The microbial load and the severity of morphological changes in the prostate during infection with various titers of uropathogens: a comparison of data from an animal model study

INTRODUCTION

Relevant studies showed that in patients with chronic bacterial prostatitis (CBP), isolates containe causative (arise from Escherichia coli) and debatable bacteria in a significant number of cases [1]. Trichieri et al. (2021) performed a retrospective evaluation of the results of the cultural study of the prostatic fluid, ejaculate, mid-stream and voided after massage urine in patients with CBP (n = 924) from seven countries. E. coli were detected in 31.0% of cases, and Staphylococcus species were revealed in 13.0% of cases [2]. The earlier publication by Stamatiou et al. (2017) provided the analysis of the bacterial study results of prostatic fluid and mid-stream urine samples. It showed that patients with primary CBP (n = 253) had E. coli in 90 isolates (69 monovariants/21 polymicrobial samples) and a mixed coagulase-negative staphylococci (CoNS) Staphylococcus haemolyticus + Staphylococcus hominis — in 71 isolates (46/25 samples, respectively). In patients with recurrent CBP (n = 137), E. coli was verified in 52 isolates (35/17 samples, respectively) and a mixture of S. haemolyticus + S. hominis, respectively, in 37 isolates (18/19 samples, respectively) [3]. Thus, in a significant number of cases, CoNS was isolated in patients with CBP. However, in the publications presented and previous studies, there is no data on the verification of anaerobic flora due to the application of a standard method of a cultural study. However, clinical studies confirmed that anaerobes significantly affected the severity of the course and expression of symptoms in patients with CBP [4]. This limitation in the application of modern nutrient media and systems of cultivation and detection of anaerobic bacteria is associated with researchers’ insufficient awareness of the pathogenic potential of this group of bacteria.

Despite a significant rate of CoNS detection in patients with prostatitis-like symptoms, the representatives of this cluster remain in the status of debatable. However, in the previous experimental series of studies, the authors established that some debatable microorganisms of the CoNS cluster (S. haemolyticus), along with representatives of anaerobic bacteria (Peptococcus niger), could cause acute inflammation in the prostate (Pr) of laboratory animals (LA) with a titer of 10³ CFU/ml, which is considered pathogenic in clinical practice during the diagnosis of bacterial prostatitis [5][6]. Similar experimental modeling of acute obstructive pyelonephritis on LAs showed that aerobic bacteria E. coli, S. haemolyticus, and anaerobic bacteria Eubacterium spp. in the generally acknowledged titer and subpathogenic infecting dose of 10² CFU/ml could cause a significant acute alteration in the cortical and medullary layers of the renal parenchyma and a pelvic wall starting from Day 1 of the follow-up [7]. Considering the data obtained in the preceding series and the results of the modeling of obstructive pyelonephritis, the researchers expanded the study design and defined further tasks on the evaluation of the microbial load and inflammatory changes in Pr at the initiation of a subpathogenic titer and early follow-up period, which corresponds to the tested spectrum of the uropathogens (E. coli, S. haemolyticus, and P. niger). Besides, the analysis of the data from the preceding cycle of the modeling of bacterial prostatitis revealed certain peculiarities in the formation of an inflammatory process in Pr in the established period of follow-up when infected with debatable strains at 10³ CFU/ml in comparison with inoculation in the respective titer of E. coli, which require detailed verification for further comparison with the parameters and changes registered during the inoculation at the titer of 10² CFU/ml.

The study aimed to evaluate the relationship between the dynamics of microbial load and the degree of pathomorphological changes in prostate tissues during infection with various uropathogens in a titer of 10³ CFU/ml; to evaluate the degree of microbial load and the severity of histological changes in prostate tissues on follow-up day 7 with transurethral infection with various uropathogens in a subpathogenic titer of 10² CFU/ml; and to perform a fundamental comparative analysis of the indicators of contamination and the severity of inflammatory changes on follow-up day 7 after the inoculation of various uropathogens in titers of 10² and 10³ CFU/ml.

MATERIALS AND METHODS

Ethical statement. The animal model study was designed and conducted according to the appropriate guidelines [8-11]. The study was approved by the Ethical Committee of the Rostov State Medical University based on the planned protocol of the experiment (protocol No. 16/17 dated October 5, 2017).

Laboratory animals. Twenty mature New Zealand rabbits, weight: 3200 – 4100 g, age: 34 –40 months.

Uropathogens. Strains E. coli (strain No. 32), S. haemolyticus (strain No. 28), and P. niger (strain No. 7) were isolated from biological materials (after the Meares-Stamey test) of patients with verified prostatitis category II (NIH-NIDDK Classification System, 1999) with an extended cultural study on 12 nutrient media according to the proposed method (patent RU 2452774 C1 No. 2011103414/10 dated January 31, 2011).

Randomization. At a post-adaptation stage (7 days), LAs were randomly distributed into 4 groups by the infecting uropathogen to avoid a selection bias (Table 1). To reduce the performance bias, the group of researchers was blinded by giving a unique identification code to each infected animal in the post-infecting stage.

The stages of experimental modeling of prostatitis, vivisection, and preparation of the biopsy core, along with the methods of cultural and histological studies of Pr tissues, were described in previous studies [5][12]. Inoculate titer of uropathogens: 10^{2, 3} CFU/ml; control – Sol. NaCl 0.9%. Follow-up period — days 1, 3, 7 and 14 in a titer of 10³ CFU/ml and day 7 in a titer of 10² CFU/ml.

To evaluate and compare the severity of pathomorphological changes revealed in Pr, the authors applied a score system of the expression of pathohistological changes in a nominal scale during the visual examination of 5 FoV at a magnification of × 200: score 0 (pts.) – lack of pathological changes (signs)/impossible to evaluate changes; 1 pts. — insignificant/mild expressed changes (signs); 2 pts. — moderate expressed changes (signs); 3 pts. — severe expressed changes (signs).

To minimize the detection bias at the stage of formation of biopsy cores, the authors encrypted them to blind the personnel involved in the cultural and histological studies.

Statistical analysis. Statistical analysis of the obtained data was performed using the software Statistica 10.2 («StatSoft Inc.», Tulsa, OK, USA) and GraphPad Prism 9 by Dotmatics («GraphPad Software Inc.», Graphpad Holdings LLC, San Diego, CA, USA). The Shapiro-Wilk and Kolmogorov-Smirnov tests did not show the normal distribution of the parameters. Thus, the descriptive statistics of the quantitative characteristics were presented as a central trend of the median (Me) and interquartile range (25 and 75 percentile); in the text, it is presented as Me [LQ; UQ]. A comparison of the variables in the independent samplings was made with non-parametric statistical methods – one-way ANOVA Kruskal-Wallis H test with Dunn’s posthoc test. The results were significant at p < 0.05. Titer values (10ⁿ CFU/ml) in the text were expressed as a decimal logarithm n (n lgCFU/ml), where n is the degree of microbial load.

RESULTS

Analysis of the cultural study results. The microbial load of Pr biopsy cores with the studied microorganisms was registered in all the observed cases in Groups 1, 2, and 3 (Table 2). In Group 4 (control), resident microflora was verified, which was represented by the taxons Propionibacterium spp., Eubacterium spp., Bacteroides spp., Peptostreptococcus spp. as mono and bicomponent microbial associations in the titers from 1 to 3 lgCFU/ml.

There were no median values of the biopsy core microbial load < 2 lgCFU/ml and > 7 lgCFU/ml defined in the tested groups. The analysis of the median parameters of bacterial contamination showed that in Group 1, in 35.0% of cases, the microbial load of biopsy cores was 5 lgCFU/ml, in Group 2, in 50.0% of cases, it was 5 lgCFU/ml, and in Group 3, in 40.0% of cases, it was 4 lgCFU/ml, respectively (Fig. 1).

The comparative evaluation of the microbial load in the Pr (1A – 1D, 3 lgCFU/ml, 1–14 days) revealed the summed intergroup differences (p = 0.022) between Group 1 (E. coli) and Group 3 (P. niger) and separate intergroup differences in Group 1 (E. coli) day 7 vs Group 3 (P. niger) day 3 (p < 0.005) and Group 1 (E. coli) day 7 vs Group 3 (P. niger) day 14 (p < 0.005). The comparison of the median parameters of the bacterial contamination of the Pr on day 7 did not reveal any significant differences in the cases of infection with titers 2 lgCFU/ml and in the data cases of microbial load verified for the test titer 3 lgCFU/ml during the same period of follow-up (Fig. 2).

Evaluation of daily changes in integrative parameters of bacterial contamination (1A – 1D, 3 lgCFU/ml) of Pr biopsy cores did not reveal any common tendency in their changes in various groups from day 1 to day 14. Thus, in Group 1, a progredient increase in the microbial load was established from day 1 to day 7 and a further decrease by day 14 to the minimal values for the infecting titer. In Group 2, the authors registered fluctuations in the median titer of bacterial contamination of Pr. From day 1 to day 3, there was an increase in bacterial contamination with a decrease to minimal values by day 7. By day 14, there was a relative increase observed. In Group 3, a gradual decrease in the microbial load on Pr tissue was observed from day 1 to day 3. The values were stable on day 7 in comparison with day 3, and a relative decrease to minimal intragroup values was observed by day 14 (Fig. 3A).

Analysis of the histological study results: infection of 3 lgCFU/ml. In Group 1 (E. coli), inflammatory changes in Pr tissues were associated with an expressed neutrophil reaction at all follow-up days with a weak eosinophil-cell reaction (combined with lymphocytic-macrophagous chemotaxis on day 14). Additionally, there was a significant reaction of the microvasculature with moderate stromal edema, expressed dystrophic and necrotic changes in the glandular epithelium, and mild ectasia of Pr glands filled with an exudative component (Fig. 4A). In Group (S. haemolyticus), alterative changes were characterized by expressed neutrophil-eosinophil-lymphocyte infiltration at the initial stages, which changed to lymphocyte-macrophage infiltration from follow-up day 7. They were also characterized by a significant reactive plethora of the microvasculature with the formation of stromal edema, ectasia of Pr glands with expressed exudation, a deposit of amyloid corpuscles/protein masses, formation of necrotic foci and dystrophic changes in the glandular epithelium, and disintegration of the interstitial tissue (Fig. 4B). In Group 3 (P. niger), the respective inflammatory transformations were characterized by the prevalence of eosinophil-macrophage cell reaction on days 1 and 3, which changed to a weak neutrophil-lymphocytic cell reaction on days 7 and 14. Additionally, there was an expressed reaction of microvasculature and significant stromal edema, ectatic changes in the prostatic glands with moderate exudation and accumulation of protein masses in the lumen, and weakly expressed diffuse necrotic changes in the glandular epithelium (Fig. 4C).

Analysis of the histological study results: infection of 2 lgCFU/ml. In Group 1 (E. coli), inoculation on day 7 after infection resulted in weakly expressed acute inflammatory changes in Pr tissues, manifested as hyperplastic transformation of tissue similar to adenomatosis with formation of papillary structures in glands, edema in the stroma and areas of its destructurization, and moderately expressed lymphohistiocytic infiltration of the interstitium (Fig. 5A). In Group 2 (S. haemolyticus), inoculation on day 7 after infection in Pr tissue resulted in signs of moderate alteration manifested as adenomatous hyperplasia of the glands with reactive hypersecretion of the mucosal component, which forms lobular structures, mild edema of the stroma and areas of its disintegration, reactive plethora of the microvasculature, and diffuse moderate lymphocytic infiltration of the interstitial tissue with the formation of focal clusters in the areas of tissue adjoining to the basal glands (Fig. 5B). In Group 3 (P. niger), the authors did not reveal any change in the glandular component on follow-up day 7 after the infection. There was insignificant stromal edema with insignificant lymphohistiocytic infiltration.

In Group 4 (control), inflammatory changes in Pr tissues were not registered.

Morphometric evaluation of inflammatory changes and comparative characteristics with the parameters of the cultural study. Based on the histologic study results, Pr inflammatory changes were evaluated by the score (points – pts), which characterized the expression of the pathomorphological transformation of tissues. They were daily stratified and grouped by the infecting agent (Fig. 6).

After inoculation of the studied titer of 3 lgCFU/ml in Group 1 (E. Coli) and Group 2 (S. Haemolyticus), the total evaluation of the changes (S68 pts. and S71 pts., respectively) demonstrated a comparable expression of pathomorphological tissue transformations registered during the follow-up period. In Group 3 (P. niger), the expression of the total pathological tissue transformations was at a relatively lower level (S60 pts.). It should be noted that the expression of inflammation on follow-up day 1 in Group 2 (S. haemolyticus) and Group 3 (P. niger) was comparable (19 pts. and 18 pts., respectively) and more significant than in Group 1 (E. coli), wherein primarily infiltrative-edematous changes in tissues were observed (14 pts.). In turn, the intergroup comparison showed that the maximal alterative changes in tissues were observed on follow-up day 3 (20 pts.) in Group 2 (S. haemolyticus), and minimal changes – on day 14 (11 pts.) in Group 3 (P. niger). The evaluation of the tendencies in the development of inflammatory changes in Pr tissue showed the differences in the dynamics of these pathological transformations at the established follow-up dates when infected with various uropathogens (Fig. 3B). In all groups, inoculation in the test titer of 3 lgCFU/ml was associated with a significant active inflammatory process in Pr tissues, which was registered on follow-up day 1. The peak changes in the experimental groups were not similar. In Group 1 (E. coli), the maximal expression of pathological transformations was observed on day 7; in Group 2 (S. haemolyticus), on day 3; and in Group 3 (P. niger), on day 1. By the follow-up day 14, in all groups, there was a tendency to resolve acute inflammatory changes, which manifested as a decrease to the minimal level of activity (compared to follow-up day 1) in Group 2 (S. haemolyticus) and Group 3 (P. niger), and epiminimal level in Group 1 (E. coli), and the appearance of the areas of proliferative inflammatory transformation (formation of the foci of loose connective tissue). Comparison of graphic tendencies that reflected the dynamic variations of median bacterial contamination values and nominal histological changes parameters (Fig. 3A, 3B) at various follow-up dates revealed a complete synchronization in the increase / decrease in both trends only in Group 1 (E. coli).

After the inoculation in the test titer of 2 lgCFU/ml on follow-up day 7 in nominal values (Fig. 6), the expression of inflammatory changes was more significant (10 b.) in Group 2 (S. haemolyticus) and minimal (4 b.) in Group 3 (P. niger), although the difference in the bacterial contamination parameters on this date between these cases was not significant for the tested uropathogens (Table 2).

DISCUSSION

The study results defined the bacterial contamination of the LA prostate in a titer of 2 and 3 lgCFU/ml of various uropathogens, which provides the valid evaluation of the microbial load and histological changes in Pr tissues. Comparative analysis of the results showed:

1. In all cases, infection in a 3 lgCFU/ml titer led to an increase in bacterial contamination compared to the initial inoculating titer (except niger on day 14). Furthermore, in all cases, acute inflammatory changes in Pr tissues were revealed. These changes tended to resolve on follow-up day 14. The most expressed and comparable by severity changes were registered in Group 1 (E. coli) and Group 2 (S. haemolyticus). They also corresponded to higher total median values of bacterial contamination in these groups throughout the follow-up period compared to Group 3 (P. niger), where the expression of changes along with the median bacterial load was observed at a lower level. Graphic trends, which reflect the changes in the daily median parameters of bacterial contamination and expression of inflammatory transformations of Pr tissues, showed the associations only in Group 1 (E. coli). It should be mentioned that in Group 3 (P. niger), the unsynchronization of the trends was observed only on day 7. The nominal evaluation of the expression of pathological changes showed a reduction of inflammation but the parameters of the microbial load remained at the stationary level in comparison with day 3. In Group 2 (S. haemolyticus), the synchronization of both trends was observed from day 1 to day 7. However, the expression of the inflammatory changes continued decreasing to day 14 in contrast to an increase in the microbial load on this date in comparison with day 7. In a similar study, a similar tendency was established in the evaluation of the simultaneity of changes of the microbial load parameters and inflammatory changes in Pr tissues infected with E. coli in an increasing titer of 3, 5, 7 lgCFU/ml. In the case of infection in a titer of 5 and 7 lgCFU/ml, the bacterial contamination did not correspond to the severity of pathomorphological transformations in tissues. At the same time, the most significant inflammation was observed in a titer of 5 lgCFU/ml in the cases of intermediate parameters of bacterial contamination. In the groups infected with a titer of 3 and 7 lgCFU/ml, inflammatory changes were respectively, despite a two-fold increase in the median parameters of Pr microbial load in the titer of 3 lgCFU/ml [12]. It should be highlighted that the peak median parameters of the Pr bacterial contamination in each group corresponded to the most significant inflammatory changes: in Group 1 (E. coli) – on day 7, in Group 2 (S. haemolyticus) – on day 3, and in Group 3 (P. niger) – on day 1. Along with this, in the study mentioned above, in groups infected with a titer of 5 and 7 lgCFU/ml E. Coli, the maximal expression of the inflammation did not correspond to the parameters of microbial load. Moreover, in the group infected in a titer of 5 lgCFU/ml E. coli, the highest nominal values of morphometry paradoxically corresponded to the minimal bacterial contamination in this group.
2. The study of the infection in a titer of 2 lgCFU/ml in all the tested groups on follow-up day 7 showed an accumulation of bacterial mass in Pr tissue with the most significant parameters in Group 2 ( haemolyticus) and the lowest – in Group 3 (P. niger). Besides, in all groups, there was a certain increase in the median parameters of the tissue bacterial contamination observed. It was more significant in Group 2 (S. haemolyticus) compared to the concentration of infecting studied. However, there was a tendency and pathological transformation of tissues that also showed to be more intensive in Group 2 (S. haemolyticus) compared to Group 1 (E. coli). This was manifested as a moderate hypersecretory activity of the Pr glandular epithelium, plethora of the microvasculature, stromal edema, and lymphocytic infiltration. In turn, in Group 3 (P. niger), apart from the stromal edema, there were no significant changes. Considering the insignificance of cell infiltration, there are no grounds to characterize it as inflammatory. The comparison with the morphometry data, established after inoculation in a titer of 3 lgCFU/ml on follow-up day 7, revealed a significant prevalence of inflammatory changes when infected with the studied titer in comparison with the infection in a titer of 2 lgCFU/ml. It was especially expressed in Group 1 (E. coli), wherein the difference between the nominal values was the most significant (19 pts. for 3 lgCFU/ml vs 7 b. for 2 lgCFU/ml). Thus, the pathomorphological changes registered in the groups were not comparable by the severity of the pathological transformations, determined during infection in a higher uropathogen titer, despite the prevalence of the median parameters of the microbial load (insignificant) on day 7 in Group 2 (S. haemolyticus) and Group 3 (P. niger) in the case of infection in a titer of 2 lgCFU/ml compared to the titer of 3 lgCFU/ml. It should be noted that in the experimental modeling of acute obstructive pyelonephritis, the causative uropathogen E. coli, debatable microorganisms from the CoNS clusters (S. haemolyticus), and anaerobes (Eubacterium spp.) in a subpathogen titer 2 lgCFU/ml caused significant infiltrative-necrotic inflammatory changes in the renal parenchyma and pelvis from day 1 and reached maximal expression on follow-up day 5 [7]. It is suggested that the severity of inflammation can be associated with complete obstruction, which results in a significant increase in the intrapelvic pressure, and thus, formation of the infected urine reflux in the renal tissues (active accumulation of pathogen and a respective increase in the bacterial load). The present study did not reveal any total obstructive changes in the Pr that would cause acute urine retention, although LA autopsies showed the signs of chronic incomplete urinary retention.

It should be noted that there are some peculiarities in the development of the inflammatory process revealed during the evaluation of the expression of pathomorphological changes. In Group 2 (S. haemolyticus) and Group 3 (P. niger), infection in a titer of 3 lgCFU/ml at the early stages of observation (day 1 and 3) led to a significant eosinophil infiltration, which was observed also in Group 1 (E. coli) but at a lower level. Eosinophils are potent inducers of the functional disorders in the inflammatory area because of the production of cytokine proteins. Furthermore, they contribute to the chemotaxis of the T-cell, which contributes to the formation of large lymphocytic-macrophage clusters adjoining the alveolar structures. In general, these changes can be evaluated as visual manifestations of the hypersensitivity reaction (types II/IV) [13]. Along with this, in Group 3 (P. niger) on day 7, there was a formation of the foci of loose connective tissue, which indicated a conversion of the active inflammatory process to the resolution stage. This character of changes can be explained by the hypereosinophilic cell-mediated response. Some studies confirmed that damaged stromal cells released stimulating factors that activated the synthesis of TGF-b, EGF, VEGF, and other factors that contribute to tissue regeneration [14-16]. It should be noted that a group of Japanese researchers revealed a similar eosinophilic cellular infiltration in biopsy cores of tissues after transurethral resection of Pr in a patient who suffered from eosinophilic granulomatosis and polyangiitis (EGPA) [17]. In turn, significantly lower formation of necrosis and destruction foci in Group 2 (S. haemolyticus) and Group 3 (P. niger) can be explained by the lack of potent E. coli damaging factors in Pr tissues and lower collateral damage caused by reduced neutrophilic reaction [18]. In all the tested groups on day 1 and 3, there was a significant development of interstitial edema associated with expressed reactive changes in the microvasculature and an increase in vascular permeability, which was maximally expressed on day 1 in Group 3 (P. niger). Along with these pathological processes in tissues, the formation of protein masses and amyloid corpuscles can be highlighted can be highlighted in the ducts and lumens of the glandular Pr apparatus. Their verification can indirectly indicate the expression and rate of interstitial edema development in surrounding tissues and the general level of damage in various tissue anatomical structures, which in combination lead to manifested congestion and impairments of prostate fluid evacuation from day 1 and day 3 in Group 2 (S. haemolyticus) and Group 3 (P. niger).

Limitations. The study limitations include a small volume of observations in the case of infection in a titer of 2 lgCFU/ml with the tested uropathogens and the lack of the developed intravitam biopsy method for LA, which would allow the researchers to evaluate the dynamics of the changes in the bacterial contamination and expression of inflammatory changes in Pr tissues in each sample without their euthanasia.

CONCLUSIONS

To sum up the presented results, the following keypoints can be highlighted:

1. E. coli, S. haemolyticus, and P. niger in a low titer of 10³ CFU/ml provoke inflammation in the Pr from the follow-up day 1. The expression of pathomorphological changes is especially manifested and comparable in the case of infection with E. coli and S. haemolyticus. Along with this, P. niger realized the maximum altering potential from follow-up day 1, on the contrary, E. coli demonstrates maximal changes from follow-up day 7. The evaluation of the dynamics of bacterial contamination and severity of inflammatory changes in the Pr tissues revealed a complete synchronous association between them on the studied dates only in the cases of infection with E. coli. In the case of infection with P. niger and S. haemolyticus, the synchronization was relative, and on days 7 and 14, unsynchronization was observed, respectively. It should be highlighted that the inflammatory process has some differences in the cases of infection with various uropathogens. Despite a similar severity of inflammation, induced by E. coli and S. haemolyticus, the character and peculiarities of the inflammatory process are comparable in S. haemolyticus and P. niger.
2. Despite a relative prevalence of the microbial load of Pr tissues with haemolyticus and E. coli over P. niger after the infection with a subpathogenic titer of 10² CFU/ml, there were differences (insignificant) in the pathogenic potential of this titer for the tested microorganisms. On follow-up day 7, S. haemolyticus and E. coli induced weak diffuse inflammation in Pr. However, it is less intensive in the case of infection with S. haemolyticus. In turn, P. niger provokes alterative changes that are difficult to interpret as acute inflammatory at this stage of observation.
3. A comparison of the severity of changes revealed on day 7 after infection with a titer of 10² and 10³ CFU/ml showed that the infecting titer of 10² CFU/ml was incomparable by the pathogenic potential with 10³ CFU/ml. In addition, it cannot induce the formation of severe pathological transformations of Pr tissues regardless of the type of infectious agent, despite a significant accumulation of bacterial mass at this follow-up period.