Abstract

It is believed that an infection is more common and runs a more protracted course in people with diabetes. In clinical practice, it is important to be aware of these associations, as the prognosis is often dependent upon prompt recognition and appropriate treatment. To show the course of brucellosis in the diabetic state, a model of Brucella melitensis infection was used in the setting of streptozotocin-induced diabetes in rat. B. melitensis infection proceeded more severely in diabetic rats and the severity of diabetes affected the prognosis. However, no association was found between B. melitensis and insulin using in vitro and in vivo experiments. Our study illustrates that B. melitensis infection in diabetes should be taken seriously.

Introduction

It is believed that an infection is more common and runs a more protracted course in people with diabetes. For instance, tuberculosis was a major cause of death among patients with diabetes mellitus before the advent of insulin therapy [1]. To our knowledge, there are no published data considering the course of Brucella infection in diabetic patients or experimental animal models on this subject. In clinical practice, it is important to be aware of such associations and of the possibility of unusual or atypical infections [2], as the prognosis is often dependent upon prompt recognition and appropriate treatment [3].

Brucella spp. can establish themselves and cause chronic disease in humans and animals. They are facultative intracellular pathogens, which can resist killing by neutrophils, replicate inside macrophages and in ‘non-professional’ phagocytes and maintain a long-lasting interaction with the host cells [4]. The spectrum of brucellosis depends on many factors, including the immune status of the host, the presence of other underlying diseases or conditions and the species of infecting organism [5].

The disease exists worldwide, especially in the Mediterranean basin, the Arabian Peninsula, the Indian subcontinent, in parts of Mexico and Central and South America [6]. Brucella melitensis infection is particularly problematic because Brucella abortus vaccines do not protect effectively against B. melitensis infection. The B. melitensis Rev.1 vaccine has not been fully evaluated for use in cattle. Thus, bovine B. melitensis infection is emerging as an increasingly serious public health problem in some countries [5].

Since the clinical picture in human brucellosis can be misleading and diabetes is a serious health problem, much attention should be paid to these two intersecting groups. Thus, the aim of the present study was to investigate whether the streptozotocin-induced diabetes affects the course of B. melitensis infection in a rat model.

Materials and methods

Animals and induction of diabetes

Wistar Albino rats (275–350 g) were housed individually and allowed free access to food and water. Animals were randomly separated into four groups: A, B, C, and the control group. In the diabetic rat trial, diabetes was induced in overnight fasted rats by intraperitoneal injection of streptozotocin dissolved in 0.05 M citrate buffer (pH 4.5) 4 days prior to the B. melitensis inoculation (Table 1); 80 mg streptozotocin per kg were administered to the rats in Groups A and C each on two consecutive days and 65 mg streptozotocin per kg to rats in Group B once. Rats in the control group received an equivalent volume of phosphate-buffered saline (PBS). Glucose levels in serum were determined at the time of killing by using glucose oxidase assay. Rats with serum glucose level of >250 mg dl−1 at the time of killing were used in the experiments.

Table 1

Induction of diabetes with streptozotocin, and results according to the groups

Groups Streptozotocin Insulin (NPH) Serum glucose (mg dl−1Mean log10 cfu/organ weight Weight gain (g) 
    Spleen Liver  
A (n=12) 80 mg kg−1 twice none 488.1±11.71 4.266±0.06 2.327±0.07 −45±8.4 
B (n=10) 65 mg kg−1 once none 403.0±10.01 3.975±0.05 1.928±0.08 −15±7.5 
C (n=8) 80 mg kg−1 twice 10 mU g−1 446.4±12.62 4.228±0.06 2.163±0.13 −33±6.4 
Control (n=20) none none 119.5±3.588 2.934±0.07 1.588±0.10 +5±5.3 
Groups Streptozotocin Insulin (NPH) Serum glucose (mg dl−1Mean log10 cfu/organ weight Weight gain (g) 
    Spleen Liver  
A (n=12) 80 mg kg−1 twice none 488.1±11.71 4.266±0.06 2.327±0.07 −45±8.4 
B (n=10) 65 mg kg−1 once none 403.0±10.01 3.975±0.05 1.928±0.08 −15±7.5 
C (n=8) 80 mg kg−1 twice 10 mU g−1 446.4±12.62 4.228±0.06 2.163±0.13 −33±6.4 
Control (n=20) none none 119.5±3.588 2.934±0.07 1.588±0.10 +5±5.3 

NPH, Neutral Protamine Hagedron insulin; cfu, colony forming unit.

Data are presented as mean ±S.E.M.

Effects of insulin treatment on B. melitensis infection

One hour prior to B. melitensis inoculation, 10 mU of Neutral Protamine Hagedron (NPH) insulin per gram of body weight were injected once intraperitoneally to the rats in Group C.

Effects of insulin on bacterial growth in liquid cultures

B. melitensis strain 16M was cultured from frozen stock on Brucella agar. This culture was used to inoculate Brucella broth, which was then grown for 48 h at 37°C. 100 µl of this starter culture containing approximately 108 organisms was transferred to separate flasks containing Brucella broth plus insulin (Humulin R, a human biosynthetic insulin; Eli Lilly, Indianapolis, IN, USA) at a concentration of 10 U ml−1. The final concentration of Brucella cells in each flask was 107 ml−1. In control cultures, human serum albumin was added to a final protein concentration equivalent of 10 U of insulin per ml. These growth experiments were performed in triplicate, and the growth rates were determined by measuring the absorbance (A550) of the cultures at 0, 8, 24, 48 and 72 h after inoculation and incubation in a shaking water bath (200 rpm) at 37°C (UV-Visible 1240 Spectrophotometer, Shemadzu).

Growth of B. melitensis in rat serum

The blood samples obtained from rats in Group A and in the control group were centrifuged (7000×g for 15 min), and the serum was removed. The complement was inactivated by placing the serum into a 56°C water bath for 30 min. To each of nine 125-ml flasks we added 5 ml of complement-inactivated serum, 5 ml of PBS, and 100 µl of a dilution of a 48 h culture of B. melitensis containing approximately 108 organisms (giving a final concentration of 107 organisms ml−1 in each flask). The first triplicate set contained 5 ml of control serum in each flask, the second triplicate set contained 5 ml of diabetic rat serum, the third triplicate set contained 5 ml of diabetic rat serum with 0.1 U of human recombinant insulin per ml in each flask. At 0, 8, 24, 48 and 72 h after the inoculation of the flasks with bacteria, 100-µl samples were removed from the flasks and the numbers of bacteria were determined.

Experimentally induced B. melitensis infection

Cells of B. melitensis strain 16 M in a logarithmic phase were suspended in a saline and suspensions were adjusted (based on live counts) to yield 2×104 to 4×104 colony forming unit (cfu). Inoculation was performed by injecting one dose of 0.5 ml saline containing 2×105 to 4×105 cfu intraperitoneally to the rats in all groups. Seven days after the challenge to B. melitensis, rats were weighed and assessed for the number of B. melitensis isolated from the spleen and the liver. Spleens and livers were aseptically removed, weighed and each organ was homogenized (Stomacher 80, Seward, UK) in 5 ml of sterile saline. Aliquots of 0.1 ml of the homogenates were diluted 10-fold in saline and plated onto Brucella agar plates to obtain a viable count. Plates were incubated at 37°C for 72–96 h and a colony count was performed. Each procedure was repeated in triplicate. The mean count was calculated and expressed as log10. If no bacterial growth was apparent after 4 days of incubation, the plates were incubated for an additional 3 days before being considered sterile.

The experiments reported in this study were carried out in accordance with the declaration of Helsinki. Ethical approval was granted by Kocaeli University Ethics Committee (Kocaeli, Turkey).

Statistical analysis

Comparative analysis between groups, of mean log10 cfu/organ weight, mean of serum glucose level, rat body weight, spleen weight, liver weight and their ratio were carried out by using the Mann–Whitney test. A P value of <0.05 was considered to be statistically significant. The relationship of mean serum glucose level (mg dl−1) and the number of B. melitensis isolated from the spleen (mean log10 cfu/tissue weight) were analyzed by using the Pearson correlation test with a level of significance of P=0.05.

Results

Effects of streptozotocin on blood glucose and body weight

At the time of killing blood was taken from both diabetic and non-diabetic rats. Serum glucose levels were determined by using glucose oxidase assay. The mean serum glucose level of rats in Group C was significantly lower (P<0.05) than that in Group A. The mean serum glucose levels of rats in both groups were also significantly higher (P<0.05) than the mean serum glucose level of rats in Group B (Table 1).

During the study period each rat was weighed periodically and weight gain or loss was calculated for all groups. The mean body weight of streptozotocin-administered rats (Groups A, B, C) was significantly (P<0.05) lower than the mean body weight of rats in the control group.

Effects of diabetes on B. melitensis infection

The data obtained from quantitative bacteriological spleen and liver cultures of diabetic and non-diabetic rats infected with B. melitensis are shown in Table 1. The number of B. melitensis isolated (mean log10 cfu/organ weight) from the spleen and the liver of rats in Group A was significantly higher (P<0.05) than in rats in Group B. The number of organisms isolated from the spleen and the liver of rats in both groups were also significantly higher (P<0.05) than that of the rats in the control group.

A statistically significant correlation (P<0.05) between the serum glucose level and the number of B. melitensis cells isolated from spleen and liver of the rats was also evident in all groups (Fig. 1).

Figure 1

Relationship of streptozotocin-induced diabetes and Brucella melitensis infection of rats. Statistically significant correlation (r=0.902, P<0.05) occurs between the amount of streptozotocin and severity of B. melitensis infection of rats.

Figure 1

Relationship of streptozotocin-induced diabetes and Brucella melitensis infection of rats. Statistically significant correlation (r=0.902, P<0.05) occurs between the amount of streptozotocin and severity of B. melitensis infection of rats.

No physical signs of infection were observed in rats after they had been challenged with B. melitensis. In order to show any possible enlargement of spleen and liver, the ratios between the organ and the weight of the body were calculated. No significant differences were found in the ratios between the spleen (P=0.204) or liver (P=0.977) and body weight among the groups.

Interaction of insulin with B. melitensis

Streptozotocin (80 mg kg−1) was administered to the rats in Groups A and C on each of two consecutive days and 10 mU of insulin NPH per gram body weight was injected intraperitoneally to the rats in Group C prior to inoculation with B. melitensis, as described above. Although we observed a slight decrease in the number of organisms recovered from the spleen and liver of rats in Group C compared to those in Group A, this difference was not statistically significant (P>0.05).

Using the standard strain B. melitensis 16M for testing the effect of insulin on growth of B. melitensis in the medium, we observed that the growth rate of this strain was not affected at a concentration of 10 U of insulin per ml.

The effect of insulin on growth of B. melitensis in the rat serum was examined with complement-inactivated serum from streptozotocin-injected rats. The serum from streptozotocin-injected rats with insulin levels restored by the addition of 0.1 U of human recombinant insulin per ml of rat serum and complement-inactivated normal rat serum. After 24, 48 and 72 h, growth in the serum from streptozotocin-injected rats (insulin-depleted) and growth in the serum from streptozotocin-injected rats with restored insulin were similar to that of the control serum.

Discussion

In this study, a model of B. melitensis infection was used in a setting of streptozotocin-induced diabetes in rats. We found that diabetic rats challenged with B. melitensis infection had a significantly higher number (P<0.05) of B. melitensis in their spleen and liver than the rats in the control group. The reason for this result might be related to the decrease in Th1-type immune response due to induced diabetic state. Moreover, the bacteria can survive in phagocytic cells and multiply to high concentrations. Thus, the differences between the number of B. melitensis in rats’ spleen and liver among the groups were referred to the adverse effects of diabetes and the characteristics of the B. melitensis infection.

The disease spectrum of brucellosis depends on the infecting organism. B. abortus and Brucella canis tend to produce mild disease with rare suppurative complications. Conversely, B. melitensis, the most common cause of brucellosis, also causes severe disease with a high incidence of serious complications [7]. In this study, however, no physical signs of infection were observed. Furthermore, the number of B. melitensis among the diabetic rats seemed not to be affected by the severity of the induced hyperglycemic state.

The animal model, which mimics human brucellosis, was developed and used to study the efficacy of various antibiotics in the treatment. The criteria for therapeutic efficacy in brucellosis animal models are: cure documented by the sterilization of the animals’ spleen or reduction of viable counts of Brucella cultured from the homogenized spleen [8]. Previous studies applying the same model mostly preferred using mice. However, in the present study we used rats since we attained to modify the method by using rats in our previous study [9]. In fact, mice are not infected in nature, and in the laboratory, infected mice show a transitory bacteremia, where bacteria are slowly eliminated some weeks after the onset of the infection [4]. In both our studies, none of the five naive rats, which were exposed to B. melitensis infection for a month, had clearance of bacteria in their spleen and liver. That is, there was no spontaneous cure among the infected rats. Moreover, we observed a steady increase in the number of B. melitensis in their spleen.

After injection of NPH insulin into the rats, which had been treated with streptozotocin, the number of B. melitensis isolated from spleens of insulin-treated rats (Group C) was reduced compared to the non-insulin-treated group (Group A), but the difference was not statistically significant (P>0.05). In order to elucidate the interaction of insulin with B. melitensis, we used two in vitro experiments, described previously [10]. However, our present study does not provide any evidence for the ability of a human hormone, insulin, to interact with B. melitensis, whereas hormones and hormone-binding proteins resembling those of vertebrates, are widespread in fungi, yeast and bacteria [10].

Our present study illustrates that B. melitensis infection in diabetes should be taken seriously. We applied a model of B. melitensis infection in the setting of streptozotocin-induced diabetes. Further use of this model may provide new insights into the therapy of humans, who may be threatened with this organism through zoonotic infections and possibly by bio-terrorism.

References

[1]
Saiki
O
Negoro
S
Tsuyuguchi
I
Yamamura
Y
(
1980
)
Depressed immunological defense mechanisms in mice with experimentally induced diabetes
.
Infect. Immun.
 
28
,
127
131
.
[2]
Rajbhandari
S.M
Wilson
R.M
(
1998
)
Unusual infections in diabetes
.
Diabetes Res. Clin. Pract.
 
39
,
123
128
.
[3]
Bartelink
M.L
Hoek
L
Freriks
J.P
Rutten
G.E
(
1998
)
Infections in patients with type 2 diabetes in general practice
.
Diabetes Res. Clin. Pract.
 
40
,
15
19
.
[4]
Dornand
J
Gross
A
Lafont
V
et al
(
2002
)
The innate immune response against Brucella in humans
.
Vet. Microbiol.
 
90
,
383
394
.
[5]
Corbel
M.J
(
1997
)
Brucellosis: an overview
.
Emerg. Infect. Dis.
 
3
,
213
221
.
[6]
Young
L.J.
(
1999
)
Brucella Species
, In:
Antimicrobial Therapy and Vaccines
  (
Yu
V.L.
Merigan
C.T.
Barriere
S.L.
, Eds.) pp.
71
93
.
Williams and Wilkins
,
Baltimore, MD
.
[7]
Koneman
E.W.
Allen
S.D.
Janda
W.M.
Schreckenberger
P.C.
Winn
W.C.
(
1997
)
Miscellaneous fastidious gram-negative bacilli
. In:
Diagnostic Microbiology
  (
Koneman
E.W.
Allen
S.D.
Janda
W.M.
Schreckenberger
P.C.
Winn
W.C.
, Eds.), pp.
395
472
.
J.B. Lippincott
,
Philadelphia, PA
.
[8]
Shasha
B
Lang
R
Rubinstein
E
(
1994
)
Efficacy of combinations of doxycycline and rifampicin in the therapy of experimental mouse brucellosis
.
J. Antimicrob. Chemother.
 
33
,
545
551
.
[9]
Yumuk
Z
Ozdemirci
S
Erden
B.F
Dundar
V
(
2001
)
The effect of long-term ethanol feeding on Brucella melitensis infection of rats
.
Alcohol Alcohol.
 
36
,
314
317
.
[10]
Woods
D.E
Jones
A.L
Hill
P.J
(
1993
)
Interaction of insulin with Pseudomonas pseudomallei
.
Infect. Immun.
 
61
,
4045
4050
.