First draft prepared by
    Dr R. Fuchs
    Ministry of Sciences
    Republic of Croatia, Zagreb, Croatia


         Spiramycin had previously been evaluated at the twelfth and
    thirty-eighth meetings of the Committee (Annex 1, references 17 and
    97). Based on the estimated concentration with no effect on the human
    gut flora, a temporary ADI of 0-5 g/kg bw was established at the
    thirty-eighth meeting, with the requirement of additional  in vivo
    studies on the effect of spiramycin on the human intestinal flora.

         This monograph addendum summarizes the data that have become
    available since the previous evaluation (Annex 1, reference 98).  


    2.1  Toxicological studies

    2.1.1  Special studies on microbiological activity

         A preliminary study of the effects of spiramycin on faecal
    coliforms and enterococci of the human gastrointestinal flora was
    carried out in mice.  In this study, a dilution of pooled faecal flora
    from healthy human volunteers was transferred anaerobically to twenty
    6-week old female germ-free mice.  A transfer of  Bacteroides fragilis
    was made prior to the human flora transfer.  Seven days after the
    human faecal flora transfer, the mice were divided into 4 groups of 5
    mice each.  Each mouse was caged individually.  Group 1 (negative
    control) received pure drinking-water.  Group 2 (positive control)
    received drinking-water containing spiramycin at a concentration of
    200 ppm for 32 days.  Groups 3 and 4 (test groups) received drinking-
    water containing spiramycin at concentrations of 0.2 or 0.4 mg/l,
    respectively (equivalent to 50 g/kg bw/day and 100 g/kg bw/day,
    respectively), for 32 days.

         Faecal samples were collected on day 0, on 10 occasions after day
    9, and a final sample on day 32.  Total counts for gram-negative
    anaerobes, gram-positive anaerobes, coliforms and enterococci were
    recorded on day 0.  These same counts were performed on the 10 samples
    taken from days 9 to 32.  Additionally, an evaluation of the degree of
    spiramycin resistance in coliforms and enterococci was performed on
    these samples.  For this purpose, coliforms were incubated and counted
    on PCB-desoxycholate agar supplemented with 512 mg spiramycin/l. 
    Enterococci were incubated and counted on Bile Esculine agar
    supplemented with 4 mg spiramycin/l.

         No effect on coliform resistance to spiramycin was reported at
    any dose level.  The percentage of enterococci resistant to spiramycin
    in the 50 g/kg bw/day group was similar to negative controls.  The
    percent of enterococci resistant to spiramycin was increased in the
    positive control and the high-dose group.  In the negative control
    group, large variations in the percentage of coliforms and enterococci
    resistant to spiramycin were reported throughout the study.  Values
    ranged from 1.2 to 28% for coliforms and from 4.7 to 55% for
    enterococci.  Therefore, the significance of the increase in the
    percentage of spiramycin resistant enterococci in test group was
    questionable (Corpet, 1992).

         A study was conducted to evaluate the effect of spiramycin on
    chicks artificially infected with  Salmonella typhimurium. Two
    groups (A and C) of twelve 15-day old chicks received feed containing
    no spiramycin and two groups (B and D) of twelve 15-day old chicks
    consumed feed containing spiramycin embonate (equivalent to 20 mg
    spiramycin base/kg feed).  Groups A and B were inoculated with 

     Salmonella typhimurium, variety Copenhagen (strain 74-928,
    resistant to nalidixic acid), 5 days after the initiation of

         The following parameters were evaluated: the number of
    salmonellae excreted per gram of faeces; the proportion of salmonellae
    resistant to 10 antibiotics commonly used in human and veterinary
    therapeutics; the degree of resistance; and the resistance spectrum. 
    These evaluations occurred at various times after the inoculation (2,
    6, 8, 10, 13, 21, 28, 35, 42, and 49 days).  At the beginning and end
    of treatment, the proportion of faecal coliforms, staphylocci and
    micrococci resistant to 10 commonly used antibiotics, the level of
    resistance, and the resistance spectrum, were determined in the two
    groups of noninfected chicks (groups C and D).  Coliform counts were
    conducted at the beginning and end of the trial on 100 strains (10
    chicks selected per group, 5 strains per chick).

         Spiramycin caused no significant increase in the relative number
    of excreted salmonellae, nor did it produce an increase in the number
    of chicks excreting salmonellae.  The proportion of salmonellae
    resistant to common antibiotics was not increased.  At the end of the
    trial, 83% of untreated chicks and 75% of treated chicks continued to
    excrete salmonella in faeces.  Twenty-seven percent of the  E. coli
    strains isolated before the beginning of treatment were found to be
    resistant to 4 antibiotics (streptomycin, tetracycline,
    chloramphenicol and sulfadiazine).  At the end of the trial,  the
    proportion of resistant  E. coli was similar in both groups (49% in
    the untreated and 34% in the treated chicks).  The salmonella
    antibiotic resistance spectrum at the end of the trial was identical
    to the one conducted at the beginning of the trial.  Also, all
    staphylococci and micrococci were sensitive to the antibiotics
    studied.  Thirty-five strains from 10 untreated chicks and 11 strains
    from 10 treated chicks were isolated.  Spiramycin intake was
    determined to be 3000 g/kg bw/day on day 20 and 1666 g/kg bw/day on
    day 70 (Benazet & Cartier, 1979). 

         An  in vitro assessment of spiramycin MICs for 9 bacterial
    species (10 or 20 strains of each species) from human gastrointestinal
    flora was conducted.  Dominant flora consisting of strictly anaerobic
    bacteria were tested at a concentration of 109 bacteria/ml.  These
    included 10 strains of  Bacteroides spp., Fusobacterium spp.,
     Bifidobacterium spp., Eubacterium spp., Clostridium spp.,
     Lactobacillus spp., and  Peptostreptococcus spp. Sub-dominant
    flora consisting of facultative aero-anaerobic and microaerophilic
    bacteria, were also tested at a concentration of 107 bacteria/ml. 
    These included 20 strains each of  Escherichia coli and
     Enterococcus faecalis. In the total 110 strains tested, the MIC
    value was >1 g/ml.  In 99 strains, the MIC value was >128 g/ml
    (Roques & Michel, 1993).


         At the present meeting, the Committee considered data from new
     in vivo and  in vitro studies on the effect of spiramycin on
    human gastrointestinal flora.

         In an  in vivo study in mice, a dilution of pooled faecal flora
    from healthy human volunteers was transferred anaerobically to germ-
    free mice.   The animals were then treated with up to 200 mg
    spiramycin/l of drinking-water for 32 days.  Increases in resistant
    microorganisms were observed at 0.2 mg/l of water, equivalent to 40
    g/kg bw/day.  Although a quantitative endpoint was identified, there
    were certain shortcomings in this study.  There were large variations
    in the number of resistant coliforms and resistant enterococci in the
    non-treated control group and high populations of resistant organisms
    in all groups before spiramycin treatment.  Moreover, in the selective
    medium used to determine the total and resistant coliforms and
    enterococci in the pooled faeces of mice, only one concentration of
    spiramycin was employed for each bacterial group.   

         The Committee also evaluated data from an  in vivo study
    performed in chickens in which the effects of the drug on  Salmonella
     typhimurium, Escherichia coli and several other microorganisms
    were studied.  The Committee concluded, however, that this study was
    of little relevance for the microbiological evaluation of the effects
    of spiramycin on human gastrointestinal flora because the micro-
    organisms investigated in this study were not of human origin.

         Studies to determine MIC values for spiramycin were conducted
    using bacterial species isolated from healthy human volunteers. 
    Dominant flora tested consisted of strictly anaerobic bacteria (109
    bacteria/ml), while the sub-dominant flora included facultative
    aerobic and microaerophilic bacteria (107 bacteria/ml).  In a total
    of 110 strains tested, all the MIC values were >1 g/ml.  These
    results confirmed those of the earlier studies evaluated at the
    thirty-eighth meeting of the Committee, performed on a limited number
    of strains.  Taking into account the results of the studies already
    evaluated at the previous meeting and new data from the  in vitro
    and  in vivo studies, the Committee was reassured of the
    microbiological safety of spiramycin.

         At the thirty-eighth meeting of the Committee a temporary ADI of
    5 g/kg bw was calculated using the following formula:

                         Concentration without effect       x   Daily faecal
    Upper limit of       on human gut flora (1 g/ml)           bolus (g)
    temporary ADI    =                                                           
                         Fraction of           x  Safety        x   Human body
                         dose bioavailable        factor            weight
                                                                    (60 kg)

                         1 x 150
                         0.05 x 10 x 60

                     =   5 g/kg bw
         The safety factor of 10 was used to cover the variability between
    individuals for all extrapolated parameters.


         In view of the additional reassurance provided by the new data,
    and as these studies covered a wide range of organisms, the Committee
    reconsidered the magnitude of the safety factor and concluded that a
    safety factor of 1 instead of 10, was appropriate.  The other
    parameters used in the previous evaluation provide a conservative
    estimate.  As a result the Committee established an ADI of 0-50 g/kg
    bw, using the above formula.


    BENAZET, F. & CARTER, J.R. (1979).  Influence of spiramycin (5 337
    R.P.) on the implantation and excretion of  Salmonella typhimurium 
    in artificially infected chicks and on the resistance of these
     salmonenella, and faecal  E. coli, Staphylococci and  Micrococci
    to common antibiotics.  Unpublished report, Rhne Poulenc RP/RD/CNG 
    No. 20 193.  Submitted to WHO by Rhne Mrieux, Toulouse, France.

    CORPET, D.E. (1992).  Effect of spiramycin residues on faecal
    coliforms and enterococci in human flora associated mice.  Unpublished
    report from Institut National Recherche Agronomique, Toulouse, France. 
    Submitted to WHO by Rhne Mrieux, Toulouse, France.

    ROQUES, C. & MICHEL, G. (1993).  Determination of minimal inhibitory
    concentrations (MIC) of spiramycin for bacterial species in the human
    gut flora.  Unpublished report Ph/93-152 from Facult des Sciences
    Pharmaceutiques, Toulouse, France.  Submitted to WHO by Rhne Mrieux,
    Toulouse, France.

    See Also:
       Toxicological Abbreviations
       Spiramycin (WHO Food Additives Series 29)
       SPIRAMYCIN (JECFA Evaluation)