When I showed Sydney Brenner the first paper claiming a physiological role for nanobacteria ("Nanobacteria: an alternative mechanism for pathogenic intra- and extracellular calcification and stone formation", Kajander and Ciftcioglu, PNAS, 95 (8274-8279), 1998), he just chuckled.  And rightly so, given the expansive claims of that and succeeding papers.  Early claims that 30 nanometer particles visible in electron microscopy experiments contained DNA were challenged when the resulting sequences were shown to be identical to those found in common bacterial laboratory contaminants.  That is, while the original work pointed to some interesting evidence, there wasn't enough meat on the bone to convince people who have been watching biology since its modern beginnings.

Work has continued, however, and now careful studies have demonstrated nano sized objects at the core of structures from human bodies, where those nano objects definitely contain DNA.  In "Evidence of nanobacterial-like structures in calcified human arteries and cardiac valves" (Am J Physiol Heart Circ Physiol 287: H1115-H1124, 2004), Miller et al examine a variety of human tissues removed during surgery and conclude that "nanometer-scale particles similar to those described as nanobacteria isolated from geological specimens and human kidney stones can be visualized in and cultured from calcified human cardiovascular tissue."

The paper describes using light and scanning electron microscopy, immunostaining, and DNA staining to characterize objects 30-150 nm in size that appear in physiological samples.  Interestingly, there is already a commercial antibody available, "8D10", that appears to recognize a ~50-kDa protein only found in tissues and cultures that were observed to contain the nanobacteria.  Moreover, simultaneous immunostaining using 8D10 and DNA staining using PicoGreen revealed that structures cultured from filtered homogenates of human aneurysm contained both protein and DNA.  The most compelling evidence from a traditional biology perspective is that the nanobacteria can be propagated in culture media.  That is, the structures are self-replicating.  Decalcified particles contained structures that appear akin to cell membranes.

By way of acknowledging alternative explanations for their data, the authors note that;

Although a unique nucleic acid sequence remains to be identified from the nanosized particles identified within human arterial tissue in the present report, it is possible that these structures may represent either a variant form of microorganisms or an unrecognized bacterial growth stage such as L-forms, cell wall-deficient bacteria, and/or defective bacteria that have been hypothesized to represent either pleuropneumonic-like organisms or Mycoplasma species, which have been detected in serum of patients with long histories of chronic diseases. They may also represent an Archaea symbiont that requires cell contact or lipids from other cells for growth.

They go to observe their data are consistent with nanobacteria as a cause of disease;

Nanobacteria derived from bovine serum are internalized by human cells and appear to be cytotoxic. Similar internalization of nanolike particles in arterial smooth muscle would be consistent with induction of apoptosis, formation of matrix vesicles, and the inflammatory basis of atherogenesis. An infectious etiology of arterial calcification is consistent with increased lesion formation in experimental models of atherosclerosis.

Note that this text implies nanobacteria may be infectious agents.  Miller et al lay out the test of this hypothesis;

...A definitive cause and effect relationship needs to be established between these nanoparticles and [pathogenesis]. For example, it will be necessary to evaluate severity of calcification and disease progression in the absence, presence and titer of nanoparticles in humans. In the experimental setting, it will require infection of a naïve animal with cultured nanoparticles and subsequent identification of the particles within arterial calcification. Definitive characterization of these unique particles will require isolation and sequencing of genetic material (DNA or RNA).

No doubt the debate over nanobacteria will continue until the above criteria are met, but the Miller paper definitely contributes significantly to the discussion.

In the end, this sort of report illustrates how naive we are about what organisms inhabit the human ecosystem.  We haven't even isolated all the viruses and "normal" bacteria that live in and on humans.  And then something strange like nanobacteria come along.  We have lots of work to do.