Introduction
The experiments reported here represent the continuation of our studies on the mechanisms by which nanoparticles resembling the so-called nanobacteria (NB) are formed and by which anomalous ectopic calcification is triggered [1]–[3]. NB are supposed to represent exotic, slow growing, pleomorphic, sub-micrometer (50–500 nm) microorganisms coated with proteins and containing carbonate hydroxyapatite (HAP) [4]–[12]. These putative microorganisms are deemed identical to the spherical forms observed earlier by Folk in geological samples and which he called ‘nannobacteria’ [4], [13], [14]. The purported NB are ubiquitous, being found not only in soils [13], [14], water [15], [16], and air [17], but also in harsh and remote environments like meteorites [18], stratosphere [19], [20], and even interstellar dust [21]. They are supposedly associated with body fluids [4]–[12], body infusion products and vaccines [4], [22] and have been described as causative or aggravating agents of numerous diseases, especially extraskeletal calcifications ([4]–[12]; see also ref. [23] for a list of such ectopic calcifications with a potential link to NB). NB have thus been implicated in kidney stone formation, calcifications, and polycystic disorders [5], [8], [9], [11], [12], [24]–[27] as well as atherosclerosis and other cardiovascular calcifications [28]–[32]. Notably, NB have been characterized from the outset as infectious agents of disease [5]–[12], [23]. Thus, while NB are now referred also as calcifying nanoparticles (CNP)—a term that more appropriately acknowledges the microbiological and biochemical uncertainties and limitations associated with NB—they continue to be viewed and heralded as infectious and transmissible agents of anomalous ectopic calcifications [6], [7], [10], [26]–[32]. In fact, their infectious nature has been deemed of cardinal importance not only in the understanding of their complex and enigmatic biological cycle [8], [9] but also in providing a basis for justifying the use of antimicrobial therapy regimens to combat supposed NB infections [33]–[35]. As agents of transmissible infectivity, NB are said to fulfill either in part or completely the classic Koch's postulates [6]–[8], [28], [36]–[39]. On these same grounds, NB have been proposed as a potential global health hazard [16], [17], [40], [41]. NB are thus unusual in that they are seen by some [25] not only as primitive symbionts and precursors of biological life but also as infectious agents of disease—extraordinary claims in themselves that have led to intense debate and skepticism. An extensive list of these and other claims made by the NB proponents has been presented earlier [42].
These same views on NB have been refuted on both epistemological [43] and experimental [1]–[3], [44]–[46] grounds. Cisar et al. [44] have provided convincing evidence that challenges the organismic claims made earlier for NB, indicating instead that NB represent lifeless HAP complexed with organic compounds; furthermore, the authors have emphasized the role of lipids like phosphatidylinositol as nucleators of HAP that in turn can reproduce the many morphologies associated with NB. Raoult et al. [46] have reached the conclusion that NB are “mineralo fetuin” complexes that they have renamed as “nanons.” Both these groups have described a protein profile associated with the putative NB grown in serum-free conditions consisting of no more than a few major bands, the main one being fetuin-A, according to Raoult et al. [46]. This simple protein profile stands in marked contrast to the much more complex organismic protein profile that one would have expected for live bacteria and that had in fact been published by other groups [7], [26]. In Cisar et al. 's view [44], the proteins associated with NB grown from the various body fluids are most likely derived from the direct binding of proteins to the growing HAP. Moreover, Raoult et al. [46] have shown that the earlier antibodies marketed as specific for NB, rather than binding to foreign proteins of microorganismic origin, detect fetuin-A instead.
Our own studies have sought to reproduce the NB phenomenology in its entirety by means of a reductionist approach using simple calcium compounds like calcium carbonate as well as other competing ions to reconstitute particles that morphologically resemble NB [1]. Our results are significant in that they demonstrate that various calcium compounds may biomimetically resemble complex biological forms like NB—findings that are consistent with earlier observations made on geological samples enriched for calcite, carbonate, and silicate phases [47]–[50], all of which are capable of biomimetically resembling living organisms. These various studies raise caution against the use of morphological criteria per se as the sole basis for implicating biological or life-related activities. In our studies, the initially amorphous calcium particles were seen in the presence of serum to rapidly bind to proteins and ions, to acquire phosphate, to crystallize, and to nucleate further aggregation of HAP [2]. Serum was further shown to inhibit paradoxically the formation of NB-like precipitations, presumably through the binding to calcium and nascent apatite by calcification inhibitors present in the serum—a protective mechanism built-in likely to prevent unwanted calcification [2]. When saturated with calcium phosphate, these same calcification inhibitors—showing high affinities for calcium, nascent apatite, or perhaps both—are thought to turn into seeds for the nucleation of HAP, that upon further phase conversion and growth, end up mimicking and displaying the many peculiar characteristics of NB, including most notably their marked pleomorphism [2]. In our hands, the proteins associated with the NB scaffold were found to include albumin, complement components 3 and 4A, fetuin-A, and apolipoproteins A1 and B100 [2], [3]. This protein profile appears to mirror the very composition of the calcium and apatite binding proteins found in the serum that in turn had been used to culture and assemble NB; their relative abundance found in the NB scaffold appears to vary directly with their respective protein concentrations seen in the feeder serum as well as with their calcium and apatite binding affinities [2]. Accordingly, in the case of NB formed in the presence of other body fluids (saliva, urine, ascites, cerebrospinal fluid, pleural effusion, and synovial fluid), the bound proteins have also turned out to consist of the most abundant calcium-binding proteins found in these fluids, including various immunoglobulin G chains for saliva, and albumin and apolipoprotein A1 for the other body fluids [2]. The NB protein composition can be further modulated by changing or removing the source of proteins comprising the milieu of the nucleating NB, indicating that their presence in the NB scaffold may be circumstantially determined by their availability in the surrounding microenvironment and by their affinity for calcium and apatite [2]. Still in this context, our own exhaustive proteomic analyses of multiple putative NB specimens have revealed only common mammalian proteins that can be entirely explained by the source of serum, body fluid, or tissue used for culturing NB [2], [3], in marked contrast to a more exotic protein profile claimed earlier for NB and comprising of prokaryotic proteins like porins and complex peptidoglycans [5], [24] as well as the bacterial translation elongation factor Tu and the molecular chaperone GroEL [26]. Furthermore, our own immunoblotting experiments have revealed that both polyclonal and monoclonal antibodies, previously claimed to be specific for NB and used in numerous studies to affirm the presence of (and a pathological implication for) NB in human tissues, in fact bind strongly to both albumin and fetuin-A from the same as well as across species [2]. This surprising finding indicates that all the past immunodetection studies used to implicate a role for NB in human pathologies may very well have detected either albumin or fetuin-A, or a combination of both, derived not only from the very fetal bovine serum (FBS) used for the culture and demonstration of NB, but possibly also from the human tissue under investigation as well. This confusing and misleading scenario now needs to be taken into account when interpreting any past immunodetection study done to ascertain the presence of NB in human tissues.
Our earlier study indicates that the purported NB can be seen as consisting essentially of protein-mineral or organic-mineral complexes formed by means of a dual inhibition-seeding mechanism involving the binding of proteinaceous and other organic calcification inhibitors to calcium and nascent apatite minerals [2]. These inhibitors function presumably within a larger homeostatic cycle of calcium sequestration and clearance [2]. Subsequent experiments have demonstrated the formation of NB-resembling calcium granulations (also referred as calcium granules in the present study) within both fetal bovine and human sera that had been overloaded with either calcium, phosphate, or a combination of both; this granular formation appears to be the result of direct chemical binding between ionic and organic moieties and it precludes the need for any prolonged incubation under culture conditions [3] that earlier had been deemed necessary for the demonstration of NB [4], [5]. By means of simple binding between serum components and calcium and phosphate, we are now able to demonstrate a plethora of NB-like morphologies that include round, laminated particles, spindles, and even films [3]. This same sequestration of calcium in the form of calcium granules is known to occur throughout nature as part of a general cycle of calcium storage, retrieval, deposition, and disposal (see ref. [3] for a more detailed discussion; see Ryall's comprehensive and insightful review [51] on calcium granules, now known to be found in organisms spanning a wide range of phylogenetic complexities).
Against these developments, the NB phenomenology may be seen to have come full circle, with the NB-like precipitations representing no more than the simple outcome of a physiologically relevant homeostatic cycle designed by nature to deal with excess calcium and apatite. As such, and perhaps even ironically, given the controversies generated by this topic, NB-like formations are no less real and they do represent structures that are verifiable and demonstrable. However, they certainly do not seem to represent self-sustained lifeforms, primitive symbionts, or even transmissible and infectious disease agents as originally proposed. Precisely because such particles can be viewed now as comprising a part of the normal cycle of calcium movement and disposal, it is important to understand what exactly determines their assembly, where they are distributed, and whether they have any pathological role after all.
Here, we seek to dissect the role of serum proteins in the assembly of NB-like formations using the in vitro reconstitution methods established through our earlier studies [1]–[3]. In our mind, it is critical to establish whether purified proteins like fetuin-A or albumin are capable of replicating the entire NB phenomenology in the absence of other organic moieties (lipids, carbohydrates, and other proteins) that may also be present in the same serum, body fluids, and tissues that had earlier been used to “grow” NB. If so, we want to know whether the simple combination of minerals and proteins can give rise to both the morphological diversity (pleomorphism) and the distinct chemical signature deemed representative of NB-like formations as well as serum calcium granulations [1]–[3].
Both albumin and fetuin-A have been linked to NB-like formations [1]–[3], [46]. We had established earlier that both proteins are among the more abundant proteins found in NB-like particles cultured from FBS or human serum (HS) [1]–[3]. This observation not only reflects the high concentration of albumin and fetuin-A in the serum, but also the fact that these proteins bind avidly to the mineral phase of NB-like particles (see refs. [1]–[3] and the references therein). In our hands, preliminary experiments done with purified proteins indicate that albumin and, even more noticeably, fetuin-A inhibit apatite formation in the presence of excess calcium phosphate (e.g. at supersaturating concentrations) but this inhibition is only transient and it is followed by subsequent mineral precipitation [2]. However, it is not clear whether fetuin-A and albumin will seed apatite when added to a metastable solution exemplified by fresh culture medium, as we seek to verify here. Given the wealth of information accumulated on both albumin and fetuin-A insofar as their roles in calcium and apatite binding and calcification are concerned and their predominance among the proteins species linked to NB-like formations, we want to further establish whether the workings of these proteins alone can explain the entire NB biology, and if so, whether there are differences seen in the presence of either metastable or supersaturated calcium phosphate solutions. Since protein-mineral reactions of this kind may be involved in similar biomineralization processes in the body, we believe that a better understanding of this type of nanoparticle assembly may ultimately help address their pathophysiological role in vivo.