© 2003 Robert A. Freitas Jr. All Rights Reserved.

Robert A. Freitas Jr., Nanomedicine, Volume IIA: Biocompatibility, Landes Bioscience, Georgetown, TX, 2003

15.3.1.2 Cell Response to Diamond Surfaces

Cellular interactions that occur at the tissue-implant interface are another important determinant of biocompatibility [521, 584-586]. For example, neutrophils, the most abundant white cells in human blood, will directly adhere to protein-coated implant surfaces, leading to inflammatory responses.

The first pioneering study of cellular response to diamond surfaces was completed by Thomson and colleagues [587] in 1991, using tissue culture plates with diamond-like carbon (DLC) coatings 0.4 microns thick. DLC is an amorphous hydrocarbon polymer with carbon bonding largely of the diamond type instead of the usual graphitic bonding [588], thus has many of the useful properties of diamond [589]. (The varying ratio of diamond type (sp³) carbons to graphite type (sp²) carbons in DLC may account for some differences in behavior exhibited by different DLCs.) DLC can be deposited near room temperature [590]. Mouse fibroblasts grown on the DLC coatings for 7 days showed no significant release of lactate dehydrogenase (an enzyme that catalyzes lactate oxidation, often released into the blood when tissue is damaged) compared to control cells. This demonstrated that there was no loss of cell integrity due to the DLC coatings. Mouse peritoneal macrophages similarly cultured on DLC also showed no significant excess release of lactate dehydrogenase or of the lysosomal enzyme beta-N-acetyl-D-glucosaminidase (an enzyme known to be released from macrophages during inflammation). Morphological examination revealed no physical damage to either fibroblasts or macrophages. This confirmed the biochemical indication that there was no toxicity and that no inflammatory reaction was elicited in vitro. Follow-up studies in 1994-95 found that murine macrophages, human fibroblasts, and human osteoblast-like cells grown on DLC coatings on a variety of substrates exhibited normal cellular growth and morphology with no in vitro cytotoxicity [591, 650]. In 2001, the same research group [4722] cultured two osteoblast-like cell lines on DLC-coated plates for 72 hours and found no adverse effects on these cells, as measured by the production of three osteoblast-specific marker proteins (alkaline phosphatase, osteocalcin, and type I collagen).

Other experiments have largely confirmed these results. For instance, human hematopoietic myeloblastic ML-1 cells and human embryo kidney cells proliferated continuously on DLC film with very high viability and no toxicity [592]. Scanning electron microscopy used to investigate the morphological behavior of osteoblasts found that these cells attached, spread and proliferated normally without apparent impairment of cell physiology when placed on DLC or amorphous carbon nitride films, whereas cells placed on silicon were able to attach but not to spread [593]. Human osteogenic sarcoma T385 cells and 1BR3 fibroblasts cultured on DLC-coated wells also showed DLC to be biocompatible [594]. The cytotoxicity study of DLC coatings by Parker and colleagues [595], employing the Kenacid Blue cytotoxicity test in vitro with 3T3-L1 mouse fibroblasts, found normal cell growth on diamond surfaces. Other tests by this team of the biocompatibility of “amorphous carbon hydrogen” using a standard cell line showed that such films are nontoxic to cells, appear to increase cell attachment, and afford normal cell growth rates [649]. Dion et al [596-598] looked at general cytotoxicity and hemocompatibility of DLC surface with 3T3 Balb/c cloned cells. Human endothelial cells isolated from placenta were also investigated as a model for differentiating cells. No negative effects due to DLC coatings were observed on the viability of cells, all of which showed normal metabolic activities. O’Leary and colleagues [654] evaluated cytotoxicity and cell adhesion of murine fibroblasts on saddle field source deposited DLC (containing less than 1% hydrogen) coating a titanium alloy surface, and found normal cell adhesion, density, and spreading on DLC. Other studies of DLC biocompatibility [656-659, 1680, 5689, 5690] have shown equally promising results.

In a study previously described (Section 15.3.1.1), Tang et al [521] examined the attachment of neutrophils to plasma-preincubated ~1 cm² 350-micron-thick CVD diamond wafers. Incubation for 1 hour with purified human neutrophils at 2 x 10⁶ cm^-3 produced an attachment rate of ~4 x 10⁹ cells/m² (~0.004 cells/micron²), about the same as for 316 stainless steel and 40% lower than for titanium, two common and well-tolerated implant materials. SEM photographs of CVD diamond wafers implanted intraperitoneally in live mice for 1 week revealed minimal inflammatory response. Interestingly, on the rougher “polished” (~1-micron features) surface, a small number of spread and fused macrophages 10-13 microns in diameter were present, indicating that some activation had occurred. However, on the smoother “unpolished” (<0.25-micron features) surface, samples were partially covered by round, non-spread (non-activated) cells, 4-7 microns in diameter, which had formed no obvious pseudopodia or cell bridges. Tang noted that “the morphology of unpolished surfaces of CVD diamond could be responsible for preventing the activation of surface-adherent cells [but] the mechanism for this differential response of phagocytic cells...is not yet understood.” If surface rugosity [599, 676, 677, 776], topography [600-602], or crystalline structure [603] can account for the differential response, then it is perhaps possible that atomically-precise diamondoid surfaces with <1 nm features – constituting much of the external surfaces of medical nanorobots and nanoorgans – could be rendered adequately macrophage-resistant (Section 15.4.3.6).

In 2002, Linder et al [4721] found that: (1) the adhesion of primary human monocytes to DLC-coated glass coverslips is not significantly enhanced in comparison to uncoated coverslips, (2) the actin and microtubule cytoskeletons of mature human macrophages show normal development on DLC, and (3) growth on DLC does not affect the activation status of human macrophages, as judged by polarization of the cell body. The researchers concluded that “it is unlikely that contact with a diamond-like carbon coated surface in the human body will elicit inflammatory signals by these cells.”

Finally, Jones et al [660, 4726] deposited DLC coatings (by plasma-assisted CVD) and other coatings on titanium substrate and tested their hemocompatibility, thrombogenicity, and interactions with rabbit blood platelets. The DLC coatings produced no hemolytic effect, platelet activation, or tendency towards thrombus formation. Platelet spreading correlated with the surface energy of the coatings (typically ~40 mJ/m² for DLC [4730]) with the lowest spreading for DLC [4726]. In general, platelet adhesion is reduced both by increased surface wettability and by the presence of platelet adhesion inhibiting proteins such as kininogen [756] which could be made available at nanorobot surfaces if required. Platelet adhesion to DLC or polycrystalline diamond surfaces has been measured experimentally as ~0.007 platelets/micron² after a 5-minute exposure to fresh human blood flowing at a wall shear rate (Section 9.4.1.1) of 50 sec^-1 [1680].

Any small object made of hydrophobic material may insert into bilayer lipid membranes. Experimental data have not been reported for diamond due to the unavailability of appropriate-size particles, but once these particles are obtained the interactions of, say, diamond nanorods with membranes can be studied and will likely show the insertion. E. Pinkhassik notes that an inadequately-controlled individual diamondoid nanorobotic arm or its protrusions might spontaneously enter the membrane of some cells, analogous to the solvation wave drive for cytopenetration proposed in Section 9.4.5.3.

Last updated on 30 April 2004