Indian Journal of Medical Microbiology IAMM  | About us |  Subscription |  e-Alerts  | Feedback |  Login   
  Print this page Email this page   Small font sizeDefault font sizeIncrease font size
 Home | Ahead of Print | Current Issue | Archives | Search | Instructions  
Users Online: 3031 Official Publication of Indian Association of Medical Microbiologists 
 ~ Next article
 ~ Previous article 
 ~ Table of Contents
 ~  Similar in PUBMED
 ~  Search Pubmed for
 ~  Search in Google Scholar for
 ~Related articles
 ~  Article in PDF (35 KB)
 ~  Citation Manager
 ~  Access Statistics
 ~  Reader Comments
 ~  Email Alert *
 ~  Add to My List *
* Registration required (free)  

 ~  Abstract
 ~  Material & Methods
 ~  Results
 ~  Discussion
 ~  References

 Article Access Statistics
    PDF Downloaded318    
    Comments [Add]    

Recommend this journal

Year : 2001  |  Volume : 19  |  Issue : 2  |  Page : 5-11

Haemagglutination, haemolysin production and serum resistance of proteus and related species isolated from clinical sources

Department of Microbiology, Government Medical College, Nagpur - 440 003, Maharashtra, India

Correspondence Address:
Department of Microbiology, Government Medical College, Nagpur - 440 003, Maharashtra, India

 ~ Abstract 

A total of 148 strains of Proteus and related species comprising of Proteus mirabilis (116), Proteus vulgaris (24), Providentia rettgeri (4), Providentia alcalifaciens (2), Providentia stuarti (1) and Morganella morganii (1), isolated from various sources, were examined for haemagglutination (HA), haemolysin production (HL) and serum resistance (SR). Maximum isolates were obtained from urine (47.30%) and pus (40.54%) and they were multidrug resistant. The sensitivity to Ciprofloxacin was 78.38%, Gentamicin: 62.84%, Cefotaxime: 29.73%, Norfloxacin: 22.97%, Tetracycline: 20.95% and Ampicillin: 6.76%. There were four commonest resistance patterns shown by 58.62% of Proteus mirabilis and 66.67% of Proteus vulgaris strains. Haemagglutination was shown by 91 (61.49%) strains, HL production in 126 (85.14%) strains and SR by 124 (83.78%) isolates. All the three i.e. HA, HL and SR were simultaneously present in 77 (52.27%) strains, any two were present in 40 (27.03%) strains and any one was positive in 30 (20.03%) strains. Thus in as many as 147 (98.32%) isolates, any one or more virulence factors were present. The virulence in commensal pathogen like Proteus is basically a multifactorial phenomenon. The presence of more virulence factors in one strain may increase its pathogenic ability. The evaluation of multiple virulence factors instead of one single parameter will be of greater help in assessing its pathogenic potential.

How to cite this article:
Mishra M, Thakar Y S, Pathak A A. Haemagglutination, haemolysin production and serum resistance of proteus and related species isolated from clinical sources. Indian J Med Microbiol 2001;19:5-11

How to cite this URL:
Mishra M, Thakar Y S, Pathak A A. Haemagglutination, haemolysin production and serum resistance of proteus and related species isolated from clinical sources. Indian J Med Microbiol [serial online] 2001 [cited 2020 May 30];19:5-11. Available from:

Proteus species is a known uropathogen especially causing urinary tract infections (UTI) in catheterized patients[1],[2],[3] and those with urinary tract abnormalities. It may lead to pyelonephritis, stones, fever and bacteraemia.[4],[5],[6] It is also a common nosocomial isolate. Mechanism of pathogenesis of these organisms is yet to be well understood and several virulence factors have been postulated. These include urease production,[6],[7] adhesion to uroepithelium,[8],[9] haemolysin production,[10],[11],[12] swarming,[13] cell invasiveness,[14] cleavage of IgG and IgA by proteolytic enzymes,[15] outer membrane proteins[16] and resistance to normal human serum.[17],[18]
It has been suggested that heavily fimbriated organisms were better able to initiate the infection.[9] Similarly production of haemolysin has been correlated with cytotoxicity for vero cells and with increased virulence in a mouse model of UTI.[19] To determine the frequency with which some of these virulence factors are present in clinical isolates of Proteus and related species, whether they coexist and whether their presence could serve as an index of virulence, the haemagglutinins, haemolysin production and serum resistance of Proteus, Providentia and Morganella isolated from various clinical specimens were studied and whether their expression vary in urinary isolates as compared to others was assessed.

 ~ Material & Methods Top

Bacterial strains
A collection of 148 strains of different species of Proteus, Morganella and Providentia isolated from various clinical specimens were examined. The strains were subdivided into species on the basis of standard biochemical reactions.[20] The strains were stocked on nutrient agar slopes for further testing.
Antibiotic sensitivity test
The antibiotic sensitivity test of each isolate was done on Mueller Hinton Agar (Hi-Media) by Kirby-Bauer disc diffusion technique[21] using the antibiotic discs of Ciprofloxacin (5 g), Gentamicin (10 g), Cefotaxime
(30 g), Norfloxacin (10 g), Tetracycline (10 g) and Ampicillin (2 g).
Bacterial suspension of each strain was tested in HA test with erythrocytes of guinea pig, rabbit, human blood group O and sheep. The bacteria were subcultured statically thrice for 48 hours in nutrient broth at 37oC,[8] then harvested by centrifugation and cell pellets were resuspended in 0.5 mL of PBS, pH 7.2. Haemagglutination was carried out by mixing 0.05 mL of bacterial suspension with 0.1 mL of erythrocyte suspension at 23oC.[8] The HA was tested against tannic acid treated and tannic acid untreated erythrocytes in presence and in the absence of mannose. Haemagglutination was defined as visible clumping erythrocytes. Mannose sensitive (MS) haemagglutination was demonstrated by agglutination of erythrocytes in absence but not in presence of 50 mM mannose. Mannose resistant Klebsiella-like (MR/K) haemagglutination was shown by agglutination of tannic acid treated erythrocytes but not by untreated erythrocytes; the reaction was not inhibited by 50 mM mannose. Mannose resistant Proteus-like (MR/P) haemagglutination was demonstrated by agglutination of untreated as well as tannic acid treated erythrocytes even in presence of 50mM mannose.
The procedure described by Mobley & Chippendale[9] was used. The bacteria were grown overnight in brain heart infusion broth (BHIB) at 37oC with aeration. Samples were taken and 0.1 mL of two folds dilution of bacterial suspensions in PBS (pH 7.2) were mixed with 0.05 mL of 3 per cent suspension of sheep RBCs in PBS (pH 7.2) and incubated at 37oC for 1 hour. The haemolytic titre was defined as the highest dilution in which no visible RBC button was observed at the bottom.
Serum Resistance
The SR assay was performed by the method described by Kumar et al.[18] Pooled normal human serum (PNHS) was taken and inactivated at 56oC for 30 minutes for the test. The test strains were grown overnight at 37oC in nutrient broth, were diluted to 104 CFU/mL in 5 mL fresh nutrient broth and incubated at 37oC for 2 hours. The cultures were centrifuged at 1500 g for 5 min and deposit resuspended in 5 mL PBS (pH 7.2). Equal volumes of (0.2 mL each) PNHS and bacterial suspension were mixed and incubated in a water bath. The viable counts were made on nutrient agar by Miles & Misra method[22] at 0, 60, 120 & 180 min. If viable count dropped to less than 1 per cent of initial value the strains were termed as sensitive and if more than 90 per cent of organisms survived after 180 minutes they were termed resistant. The score obtained between 1 and 90 percent was considered intermediate.

 ~ Results Top

Out of 148 strains, 116 were Proteus mirabilis, 24 Proteus vulgaris, 4 Providentia, 2 Providentia alcalifaciens and one each of Morganella morganii and Providentia stuartii. Their distribution of isolation from different clinical specimens is shown in [Table - 1]. The urease production was uniformly present in all the isolates of P. mirabilis and P. vulgaris.
The antibiotic profile of all the isolates is shown in [Table - 2]. There was no difference of antibiotic sensitivity profile amongst urinary and other isolates of P. mirabilis and P. vulgaris. In P. mirabilis isolates 21 different antibiotic patterns were observed while in P. vulgaris eight patterns were seen using the six antibiotics. The four commonly observed patterns of both are shown in [Table - 3]. It was interesting to see that the same four resistant patterns occurred most frequently for both P. mirabilis andP. vulgaris.
The results of haemagglutination, haemolysin test and serum resistance are depicted in [Table - 4]. The haemagglutination was observed in 73 strains of P. mirabilis and 14 strains of P. vulgaris. Mannose sensitive haemagglutination was produced by 12 strains of P. mirabilis, 7 of P. vulgaris and one each of P. rettgeri and P. alcalifaciens. The strains that were mannose sensitive, haemagglutination positive also produced other haemagglutinins. Haemolysin production was observed in all but 22 strains. Isolates produced measurable haemolytic activity with titres ranging from 1:2 to 1:512. There was no significant difference between the haemolytic titres of urinary isolates and isolates from other sources. A total of 124 strains included in the study were also found to be serum resistant and only 24 were serum sensitive. Only two strains exhibited intermediate pattern but on repeat testing they were found to be serum resistant.
The presence of the three factors-haemagglutination, haemolysin production and serum resistance-either alone or in combination is shown in [Table - 5]. Except one strain of Proteus mirabilis isolated from urine all other isolates possessed either one or more of these three virulence factors.

 ~ Discussion Top

The clinical isolates included in the present study mostly comprised of P. mirabilis and P. vulgaris and most frequently isolated from urine and pus specimens with very few isolates of related species. Proteus and related species are known pathogens that can commonly cause infections in hospitalised patients. The isolates were obtained from the patients admitted in the hospital; hence they are more or less likely to be the hospital strains only. Whether these strains do possess some peculiar characteristic or exhibit some specific virulence trait is not well established. The antibiotic sensitivity profiles can provide us some clues regarding the proximity of different isolates to each other. In the present study all the isolates were more sensitive to Ciprofloxacin (78.38%) and then to Gentamicin (62.84%); otherwise they were fairly resistant to Cefotaxime, Norfloxacin, Tetracycline and Ampicillin. Out of the 64 possible resistant patterns that can be obtained using the six antibiotics, P. mirabilis isolates of the present study displayed only 21 patterns and P. vulgaris only eight; of which only four resistant patterns accounted for 58.62% P. mirabilis isolates and 66.67% for P. vulgaris isolates. It is interesting to note that these four commonly encountered patterns for P. mirabilis and P. vulgaris were same [Table - 3] pointing towards their proximity to each other in a hospital. Similarly there was no difference in sensitivity patterns between urinary isolates and isolates from other sources indicating their common source of existence i.e. within the hospital.
In this study three virulence factors viz, haema-gglutination, haemolysins and serum resistance were examined. Barring one urinary isolate of P. mirabilis, all other isolates did display at least one of these three factors. There was no difference between production of either or all of these virulence factors between urinary and other isolates. Since all these isolates of the present study are obtained from clinical specimens, they are more likely to display the virulence factors in contrast to the environmental ones.
Adhesion mechanisms have been recognized as relevant factors for development of infection, especially urinary tract infection[9]. The strains of Proteus, Providentia and Morganella included in the present study, produced at least three different HAs (MS, MR/K & MR/P) that may be present simultaneously. The role of haemagglutinins of Proteus in pathogenesis have been well recognized[9] although the distribution of HAs and fimbria in these species is more complex.
Haemolysin has been considered to be an important virulence factor by some authors[9],[14] while others could not observe significant difference in virulence[23]. Zunino et al[24] suggested that haemolysin is not essential during early infection but this factor is important at late stages of infection. As many as 85.14% strains, in the present study, produced haemolysins. There was no significant difference in the haemolysin activity of urinary isolates and isolates from other sources together.
The resistance to bactericidal activity of normal human serum is considered to be an important virulence factor of Proteus species[18]. In the present study as many as 124 out of 148 (83.78%) isolates exhibited serum resistance. This virulence factor may contribute to the invasiveness of the bacteria.
There was coexistence of all these three factors in more than half (52.03%) the clinical isolates of Proteus and related species, while presence of two of the three virulence factors was evident in more than half of the remaining strains (27.03%) and except one isolate, all others had at least one of these three factors.
A large number of virulence factors of Proteus have been documented. Of the different virulence factors documented for Proteus, urease production is a major virulence factor of the species itself and cannot be used as an index of virulence as all P. mirabilis and P. vulgaris strains produce this enzyme[9]. Similarly swarming too is constantly associated with these two species. The multi drug resistance of these strains may further contribute to their survival and the virulence. Various virulence factors may act independently or their actions may be complementary to each other. Intimate contact of bacterial cell with epithelium is probably essential for cytotoxic action. Therefore in development of Proteus pyelonephritis the essential step will be to adhere to kidney epithelium, which may then potentiate the haemolysin as well as urease activities.
The pathogenic ability and the presence of virulence factors amongst Proteus and the related species are alike. The virulence in commensal pathogens like Proteus and these species is basically a multifactorial phenomenon. The presence of virulence factors in more numbers in a particular strain is likely to increase its pathogenic ability. The evaluation of a particular single virulence factor to assess its pathogenic potential may result in an erroneous impression.
Greater attention is usually paid to the uropathogenic potential of Proteus species but the strains isolated from other clinical material also are likely to display the same virulence markers and they are more likely to be causative agents than the harmless colonizers especially if the infection is acquired in the hospital. 

 ~ References Top

1.Mobley HLT, Warren JW. Urease positive bacteriuria and obstruction of long term urine catheters. J Clin Microbiol 1987; 25: 2216-2217.  Back to cited text no. 1    
2.Warren JW, Tenney JH, Hoopes JM, Muncie HL, Anthony WC. A prospective microbiologic study of bacteriuria in patients with chronic indwelling ureteral catheters. J Infect Dis 1982; 146: 719-723.  Back to cited text no. 2    
3.Gleckman R, Blagg N, Hibert D, Hall A, Growley M, Pritchard A, Warren W. Symptomatic pyelonephritis in elderly men. J Am Geriatr Soc 1982; 30: 690-693.  Back to cited text no. 3    
4.Stamm WE, Martin SM, Bennett JV. Epidemiology of nosocomial infections due to Gram-negative bacilli: aspects relevant to development and use of vaccine. J Infect Dis 1977; 136 (suppl): S151-S160.  Back to cited text no. 4    
5.Senior BW. The special affinity of particular types of Proteus mirabilis for the urinary tract. J Med Microbiol 1979;12: 1-8.  Back to cited text no. 5    
6.Mobley HLT, Hausinger RP: Microbial ureases: Significance, regulation and molecular characterization. Microbiol Rev 1989; 53: 85-108.  Back to cited text no. 6    
7.Jones BD, Mobley HLT. Genetic and biochemical diversity of ureases of Proteus, Providentia and Morganella species isolated from urinary tract infection. Infect Immun 1987; 55: 2198-2203.  Back to cited text no. 7    
8.Old CD, Adegbola RA. Haemagglutinins and fimbriae of Morganella, Proteus and Providentia. J Med Microbiol 1982; 15: 551-564.  Back to cited text no. 8    
9.Mobley LT, Chippendale GR: Haemagglutinin, urease and haemolysin production by Proteus mirabilis in clinical sources. J Infect Dis 1990; 161: 525-530.  Back to cited text no. 9    
10.Welch RA. Identification of two different haemolysin determinants in uropathogenic Proteus isolates. Infect Immun 1987; 55: 2183-2190.  Back to cited text no. 10    
11.Koronakis V, Cross M, Senior B, Koronakis E, Hughes C. The secreted haemolysins of Proteus mirabilis, Proteus vulgaris and Morganella morganii are genetically related to each other and to the alpha-haemolysin of Escherichia coli. J Bacteriol 1987; 169 : 1509-1515.   Back to cited text no. 11    
12.Senior BW, Hughes C. Production and properties of haemolysins from clinical isolates of the Proteae. J Med Microbiol 1987; 24 : 17-25.  Back to cited text no. 12    
13.Allison C, Coleman N, Jones PL, Hughes C. Ability of Proteus mirabilis to invade human uroepithelial cells is coupled to motility and swarming differentiation. Infect Immun 1992; 60: 4740-4746.   Back to cited text no. 13    
14.Peerbooms PGH, Verwieg AMJJ, Mac Laren DM. Vero cell invasiveness of Proteus mirabilis. Infect Immun 1984; 43: 1068-1071.  Back to cited text no. 14    
15.Loomes LM, Senior BW, Kerr MA. A proteolytic enzyme secreted by Proteus mirabilis degrades immunoglobulins of the immunoglobulin A1 (IgA1), IgA2 and IgG isotypes. Infect Immun 1990; 58 : 1979-1985.  Back to cited text no. 15    
16.Moayeri N, Collins CM, O'Hanley P. Efficacy of a Proteus mirabilis outer membrane protein vaccine in preventing experimental pyelonephritis in a BALB/C mouse model. Infect Immun 1991; 59 : 3778-3786.   Back to cited text no. 16    
17.Larsson P, Olling S. O antigen distribution and sensitivity to the bactericidal effect of normal human serum of Proteus strains from clinical specimens. Med Microbiol Immunol 1977; 163: 77-82.  Back to cited text no. 17    
18.Kumar S, Mathur MD. Sensitivity to the bactericidal effect of human serum of Proteus strains from clinical specimens. Indian J Pathol Microbiol 1997; 40: 335 -339.  Back to cited text no. 18    
19.Senior BW, Albrechtsen M, Kerr MA. Proteus mirabilis strains of diverse type have IgA protease activity. J Med Microbiol 1987; 24: 175 - 180.  Back to cited text no. 19    
20.Crichton PB. In Mackie & Mc Cartney, Practical Medical Microbiology. 14th Ed. Collee JG, Marmion BP, Fraser AG, Simmons A eds. Churchill Livingstone, New York 1996: 361.  Back to cited text no. 20    
21.Bauer AW, Kirby WMM, Sherris JC, Turck M. Antibiotic susceptibility testing by a standardized single disc method. Am J Clin Pathol 1966; 45: 493 - 496.   Back to cited text no. 21    
22.Miles AA, Misra SS, Irwin JO. The estimation of the bactericidal power of the blood. J Hyg (Cambridge) 1938; 38: 732-749 .  Back to cited text no. 22    
23.Swihart KG , Welch RA. The Hpm A haemolysin is more common than Hly A among Proteus isolates. Infect Immun 1990; 58: 1853-1860.  Back to cited text no. 23    
24.Zunino P, Piccini C, Fajardo CL. Growth: Cellular differentiation and virulence factor expression by Proteus mirabilis in vitro and in vivo. J Med Microbiol 1999; 48: 527 -534.  Back to cited text no. 24    
Print this article  Email this article
Previous article Next article


2004 - Indian Journal of Medical Microbiology
Published by Wolters Kluwer - Medknow

Online since April 2001, new site since 1st August '04