Growth of Salmonella enteritidis in artificially contaminated eggs: the effects of inoculum size and suspending media

https://doi.org/10.1016/S0168-1605(01)00540-2Get rights and content

Abstract

Growth profiles of two isolates of Salmonella enteritidis phage type (PT) 4 inoculated into either the albumen of whole shell eggs or into separated albumen were found to be markedly affected by the size of the inoculum and the composition of the medium used to suspend the cells prior to inoculation. Using our model with an inoculum of two cells, multiplication of the Salmonella was not seen in 93% of eggs held at 20 °C for 8 days. In approximately 7% of eggs, however, growth occurred during the 8 days of storage.

If the inoculum equaled or exceeded 25 cells per egg when eggs were subsequently stored at 20 °C, or 250 cells per egg when eggs were stored at 30 °C, high levels of growth of Salmonella in the egg occurred significantly more frequently than when the inoculum was two cells. High levels of growth were also seen more frequently if the inoculum was suspended in buffered peptone water or maximal recovery diluent rather than in phosphate buffered saline. Growth of Salmonella in separated albumen occurred very infrequently (1.1% of samples) at low inoculum levels and did not become significant until the inoculum was 250 cells or greater. Growth in the albumen was unaffected by the composition of the suspending medium. Provided that the inoculum was approximately 2 cells per egg and the bacteria were suspended in PBS, observed growth profiles of S. enteritidis inoculated into the albumen of whole eggs resembled those in naturally contaminated eggs.

Introduction

Human infection with phage types (PTs) of Salmonella enteritidis continues to be of major international importance. In the UK, as elsewhere, contaminated eggs have been identified as vehicles of infection (Coyle et al., 1988), although many outbreaks have resulted from the misuse or mishandling of eggs. In the last two years, infection rates of humans in the UK have fallen, almost certainly as a result the fact that most eggs on sale now come from vaccinated egg laying flocks (Anon., 2000), although eggs from non-vaccinated flocks continue to be sold.

Investigations in the UK with naturally contaminated eggs have demonstrated that S. enteritidis is capable of multiplication in a proportion of eggs following storage at ambient (approximately 20 °C) temperature (Humphrey et al., 1991). As the production of Salmonella-positive eggs by either naturally or artificially infected hens is infrequent Humphrey et al., 1989a, Gast and Beard, 1992, and may occur over a limited time, artificially inoculated eggs have been used to examine the effect of factors such as storage conditions and time on the proliferation of Salmonella in eggs (Humphrey and Whitehead, 1993). One such study employed initial contamination levels of 104 cells per egg (Hammack et al., 1993), even though the examination of eggs from naturally infected hens has shown that fresh eggs generally contain less than 20 cells of S. enteritidis Humphrey et al., 1989b, Humphrey et al., 1991, Mawer et al., 1989, and those from artificially infected hens levels of 10 cells/ml egg contents or below (Gast and Beard, 1992). In general, and particularly during outbreak investigation, the eggs referred to in the above studies were not examined immediately after lay. Gast and Holt (2000) confirmed the earlier observation of Humphrey (1993) that in artificially contaminated eggs the inoculum increased approximately tenfold during the first 24 h post-inoculation. This suggests that naturally contaminated eggs may receive approximately two cells when in the hen's reproductive tissues. Therefore, in this work, we have used a low inoculum level of two cells, as being representative of the number of cells in an egg at the time of lay.

Other studies on artificially contaminated eggs have used Salmonella cells suspended in culture medium (Braun and Fehlhaber, 1995). This could disturb the bacteriostatic properties of the albumen, which cannot typically support the growth of Salmonella (Baron et al., 1997) due to an absence of available iron, which is essential for bacterial growth (Schade and Caroline, 1944). Growth patterns of Salmonella in the above experiments often do not match those observed in naturally contaminated eggs, in that they suggest that invasion of the yolk by Salmonella and subsequent rapid growth occurs in almost all inoculated eggs. Data from the study of naturally contaminated eggs do not support these observations, although investigations of two outbreaks suggested that some apparently fresh eggs could carry high numbers of S. enteritidis Taylor et al., 1984, Morgan et al., 1994.

As growth of the inoculum occurs in some eggs during storage and not in others (Humphrey et al., 1991), it is probably more meaningful to look at the frequency of occurrence of growth of the bacterial inoculum within an egg rather than calculate the mean bacterial count of a group of eggs (Gast and Holt, 2000). The major site of bacterial growth within the egg is the yolk Board, 1966, Humphrey and Whitehead, 1993. In this environment, Salmonella will grow to high level (>106 cells), whilst within the albumen it will grow much more poorly (Gast and Holt, 2000). This means that any attempt to obtain mean levels of growth in terms of numbers of cells from a group of eggs in which yolk contamination occurs infrequently, as is the natural situation, will mask the production of the occasional egg containing very high numbers of bacteria. Thus, in this paper, we have expressed results as the frequency of occurrence of a highly contaminated egg in the manner of Board (1966). For the purposes of risk assessment, it is important to calculate the frequency with which a hazardous event occurs, in this case this is the growth of initially low numbers of Salmonella in an egg to a high number of bacteria able to cause human infection. The method of analysis used in the work gives data on the likelihood of growth occurring in an egg when different numbers of cells are deposited, or when eggs are stored under different temperature conditions.

Data from the previously mentioned egg contamination studies have been used to decide advisory and regulatory temperature and time limits for egg storage. There is a need to develop a model for artificial egg contamination that closely mirrors the natural situation. Variables such as inoculum size and the medium used to suspend the Salmonella cells prior to their introduction into the egg should be examined to see what effect these have on bacterial growth within the egg. This paper reports on the effects of the media in which cells are suspended and inoculum size on growth profiles of S. enteritidis PT4 isolates in whole shell eggs and separated albumen. A model of egg contamination is presented which allows accurate quantification of the risk which Salmonella-contaminated eggs pose to the consumer when stored under different conditions.

Section snippets

Eggs

Eggs were obtained from a Salmonella-free local caged production unit within 6 h of lay, from hens which had not been vaccinated against Salmonella. Eggs were inspected on arrival at the laboratory and any found to be cracked or contaminated with faeces were discarded. The weight of eggs used was between 50 and 75 g. Experimental work was started on the day of egg collection.

Preparation of inocula

The study used two isolates of S. enteritidis PT4, laboratory identification codes E and G. Strain E was isolated from a

Effect of PT4 strain on observed growth profiles

Irrespective of inoculum size or suspending media, no significant differences were observed in the growth patterns of the two PT4 isolates. Thus, in experiments using the same inoculation conditions, there was no significant difference between the contamination levels shown in batches of eggs inoculated with either isolate (p>0.05). The similarity between the behaviour of the PT4 strains meant that a decision was made to combine data from experiments using either isolate.

Size of inoculum

The average number of

Discussion

Current available evidence suggests that eggs contaminated with Salmonella in vivo contain low numbers of cells when fresh. The cells are suspended in a medium with a low concentration of free iron, as iron is tightly bound to transferrins. If growth profiles of artificially contaminated eggs are to parallel the natural situation, it is important that Salmonella cells are introduced in a way which mimics, as far as possible, the natural condition. S. enteritidis is able to grow in the contents

Acknowledgements

The authors would like to thank the British Egg Marketing Board Research and Education Trust, the Worshipful Company of Poulters, the British Egg Industry Council and MAFF for funding aspects of this work, and Mrs. Georgie Broom for typing the manuscript.

References (27)

  • Anon

    Trends in selected gastrointestinal infections: 1999

    Communicable Dis. Rep.

    (2000)
  • R.G. Board

    The course of microbial infection of the hen's egg

    J. Appl. Bacteriol.

    (1966)
  • C.E. Clay et al.

    Growth of Salmonella enteritidis in artificially contaminated hens' shell eggs

    Epidemiol. Infect.

    (1991)
  • Cited by (0)

    View full text