ABSTRACT
The intestinal flora of Lucilia larvae consists mainly of non-lactose-fermenting, gram-negative bacilli which do not liquefy gelatine. Proteolytic organisms are not present in the gut in significant numbers, but they occur in blown meat.
A method of rearing larvae aseptically is described. The eggs are sterilised by treatment with c 1 per cent, mercuric chloride solution and the larvae reared on heated brain mush, sterility being tested by inoculating aerobic and anaerobic media. The absence of symbionts transmitted inside the egg has been concluded from an examination of stained smears and sections of “sterile” larvae.
When larvae are reared aseptically on sterilised brain, the reaction of the gut contents is normal, tryptase is present in the intestine and excreta, and the growth rate is almost the same as in the presence of bacteria.
It is, therefore, concluded that micro-organisms play no part in intestinal digestion.
Sterile larvae excrete ammonia, but the amount is insufficient to make the food alkaline until the third or fourth day of larval growth. With infected cultures the reaction is distinctly alkaline on the second day.
The ammonifying bacilli isolated from normal larvae are probably responsible for the rapid appearance of ammonia in blown meat.
INTRODUCTION
The first paper of this series (Hobson, 1931) dealt with certain aspects of digestion in blow-fly larvae; the anatomy and histology of the gut, the reaction of the gut contents and the digestive enzymes present. As blow-fly larvae cannot ingest solid particles of food, digestion of meat involves two phases : (i) the external digestion or liquefaction of the meat, (ii) the internal digestion or hydrolysis and absorption of the ingested food. The present paper concerns the part played by bacteria in the internal digestion. It includes an examination of the normal flora and an investigation of larvae reared aseptically on heated brain mush ; this is a soft food which the larvae can ingest with little or no predigestion. The same pure strain of Lucilia sericata Mg. has been used throughout.
OBSERVATIONS AND EXPERIMENTS
(i) The intestinal flora
No comprehensive examination of the bacteriology of blow-fly larvae appears ever to have been made ; however, the experiments of various workers on the survival of pathogenic organisms in muscid larvae provide some data on the natural flora. Their results, which have been summarised by Graham-Smith (1913), show that the intestine contains large numbers of bacilli which grow well on MacConkey’s medium and do not ferment lactose or liquefy gelatine ; on a less selective medium there develop other types which have not been examined. An investigation was, therefore, made of the flora of Lucilia larvae with the object of confirming these results and especially of determining whether proteolytic bacteria are present.
For preliminary information, serial sections and smears from different parts of the body were stained and examined. In larvae feeding on decomposing meat the gut was found to contain large numbers of micro-organisms, but in larvae placed for a few hours on fresh food the numbers showed a striking decrease. The most abundant were small gram-negative rods and cocci, also gram-positive cocci ; yeasts were often present but in small numbers. There was no evidence of micro-organisms occurring in the tissues or haemocoel fluid.
In order to study the cultural properties of the intestinal flora, the gut contents were plated out and the most abundant forms isolated. The food supplied was raw beef, fresh at the time of oviposition, but liquefied and decomposing when the larvae were removed for examination. The material was taken from the posterior part of the mid-gut where digestion occurs, also from the hind-gut. Contamination with the contents of the crop and anterior part of the mid-gut was avoided so as to exclude recently ingested organisms. The larvae were washed and flamed in alcohol, the guts dissected out intact with sterilised needles and the contents of the posterior region opened into sterile saline. Suitable dilutions were plated on nutrient agar and after incubation single colonies were picked off, subcultured on agar slopes and examined. Most of the colonies developing on agar were found to be gram-negative non-liquefying bacilli, gelatine liquefaction being tested by incubating stabs for a month at 23° C. Gram-positive staphylococci were occasionally found, some of which liquefied gelatine slowly. Since only a small proportion of the colonies on the plates could be examined, dilutions of the gut contents were inoculated directly into gelatine. Shakes showed little or no liquefaction after a week at 23° C., and comparatively few liquefying colonies developed in 4 days on plates. Anaerobes were not abundant in the intestine, since inoculations of deep glucose agar tubes showed poor growth at the bottom of the medium. The intestinal flora of Lucilia larvae, therefore, consists mainly of non-proteolytic aerobic bacilli. However, forms which liquefy gelatine were isolated from meat on which larvae were feeding.
The gram-negative bacilli which prevail in the intestine were found to be the non-lactose-fermenting types described by previous workers. The cultures already isolated from agar plates were examined by the methods used by Graham-Smith (1913) for classifying the non-lactose-fermenting bacilli from house-flies. To ensure that the cultures were pure, they were first replated on MacConkey’s medium and subcultured on agar slopes from single colonies. The liquefiers isolated from blown meat were also tested.
The results are shown in Table I; Nos. 1, 2 and 3 were isolated from the intestine, Nos. 4 and 5 from meat on which larvae were feeding. No. 1 is probably the organism which Ledingham (1911) found in Musca larvae and referred to as Bacillus “A.” According to Graham-Smith, Bacillus “A” frequently occurs in house-flies and non-lactose-fermenting bacilli are common in blow-fly larvae. No. 5 was identified as Proteus vulgaris from its cultural reactions and spreading habit on agar plates. The liquefiers found on blown meat are very similar to those isolated by Ledingham from Musca larvae ; he described a strain of Proteus vulgaris similar in its reactions to No. 5 and a liquefier which differed from No. 4 only in fermenting mannite. The results of this investigation are, therefore, in good agreement with the previous evidence on the bacteriology of muscid larvae.
A remarkable feature of the intestinal flora of Lucilia larvae is its extreme limitation. Non-lactose-fermenting bacilli were constantly found in large numbers, other forms occurring only occasionally and in small numbers. Undoubtedly these bacilli are well adapted to the conditions in the intestine ; nevertheless the absence of common saprophytic organisms is surprising in view of the nature of the diet. These findings refer to the posterior part of the gut. Probably the acid reaction (about pH 3·2) in the middle region kills many of the ingested bacteria, only organisms favoured by the environment developing in the subsequent alkaline region. However, this will not explain the absence of spore-forming organisms. Duncan (1926) described a thermostable bactericide which occurs in the gut of certain insects, including house-flies, and is especially toxic to spore-forming organisms such as B. subtilis. Preliminary tests for the presence of this substance in Lucilia larvae have so far yielded inconclusive results.
With regard to the part played by bacteria in digestion, the predominance of non-liquefying organisms suggests that the natural flora does not form proteolytic enzymes in the intestine. In the posterior region of the mid-gut the contents are alkaline, about pH 8-o, and the larval tryptase is adapted for working at this reaction (Hobson, 1931). Some of the bacilli isolated from the intestine can deaminate protein as they make milk strongly alkaline (Table I); however, it will be shown later that the absence of the natural flora does not affect the reaction in the digestive region of the gut. The proteolytic forms isolated from the food probably accelerate the liquefaction of meat and the ammonia-forming organisms, by making the food alkaline, may increase the activity of the tryptase which is excreted by the larvae.
(ii) Method of rearing Lucilia larvae aseptically
Wolhnan (1911, 1919) reared sterile Calliphora larvae on autoclaved brain, the eggs being sterilised with weak mercuric chloride solution or 10 volume hydrogen peroxide. His method has been slightly modified and applied to Lucilia larvae.
Sterilisation of the eggs
In preliminary tests on various methods of sterilising Lucilia eggs, treatment with o-i per cent, mercuric chloride solution proved the most satisfactory. Immersion for 15 minutes produced surface sterility but only a small proportion of the treated eggs hatched. Further investigation showed that no mortality occurred if the eggs were not sterilised until nearly ready to hatch.
The following procedure was finally adopted.
In order to obtain eggs of known age, fresh pieces of meat are placed in the cage and examined at hourly intervals. The eggs are placed in 0·5 per cent. Agral I solution, carefully separated with camel-hair brushes and left immersed in the liquid for about 24 hours, the exact time depending upon the temperature. They are then transferred to a covered glass evaporating basin, washed by decantation six times with sterile 0·5 per cent. Agral I solution, treated for 15 minutes with 0·1 per cent, mercuric chloride solution and finally washed three times with Agral I solution. After the last wash liquid has been decanted off, the basin is inverted. The eggs adhere to the under surface of the glass and are transferred to the medium in a loopful of sterile saline on a platinum wire. All the apparatus used is previously sterilised and the operations are carried out with aseptic precautions.
With this method about 90 per cent, of the eggs hatch. Usually batches of 10 to 20 eggs are treated at a time, but as many as 40 eggs can be sterilised together. If necessary, recently laid eggs can be sterilised on the same day by first incubating them for 6 hours at 37° C., the eggs being suspended in the condensation liquid under the top of a Petri dish containing water. Lucilia eggs will hatch under water, but the incubation periods at different temperatures are then longer than the corresponding values given by Wardle (1930). A dilute solution of Agral I (a proprietary article) was chosen for washing the eggs because it has a very low surface tension, does not form a precipitate with the mercuric chloride solution and is not toxic to the larvae or eggs.
Growth on heated brain
The medium used for rearing sterile larvae was sheep’s brain, which was prepared and sterilised as follows. A fresh brain was minced, thoroughly mixed, placed in plugged test-tubes and sterilised by steaming twice for hours on successive days. The liquid which separated after heating was removed, either by placing cotton-wool under the food, or by stirring the water and soft coagulum into a paste with a sterile glass rod after the first steaming. Sterile larvae grow well on this medium and attain the normal maximum size in 4 or 5 days. If transferred to a dry tube, fully grown sterile larvae will pupate and produce blow-flies which appear perfectly normal.
Method of testing sterility
The usual procedure for testing the sterility of cultures was to inoculate agar slopes and deep glucose-agar tubes with smears of the food ; contamination, when present, usually became apparent after 24 hours’ incubation at 37° C. The sterility of larvae used for important experiments was also confirmed by examining stained smears of the gut contents, as occasionally yeasts were found which did not develop on nutrient agar. A careful search was also made for the presence of symbionts carried within the egg.
Examination of Lucilia larvae for inherited symbionts
Many insects harbour symbiotic micro-organisms, which are transmitted through the egg from one generation to the next. In some cases the symbionts develop only in certain cells of the host and often they are difficult to cultivate on artificial media. “Sterile” Lucilia larvae were, therefore, examined for micro-organisms derived from the egg. Sections and smears were prepared and stained in various ways, Giemsa’s method being used to demonstrate Rickettsia. Since no micro-organisms were observed in these preparations, either intracellular or free living, it is concluded that inherited symbionts are absent. Although negative evidence of this nature cannot afford final proof, the following considerations support this conclusion.
With the exception of the Rickettsia group, the symbionts described in insects are not usually difficult to stain and recognise, and, although many workers have studied the histology of muscid larvae and pupae, well-defined intracellular symbionts have not been described. Rickettsiae are notoriously difficult to detect, but symbiosis in this group is confined mainly to blood-sucking arthropods. Hertig and Wolbach (1924) examined various insects for rickettsia-like organisms and obtained negative results with flies (Stomoxys and Musca).
It may be noted that Bogdanow (1906, 1908) claimed that the contents of blowfly eggs (Calliphora) are infected with bacteria. He treated batches of eggs with mercuric chloride solution and found the larvae from these eggs to be infected with micrococci which grew readily on artificial media. He did not prove the surface sterility of the eggs and later found about 60 per cent, of the larvae sterile, when eggs were treated and hatched separately. Bogdanow’s results may have been due to incomplete sterilisation of the eggs, since Wollman (1911) obtained sterile Calliphora larvae from eggs treated in large batches. It is not possible to identify Bogdanow’s “micrococci” from his description, but in their morphology and cultural reactions on gelatine and milk they appear to resemble the non-lactose-fermenting bacilli described from Lucilia larvae.
(iii) Digestion in sterile larvae
For these experiments larvae were reared aseptically on heated brain by the method already described and only cultures of proven sterility were employed. The following points were investigated : the reaction of the gut contents and medium, the digestive enzymes and the growth rate.
The reaction in the medium and alimentary tract
The reaction in different parts of the gut was determined by transferring young aseptic larvae to sterile nutrient agar containing various indicators. When the gut was seen to contain the indicator, the larva was dissected and examined, the tube being incubated to test for infection. The reaction in the crop and anterior region of the mid-gut has been assumed to be the same as that of the food, since this was found to be the case with non-sterile larvae (Hobson, 1931). The results are collected in Table II and compared with the corresponding values for larvae reared on decomposing meat. The result given for the sterile food was obtained with cultures containing small numbers of young larvae (2 to 3 days old), the residues being extracted with water and the pH estimated colorimetrically.
Table II shows that the reaction of the food in the mid-gut is the same in the presence and absence of bacteria, except in the anterior segment. Bacteria, therefore, do not produce the acidity in the middle region of the mid-gut, or the alkalinity in the posterior region. On the other hand, the formation of ammonia in blown meat is probably due in part to bacterial action.
When small numbers of larvae from unsterilised eggs were reared on heated brain, the reaction of the food was found to be alkaline, about pH 8 · 0, on the second day of larval growth, the original pH being 6 · 3. With sterile larvae, the reaction of the food depended on the ratio of larval weight to amount of residual food. With small numbers of sterile larvae, the food remained acid during the first three days of growth and became faintly alkaline on the fourth or fifth day, when growth was nearly complete. With a high concentration of sterile larvae, the food became alkaline on the third day and the residue at the end of growth was strongly ammoniacal, a pH of 8 · 2 being found in one case. That the excreta of sterile larvae are alkaline and contain ammonia was shown by testing with a loopful of Nessler’s solution on a platinum wire ; if the food was still acid, the drop only turned yellow when placed near a larva defecating on the side of the tube.
These results appear to reconcile the conflicting views of Weinland (1906,1907) and Bogdanow (1906, 1908) on the origin of ammonia in blown meat. Weinland concluded that Calliphora larvae themselves produce ammonia, protein being deaminated in the tissues. He found that ammonia formed 69 to 82 per cent, of the nitrogen in the excreta and that starved larvae ammonified fresh meat rapidly, the exact time not being stated. Bogdanow (1906) suggested that bacterial action was responsible for the ammonia in Weinland’s experiments. He attempted to rear larvae aseptically, but the meat residues were generally contaminated with a micrococcus and smelt of ammonia. The micrococci were constantly present in blown meat under normal conditions and Bogdanow regarded them as the source of ammonia. These results were criticised by Weinland as inconclusive and, although Bogdanow (1908) succeeded later in obtaining larvae free from micro-cocci, he did not state whether or not the residues from sterile larvae contain ammonia or show that the micrococci form ammonia from protein. However, Wollman (1911) noted that the food residues were odourless and free from ammonia when Calliphora larvae were reared aseptically on autoclaved meat. In my experiments, ammonia could readily be detected in the excreta of sterile Lucilia larvae and in the residues at the end of growth, but tests on aqueous extracts of the food gave inconclusive results when made during the early stages of growth. Wollman’s negative result may have been due to the fact that he examined the food residues and not the larvae. He did not state what amount of meat was supplied per larva and whether he tested water extracts or the air in the tube. When a sufficient excess of food is present, the reaction remains acid and ammonia is not evolved.
I have confirmed the presence of ammonia in sterile cultures of Lucilia larvae several times, both with meat and brain as food, the absence of micro-organisms being proved by inoculation and examination of smears. Weinland was, therefore, correct in his assertion that blow-fly larvae themselves produce ammonia, although he did not disprove bacterial action by aseptic experiments. Bogdanow was not entirely wrong in attributing the ammonia to micro-organisms, as they normally form a part of the ammonia in blown meat and are probably responsible for the initial neutralisation of the lactic acid. Lucilia larvae have been found to increase from 0-05 to about 5 mg. during the first 2 days of growth and reach 60 mg. on the fourth day (cf. Fig. 1, p. 136). Hence, unless the proportion of larvae is excessive, the amount of ammonia excreted will probably not become appreciable until the third day of growth.
Digestive enzymes
The following enzymes were found in Lucilia larvae which had been reared on unsterilised meat (Hobson, 1931): amylase in the salivary gland and tryptase, peptidase and lipase in the mid-gut, the proteolytic enzymes persisting in the excreta. As tryptase is the most important digestive enzyme, its presence has been confirmed in sterile larvae. Ten half-grown sterile larvae were dissected, the contents of the mid-gut emulsified in 1 c.c. of water, and the emulsion added to an equal volume of 10 per cent, gelatine. The mixture was made just alkaline to phenol red and incubated at 37° C. in the presence of thymol. At intervals aliquots were removed and digestion measured by formol titration. Active digestion of the gelatine occurred, a boiled control being inactive. Sterile larvae therefore, secrete tryptase in the mid-gut; their excreta also contain this enzyme, as the following experiment shows. A residue from the growth of sterile larvae was shaken up with an equal volume of water and the mixture added to twice its volume of 10 per cent, gelatine containing thymol. After incubation for 1 hour at 37° C., the gelatine remained liquid on cooling, a boiled control treated in the same way solidifying readily. This experiment confirms the result of Wollman (1922), who found a proteolytic enzyme in the residues from the aseptic growth of blow-fly larvae.
Growth on heated brain
In order to measure the growth rate of sterile Lucilia larvae on heated brain mush, a series of cultures was reared under similar conditions. Six or seven larvae were grown in each tube on an excess of food, the temperature being 23° C. and the relative humidity 75 per cent. At intervals all the larvae in a tube were removed and weighed individually, the residue being tested for sterility. The treatment of the controls was identical except that the medium was infected from blown meat 24 hours before the hatching of the larvae. The results are collected in Table III, the growth rate being shown in Fig. 1. It may be stated that, under comparable conditions, growth is no better on raw meat (beef) than on steamed brain which has been reinfected.
Table III and Fig. 1 show that the presence of bacteria improves the development of larvae on sterilised brain, but the difference is not marked. By interpolation from Fig. 1, the times required to reach 30 mg., half the final weight, were calculated to be 74 and 67 hours for the sterile and infected series respectively. The growth rate on heated brain is decreased, therefore, only 10 per cent, by asepsis. Furthermore, individual variation was not greater in the sterile series. Retarded forms were not found and the mean weight of the larvae at the end of feeding was approximately normal.
These results confirm the observations of Wollman (1919) on the aseptic growth of blow-fly larvae (Calliphord) on autoclaved brain. He found growth no worse under these conditions than on infected raw meat, development being measured by the length of the larvae after fixation.
The growth of sterile larvae on heated food may involve other factors than digestion, but these are probably avoided by using brain mush. This food remains soft when heated and larvae can probably ingest it unchanged. As an organ, brain has a high vitamin content which should not be affected by the method of sterilising, intermittent heating at 100° C. for about 3 hours in all. Wollman (1919 a) concluded from feeding trials on rats that autoclaving for 45 minutes at 130° C. does not completely destroy the vitamin content of brain. He suggested that blow-fly larvae store and concentrate vitamin present in their food, since the growth of rats on autoclaved brain and rice was improved by adding to the diet small numbers of larvae which had been reared aseptically on autoclaved brain.
DISCUSSION
(i) The intestinal flora and digestion
The presence of tryptase in sterile Lucilia larvae and their food residues shows that the cells of the mid-gut secrete the enzyme in excess. For, in normal larvae, this was the only tissue found to contain tryptase; also, the identity of the enzymes in the mid-gut and excreta was inferred from the similarity of their pH activity curves (Hobson, 1931). The results of the bacteriological examination bear out this conclusion, since liquefying organisms have not been found in the intestine in appreciable numbers. The proteolytic activity of the gut contents and excreta is not, therefore, augmented by bacterial synthesis in the intestine. Since, also, the gut contents in the digestive region have the normal alkaline reaction in the absence of bacteria, it is concluded that micro-organisms play no part in the digestion of food in the gut. This has been confirmed by showing that larvae grow at almost the normal rate when reared on heated brain mush under aseptic conditions.
(ii) Formation of ammonia
The only abnormality observed in cultures of sterile larvàe was the absence of alkalinity in the food during the early stages of growth. Although the excreta of sterile larvae have been shown to contain free ammonia, the amount produced by young larvae is not sufficient to make the food alkaline at first. The ammonia-forming bacilli isolated from the larvae may play an important part in the lique-faction of meat, since the pH of fresh rigor muscle is approximately 5·8 and the tryptase excreted by the larvae acts best at an alkaline reaction. This question is being further investigated, also the rôle of the liquefying organisms found in blown meat.
Acknowledgements
I wish to record my thanks to Dr P. A. Buxton for his active interest in this work and hospitality in extending to me the facilities of the Entomological Department. My appreciations are also due to Dr V. B. Wigglesworth for suggestions and helpful criticism, to Dr J. T. Duncan for advice on the bacteriological work, and to Prof. W. W. C. Topley for the assistance received from the Bacteriological Department, which has supplied all the media used. I am indebted to the Empire Marketing Board for a grant which entirely finances this work.