Noninvasive gene transfer to the lung for systemic delivery of therapeutic proteins

Abstract

This study evaluates the use of vectors based on adeno-associated viruses (AAVs) to noninvasively deliver genes to airway epithelial cells as a means for achieving systemic administration of therapeutic proteins. We intranasally delivered AAV vectors to mice in which the same AAV2 genome encoding a cellular marker was packaged in capsids from AAV1, 2, or 5 (AAV2/1, AAV2/2, or AAV2/5, respectively). Gene expression levels achieved in both airways and alveoli were higher with AAV2/5 than with AAV2/1 and were undetectable with AAV2/2. The same set of vectors encoding a secreted therapeutic protein, erythropoietin (Epo), under the control of a lung-specific promoter (CC10) was intranasally delivered to mice, resulting in polycythemia with the highest levels of serum Epo obtained with AAV2/5 vectors. After a single intranasal administration of this vector, secretion of Epo was documented for 150 days. Similarly, intranasal administration of an AAV2/5-CC10-factor IX vector resulted in secretion of functional recombinant protein in the bloodstream of hemophiliac, factor IX-deficient mice. In addition, we demonstrate successful readministration of AAV2/5 to the lung 5 months after the first delivery of the same vector. In conclusion, we show that intranasal administration of AAV vectors results in efficient gene transfer to the lung only when the vector contains the AAV5 capsid and that this noninvasive route of administration results in sustained secretion of therapeutic proteins in the bloodstream.

Figure 1

Transduction of murine lung with vectors based on various AAV serotypes encoding hPALP. Vectors were intranasally administered undiluted (3 × 10¹¹ GCs per mouse, upper panels) or diluted 1:5 (6 × 10¹⁰ GCs per mouse, lower panels) to immunocompetent animals (n = 5 per group). (a) Histochemical detection of PALP in the lung 4 weeks after vector delivery. Positive cells are present in both alveolar and airway cells when AAV2/5 and AAV2/1 are given. No transduction is observed with AAV2. (b) Quantification by ELISA of hPALP activity in lung homogenates 4 weeks after vector delivery. The specific serotypes and dilution factors used are shown at the bottom. Values shown are mean ± SD. Equivalent results were obtained with the chemiluminescent assay (not shown).

Figure 2

Serum Epo (a) and hematocrit (b) levels following intranasal administration of vectors based on various AAV serotypes in mice. We intranasally administered 3 × 10¹¹ GCs per mouse of AAV2/1-, AAV2/2-, or AAV2/5-CC10-rhEpo 4 weeks before the measurements. Control animals received the same set of vectors containing lacZ. Values shown are mean ± SD.

Figure 3

Serum Epo (a) and hematocrit (b) levels over time following a single intranasal administration of 3 × 10¹¹ GCs per mouse of AAV2/5-CC10-rhEpo (black bars) or AAV2/5-CC10-lacZ (white bars). Values shown are mean ± SD (at 120 and 150 days, all but two animals had died due to effects of severe polycythemia).

Figure 4

FIX secretion from FIX^–/– murine lung transduced with AAV2/5. (a) Circulating FIX levels over time following viral delivery. White bars correspond to animals that received AAV2/5-CC10-lacZ, black bars to animals that were given AAV2/5-CC10-cFIX. (b) aPTT levels following intranasal delivery of AAV2/5-CC10-lacZ (white bars) or AAV2/5-CC10-FIX (black bars). Values shown are mean ± SD.