By David N. Leff

Like a cerebral chastity belt, the blood-brain barrier (BBB) locks out pathogens and other unwanted particles ¿ as well as useful therapeutic drugs. But love laughs at locksmiths.

It turns out that besides that heavily guarded front door into the central nervous system (CNS), there¿s a back-door barricade as well. The barroom bouncer at the main portal is p-glycoprotein (Pgp), a well-known molecular transporter of many substances across cell-surface membranes. At the rear entrance, guarding the brain¿s choroid plexus (CP), stands Pgp¿s strong-arm partner, MRP, which stands for multi-drug-resistance-associated protein.

Right now, a covey of major pharmaceutical and biotechnology companies have candidate inhibitors of Pgp, co-administered with frontline cancer chemotherapy drugs, in Phase II clinical trials. Those high-potency, non-toxic, two-drug combos aim to muscle Pgp aside so the antitumor agents can penetrate cancerous cells, and there induce their effect.

That multi-drug resistance strategy, suggests a ¿multi-threat¿ scientist at Washington University, in St. Louis, may be neat and plausible, but possibly insufficient.

He is radiologist and cancer cell biologist David Piwnica-Worms, who holds professorships at the university¿s departments of radiology, molecular biology and pharmacology.

¿Where it¿s desired to get drugs into the CNS,¿ he told BioWorld Today, ¿we would propose that you need to block both the Pgp and the MRP. Let¿s say you just inhibited Pgp with one of these drugs now under development by Eli Lilly or Glaxo, or Novartis or Vertex. Many of the chemotherapeutics and anti-psychotics are recognized by both MRP and Pgp. So, if you block Pgp alone, stuff could come in on the blood-brain-barrier side, but then get pumped right out by MRP on the CP side. We suggest that you inhibit both Pgp and MRP, to allow full penetration of the drug into the brain. That simple concept wasn¿t clear until we found those two transporters on the CNS.

¿[In the brain,] the Pgp and MRP would be playing an appropriate role in helping to protect the CNS from the anticancer protocol,¿ Piwnica-Worms said. ¿Conversely, where it¿s desired to get drugs, such as HIV protease inhibitors, into the brain ¿ which some suggest may be a latent site of harboring viral infection ¿ you¿d want to get the agent in there. The CNS seems to be a sanctuary for HIV replication. It¿s thought to correlate with AIDS dementia and CNS lymphomas. The connection is not proven, but it¿s certainly suspicious.

Enter HIV Protease Inhibitors Into Brain

¿In that case,¿ he continued, ¿these two molecular transporters may be working against you. It¿s very hard to treat virus in the brain, because many of the antiviral drugs in use are recognized by these transporters. Therefore, you can¿t get adequate levels into the CNS. Interestingly, there¿s some new data with the HIV protease inhibitors being recognized by Pgp, but no direct evidence about their role in MRP. We have some unpublished data that there may be some interaction there too. So if you could selectively inhibit both Pgp and MRP, it might be desirable for improving drug delivery to the CNS.¿

The brain¿s back-door blood barrier, the choroid plexus (CP), is also known as the blood-cerebrospinal barrier. Its main function is to pull certain substances out of neighboring blood vessels to generate the cerebrospinal fluid (CSF) that bathes nerve cells throughout the brain ventricles and spinal cord.

In both the CP and BBB, the actual physical barricades can be thought of as riot policemen locking arms to control an unruly crowd.

¿There are three levels of protection at both the BBB and CP barriers,¿ Piwnica-Worms explained. ¿Two different sites with similar function. For the BBB, the capillary endothelial cells lining the brain structures are aligned right next to each other by seamless tight junctions between the cells. In addition, the glial cells that line up behind these capillary endothelial cells send out pseudopods that appear to be another physical barrier. These are little projections, almost like a row of soldiers with their feet standing on the ground, right next to each other.

¿By analogy,¿ Piwnica-Worms said, ¿the same thing is happening on the CP side of the brain. But the CP has leaky capillaries, very different from those in the rest of the brain. These fenestrated capillaries are capped by an epithelial layer of the CP, which has tight junctions on its surface facing the cerebrospinal fluid side of the CNS. Those line up and fit together like bricks in a road to provide that barrier.

¿It¿s becoming clear,¿ he said, ¿that there is differential expression of Pgp, MRP and other molecular transporters on the blood side and the CSF side of these CP epithelial cells. They regulate the transport of amino acids, nutrients, vitamins, whatever ¿ as well as useful drugs ¿ across that epithelial layer.¿

To elucidate this barrier mechanism, and how it might be circumvented, Piwnica-Worms, wearing his radio-imaging hat, performed an in vivo human experiment. His model drug consisted of a membrane-permeable radiopharmaceutical, technetium-99, labeling a molecule that is transported by both Pgp and MRP.

¿This radiotracer,¿ Piwnica-Worms said, ¿is detectable by external, noninvasive imaging because of its central technetium-99 atom. This emits a gamma photon by which a standard nuclear medicine camera lets us map the spatial location of the radiopharmaceutical in our patient¿s body. So, we injected it into a vein, and the blood circulation distributed the imaging tracer throughout the body. For our brain scan, the camera centered over the patient¿s head, and rotated every 30 seconds, like CAT scans or MRI studies.

¿We were able to map the technetium isotope localized to the CP tissue, and not in the adjacent cerebrospinal fluid,¿ he said. ¿Knowing that, in the absence of molecular transporters, the radiopharmaceutical is freely diffusible across cell membrane bi-layers and through cells, when we saw that its image was concentrated, we knew that something was stopping its penetration into the CP.¿

Piwnica-Worms is senior author of a paper in the current Proceedings of the National Academy of Sciences (PNAS), dated March 30, 1999. It is titled: ¿Choroid plexus epithelial expression of MDR1 P glycoprotein and multi-drug resistance-associated protein contribute to the blood-cerebrospinal-fluid drug-permeability barrier.¿ He holds a U.S. patent covering the use of radiolabeled compounds to image multi-drug resistance phenotypes in vivo, on which Mallinckrodt Inc., of St. Louis has an option.