Background
Leptospirosis is a worldwide bacterial zoonosis caused by several species of invasive spirochetes belonging to the genus
Leptospira. It affects humans in both rural and urban areas, particularly in developing countries with warm and humid climate [
1‐
3]. Water contaminated by urine from animal reservoirs is the main source of human infection, usually through cut or abraded skin. Leptospirosis is characterized by a broad spectrum of clinical manifestations, ranging from subclinical infection to Weil's syndrome, a severe and potentially fatal disease characterized by hemorrhage, acute renal failure and jaundice [
4]. Deaths may occur in less than 72 h after the advent of respiratory signs and symptoms such as severe hemorrhage of lungs, which usually appear between the fourth and the sixth day of disease [
5].
The use of experimental models remains a critical component for elucidating pathogenesis of leptospirosis. Young guinea pigs and hamsters are the most commonly used experimental models for acute leptospirosis [
6]. The intraperitoneal (i.p.) inoculation route is the most widely applied infection route by producing a lethal infection in experimental animals and mimicking the clinical symptoms of severe leptospirosis in humans [
7‐
10]. However, this route of infection does not reflect real conditions encountered during natural infection, because leptospires are believed to enter the host via mucous membranes or abrasions of the skin. It has been a long time for researchers to challenge animals through alternative routes to mimic natural entry of leptospires into hosts. Even about one century ago, Ido and his colleagues attempted to reproduce natural conditions by conveying the leptospries directly to the guinea pig by the bite of rat (carrier of leptospires). The results indicated that leptospirosis is rarely transmitted by the bite of rat [
11]. Since then, different infection routes such as conjunctival (c.j.) and subcutaneous (s.c.) have been employed in canine, horse, hamster and guinea pig, and resulting in acute leptospirosis in inoculated animals [
11‐
16]. By using infection routes different from the classic i.p. inoculation, these studies contributed to the elucidation of pathogenesis of leptospirosis in experimental animals. However all these methods bypassed the epidermis of host, the entry route and mode of leptospires directly via epidermis have been poorly studied as of today. There is still little data on how the leptospires interact with the epidermis and if the inoculated leptospires can penetrate skin and disseminate in host subsequently.
In this study, we examined the ability of leptospires to produce infections in guinea pigs when applied to damaged or undamaged skin. The results showed that infection with virulent leptospires, using abraded skin inoculation route of infection, produced typical leptospirosis in guinea pigs, whereas there were no symptoms in guinea pigs through shaved-only skin. The availability of this novel model will enable understanding of the pathogenesis of leptospirosis, as well as to study cutaneous barriers and epidermal interactions with this organism.
Discussion
Intraperitoneal injection is the most widely used infection route in experimental leptospirosis studies [
9,
24]. It reproduces the processes of the human leptospirosis in the animal models using a very easy way. However, intraperitoneal injection does not reflect the natural transmission of the pathogen. Leptospires are thought to enter the human body via cuts or abrasions in the skin. The entry of leptospires directly via epidermis has been poorly studied. Some reports of clinical leptospirosis cases have clearly identified the initial cutaneous injury [
25], others have not noted such a preexistent lesion [
26,
27]. It is not known whether the organism can penetrate intact skin or abraded skin. In this study, we established a guinea pigs leptospirosis model using epicutaneous inoculations route, to gain a better understanding of host-pathogen interaction and the pathogenesis of leptospirosis.
In this study, guinea pigs were inoculated with leptospires onto either shaved-only or abraded skin. Guinea pigs with abraded skin displayed clinical signs of leptospirosis. In contrast, lesions were not detected in the shaved-only animals which were inoculated the same amount of virulent L. interrogans strain Lai. These data confirmed that the intact keratinocyte layer is a very efficient barrier against leptospires, and intact skin can prevent the infiltration of leptosipres to the host.
It should be noted that the
L. interrogans strain Lai used in this research was originated from a female patient who died of pulmonary hemorrhage after an infection with this organism, which had previously been studied in a variety of animal models and found to develop a typical leptospirosis in guinea pigs with intraperitoneal route [
9,
28,
29]. The inoculation dose was referred to the previous guinea pig model reported in our lab [
9].
Our study here showed that infection with the
L. interrogans strain Lai using abraded skin inoculation route of infection produced a lethal infection in guinea pigs that mimicked the clinical characteristics of severe leptospirosis in patients, as described elsewhere [
5,
30,
31]. The main clinical signs were serious pulmonary hemorrhage, jaundice, retroperitoneal hemorrhage and renal hemorrhage.
Our data showed that virulent leptospires can rapidly (within 2 h) penetrate the abraded epidermis and enter the dermis; at some point within 2 h p.i., the invading organisms also distribute to blood. Attachment to host cells and host extracellular matrix (ECM) components is likely the necessary step for leptospires to penetrate, disseminate and persist in mammalian host tissues. Consistent with the ability of
L. interrogans to migrate through host tissues, a wide range of adhesion molecules were discovered in these organisms that may facilitate this process [
32‐
34]. It has been reported that many leptospiral proteins, including LigA/B, Lsa21, Lsa27, LenA to F, LipL32, OmpL37, TlyC and LipL53, have affinity for ECM and cell surface in vitro [
32‐
41]. Some of these proteins, such as OmpL37, were reported to have the strong binding affinity for skin and aorta elastin, and might facilitate the attachment of leptospires to elastin-rich inner layer of the skin as well as vascular structures [
37].
It is evident that leptospires penetrate abraded skin and quickly establish a systemic infection by crossing tissue barriers. It is likely that
L. interrogans can move through the tissue barrier by association with blood vessels, because leptospires were detected aggregated around the capillaries in muscular layer and peritonaeum in this study. It was thought that leptospires, like other spirochaetes, spread through intercellular junctions [
42]. However, they have been shown to efficiently enter host cells in vitro [
43‐
45]. Previous work accomplished by Martinea-Lopez
et al. demonstrated that
L. interrogans can disrupt the dynamics of the actin cytoskeleton in the human microvascular endothelial cell line and rapidly translocate through the cell layers [
46]. Other studies in human leptospirosis have shown that leptospiral antigens were detected in the cytoplasm of the endothelial cells of septal capillaries [
30]. Consistent with these results, leptospires were detected intracellularity in the vascular endothelial cells and disruptions of vascular basement membrane were also observed in this study. Our findings suggested that leptospires cross endothelial barrier and cause heamatogenous dissemination by pass through the endothelial cell cytoplasm.
When
L. interrogans strain Lai was inoculated on the abraded skin, localized changes around the inoculated site were detected. All of the guinea pigs showed hemorrhage at the dermis around the site-inoculation before the appearance of internal organs hemorrhage. Skin hemorrhage was rarely reported in animals infected experimentally through the i.p. route, and little attention has been called for. The mechanism of hemorrhage caused by leptospirosis has not been elucidated yet. Factors contributing to the hemorrhage might involve direct action of toxins and autoimmune process. Nicodemo and coworkers detected the intact leptospires in capillary endothelial cells, indicating the lung injury is directly triggered by leptospires and/or by their toxic products [
30]. Another study demonstrated the deposition of antibodies and complement along the alveolar basement membrane of infected guinea pigs, indicating pulmonary hemorrhage might be led by autoimmune process [
8]. Our data showed that abundant leptospires were detected in the dermis and subcutaneous tissue of hemorrhagic area and were rarely detected in adjacent none hemorrhagic areas, confirming the high burden of leptospires in the dermis is an important factor to cause hemorrhage. Humoral immune response seems not be associated with the pathogenesis of skin hemorrhage, as dermis hemorrhage developed as early as 8-24 h p.i.. Further examination of the local hemorrhage may give a clue to understand the mechanism of hemorrhage in this disease.
Hemorrhage in the skin is produced as one of the general symptoms in clinical cases [
4]. However, Hemorrhage localized at infected site was rarely recognized clinically. Local hemorrhage in our experiment model might caused by high dose inoculation of leptospires. When in nature infection, it seems like that low dose leptospires in the cuts or abrade skin will not cause skin hemorrhage until large amount of pathogen proliferated in the circulation, and then extensive skin hemorrhage will be produced.
Recently, Lourdault and his colleagues compared different routes (i.p., c.j. and s.c. inoculation) of infection and the dissemination of leptospries in blood and tissues of guinea pigs using multiple methods including real-time PCR [
16]. The results showed infected guinea pigs developed similar physical signs and pathological changes after i.p., s.c. and c.j. inoculation with leptospires, and the bacterial burden in tissues and histopathology revealed no major differences between the three routes of infections [
16]. In the guinea pigs with abraded skin inoculation, our real-time PCR results showed that the bacteraemia peaked at 96 h p.i. and then quickly decreased at 144 h p.i., which were consistent with the result of i.p. inoculated guinea pigs or s.c. inoculated hamsters reported by Lourdault and Truccolo respectively [
15,
16]. It is interesting to note that the high leptospires burden (3 × 10
5 leptospires ml
-1) detected in the blood at 2 h p.i., and then quickly dropped by 1 log at 8 h p.i.. It is speculated that high dose (5 × 10
8) leptospires inoculation cause a rapid flood of leptospira from the inoculated site to the bloodstream, then the majority of the leptospires were cleared by the innate immune system in the following several hours. As pathogenic
Leptospira were reported to be able to survive, and be more resistant to the action of the complement system [
47‐
49], polymorphonuclear neutrophils (PMNs), which constitute the largest population of intravascular phagocytes, are expected to play an important role in leptospiral clearance. It was reported that PMNs are able to kill pathogenic strains of
Leptospira by oxygen dependent and independent mechanisms [
50]. However, some experimental models showed that phagocytosis of pathogenic
Leptospira by neutrophils and macrophages is only effective if this pathogen is opsonized by specific IgG [
51‐
53]. Further investigations on PMNs activation and elimination of pathogenic leptospires are required to elucidate the establishment of innate immune responses in leptospirosis.
The traditional intraperitoneal inoculation is easy to handle and allows reproducible amounts of leptospires to be introduced. It is still the most widely used model to study the systemically infection of leptospirosis. However, there are some shortages of i.p. or other non-epicutaneous routes when apply on the pathogens causing infection through skin. In study performed by Bischof and colleagues, the subcutaneous injection of
B. anthracis (Sterne strain, which lacks the pX02 capsule plasmid) caused lethal infection in C57BL/6 mice, while quite resistant to epicutaneous inoculation of
B. anthracis onto abraded skin [
20]. This study suggested that our epicutaneous inoculation model would be an alternative way to apply the characterizations of
Leptospira mutants that are deficient in protein with binding affinity for skin.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
YZ, PH and XCJ designed the research project. YZ and XYZ coordinated the leptospira culture. YZ and XLL participated in developing a guinea pigs model of leptospirosis and pathology experiments. PH and XLL carried out the real-time PCR experiments. XCJ and HLY examined tissue samples. PH, HLY and XKG drafted the manuscript. All authors contributed to the writing and preparation of the manuscript. All authors read and approved the final manuscript.