermore, some pathogenic bacteria such as Staphylococcus spp and Bacillus spp have been recorded in dry cassava flour (Ogori and Gana, 2013). Therefore, application of unhygienic moist cassava flour into R. differens collection drums may be an important entry point for the microbes. For example, pathogenic bacteria (Staphylococcus sp and B. thuringiensis) and the fungus (Rhodotorula dairenensis) were only recorded in whole R. differens samples from the collection drums but not in samples from the wild. Also, Serratia marcescens–an opportunistic pathogen (Khanna et al., 2013), could have probably gained entry into R. differens through injuries sustained when insects struggle to escape from collection drums or from manual handling by humans. The study further recorded a marked

27 variation in the number and distribution of bacterial and fungal species and/strains at different sampling points along the value chain of R. differens. This could possibly be attributed to adaptive dynamic shifts in response to changes in niche/community composition (Levy and Borenstein, 2013).

The bacterium B. cereus was recorded at all R. differens sampling points and is reportedly associated with many insect taxa (Khanna et al., 2013). Ruspolia differens trapped with other by-catches could have been contaminated by insects containing this microbe and possibly upon contact with soil. The presence of B. cereus in deep fried ready-to-eat R. differens may be attributed to their ubiquitous nature (Podschun et al., 2001) coupled with unhygienic practices which predispose the insects to such contaminants. Ssepuuya et al. (2016) reported that R. differens vended in transparent polythene bags/plastic buckets were often left open to attract buyers, which exposes the insects to environmental contaminants. Klunder et al. (2012) further reported that some Bacillus spp are resistant to heat treatment and hence could be found in ready-to-eat-foods (for example deep-fried R. differens), causing food spoilage or health problems to consumers. Bacillus species among other pathogenic microorganisms have also been recorded in other edible insects (Banjo et al., 2006).

Acinetobacter baumannii and T. asahii were recorded only in samples from distribution points but not in samples from the wild and collection drums suggesting that they could have been picked up during transportation. Also, Roseomonas sp was recorded only in plucked samples suggesting that it could have been introduced during plucking, possibly upon contact with contaminated soil/water. Sandoe and Malnick, (1997) reported the presence of Roseomonas sp in soil/water and so contamination of R. differens is possible upon contact during the plucking process. The risk of food contamination with pathogens increases with long chain of food handlers (Woh et al., 2017). Also, the pathogenic bacteria S. marcescens and Staphylococcus spp were detected in R. differens samples from all sampling points except in wild-collected insects. This could be attributed to contamination arising from handling by different people (some of whom may not practice proper personal hygiene) at the different points of the food handling chain (Banjo et al., 2006).

The study further revealed that only one fungal species (P. laurentii) was isolated exclusively from wild-caught R. differens. Papiliotrema laurentii is ubiquitous and prevalent in the soil rhizosphere, where it assimilates a wide range of hexoses, pentoses and organic acids

(Leguina et al., 2019). Other microbes introduced at trapping points and along the subsequent points of the value chain could have caused competition and hence extinction of P. laurentii.

5.2 Bacterial and fungal safety levels in/on Ruspolia differens in Uganda There were varying levels of bacterial and fungal loads recorded at the different sampling points and hence differing levels of safety concerns. The bacterial counts along the value chain were significantly higher in whole R. differens samples from distribution points than in plucked samples in the markets, and whole samples from collection drums (at trapping points) and the wild.

Potential sources of microbial contamination of R. differens at the distribution points are market environment, confinement in bags and human handling. Based on field observations, hygiene at the distribution points was generally poor as compared to other points of the value chain. The distribution points had limited space and were littered with garbage which was left to accumulate before being removed by the concerned authorities. Similar observations including poor personal hygiene among food handlers along highway markets in Uganda were also reported by Bagumire and Karumuna (2017). Oranusi and Braide (2012) and Musundire et al. (2016) reported that primitive and unhygienic methods of handling, packaging and storage of edible insects cause re-contamination. The significant decline in bacterial counts in raw R. differens samples after plucking in the market could be attributed to extraneous contamination from the environments and handlers, which could most likely settle on appendages which are easily removed during plucking.

Unlike bacteria, the fungal counts in raw R. differens samples at different sampling points along the value chain were not statistically different. Fungal growth turn-over rates take hundreds of days unlike bacteria with turn-over rates of hours to few days (Rousk and Bååth, 2011). Therefore, it is unlikely for the fungal populations to change drastically from the time R. differens is moved from the wild to the traps, to the time they are plucked in the market, which takes between 1-2 days. The use of relative abundance of colony forming units along the value chain of R. differens may therefore not suffice in determining high risk points for fungal contaminants.

The bacterial and fungal counts in deep fried ready-to-eat R. differens were 3- and 2-fold lower, respectively, than in raw insects at all sampling points. This could be attributed to the

29 efficiency of deep frying in reducing microbial loads in edible insects (Gatheru et al., 2019). The deep fried R. differens had lower bacterial and fungal counts than the standard maximum limits recommended for edible insects in Uganda and Kenya ( UNBS, 2019; KEBS, 2020). Deep fried ready-to-eat R. differens sold in Ugandan markets is therefore safe for human consumption provided that they are handled in ways that do not expose them to re- contamination with microbes.

On the other hand, the bacterial and fungal counts in raw R. differens were higher than the maximum standard limits recommended for edible insects suggesting that the consumption of raw R. differens as practiced by some people in Uganda (Agea et al., 2008, P. Akite, pers. comm) is unsafe. This could be attributed to the natural interactions of some microbes with insects (Khanna et al., 2013), which are destroyed through processing by deep-frying. A good example is the case of the tachinid fly larvae that were found in grasshopper samples from Busega market by Dr. Panta Kasoma (P. Akite, pers. Comm). These flies are hairy and if they are found in areas with poor sanitary conditions, they are likely to carry certain pathogens into their host insect species. This would indirectly be transmitted to the humans who may consume these insects raw as is the case reported for the long-horned grasshopper. Also, the high nutrient and moisture content in raw R. differens (Kinyuru et al., 2010) provides a conducive atmosphere for microbial infestation, growth and survival (Klunder et al., 2012). Other studies for example by Kaya and Vega (2012) and Klunder et al. (2012) reported wild harvested edible insects to be highly contaminated with microorganisms as compared to those farmed in controlled environments.

The study findings further revealed that there was no detectable difference in fungal counts in R. differens collected from Masaka and Kampala during both swarming seasons. However, the bacterial counts in R. differens samples from Masaka were significantly higher than in those from Kampala during the first swarming season only. The first rainy season in Uganda is normally wetter than the second season (Nsubuga et al., 2014), which may account for the different results in the two swarming seasons. This finding partly corroborates the report by Bagumire and Karumuna (2017) that microbial loads in ready-to-eat meat sold along high ways in Uganda during the wet season were higher than in the dry season. The specific reasons for the higher bacterial counts in R. differens samples from Masaka than those from Kampala in the first swarming season remain unknown.

5.2.1. Microbial loads in R. differens samples from other districts sold at the distribution points in Kampala district In addition, the study further established that bacterial and fungal counts in whole R. differens samples at distribution points in Kampala were not influenced by their districts of origin. Ssepuuya et al. (2019) reported that R. differens traders in Uganda normally buy and package freshly harvested insects in bags and transport them by public means to distribution points. The similarity in packaging, storage and transportation practices across traders may account for the observed similar level of bacterial and fungal contamination of raw whole R. differens, irrespective of the district of origin. Charpe et al. (2019) also reported that microorganisms have a specific range of extrinsic conditions under which they interact with intrinsic factors favoring microbial growth and survival.

5.3 Health implications of the pathogenic bacteria and fungi detected in R. differens. The detection of pathogenic microorganisms in edible R. differens shows that their use as food (especially when raw) poses health risks to humans and other animals. The pathogenic bacteria such as Staphylococcus sciuri, Bacillus cereus, Serratia marcescens and Acinetobacter baumannii are liable to cause various infectious diseases in humans. Acinetobacter baumannii for example is reported to have evolved to become a highly troublesome pathogen in hospitals worldwide, causing a range of infectious syndromes, including ventilator-associated pneumonia (VAP) and catheter-related bloodstream infection (Peleg et al., 2015).

Also, mycotoxicoses and mycoses are the main ways in which human health is affected due to infection with fungi and contamination with mycotoxins (Atanda et al., 2013). Mycotoxicoses can enter the blood stream, lymphatic system and inhibit protein synthesis, damage macrophage systems and increase sensitivity to opportunistic infections (Atanda et al., 2013). Symptom appearance of pathogenic microbes depends on the level of contamination, duration of exposure, type of microbe(s), degree of combinations, individual difference and the physiological status of the victim (Peleg et al., 2015). Some microbes have the potential for both acute and chronic health effects through ingestion, skin contact and inhalation.

CHAPTER SIX: CONCLUSIONS AND RECOMMENDATIONS 6.1 Conclusions i. Ruspolia differens obtained along the value chain in Uganda harbours seven species each of bacteria and fungi and microbe species are more diverse at trapping points.

ii. The pathogenic bacteria S. sciuri, A. baumannii, and S. marcescens recorded in traded R. differens but not in wild-caught whole insects indicate that there is post-harvest contamination along the value chain. Distribution points and trapping points are the observed hotspots for microbial contamination of R. differens. iii. Bacterial and fungal counts in deep-fried ready-to-eat R. differens are ~3- and 2-fold lower, respectively, than in raw R. differens. The counts of these microbes in deep- fried R. differens are within the recommended food safety limits, thus safe for human consumption. Also, bacterial and fungal counts in R. differens samples at the distribution points in Kampala are not influenced by their districts of origin.

6.2 Recommendations 1. There’s need for further investigation of specific handling practices at the distribution and trapping points which cause contamination of R. differens.

2. Also, elucidation of the possible cause(s) of higher bacterial counts in R. differens samples from Masaka than those from Kampala in the first season only should be carried out.

3. The general public should be sensitized about the dangers of eating raw R. differens. This should involve having training programs for personnel at local levels to inspect and monitor hygiene and to initiate sensitization programs at various levels along R. differens value chain.

4. Although bacterial and fungal counts in the deep fried ready-to-eat R. differens were within the standard recommended limits for edible insects, the presence of the bacterial pathogen B. cereus is a safety concern that needs to be addressed. Also, the vending practices of these ready-to-eat insects exposes them to other risks of environmental contaminants and as such, it’s imperative for harvesters and all other

dealers along the value chain of R. differens to observe good harvesting and post- harvest handling practices that eliminate re-contamination.

5. Further research on materials smeared in drum walls to prevent escape of the catch and for economically feasible innovations to store ready-to-eat insects in vending places taking into consideration their shelf life should be undertaken. Also, more research is required to examine the effect of other factors like packaging procedure, duration of sale and hygienic practices of handlers on microbial loads in deep fried R. differens.

6. The study findings also highlight a need for standards to regulate hygienic measures along the handling chain of R. differens especially at commercial distribution points and trapping points. Additionally, sanitary conditions around marketing areas should be improved to keep away potential parasitoids.

REFERENCES Adwan, G. and Salama, Y. (2016). Microbial contamination of environmental surfaces An- Najah National University Setting. Journal of Scientific Research and Reports. 10(2): 1– 9. https://doi.org/10.9734/JSRR/2016/23098 Agea, J.G., Biryomumaisho, D., Buyinza, M. and Nsubuga, G.N. (2008). Commercialization of Ruspolia ntidula (Nsenene grasshoppers) in central Uganda. African Journal of Food Agriculture Nutrition and Development. 8: 319–332. Atanda, O., Makun, A. H., Ogara, M. I., Edema, M., Idahor, O. K. and Eshiett, M. Oluwambamiwo, F. (2013). Fungal and mycotoxin contamination of Nigerian foods and feeds: Mycotoxins and food safety in developing countries. INTECH, Nigeria. https://doi.org/10.5772/55664 Ayieko, M.A., Kinyuru, J.N., Ndong’a, M.F. and Kenji, G.M. (2012). Nutritional value and consumption of black ants (Carebara vidua Smith) from the lake Victoria region in Kenya. Advanced Journal of Food Science and Technology. 4: 39–45. Bagumire, A. and Karumuna, R. (2017). Bacterial contamination of ready-to-eat meats vended in highway markets in Uganda. African Journal of Food Science. 6: 160–170. https://doi.org/10.5897/AJFS2016.1550 Banjo, A.D., Lawal, O.A. and Songonuga, E.A. (2006). The nutritional value of fourteen species of edible insects in southwestern Nigeria. African Journal of Biotechnology. 5: 298–301. https://doi.org/10.5897/AJB05.250 Belluco, S., Losasso, C., Maggioletti, M., Alonzi, C.C., Paoletti, M.G. and Ricci, A. (2013). Edible insects in a food safety and nutritional perspective: A critical review. Comprehensive Reviews in Food Science and Food Safety 12: 296–313. https://doi.org/10.1111/1541-4337.12014 Bokulich, N.A., Lewis, Z.T., Boundy-Mills, K. and Mills, D.A. (2016). A new perspective on microbial landscapes within food production. Current Opinion in Botechnology. 37:182–189. https://doi.org/10.1016/j.copbio.2015.12.008 Chakravorty, J., Ghosh, S. and Meyer-Rochow, V. B. (2011). Practices of entomophagy and entomotherapy by members of the Nyishi and Galo tribes, two ethnic groups of the state of Arunachal Pradesh (north-east India). Journal of Ethnobiology and Ethnomedicine, 7(5): 1–14. https://doi.org/10.1186/1746-4269-7-5 Charpe, A.M., Sedani, S.R., Murumkar, R.P. and Bhad, R.G. (2019). Effect of temperature on microbial growth in food during storage. Multilogic in Science. 8: 56–58. Colman, D.R., Toolson, E.C. and Takacs-Vesbach, C. (2012). Do diet and taxonomy

influence insect gut bacterial communities? Molecular Ecology. 20: 5124–5137 Doi.10.1111/j.1365-294X.2012. 05752.x Cummings, B. (2007). Pour plate method. Preparing serial dilutions. Pearson education Inc publishers. https://www.google.com/url?sa=i&url=https%3A%2F%2Fmicrobeonline.com%2Fpour- plate-method-principle-procedure-uses Dobermann, D., Swift, J.A. and Field, L.M. (2017). Opportunities and hurdles of edible insects for food and feed. Nutrition Bulletin. 42: 293–308. https://doi.org/10.1111/nbu.12291 Dos Santos, T.T., De Oliveira, D.P., Cabette, H.S.R. and De Morais, P.B. (2019). The digestive tract of Phylloicus (Trichoptera: Calamoceratidae) harbours different yeast taxa in Cerrado streams, Brazil. Symbiosis 77: 147–160. https://doi.org/10.1007/s13199- 018-0577-9 Durst, P.B., Johnson, D.V., Leslie, R.N.and Shono, K. (2010). Edible forest insects: humans bite back. FAO Public, Bangkok, Thailand 978-92-5-106488-7.EFSA. (2015). Risk profile related to production and consumption of insects as food and feed EFSA Scientific Committee. EFSA Journal, 13(10): 4257. https://doi.org/10.2903/j.efsa.2015.4257 EFSA. (2015). Risk profile related to production and consumption of insects as food and feed EFSA Scientific Committee. EFSA Journal, 13(10): 4257. https://doi.org/10.2903/j.efsa.2015.4257 Engels, P. and Moran, N.A. (2013). The gut microbiota of insects- diversity in structure and function. Federation of European Microbiology Society Reviews. 37: 699–735. FAO/WHO. (2002). Codex alimentarius commission. Joint FAO/WHO food standards programme codex coordinating committee for the near east: Second session. http://www.fao.org/fileadmin/user_upload/gmfp/docs/Codex%20working%20paper%20 on%20elaboration%20of%20a%20regional%20standard%20for%20microbiological%20 levels%20in%20foodstuffs%20(CX-NEA%2003-16)%201.pdf Gallo, M. (2018). Novel Foods : Insects - Safety Issues. In Encyclopedia of food security and sustainability. Elsevier. https://doi.org/10.1016/B978-0-12-812687-5.22134-5 Garofalo, C., Milanović, V., Cardinali, F., Aquilanti, L., Clementi, F. and Osimani, A. (2019). Current knowledge on the microbiota of edible insects intended for human consumption: A state-of-the-art review. Food Research International. 9: 1-89. https://doi.org/10.1016/j.foodres.2019.108527 108527

Garofalo, C., Osimani, A., Milanović, V., Taccari, M., Cardinali, F., Aquilanti, L., Riolo, P., Ruschioni, S., Isidoro, N. and Clementi, F. (2017). The microbiota of marketed processed edible insects as revealed by high-throughput sequencing. Food Microbiology. 62: 15–22. Gatheru, J.W., Khamis, F.M., Ombura, F.L.O., Nonoh, J., Tanga, C.M., Maina, J., Mohamed, S.A., Subramanian, S., Ekesi, S. and Fiaboe, K.K.M. (2019). Impact of processing methods on microbial load of reared and wild-caught edible crickets (Scapsipedus icipe and Gryllus bimaculatus) in Kenya. Journal of Insects as Food and Feed. 5: 171–183. Giovanni, B. L. (2005). Chapter 9: Minilivestock consumption in the ancient near east: The case of locusts. Science Publishers, Inc. Enfield, NH, USA Printed in India. Gusmão, D.S., Santos, V.A., Marini, D.C., De Souza, E.R., Peixoto, A.M., Bacci, M.J., Berbert-molina, M.A. and Lemos, A.J.F. (2007). First isolation of microorganisms from the gut diverticulum of Aedes aegypti ( Diptera : Culicidae ): new perspectives for an insect-bacteria association. Memorias do Instituto Oswaldo Cruz. 102: 919–924. Jang, Y. (2011). Male responses to conspecific advertisement signals in the field cricket Gryllus rubens (Orthoptera : Gryllidae). PLoS ONE, 6(1): 1–8. https://doi.org/10.1371/journal.pone.0016063 Jongema, Y. (2015). List of edible insects of the world. Wageningen University of Research, Wageningen, the Netherlands. Available at: http://tinyurl.com/zb9u3b9. Kaaya, A.N. and Eboku, D. (2010). Mould and aflatoxin contamination of dried cassava chips in Eastern Uganda: Association with traditional processing and storage practices. Journal of Biological Sciences. 10: 718–729 KEBS, (2020). Edible insect's products—Specification DKS 2922-1:2020: ICS 67.120; Kenya Bureau of Standards: Nairobi, Kenya. Kelemu, S., Niassy, S., Torto, B., Fiabo, K., Affogon, H., Tonnang, H., Maniania, N. K. and Ekesi, S. (2015). African edible insects for food and feed : inventory , diversity, commonalities and contribution to food security. Journal of Insects as Food and Feed. 1: 103–119. https://doi.org/10.3920/JIFF2014.0016 Khanna, A., Khanna, M. and Aggarwal, A. (2013). Serratia marcescens - A rare opportunistic nosocomial pathogen and measures to limit its spread in hospitalized patients. Journal of Clinical and Diagnostic Research. 7: 243–246. https://doi.org/10.7860/JCDR/2013/5010.2737 Kinyuru, J.N., Kenji, G., Muhoh, S.N. and Ayieko, M. (2010). Nutritional potential of longhorn grasshopper (Ruspolia differens) consumed in Siaya district , Kenya. Journal

of Agriculture, Science and Technology. 12: 32–46. Klunder, H. C., Wolkers-rooijackers, J., Korpela, J. M. and Nout, M.J.R. (2012). Microbiological aspects of processing and storage of edible insects. Food Control. 26: 628–631. https://doi.org/10.1016/j.foodcont.2012.02.013 Leguina, A.C.D.V., Barrios, A.C., Soro, M.D.M.R., Lacosegliaz, M.J., Pajot, H.F., de Figueroa, L.I.C. and Nieto-Peñalver, C.G. (2019). Copper alters the physiology of tomato rhizospheric isolates of Papiliotrema laurentii. Scientia Horticulturae. 243: 376- 384. Leonard, A., Khamis, F.M., Egonyu, J.P., Kyamanywa, S., Ekesi, S., Tanga, C.M., Copeland, R.S. and Subramanian, S. (2020). Identification of Edible Short-and Long-Horned Grasshoppers and Their Host Plants in East Africa. Journal of Economic Entomology. https://doi: 10.1093/jee/toaa166 Levy, R. and Borenstein, E. (2013). Metabolic modeling of species interaction in the human microbiome elucidates community-level assembly rules. Proceedings of the National Academy of Sciences 110: 12804–12809. Masaka District Local Government. (2020). The official website of Masaka district. https://masaka.go.ug/content/geographical-features Matojo, N. D. and Hosea, K. M. (2013). Phylogenetic relationship of the longhorn grasshopper Ruspolia differens Serville (Orthoptera: Tettigoniidae) from northwest Tanzania based on 18S ribosomal nuclear sequences. Journal of Insects. 2013: 1–5. https://doi.org/10.1155/2013/504285 Matojo, N. D. and Njau, M. A. (2010). Plasticity and biosystematics of swarming of the conehead Ruspolia differens serville (Orthoptera: Conocephalidae). International Journal of Integrative Biology. 9(2): 97–103. Megido, C. R., Desmedt, S., Blecker, C., Drugmand, D., Haubruge, É. Y., Alabi, T. and Francis, F. (2017). Microbiological Load of Edible Insects Found in Belgium. Journal of Insects. 8(12): 1–8. Doi:10.3390/insects8010012 Megido, C. R., Sablon, L., Geuens, M., Brostaux, Y., Alabi, T., Blecker, C., Drugmand, D., Haubruge, É. and Francis, F. (2014). Edible insects acceptance by belgian consumers: Promising attitude for entomophagy development. Journal of Sensory Studies, 29, 14– 20. https://doi.org/10.1111/joss.12077 Mmari, M.W., Kinyuru, J.N., Laswai, H.S. and Okoth, J.K. (2017). Traditions, beliefs and indigenous technologies in connection with the edible longhorn grasshopper Ruspolia differens (Serville 1838) in Tanzania. Journal of Ethnobiology and Ethnomedicine. 13:

1–11. https://doi.org/10.1186/s13002-017-0191-6 Musundire, R., Osuga, I.M., Cheseto, X., Irungu, J. and Torto, B. (2016). Aflatoxin contamination detected in nutrient and anti-oxidant rich edible stink bug stored in recycled grain containers. PLoS ONE 11:1–16. https://doi.org/10.1371/journal.pone.0145914 NEMA. (2014). National state of the environment report for Uganda 2014: harnessing our environment as infrastructure for sustainable livelihood and development. NEMA, Kampala, Uganda. http://www.nemaug.org Ng’ang’a, J., Imathiu, S., Fombong, F., Ayieko, M., Vanden Broeck, J. and Kinyuru, J. (2019). Microbial quality of edible grasshoppers Ruspolia differens ( Orthoptera : Tettigoniidae ): From wild harvesting to fork in the Kagera Region , Tanzania. Journal of Food Safety. 39: 1–6. https://doi.org/10.1111/jfs.12549 Nsubuga, F.W.N., Botai, O.J., Olwoch, J.M., Rautenbach, C.J., Bevis, Y. and Adetunji, A.O. (2014). The nature of rainfall in the main drainage sub-basins of Uganda. Hydrological Sciences Journal. 59: 278–299. https://doi.org/10.1080/02626667.2013.804188 Nyangena, N. D., Mutungi, C., Imathiu, S., Kinyuru, J., Affognon, H., Ekesi, S., Nakimbugwe, D. and Komi K. M. F. (2020). Effects of traditional processing techniques on the nutritional and microbiological quality of four edible insect species used for food and feed in East Africa. Journal of Foods. 574: 1-16. doi:10.3390/foods9050574 Ogori, A. and Gana, J. (2013). Microbiological loads of road side dried cassava flour from cassava balls and chunks. American Open Journal of Agricultural Research. 1:24–39. http://acascipub.com/Journals.php Okia, C.A., Odongo, W., Nzabamwita, P., Ndimubandi, J., Nalika, N. and Nyeko, P. (2017). Local knowledge and practices on use and management of edible insects in Lake Victoria basin , East Africa. Journal of Insects as Food and Feed. 3: 83–93. https://doi.org/10.3920/JIFF2016.0051 Omara, T., Nassazi, W., Omute, T., Awath, A., Laker, F., Kalukusu, R., Musau, B., Nakabuye, B.V., Kagoya, S., Otim, G. and Adupa, E. (2020). Aflatoxins in Uganda: An Encyclopedic Review of the Etiology, Epidemiology, Detection, Quantification, Exposure Assessment, Reduction, and Control. International Journal of Microbiology. 2020: 1-18. https://doi.org/10.1155/2020/4723612 Omoya, F. O. and Akinyosoye, F. (2013). Evaluation of the potency of some entomopathogenic bacteria isolated from insect cadavers on Anopheles arabiensis Giles (Order: Dipthera; Family: Culicidae) mosquito larvae in Nigeria. African Journal of

Microbiology Research. 41: 4877–4881. https://doi.org/10.5897/ajmr12.1926 Osimani, A., Garofalo, C., Milanović, V., Taccari, M., Cardinali, F., Aquilanti, L., Pasquini, M., Mozzon, M., Raffaelli, N., Ruschioni, S. and Riolo, P. (2017). Insight into the proximate composition and microbial diversity of edible insects marketed in the European Union. European Food Research and Technology. 43:1157-1171. Oranusi, S. and Braide, W. (2012). A study of microbial safety of ready-to-eat foods vended on highways: Onitsha-Owerri, south east Nigeria. International Research Journal of Microbiology. 3: 66–71. Peleg, Y. A., Hogan, D. and Mylonakis, E. (2015). Medically important bacterial – fungal interactions. Nature Reviews Microbiology. 8: 340–350. https://doi.org/10.1038/nrmicro2313 Podschun, R., Pietsch, S., Höller, C. and Ullmann, U. (2001). Incidence of Klebsiella species in surface waters and their expression of virulence factors. Applied and Environmental Microbiology. 67: 3325–3327 https://doi.org/10.1128/AEM.67.7.3325-3327.2001 Public Health England. (2019). Staining procedures. UK Standards for Microbiology Investigations. TP 39 Issue 3. https://www.gov.uk/uk-standards-for-microbiology- investigations-smi-quality-and-consistency-in-clinical-laboratories R Core Team. (2019). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/. Rinttilä, T., Kassinen, A., Malinen, E., Krogius, L. and Palva, A. (2004). Development of an extensive set of 16S rDNA-targeted primers for quantification of pathogenic and indigenous bacteria in faecal samples by real-time PCR. Journal of Applied Microbiology, 97, 1166–1177. https://doi.org/10.1111/j.1365-2672.2004.02409.x Rousk, J. and Bååth, E. (2011). Growth of saprotrophic fungi and bacteria in soil. FEMS Microbiology Ecology. 78: 17-30 https://doi.org/10.1111/j.1574-6941.2011.01106.x Rumpold, B. and Schlüter, O. (2013). Potential and challenges of insects as an innovative source for food and feed production. Innovative Food Science and Emerging Technologies, 17, 1–11. https://doi.org/10.1016/j.ifset.2012.11.005 Sandoe, J.A.T. and Malnick, H. (1997). A case of peritonitis caused by Roseomonas gilardii in a patient undergoing continuous ambulatory peritoneal dialysis. Journal of Clinical Microbiology. 35: 2150–2152. Schabel, H.G. (2008). Forests insects as food: a global review. In: Durst, B.P., Johnson D.V., Leslie R.N., Shono K, editors. Forest insects as food: humans bite back. Proceedings of a workshop on Asia-Pacific resources and their potential for development. RAP

Publications, Chiang Mai, Thailand, pp 1-4. Sengendo, F., Subramanian, S., Chemurot, M., & Tanga, C. M. (2021). Efficient harvesting of safe edible grasshoppers: Evaluation of modified drums and light-emitting diode bulbs for harvesting Ruspolia differens (Orthoptera: Tettigoniidae) in Uganda. Journal of Economic Entomology, 20(20), 1–8. https://doi.org/10.1093/jee/toab025 Siulapwa, N., Mwambungu, A., Lungu, E. and Sichilima, W. (2014). Nutritional value of four common edible insects in Zambia. International Journal of Science and Research, 3(6): 876–884. Ssepuuya, G., Aringo, R.O., Mukisa, I.M. and Nakimbugwe, D. (2016). Effect of processing, packaging and storage-temperature based hurdles on the shelf stability of sautéed ready- to-eat Ruspolia nitidula. Journal of Insects as Food and Feed. 2: 245–253. https://doi.org/10.3920/JIFF2016.0006 Ssepuuya, G., Wynants, E., Verreth, C., Crauwels, S., Lievens, B., Claes, J., Nakimbugwe, D. and Van Campenhout, L. (2019). Microbial characterisation of the edible grasshopper Ruspolia differens in raw condition after wild-harvesting in Uganda. Food Microbiology. 77: 106–117. https://doi.org/10.1016/j.fm.2018.09.005 Ssepuya, G., Mukisa, I.M. and Nakimbugwe, D. (2017). Nutritional composition, quality, and shelf stability of processed Ruspolia nitidula (edible grasshoppers). Food Science and Nutrition. 5: 103–112. https://doi.org/10.1002/fsn3.369 Stoops, J., Crauwels, S., Waud, M., Claes, J. and Lievens, B. (2016). Microbial community assessment of mealworm larvae (Tenebrio molitor) and grasshoppers (Locusta migratoria migratorioides) sold for human consumption. Food Microbiology. 53: 122– 127. Thompson, J.D., Gibson, T.J., Plewniak, F., Jeanmougin, F. and Higgins, D.G. (1997). The CLUSTAL X windows interface: Flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Research. 25: 4876–4882. https://doi.org/10.1093/nar/25.24.4876 Uganda Bureau of Statistics. (2014). National population and housing census 2014 - Main report. In Uganda bureau of statistics. https://doi.org/10.1017/CBO9781107415324.004 UNBS. (2019). Draft Uganda Standards: Edible insect's products—Specification DUS 2146:2019; Uganda Bureau of Standards: Kampala, Uganda. UNBS. (2019). Draft Uganda Standard: Minced meat-specification. https://members.wto.org/crnattachments/2019/SPS/UGA/19_1687_00_e.pdf Van Huis, A., Van Itterbeeck, J., Klunder, H., Mertens, E., Halloran, A., Muir, G. and

Vantomme, P. (2013). Edible insects: future prospects for food and feed security. FAO Forestry Paper 171. FAO, Rome, Italy. Vandeweyer, D., Crauwels, S., Lievens, B., and Van Campenhout, L. (2017). Metagenetic analysis of the bacterial communities of edible insects from diverse production cycles at industrial rearing companies. International Journal of Food Microbiology. 2017: 11-18. White, T.J., Bruns, T., Lee, S. and Taylor, J. (1990). Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In: Innis, M.A., D.H. Gelfand, J.J. Sninsky, and T.J. White (eds.) PCR Protocols: A Guide to Methods and Applications. Academic Press, New York, USA, pp. 315-322. https://doi.org/10.1016/b978-0-12- 372180-8.50042-1 World Health Organization (WHO). (2018). Food security and nutrition in the world, building climate resilience for food security and nutrition. FAO., Rome, Italy. Woh, P.Y., Thong, K.L., Lim, L.A., Behnke, J.M., Lewis, J.W. and Zain, S.N.M. (2017). Microorganisms as an indicator of hygiene status among migrant food handlers in Peninsular Malaysia. Asian Pacific Journal of Public Health. 29: 599–607. https://doi.org/10.1177/1010539517735856

APPENDICES Appendix 1: Diversity of microbe species isolated from Ruspolia differens at different sampling points along the value chain in Uganda. Microbe Sampling point % GeneBank Sample identity ID Accession Wild Collection Markets Markets Markets Number vegetation drums (whole) (plucked) (Fried)

Bacteria Gram positive Staphylococcus sp. Strain - + - - - 94.7 MG251638.1 MT925937 Bacillus thuringiensis - + - - - 95.2 FJ236808.1 MT925938 Bacillus thuringiensis - + - - - 96.7 MF138122.1 MT925939 Staphylococcus sciuri - + - - - 94.6 MH921990.1 MT925933 Staphylococcus sciuri - + - - - 94.7 MH921990.1 MT925934 Staphylococcus sciuri - - + - - 92.1 KF312211.1 MT925935 Staphylococcus sciuri - - - + - 96.6 MT527534.1 MT925936 Bacillus cereus + - - - - 98.8 MK418365.1 MT925940 Bacillus cereus + - - - - 98.2 MK418365.1 MT925941 Bacillus cereus - + - - - 96.9 KY622853.1 MT925942 Bacillus cereus - - + - - 98.3 MN524141.1 MT925943

Bacillus cereus - - - + - 93.9 KM434374.1 MT925944

Microbe Sampling point % GeneBank Sample identity ID Accession Wild Collection Markets Markets Markets Number vegetation drums (whole) (plucked) (Fried)

Bacillus cereus - - - - + 91.1 MT078632.1 MT925945 Gram negative Serratia marcescens - + - - - 96.7 CP053286.1 MT925946 Serratia marcescens - + - - - 99.3 CP053286.1 MT925947 Serratia marcescens - - + - - 95.8 MN197739.1 MT925948 Serratia marcescens - - + - - 91.2 MF399283.1 MT925949 Acinetobacter baumannii - - + - - 99.0 MK840990.1 MT925950 Roseomonas sp - - - + - 97.6 AF533359.1 MT925932 Fungi

Basidiomycota Rhodotorula muscilaginosa + - - - - 98.6 MG211388.1 MT921073 Rhodotorula muscilaginosa - + - - - 95.7 MT465994.1 MT921074 Rhodotorula muscilaginosa - - + - - 99.4 MG969795.1 MT921075 Rhodotorula dairenensis - + - - - 98.2 MK267618.1 MT921055 Rhodotorula dairenensis - - - + - 98.3 MK267618.1 MT921056 Papiiliotrema laurentii + - - - - 96.5 MN660276.1 MT921057

Microbe Sampling point % GeneBank Sample identity ID Accession Wild Collection Markets Markets Markets Number vegetation drums (whole) (plucked) (Fried)

Trichosporon asahii - - + - - 96.1 KY985259.1 MT921058 Trichosporon asahii - - + - - 99.6 KY985259.1 MT921059 Trichosporon asahii - - - + - 98.5 KM822862.1 MT921060 Trichosporon asahii - - - + - 99.4 KM822862.1 MT921061 Trichosporon asahii - - + - - 98.4 MT482659.1 MT921062 Trichosporon asahii - - + - - 98.6 MT482659.1 MT921063 Ascomycota Clavispora lusitaniae + - - - - 97.8 MT180904.1 MT921064 Clavispora lusitaniae - + - - - 97.7 KP131842.1 MT921065

Clavispora lusitaniae - - + - - 99.4 CP039622.1 MT921066 Clavispora lusitaniae - - - + - 98.7 MN394875.1 MT921067 Clavispora lusitaniae - - + - - 96.2 LC413208.1 MT921068 Pichia kudriavzevii - - - + - 97.9 MT136539.1 MT921069 Pichia kudriavzevii - - - + - 99.1 MT136539.1 MT921070 Pichia kudriavzevii - - - + - 92.9 MT321167.1 MT921071 Pichia kudriavzevii - - + - - 93.9 KM396282.1 MT921072

Microbe Sampling point % GeneBank Sample identity ID Accession Wild Collection Markets Markets Markets Number vegetation drums (whole) (plucked) (Fried)

Trichoderma asperellum + - - - - 97.5 MH215547.1 MT921046 Trichoderma asperellum + - - - - 96.7 MH215547.1 MT921047 Trichoderma asperellum - + - - - 93.6 MH215547.1 MT921048 Trichoderma asperellum - + - - - 98.7 MH215547.1 MT921049 Trichoderma asperellum - + - - - 96.4 MH215547.1 MT921050 Trichoderma asperellum - - + - - 99.2 MH215547.1 MT921051 Trichoderma asperellum - - - + - 96.4 MH215547.1 MT921052 Trichoderma asperellum - - - - + 97.2 MH215547.1 MT921053 Trichoderma asperellum - - - + - 96.3 MH215547.1 MT921054

Similar names of microbe species with the same GeneBank ID indicate the same strain of species. Similar names of microbe species with different GeneBank ID indicate different strains of species. % identity indicate the percentage similarities of the species identified to those in the GeneBank.

Appendix 2: Species of bacteria and fungi isolated from R. differens samples from Kampala and Masaka districts. Sampling site Species/strain Kampala Masaka Bacteria Staphylococcus sciuri - + Staphylococcus sp. Strain - + Bacillus thuringiensis strain - + Bacillus thuringiensis strain - + Bacillus cereus + - Bacillus cereus + - Bacillus cereus - + Bacillus cereus - + Bacillus cereus - + Bacillus cereus - + Serratia marcescens + + Serratia marcescens + - Serratia marcescens - + Serratia marcescens - + Acinetobacter baumannii + - Roseomonas sp - +

Fungi Rhodotorula muscilaginosa + - Rhodotorula muscilaginosa + - Rhodotorula muscilaginosa - + Rhodotorula dairenensis + - Rhodotorula dairenensis - + Papiiliotrema laurentii - + Trichosporon asahii + - Trichosporon asahii + - Trichosporon asahii + - Trichosporon asahii + -

Trichosporon asahii + - Trichosporon asahii - + Clavispora lusitaniae + + Clavispora lusitaniae + - Clavispora lusitaniae - + Clavispora lusitaniae - + Pichia kudriavzevii + + Pichia kudriavzevii + - Pichia kudriavzevii + - Pichia kudriavzevii - + Pichia kudriavzevii - + Trichoderma asperellum isolate + + Trichoderma asperellum isolate + - Trichoderma asperellum isolate + - Trichoderma asperellum isolate + - Trichoderma asperellum isolate + - Trichoderma asperellum isolate - + (+) = Present, (-) = Not detected

Appendix 3: Trapping of Ruspolia differens using local drums in Uganda.

Appendix 4: Whole R. differens at trapping points being packaged for transportation

Appendix 5: Handling processes involved in Ruspolia differens while in transit and at commercial distribution points in Kampala

Appendix 6: Laboratory analyses involved during DNA extraction

Sample preparation before DNA extraction Pure cultures of microbes

Quantification of DNA using nanodrop Gel electrophoresis Appendix 7: Samples of extracted DNA of the two microbes

Data on percentage occurrence of different bacterial/fungal species at different points of the value chain were compared using Kruskal Wallis test

Bacterial DNA Fungal DNA